Case Study

CASE STUDY RESEARCH:
Theory Methods Practice

CASE STUDY RESEARCH:
Theory Methods Practice
by
Arch G. Woodside
Boston College, USA
United Kingdom North America Japan

India Malaysia China
Emerald Group Publishing Limited
Howard House, Wagon Lane, Bingley BD16 1WA, UK
First edition 2010

Copyright r 2010 Emerald Group Publishing Limited
Reprints and permission service

Contact: [email protected]
No part of this book may be reproduced, stored in a retrieval system, transmitted in any
form or by any means electronic, mechanical, photocopying, recording or otherwise
without either the prior written permission of the publisher or a licence permitting
restricted copying issued in the UK by The Copyright Licensing Agency and in the USA
by The Copyright Clearance Center. No responsibility is accepted for the accuracy of
information contained in the text, illustrations or advertisements. The opinions expressed
in these chapters are not necessarily those of the Editor or the publisher.
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
ISBN: 978-1-84950-922-0
To my two Marthas:
Martha Elizabeth Woodside and Martha Jane Woodside
Contents
Preface ix
Acknowledgments xiii
1. Building Theory from Case Study Research 1

2. Bridging the Chasm between Survey and Case Study Research 17
3. Storytelling Theory and Research 41
4. Creating Visual Narrative Art for Decoding Stories 85
5. Subjective and Confirmatory Personal Introspection 107
6. Overcoming the Illusion of Conscious Will and Self-Fabrication 129
7. Using the Forced Metaphor-Elicitation Technique (FMET) in
Subjective Personal Introspections about Self 143
8. Surfacing Executives Interpretations of Self and the Roles of
Co-Workers in Enacting Front and Back Stage Strategies 157
9. Personal Exchanges, Social Behavior, Conversation Analysis,
and Face-To-Face Talk 189
10. Constructing Thick Descriptions of Marketers’ and Buyers’ Decision
Processes in Business-to-Business Exchange Relationships 207
11. Case Study Research on Means-End Laddering Chains 225
12. Building in Degrees of Freedom Analysis in Case Study
Research: Empirical Positivistic Testing of Data to
Alternative Theories 241
13. Applying the Long Interview in Case Study Research 263
14. Tipping-Point Modeling in Case Study Research 291
15. Participant Observation Research in Organizational Behavior 321
viii Contents
16. Systems Thinking and System Dynamics Modeling 343

17. Fuzzy Set Social Science and Qualitative Comparative Analysis 359
18. Conclusions: Principles for Doing Case Study Research 397
References 409
Subject Index 435
Preface
The need for a source offering broad and deep coverage of theory, methods, and
practice in case study research is the central premise for this book — Case Study
Research or CSR). CSR’s principal objectives include offering nitty–gritty details of
processes (steps) in building theory and designing, implementing, and evaluating a
broad range of case study research methods — coverage and depth that you will not
find elsewhere in one source.
CSR includes comparing the criticisms and strengths of case study research with
theory and methods that rely on matrix–algebra-based hypotheses testing. CSR
includes introductions and details within lengthy examples of using qualitative
comparative analysis (QCA) with available software (fsqca.com) as well as an in-
depth treatment on building in degrees-of-freedom (DOF) in case study
research. QCA is a Boolean-algebra-based approach for formal testing of the
accuracy of complex statements of contingent relationships among recipes of
antecedent conditions in predicting outcome conditions. Building in DOF
analysis is often discussed in case study research but not to the depth that you
find in this book — formal tests of hypotheses are possible in case studies using DOF
analysis as well as other methods (e.g., latency response methods and system
dynamics modeling).
Students and scholars — recognizing the limited ability of informants to
accurately report their own (mostly unconscious) thinking and doing processes —
and the problems relating to self-editing biases of informants while answering
questions — and the biases in the questions that researchers frame — frequently seek
alternatives to using fixed-point survey response instruments and collecting verbal-
only responses. Case study researchers worry about (insist on) achieving high
accuracy in understanding, explaining, and predicting thinking and doing processes.
Substantial evidence supports the view:
Most thinking occurs unconsciously

Humans have limited access to their own thinking-doing processes
More than one person affects the process under examination.
Consequently, case study researchers frequently find the use of one-shot interviews
with one person for each of 50–300 organizations (firm, family, or government
x Preface
organization) to be too low-grade in accuracy to be acceptable. Too much nuance is

missing, too much reality remains unknown, in studies that rely on one-shot
interviews. Chris Rock’s (American comedian and commentator) insight auto-
matically comes-to-mind here — ‘‘When you meet someone for the first time, you are
not meeting that person, you are meeting his representative.’’
Explicit consideration of four dimensions is relevant for theory building and
testing of processes involving humans:
Time: micro-seconds, minutes, hours, days, weeks, months, decades, centuries

Thinking: unconscious only, unconscious and conscious, complex problem-solving
People: one person, group; intergroup, nation, East–West and native cultures
Context: home, work, travel; first-time versus repeat; weather, noise level, odors.
Figure 1 shows two of these four dimensions: time and thinking. Figure 1 includes
superimposing several prevalent case-study research methods in these methods most
relevant time-thinking locations. For example, according to Jung’s (1916/1959)
archetypal theory human memory include genetic primal forces that affect automatic
responses to different context without conscious thought — behavioral and thinking
outcomes of responses learned over thousands of years.
Figure 1 includes 3 levels of thinking with unconscious thinking covering about 50
percent of the space, conscious thinking 40 percent, and meta-conscious thinking
(i.e., thinking about thinking) about 10 percent of the space. The space allocations
reflect the evidence that most thinking occurs unconsciously and humans
infrequently engage in meta-thinking issues — an aside: Gigerenzer (2008) may be
the most insightful scholar currently working on examining meta-thinking issues.
Note that Figure 1 attempts to show communication flows between the various
levels of thinking. While recognizing that conscious and unconscious thinking and
thoughts occur separately is useful theoretically, much thinking likely includes some
bits of both conscious and unconscious thoughts (Evans, 2008).
‘‘Go deep!’’ Go deep by both going into the field — real-life contexts and by
learning a variety of case study theories and methods is the advice and direction that
this book directs you to follow. This suggestion is a corollary to Weick’s (1979)
famous suggestion, ‘‘Complicate yourself !’’
the complicated individual can sense variations in a larger environ-

ment, select what need not be attended to, what will not change
imminently, what won’t happen, and by this selection the individual is
able to amplify his control variety. He safely (that is, insightfully)
ignores that which will not change, concentrates on that which will,
and much like the neurotic psychiatrist is able to anticipate significant
environmental variation when and where it occurs. Complicated
observers take in more. They see patterns that less complicated people
miss, and they exploit these subtle patterns by concentrating on them
and ignoring everything else. (Weick, 1979, p. 193, italics in the
original)
Consciousness
Meta-conscious
System Dynamics (Chapter 16)

Building degrees of freedom TAT
Some “unintentional” (Campbell 1975) (McClelland
d QCA (Chapter 17)
participant observation 1988)
responses
(Van Maanen 1979) Conscious Subjective
Personal
Long Interview Method Introspection
(McCracken 1988) (Chapter 5)
Automatic retrievals c
(Woodside &
Trappey 1992a)
Conscious and Unconscious Time
Micro Seconds Minutes Hours Days Weeks Months Years Decades Centuries 10,000 + years
Means-
End
Chain
FMET (Chapter
(Chapter 7) 11) b
Latency response Storytelling theory and research
(Brunel, et al. 2004) (Chapter 3) Confirmatory
Personal
Introspection
ZMET (Chapter 5)
(Zaltman
2003)
fMRI (brain-
scanning) a Archetypal theory
(Cabeza and and research
Nyberg 2000) Unconscious (Carl Jung 1916/1959;
Chapter 4)
Genetically coded unconscious
Figure 1: Time and thinking property spaces.

Notes: a–b–c–d, Communication linkages among unconscious and conscious thinking; QCA, qualitative comparative analysis;
FMET/ZMET, forced/Zalatman metaphor elicitation techniques; TAT, thematic apperception test.
Preface
xi
xii Preface
CSR offers deep coverage of 14 case research methods; the closing chapter
offers 12 specific principles to implement that serve to increase accuracy of
what is happening and what will happen in real-life contexts-processes involving
thinking and behavior by humans. Useful tools for going deep and for complicating
yourself!
Acknowledgments
Working with Ph.D. students and colleagues on case study research projects across
several decades inspires this book. I very much appreciate the hard work and insights
these former students and now colleagues provide. Elizabeth Wilson, James Taylor,
Gabriel Perez, Gerry Waddle, Sylvia von Wallpach, Drew Martin, Marcia Sakai,
Shih-Yuh Hsu, Roger Marshall, Dehuang Ning, Marylouise Caldwell, Roger March,
Inja Ahn, Samir Gupta, Wenhsiang Lai, Wim Biemans, Eva Thelen, Günther
Botschen, Hans Mühlbacher, Kyung Hoon Kim, Ken Miller, Suresh Sood, William
Davenport, and Carol Megehee are gifted scholars and I very much appreciate their
working with me — in some cases on several research studies.
Special thanks to Carol Megehee for reading and correcting page proofs of the 18
chapters. ‘‘Master Wordsmith’’ and scholar are appropriate titles for Professor
Megehee.
I appreciate the time made available to me by Gerald Smith, Head of Department
of Marketing, and Dean Andy Boynton, Boston College, to complete this book.
Your support enabled closure of this project in 10 rather than 12 to 14 years. Thank
you both.
Claire Ferres and Rebecca Forster of Emerald Publishing provided excellent and
timely guidance during all phases in preparing this book for publication. Their hard
work and encouragement is much appreciated and will be long remembered.
The assistance by Maureen Preskenis in revising and editing the reference pages is
much appreciated.
Very special thanks to the informants whose stories this book tells. Thank you for
sharing your life experiences, interpretations, and evaluations. Hopefully, each
informant participating in the studies that this book presents can still see her or his
own tree among the forest of case study reports among the descriptions and
interpretations in the chapters.
Chapter 1
Building Theory from Case Study Research
Synopsis
This chapter provides a new definition for case study research (CSR).
Achieving a deep understanding of processes and other concept variables,
such as participants’ self perceptions (an ‘‘emic view’’ of what’s happening and
‘‘why I did what I did’’) of their own thinking processes, intentions, and
contextual influences, is identified as the principal objective of CSR. Using
multiple methods to ‘‘triangulate’’ (i.e., confirm and deepen understanding by
using multiple sources all focusing on the same process/event) within the same
case is described.
This chapter describes core criticisms made by case study researchers of large
sample surveys. A need exists for a paradigm shift in research on organizational
behavior (including modeling the history of new product performance). The
chapter outlines the significant weaknesses of CSR as seen by other researchers.
The chapter examines Senge’s (1990) core propositions related to the ‘‘mental
models’’ of decision participants. Details illustrate the use of specific research
methods for case studies to achieve different research objectives and the
combination of objectives. Finally, the chapter illustrates basic concept
variables in case studies and briefly reviews twelve propositions relevant in
many case research studies. This chapter reviews classic and recent contribu-
tions to the literature of CSR.
Introduction: Achieving a Broad Perspective When

Defining Case Study Research
CSR is an inquiry that focuses on describing, understanding, predicting, and/or

controlling the individual (i.e., process, animal, person, household, organization,
group, industry, culture, or nationality). This definition is intentionally broader than
the definition that Yin (1994, p. 13) proposes:
A case study is an empirical inquiry that investigates a contemporary

phenomenon within its real life context, especially when the boundaries
between phenomenon and context are not clearly evident.
2 Case Study Research: Theory Methods Practice
For a given study, focusing the research issues, theory, and/or empirical inquiry
on the individual (n ¼ 1) is the central feature of CSR. As Skinner notes (1966, p. 21),
‘‘y instead of studying a thousand rats for one hour each, or a hundred rats for ten
hours each, the investigator is likely to study one rat for a thousand hours.’’ This
view is not intended to imply that CSR is limited to a sample of n ¼ 1. The reporting
of several case studies in one inquiry is possible when the inquiry is to estimate the
size of an effect (i.e., the strength of a relationship between two variables) rather than
to generalize to a population. For example, meta-analyses (e.g., Hunter, Schmidt, &
Jackson, 1982) provide tools for estimating strengths of relationships (i.e., effect
sizes). Also, reports of multiple case studies are available in organization science (e.g.,
Nutt, 1998) involving business-to-business contexts. In the marketing literature,
Howard and Morgenroth (1968) illustrate transforming the research context in one
supply chain from n ¼ 1 to n W30 by examining alternative thought/action routes
taken separately, but seemingly similar, decisions that include five principal parties: a
senior decision-maker, a regional manager, a local distributor, and two sets of
competitors.
A key point to our definition is that CSR is not limited to contemporary
phenomenon or real-life contexts, especially when boundaries between phenomenon
and context are not clearly evident. Digging up the bones of the U.S. President
Zachary Taylor in 1996 to determine if he was assassinated is an example of CSR;
B.F. Skinner’s experiments in controlling the behavior of his infant daughter are an
example of CSR. The defining feature of CSR lies in the supreme importance placed
by the researcher on acquiring data resulting in describing, understanding,
predicting, and/or controlling the individual case.
Why Case Study Research is Useful

A substantial amount of psychological, management, and marketing research focuses
on the decisions and the behaviors by individuals and groups within and between
organizations (Woodside, 1992; Woodside & Wilson, 2000). The most frequently
used research method in the field involves sending a mail survey of mostly closed-
ended questions covering 10–20 research constructs. The request usually made is that
the questionnaire be completed by one person per firm, without comparing any other
person’s answers. The reported response rates for such studies typically range from
8 to 30 percent.
This dominant logic assumes that the responding individual is willing to report her
own thinking process, the thinking processes of others involved in the decision
process, and the sequence of events that occurred over several days, weeks, months,
or years. The dominant research paradigm assumes that the research constructs (e.g.,
role ambiguity, trust, closeness of supervision) measured on fixed-point scales
provide the nuance necessary for capturing the thinking/doing processes under study.
Yet the scientific literature on thinking concludes that approximately 95 percent of
thought is subconscious (Wegner, 2002; Zaltman, 2003) and that people have only
Building Theory from Case Study Research 3
limited access to their own thinking processes, not to mention the thinking processes
of others. Consequently, research methods attempting to measure ongoing thinking
(e.g., Van Someren, Barnard, & Sandberg, 1994) and thinking by the same person
using multiple interviews over several weeks (e.g., Cox, 1967; Cyert, Simon, & Trow,
1956; Witte, 1972; Woodside & Wilson, 2000), methods to bring up subconscious
thinking (e.g., Schank, 1999; Fauconnier, 1997), and interviewing the multiple
participants involved in the thinking/doing under study (e.g., Biemans, 1989) not
only are particularly useful steps, but also they become mandatory if we really want
to achieve deep understanding in research on thinking/doing processes in industrial
marketing.
‘‘I Hate Lying Like That’’
The operational constructs using closed-ended responses developed by researchers

fail to uncover the deep nuances and dynamic interactions between thoughts and
actions within and between individuals that occur within industrial marketing
contexts. The following story illustrates such nuances that CSR can capture in ways
unlikely to be captured by closed-ended mail survey responses. The story involves a
sales call made by a representative of an industrial distributor of copiers and printing
equipment (this sales call was overheard by the author who rode in the same vehicle
with the sales rep). During the selling/buying discussion involving the new purchase
requirements, the customer mentioned that the copier purchased recently from the
sales rep was broken again. Both the sales rep and the customer mentioned that the
copier had needed a service technician to repair it almost every week since it was
installed six weeks before. The sales rep responded to the customer’s concern by
saying, ‘‘I’m sorry you’ve experienced so many problems with your new copier. We
will get to the bottom of the situation. It’s a fine piece of equipment and we will solve
the problem so it doesn’t keep coming up.’’ After getting back in his car, the sales rep
remarked to the researcher, ‘‘The copier is a piece of shit; I really hate lying like that
[to a customer]. It’s really going to hurt my relationship with the guy.’’ The sales rep
elaborated that a competing distributor carried a line of copiers that were far
superior in performance and reliability compared to his product line.
Three-Person and Five-Way Mental Processing in Industrial Marketing
Most studies in consumer, business, and industrial marketing usually focus on only
one of five mental processes, that is, verbalized thoughts. Figure 1 depicts such
thoughts as Level 1 thinking. The other four levels shown in Figure 1 include the
following mental processes.
Level 2 mental processing includes conscious editing of thoughts surfacing from
unconscious processing, spreading, and combining of thoughts held in conscious
processing. These thoughts include thoughts heard by both the person verbalizing
2
Researcher
Mental Processing levels:
3 5
1.Verbalized thoughts
2.Conscious editing of thoughts surfacing from subconscious and
mixing/spreading with thoughts verbalized and heard
3.Thoughts surfacing and vice versa being stored in subconscious
4.Unconscious processing between parties 1
5.Unconscious processing within the individual
4
1
2 2
3 3
4
5 5
Thinker, Decision-Maker, Thinker, Decision-Maker,

In Marketer Organization In Customer Organization
Figure 1: The multiple mental processes in research on industrial marketing–buying

thinking.
and hearing thoughts from another person. Level 2 processing requires much more
cognitive effort because of the attempt to handle three-way incoming thoughts from
the unconscious, the person’s own verbalizations, and the thoughts being received
from the other person. ‘‘How do I know what I think until I hear what I’ve said?’’
(see Weick, 1995) is a question that reflects the idea that a person interprets her own
thoughts after verbalizing them.
Level 3 mental processing is the surfacing of unconscious thoughts into conscious
processes (i.e., ‘‘spreading activation’’ of concepts held in ‘‘working memory’’ as well
as moving of some thoughts involved in conscious processing into unconscious
storage). Level 3 processing is automatic. An individual is often unaware of how the
thoughts came to the surface or what process occurred that moved her conscious
mind to focus on a new topic.
Level 4 mental processing includes unconscious processes between two or more
persons that do not surface into conscious processes. Each person’s nonverbal
communications influence, and may attract or repel, the other person in ways
unrecognized by both. ‘‘I don’t know why, but I don’t trust that guy’’ is a verbal
commentary of Level 4 processing.
Level 5 processing reflects a spreading activation within the unconscious of an
individual. This includes completing automatic thought and action routines without
surface recognition of the process. Level 5 mental processing may lead to behaviors
that the individual is unable to recognize or report performing (Bargh, 2002; see
Bargh, Chen, & Burrows, 1996, for an empirical study that relates to this
observation), and behaviors not done that the individual reports doing (see
Woodside & Wilson, 2002a).
Figure 1 depicts that the five mental processes also occur for the researcher as the
researcher attempts to observe and interpret the mental processes involved between
the two principals. Figure 1 illustrates the researcher’s limited ability to understand
all five processes occurring for the two principals, as well as within the researcher
herself.
Thus, the researcher’s perspective of her five mental processes would benefit
from explicit discussion and surfacing efforts in the form of introspections (see
Wallendorf & Brucks, 1993, for recommendations on how to improve introspection
studies in consumer research). The dominant theory in-use for research on industrial
marketing/buying processes relies on the assumption most likely held implicitly, and
not explicitly, by researchers that a deep understanding of such processes can be
acquired using answers from direct questioning across many firms of one respondent
per firm within a marketing or buying organization. Researchers rarely stop to ask
introspectively, what are my unconscious processes that are influencing the design
and execution of my study? How can I surface unconscious thoughts held by me?
Is acquiring informants’ answers to closed-ended questions enough for a deep
understanding of the thinking and doing processes that I am studying?
The process of answering questions always involves a degree of introspection and
‘‘auto-driving’’ (see Heisley & Levy, 1991) by an informant. The person answering
questions must retrieve some bits of information stored in long-term memory,
organize and edit the bits, and create a verbal or written response in a form that she
believes that the researcher is able to understand. If the findings from research in the
mental processing literature are accurate that most mental processing is unconscious
and informants have very limited ability in surfacing unconscious thoughts, then
acquiring a deep understanding of industrial marketing/buying processes from
conscious responses to direct questions from one respondent using a single
questionnaire must be supplemented by using alternative data collection methods.
Auto-driving indicates that the interview is ‘‘driven’’ by informants who are seeing
and hearing their own behavior. Auto-driving addresses the obtrusiveness and
reactivity inherent in consumer behavior research by explicitly encouraging
consumers to comment on their consumption behavior as ‘‘y photographs and
recordings represent it’’ (Heisley & Levy, 1991, p. 257). However, auto-driving relates
implicitly to all informants’ attempts to retrieve, organize, edit, and report answers to
questions. Asking the informant to collect, organize, and describe photographs of
themselves or to use other pictures (e.g., via Zaltman’s metaphor elicitation
technique, ZMET) to describe a context or themselves embodies explicit auto-
driving tools that can be useful for bringing up unconscious processes (e.g.,
Christensen & Olson, 2002). The researcher observing a marketer/buyer meeting, and
subsequently asking one of these two parties to describe the meeting that just
occurred, is another example of auto-driving.
Deep Understanding: The Principal Objective of Case Study Research
Any combination of the following purposes may serve as the major objective of CSR:
description, understanding, prediction, and control. However, we propose that deep
understanding of the actors, interactions, sentiments, and behaviors occurring for a
specific process through time should be seen as the principal objective by the case
study researcher. Deep understanding in CSR includes: (1) knowledge of
‘‘sensemaking’’ processes created by individuals (see Weick, 1995) and (2) systems
thinking, policy mapping, and systems dynamics modeling (e.g., Hall, 1991) — what
might be labeled appropriately as meta-sensemaking.
Sensemaking is how the individual (i.e., person, group, and/or organization) make
sense of stimuli. Sensemaking foci include: (1) focusing on what they perceive;
(2) framing what they perceive; (3) interpreting what they have done, including how
they solve problems and the results of their enactments (including the nuances and
contingencies in automatic and controlled thinking processes). Because gaining
‘‘thick description’’ (see Geertz, 1973b, pp. 5–6; Sanday, 1979; Arnould &
Wallendorf, 1994) can be restricted to varying levels of depth and detail, thick
description alone is not enough. The resulting data and information from a thick
description may focus on surface details only, for example, describing the physical
characteristics of the environments, actors, and their conversations. To learn (1) the
subjective significance of persons and events occurring in a case study and (2) the
linkages and underlying (or, influence) paths among concept variables identified in a
case requires deep understanding.
Research Steps Required to Achieve Deep Understanding
Achieving deep understanding in CSR usually involves the use of multiple research
methods across multiple time periods (i.e., triangulation; see Denzin, 1978).
Triangulation often includes: (1) direct observation by the researcher within the
environments of the case, (2) probing by asking case participants for explanations
and interpretations of ‘‘operational data’’ (Van Maanen, 1979), and (3) analyses of
written documents and natural sites occurring in case environments (Figure 2).
The category of operational data includes spontaneous conversations of
participants in a case, activities engaged in and observed by the researcher, and
documents written by the participants. ‘‘Presentational data’’ are the appearances
and answers to inquiries that informants strive to establish and maintain ‘‘in the eyes
of the fieldworker, outsiders and strangers in general, work colleagues, close and
intimate associates, and to varying degrees, themselves’’ (Van Maanen, 1979, p. 542).
Data in this category [presentational] are often ideological, normative,

and abstract, dealing far more with a manufactured image of idealized
doing than with the routine, practical activities actually engaged in by
members of the studied organization. In short, operational data deal
Transition: Transition:
T1 hours, days, T2 hours, days, T3
weeks, months, weeks, months,
Interview year Interview year Interview
participants participants participants
Observe Document Observe Document Observe Document

participants’ collection participants’ collection participants’ collection
actual behavior and analyses actual behavior and analyses actual behavior and analyses
in real time in real time in real time
Figure 2: Triangulation in CSR. Note: Showing only three time periods is arbitrary; the key point: the case study researcher
often prepares written narratives of his or her interviews, direct observations, and document analyses; then, these narratives are
presented to selected participants in the following time period to verify that the narratives include the details reported,
observed, and found in the previous time period. For examples, see Nutt (1993) and Howard and Morgenroth (1968).
Building Theory from Case Study Research
7
with observed activity (the behavior per se) and presentational data
deal with the appearances put forth by informants as these activities
are talked about and otherwise symbolically projected with the
research setting. (Van Maanen, 1979, p. 542)
Gaining deep understanding often includes research to learn the ‘‘mental models’’
(Senge, 1990; Huff, 1990) of the participants. A mental model is the set of
propositions a participant in a case understands to be reality — that is, an accurate
portrayal of the causes, events, and outcomes relevant in the case. Each person
studied in a case has a set of related but unique mental models describing
1. The ‘‘typical’’ steps (i.e., persons, conversations, behaviors, and events) that occur
in the process being studied by the researcher.
2. The steps that should occur in the typical process (i.e., the participant’s normative
mental model).
3. What actually occurred in a given process, for example, the most recent process
completed or a completed process ‘‘strategically’’ important for the organization.
4. The participant’s perceptions of how another specific person or others in the
organization, in general, understand the details of the process being examined.
Senge (1990) makes a number of telling points about decision-makers’ mental

models. Here are three of his points particularly worth noting for CSR. First, the
mental model of any one person interviewed in a case study often fails to match
closely with the direct observations made by the researcher or other persons
interviewed (taken individually); the perceptions and beliefs both expressed by the
interviewee and formed by the researcher from direct observation are likely to miss
important details and depth of understanding. Second, mental models are rarely
made explicit and tested by participants in the case; they are formed tacitly;
participants often assumed their mental models to accurate views of: (1) what has
occurred and (2) why it has occurred (or what should and should not occur) and why.
Third, important feedback relationships among variables (i.e., ‘‘loops,’’ such as
increases in variable A leads to decreases in variable B that leads to further increases
in variable A) go unrecognized by the participants in the case being studied. Hall
(1978, 1984) provides the detailed illustrations of all three of these points.
The mental model of a participant in a case study is an emic representation of
reality. For example, an emic report is the verbatim ‘‘sensemaking’’ comments made
by the individual under study in a case. The interpretation of the same process
provided by the case study researcher is an etic representation of reality.
Etic representation in CSR often includes description and explanation of emic
meaning as well as building composite accounts of the process based on data from
triangulation. Thus, the collection of emic mental model accounts from interviews
represents one set of data for the case study researcher.
Collecting operational, as opposed to presentational, data is a core strength of
CSR. A core proposition within the mental models of most case study researchers
is that operational data often vary widely from presentational data. Thus, case
researchers seek a deep understanding by directly observing in ‘‘real time’’

(Arnould & Wallendorf, 1994) and (when possible) asking case participants, ‘‘What
exactly is happening right now? What were the triggering events leading up to
what happened? What is the meaning of what just happened to the case participants?
What is going to happen next because of what has just happened?’’
A distinguishing belief embraced by case study researchers (often implicitly) is that
participant verbal reports of conversations, behaviors, and events distort and fail to
include details necessary for deep understanding of the processes under study. ‘‘Rich
as I believe these [respondent] interviews are, they are frozen in time, individual
statements only vaguely anchored in the social and historical context that created
them’’ (Wolf, 1990, p. 351). As Arnould and Wallendorf (1994) conclude, ‘‘Because
of the inherent inconsistencies and ellipses in oral reports, verbal data alone are not
regarded as sufficient for developing ethnographic interpretation. The preferred
corrective for these limitations is combining verbal report data with data from long-
term participant observation in cultural context.’’ If long-term participation is
impossible, the alternative includes adopting multi-method procedures, for example,
interviewing all parties participating in conversations and events under study, and
the collection of documents and additional unobtrusive measures (see Webb &
Weick, 1979; Webb, Campbell, Schwartz, & Sechrest, 1966).
The Core Criticisms of Large Sample Survey Research

and Case Study Research
The core criticisms made by case study researchers of large sample surveys (i.e.,
n W 100) of one person in a household, informal group, or organization include:
(1) The failure to confirm reported conversations, behaviors, and events, independent
from the one person surveyed. (2) The failure to collect the necessary detail for
gaining deep understanding of the mechanics and reasons embedded in the processes
examined. These criticisms are countered by critics (i.e., researchers using large
samples persons in identified populations) to CSR by a core criticism of their own:
CSR results are not generalized to a population, the particular case included in a
given case study is so unique that it represents a one-off context.
Briefly, the following observations relate to this debate. First, we advocate
adopting the view (i.e., mental model) that any one respondent is severely limited in
reporting the details necessary to learn to deeply understand the process being
studied — some use of triangulation of methods and multiple informants is necessary
to confirm and deepen information. Second, the objective of CSR is not to generalize
findings to a population but to probe theory (i.e., one or more explicit mental models
related to the processes being examined; see Campbell, 1975; Yin, 1994).
Third, the criticism can be directed at any one study as being idiosyncratic in its
selection of population, data collection procedures, data handling and analysis, and
selection of subjects for study from the population; labeling a study as being
idiosyncratic is one step to concluding that the data collection procedures used and
findings made cannot be replicated — a false conclusion given that no one study can
be replicated perfectly. Fourth, several case studies can be completed and fixed
samples of cases may be drawn; case studies are not limited to n ¼ 1; multiple cases,
or multiple behaviors and events within one case study, can be examined to deepen
understanding of patterns and contingencies related to theory (see Nutt, 1993, for an
example of a large sample case study and McCracken, 1988, for a defense of multiple
case sampling for identifying patterns across cases).
The objectives here do not include attacking large-sample, one person per
household or organization, one-time survey research studies. However, the
substantial amounts of respondent reporting of events that did not occur, and the
absence of reporting events that did occur in such studies (see Farley & Howard,
1975), as well as the absence in such studies of details necessary for deep
understanding of processes being studied, are additional motivators for adopting
CSR methods.
In an essay on ‘‘Organizational Performance as a Dependent Variable,’’ March
and Sutton (1997, p. 702) bemoan the fact that the bulk of research on identifying the
causes of organizational performance rely on cross-sectional data and retrospective
studies: ‘‘These studies may actually tell us less about the determinants of
performance than about the ways performance information affects memory,
cognitive processing, and story telling.’’ Retrospective bias may be the telling
weakness of most empirical studies on measuring the performance of new product
introductions specifically and, in general, on most studies measuring other areas of
organizational performance.
Performance information itself colors subjective memories, percep-

tions, and weightings of possible causes of performance. Informants
exist in a world in which organizational performance is important.
That world is filled with widely believed conventional stories about the
causes of good and poor performance. As a result, retrospective
reports of independent variables may be less influenced by memory
than by a reconstruction that connects standard story lines with
contemporaneous awareness of performance results. (March & Sutton,
1997, p. 701)
March and Sutton (1997) also fault theory building that includes not viewing
organizational performance as an independent variable: ‘‘y the theoretical ideas and
analytical models that are normally used [e.g., by the majority of organizational and
inter-organizational researchers] ignore a variety of feedback loops that are likely to
be important.’’ Organizational behavior as a series of feedback loops is a suggestion
stressed by Senge (1990) as one central for shifting research paradigms from linear
thinking to system thinking. The importance of building and testing complex models
is critical to capture the impact feedback loops on performance demonstrated
empirically by Hall (1976, 1984).
The crucial point here: deep understanding of the multiple perceived realities that
occur through time in organizations and households requires the use of multiple data
collection methods across several time periods. Meta-sensemaking — the researcher’s

pursuit of a vision of reality lying outside the social beliefs of one person interviewed
per organization or household — requires that additional data be collected (e.g.,
interviews of other persons involved in the behavior being examined; direct
observation; and the analysis of documents and other unobtrusive measures).
Data Collection and Analysis Methods Useful for Case Study Research
While the literature often CSR associates with using qualitative research methods, we
advocate viewing CSR as not being restricted to one set of research methods.
Quantitative methods, including statistical hypotheses testing, are appropriate for
many CSR studies. Also, the value of most CSR reports may be enhanced
considerably by using multiple tools, both qualitative and quantitative methods, in
the same study.
The value of most CSR reports increases with the use of dissimilar, multiple
research methods and the inclusion of multiple study objectives (e.g., see Pettigrew,
1995). One of our objectives for this book is to provide insights for achieving useful
descriptions and explanations and to go beyond these objectives — to describe the
additional, possible objectives of predicting and controlling case study behavior.
Theory Building and Theory Testing Using Case Study Research
The relevant literature often associates CSR with theory building versus theory
testing (Dyer & Wilkins, 1991; Eisenhardt, 1989). However, examples of theory
testing reports using CSR are available. The quality of a CSR report often may be
increased dramatically by designing the study to include both theory building and
theory testing (e.g., see Howard & Morgenroth, 1968; Gladwin, 1989). Consider
adopting the broader view: CSR is often appropriate for both theory building and
theory testing. This book describes several examples of successfully doing theory
building and testing by CSR scholars.
The Objectives of Case Study Research

CSR is appropriate for several research objectives: description, explanation,
prediction, and control of the individual process, animal, person, household, group,
or organization. Thus, we advocate that CSR is often appropriate for several
research objectives going beyond description and explanation. Description in CSR is
the attempt to answer who, what, where, when, and how questions. Explanation in
CSR is the attempt to answer the why question. Sometimes CSR explanations
include reports provided by: (1) the direct participants in the case; (2) informed third-
party observers to the case; and (3) the case study researcher. Prediction in CSR
includes forecasting near term and/or long-term psychological states, behaviors, or

events that will follow within the individual case and/or similar cases. Control in CSR
includes attempts to influence the cognitions, attitudes, and/or behaviors occurring in
an individual case. Control is a relevant objective in experimental studies of single
cases (see Hersen & Barlow, 1976), for example, in studying the efficacy of alternative
methods for achieving behavioral changes desired by: the participants (e.g., subjects,
clients, or patients) in a case; an organization (e.g., a product/service marketer,
a government lobbyist); and a non-profit organization or society (e.g., a department
of social work, a school or university).
Each of these four research objectives can be viewed beneficially as orthogonal to
the other three objectives (Figure 3). Thus, we advocate embracing the mental stretch
that case description is possible without explanation, and explanation without
description is also possible. Also, every possible four-way combination shown in
Figure 3 occurs in CSR.
Let us concentrate on a few of the cells in Figure 3 to demonstrate the possibilities.
Cell 1 is the null CSR report: no information relevant to describing, explaining,
predicting, or controlling is included in the study; an abstract artist rendering of
a case study is illustrative of cell 1.
Cell 7 includes attempts to ‘‘build-in degrees-of-freedom’’ (Campbell, 1975) in a
theory of behavior that may be relevant to a given category of cases. Creating a set
of 10, 20, or 30 propositions that the case study researcher proposes as typical of
decision-making and behavior for a given case theory is illustrative of building-in
degrees-of-freedom. Such theory building may be content-free of a specific case, that
is, the theory may be formed deductively.
Weick’s (1969; also see Weick, 1979; Bougon & Komocar, 1990) dynamic causal
modeling of organization and change is an example of a content-free theory for
Prediction
Objectives No Yes
Control
Description Explanation No Yes No Yes
1. Abstract 2. 3. 4.
No No (Art)
5. 6. 7. Building-in 8. Dynamic
degrees of causal
No Yes
freedom modeling
9. Naïve 10. 11. 12.

Yes No observation
13. Participant 14. Action 15. Ethno- 16. Clinical

observation research graphic (psychology,
Yes Yes
decision tree psychiatry)
modeling
Figure 3: Classifying CSR by research objectives.

explaining, predicting, and controlling case behavior without starting with a

description of a particular case. Several propositions are included in dynamic causal
modeling that these researchers believe useful for achieving effective, long-lasting
change (i.e., control) in a social system (i.e., a given case). Here is an example
proposition:
A social system’s identity nodes and loops are typically over

determined by the pattern of the whole and are almost impossible to
change directly (e.g., Warwick, 1975) or in a piecemeal fashion (e.g.,
Miller & Friesen, 1982). Thus, within a holistic approach, when the
nodes and loops of interest to strategic change coincide with the social
system’s identity nodes and loops, the solution to strategic [long-
lasting] change is indirect. The solution is to focus change efforts on
peripheral loops rather than on those directly responsible for system
identity. (Bougon & Komocar, 1990)
Cell 9 is a description without explanation, prediction, or control: a naı̈ve report

of events in a case by a reporter totally unfamiliar with what is occurring in the case.
Andy Griffin’s (a U.S. actor/comedian) humorous portrayal of a backwoodsman
reporting the first-time observance of an American football game illustrates cell 9.
Representative of cell 13 CSR, most participant observation studies include the
objectives of providing thick descriptions and deep explanation of the processes and
events occurring within a specific case; developing models to predict outcomes or
future events and designing change strategies to influence (i.e., control) case behavior
are not primary objectives in such studies. Qualitative inquiry is an example of a
scientific periodical with a primary focus on thick description and deep under-
standing in CSR.
Participatory action research reports are representative of cell 14 CSR because the
objectives of such studies include thick description, deep understanding, and
attempts to influence the design and outcomes of behaviors occurring in a case,
without attempting to build predictive models for estimating values of proposed
dependent variables (e.g., see Whyte, 1990, 1991).
Ethnographic decision tree modeling (EDTM) does not include attempts to
influence outcomes by the researcher but do include model building to predict
estimates of values of specific dependent variables. Thick description and deep
understanding provide the foundation for EDTM; thus, EDTM is illustrative of cell
15 in Figure 3. The Journal of Contemporary Ethnography is an example of a scientific
periodical with a primary focus on such research reports.
Applied theory development in social work and clinical psychology/psychiatry
illustrates cell 16 CSR — the aims of the researcher include thick description, deep
understanding, prediction of outcomes, and control of behavior exhibited in a
specific case. The Journal of Applied Behavioral Analysis, the Journal of Clinical
Psychology, and Behavior Therapy are scholarly publications related to this category
of CSR.
Suggestions Regarding Selection of Case Study Research Objectives
The aim here does not include the claim that moving away from cell 1 toward cell 16
is always best. We do suggest greater awareness of the possibilities of planning to
accomplish multiple objectives in CSR. Also, different CSR tools (i.e., research
methods) are relevant for achieving different objectives. For example, EDTM is
useful in particular for building theory for predicting outcomes occurring naturally in
cases and action research is useful in particular in designing strategies to change
behaviors and outcomes in cases. Thus, skill building in learning research tools
relevant for case studies across a wide range of objectives should complement your
training in advanced CSR.
Core Propositions in Case Study Research

Several core propositions in CSR are summarized visually in Figure 4. Briefly, 12 of
these propositions are described here. (1) Time is recognized explicitly in modeling
behavioral processes in CSR. For example, in the studies of n ¼ 1, the possibility of
variability in responses (i.e., events or outcomes) is built into the study by observing
behavior of the respondent across several time periods.
(2) In many case studies, multiple individuals participate in different conversations
and behaviors within one time period in the case. Conversational analysis is the
E1
E6
E2
E1 E4
E3
B4 E3
B2
E5
B1
G1
G1 B5
SB1 SB3 B6
I1
I1
I2
I2
Concept variable:
I = individual I3 I4
G = group
G2 G2
= conversation G2 G2
= path of events I6
SB8
I5 I5 I7 I6
E = event I5 I5 I8
SB2 SB4 SB6
B = behavior I7
T = time
SB=sentiments and G3 G3
beliefs I10 I10
I9 SB7
SB5 I9
I11
T1 T2 T3 T4
Figure 4: Concepts and propositions in CSR. Source: Original figure but relates
to Calder (1977, Figure 2, p. 198).
primary focus of many case studies. (3) Individuals are members of identifiable
households, groups, or organizations. (4) Much like actors appearing in different
scenes in a play, different individuals in the same group may participate in
conversations and behaviors in different time periods; for example, note in Figure 4
that individual 6 is found in conversations in T1 and T3.
(5) When examined deeply, most cases involve three or more informal groups or
organizations that affect the process and outcomes under study. The involvement of
‘‘third-parties’’ in interorganizational case studies is the focus of several studies in
supply-chain management (e.g., see Biemans, 1989).
(6) Identifiable individuals and groups engage in identifiable behaviors leading to
identifiable events (i.e., outcomes). (7) Specific events influence the occurrence of
other events. (8) Some events are repeated, for example, E1 to E2 to E1 in Figure 4.
(9) The presence of certain events (e.g., E5) changes the influence of another event;
for example, E6 occurs in T4 following E1, given that E5 has occurred. Thus, CSR
and theory building often includes contingency propositions of complex relation-
ships. (10) Not all members of a group communicates with every other member in the
same group; for example, in Figure 4 I1 talks with I2 and I3 in Group 1 in T1, but I1
does not talk with I2. (11) Participation in the case of identifiable groups occurs only
in a limited number of time periods; for example, G1 is found in T1 and T2, and G3
is found in T2 and T3. (12) Conversational contacts within a group may increase or
decrease from one period to the next within a case; for example, witness the increase
in contacts in G3 between T2 and T3 in Figure 4.
Figure 4 also includes sentiments and beliefs (SB). Sentiments and beliefs are
relevant to individuals and to groups. Both individuals and groups have unconscious
and conscious opinions as to what is true or false and good or bad. SB research
includes findings that some sentiments and beliefs change through time for
individuals and groups while other sentiments and beliefs are static. How behavior
affects SBs and how SBs affect behavior are topics of both theory and empirical
examination in case study research (see Homans, 1974). Events in Figure 4 include
decisions, performance outcomes, and revelatory incidents — Figure 4 does not
include the attempt to distinguish among these three categories of events.
Different streams of CSR focus on different concept variables that appear in
Figure 4. For example, policy mapping is the attempt to diagram, explain, and
predict recurring relationships among events in a case study (e.g., Hall, 1976, 1984,
1991; Howard & Morgenroth, 1968). Decision systems analysis is the attempt to
diagram and explain (but not to predict) relationships among non-recurring events in
a case study (e.g., Howard & Morgenroth, 1968).
The CSR methods appropriate for a given study depend on the nature of the
process being examined, as well as the interests of the researcher. For processes
being repeated with adjustments, such as managing a newspaper business or
pricing gasoline, policy mapping is an appropriate research tool. For one-off
processes, such as an individual or organization adopting a new technology, decision
systems analysis and EDTM are appropriate research tools. While many
different CSR methods are available, all include the recognition of the core concepts
in Figure 4.
Summary
CSR is an inquiry focusing on describing, understanding, predicting, and/or

controlling the individual (i.e., process, animal, person, household, organization,
group, industry, culture, or nationality). Any combination of the following purposes
may serve as the major objective of CSR: description, understanding, prediction, or
control. However, that deep understanding of the actors, interactions, sentiments,
and behaviors occurring for a specific process through time is the principal objective
by the case study researcher. The researcher should consider using explicit auto-
driving tools to aid in bringing-up unconscious mental processes among informants.
A mental model of a process provided by a participant interviewed in a case study
is an emic representation of reality. The interpretation of the same process provided
by the case study researcher is an etic representation of reality. Etic representation in
CSR often includes description and explanation of emic meaning as well as building
composite accounts of the process based on data from triangulation. Triangulation
includes: (1) direct observation by the researcher within the environments of the case,
(2) probing by asking case participants for explanations and interpretations of
operational data, and (3) analyses of written documents and natural sites occurring
in case environments.
Core criticisms made by case study researchers of large sample surveys consisting
of interviews of one person per household, informal group, or organization include:
(1) the failure to confirm reported conversations, behaviors, and events and (2) the
failure to collect the necessary detail for gaining deep understanding of the mechanics
and reasons embedded in the processes examined.
Core variables in CSR include individual and group behaviors through time
resulting in a sequence of paths of events (decisions, performance outcomes, and
revelatory incidents). Beliefs and sentiments held by individuals and groups are
additional core variables sometimes studied in CSR. No one CSR method is
appropriate for all studies.
Chapter 2
Bridging the Chasm between Survey

and Case Study Research
Synopsis
Chapter 2 describes how behavioral science research methods that management

and marketing scholars apply in studying processes involving decisions
and organizational outcomes relate to three principal research objectives:
fulfilling generality of findings, achieving accuracy of process actions and
outcomes, and capturing complexity of nuances and conditions. The chapter’s
unique contribution is in advocating and describing the possibilities of
researchers replacing Thorngate’s (1976) ‘‘postulate of commensurate complex-
ity’’ — it is impossible for a theory of social behavior to be simultaneously
general, accurate, and simple and as a result organizational theorists inevitably
have to make tradeoffs in their theory development — with a new postulate of
disproportionate achievement. This new postulate proposes the possibilities
and advocates the building and testing of useful process models that achieve
all three principal research objectives. Rather than assuming the stance that
a researcher must make tradeoffs that permit achieving any two, but not all
three, principal research objectives as, Weick (1979) clock analogy shows, this
chapter advocates embracing a property space (a three-dimensional box rather
than a clock) view of research objectives and research methods. Tradeoffs
need not be made; having-your-cake-and-eating-it-too is possible. The chapter
includes a brief review of principal criticisms that case study researchers
often express of surveys of respondents using fixed-point surveys. Likewise,
the chapter reviews principal criticisms of case study research studies that
researchers who favor the use of fixed-point surveys express.
Introduction
This chapter provides a new view of planning process research to achieve three
principal objectives (generality, accuracy, and complexity/coverage) rather than
adopting the view that the researcher must make tradeoffs as to which two of the
three objectives to accomplish. The chapter proposes replacing Thorngate’s (1976)

‘‘postulate of commensurate complexity’’ with a postulate of disproportionate
achievement as the guiding principle in planning research designs relating to case
study research.
After this brief introduction, section two reviews the principal criticisms by case
study researchers of fixed-point surveys. Section three reviews the principal criticisms
of empirical positivists of case study research. Section four describes how to go about
building bridges across the chasm in thinking and research methods to accomplish
all three objectives — all relevant for both empirical positivism and interpretive
research. Section five offers specific recommendations for planning research designs
to achieve generality, accuracy, and complexity/coverage. Finally, section six offers
conclusions and limitations.
Case Study Researchers’ Principal Criticisms of Fixed-Point Surveys
Case study researchers inherently find poor meals in fixed-point survey research
reports for four principal reasons. The first follows from this observation: such
surveys require informants to retrieve information explicitly (often stored in stories
and interpretations of events in long-term memories; see Schank, 1990) and transform
this information to responses using scales that usually range for low (1) to high (7) or
from strongly disagree ( 3) to strongly agree (+3). Given the substantial evidence
that most thinking is done implicitly (unconsciously) and informants have very
limited abilities to retrieve implicit thoughts explicitly (Wegner, 2002; Wilson, 2002;
Zaltman, 2003) and explicit thoughts are highly biased in favor of the nurturing the
informants’ self-egos (Argyris, 1982; Coopey, Keegan, & Emler, 1997; Laughlin,
1970), fixed-point self-report surveys on causes and processes lead to data that
provide inaccurate and meager information relating to the process under study.
The lack of meaningful policy implications from findings of five-point or seven-point
scale metrics relates to this point; the variable-level implications from such studies
lack specificity to real-life policy contexts.
Second, data from fixed-point surveys are often the core inputs to
performing structural equation models that predict models of dependent constructs
(themselves measured in fixed-point survey questions) that often include variables
such as trust, perceived quality, intention-to-buy, and satisfaction — with external
validity tests of the connection of such constructs to actual behavior absent from
such studies. Real-life participants engaging in thinking processes do not think
implicitly or explicitly using fixed-point responses even when they report doing so
(Woodside & Wilson, 2000); process participants think both implicitly and explicitly
using simple-to-complex recognition and evaluation heuristics (Gladwell, 2005;
Gigerenzer, 2001, 2007; Hogarth, 1987, 2001) that usually include a few different
combinations of causal conditions that lead to a given outcome event (e.g., purchase)
or alternative outcome (e.g., not purchase). The use of fixed-point measures
fails to capture the real-life outcomes of interest in behavioral science research.
Bridging the Chasm between Survey and Case Study Research 19
Third, correlation analysis, multiple regression, and structural equation modeling

(SEM) assume symmetrical relationships between an independent variable (either a
direct or an interaction expression) with a dependent variable — high values on the
independent measure relates to high values of the dependent measure and low values
on the independent relates to low values on the dependent measure. In real life,
relationships tend to be asymmetrical rather than symmetrical — the presence of a
specific combination of causal conditions results in sufficiency (but not necessity)
of a specific outcome with the causal conditional statement making no prediction
about the absence of the specific causal combination. Thus, low scores for a causal
combination of antecedent conditions usually relate to both low and high scores on
the outcome condition under study. For example, in asymmetrical associations
a study may predict and find that the presence of combination of three specific
levels for different antecedents in a causal recipe results in a high score for the
outcome condition but the same specific outcome both occurs and does not occur
across cases with the absence of the antecedent causal recipe. To illustrate this point,
this chapter includes a specific example of such findings in a business-to-business
purchasing study.
The nature of asymmetrical associations indicates that more than one route
operates that lead to a given outcome condition (e.g., purchase of a new product
design). Consequently, theory and empirical testing need to include consideration of
the alternative routes to a given outcome condition — one multiple regression model
(i.e., algebraic statement is insufficient in describing or understanding the multiple
causal recipes resulting in a given outcome).
Rather than reporting one to three multiple regression models or structural
equation models that are typically available in fixed-point survey-based research
reports, researchers need to consider and describe the alternative paths that are
particularly relevant for subsets of informants using methods that do not assume
symmetrical relationships.
Fourth, statistical tests of relationships build on linear additive models that aim
to measure the unique contribution of each independent variable to explaining the
variation in a dependent variable. Such models focus on providing an analysis of
net effects and are ill-equipped for measuring alternative complex causal recipes
resulting in the same outcome condition (e.g., purchase). Classical empirical
positivistic studies have severe limits in reporting and interpreting interaction effects
among three or more variables; such interpretations focus on estimating indirect
effects on a dependent variable above the direct effects of each independent variable
rather than on interpreting the relevancy of different forms of the interaction takes
on that affect the dependent variable. Conditional fuzzy set statements of alternative
causal recipes (that this chapter describes later) accurately express the various
routes to a given outcome while combinations of main and interaction effects in one
algebraic expression do not.
Case study researchers bemoan the lack of accuracy in representation and
prediction of the principal findings of researchers using classical empirical positivistic
tools. Ragin (1997) provides additional details of case research criticisms of classical
statistical hypotheses testing of associations.
Criticisms of Case Study Research by Empirical Positivists
Researchers applying classical empirical positivistic methods (e.g., correlation,

analysis of variance, regression analysis, and SEM using fixed-point survey data)
inherently find poor meals in case study research reports for four principal reasons.
First, many case study research reports include no explicit steps to create and test
theory. Empirical positivists express the view that using ‘‘direct research’’ (i.e.,
proposing explicitly to enter the field to collect data with no prior theory and that
theory follows data collection, Mintzberg, 1979) when entering the field (specific
process contexts) to observe and explain what is occurring and to build theory
inductively is unsound and impossible — the researcher carries personal cultural
value configurations implicitly into the field that affect her interpretations and
judgments and statements of conducting direct research without prior explicit theory
are inherently naı̈ve. The first reason stresses the failures to test associations formally
and failure to state theory a priori and to generalize to theory and other contexts as
telling weaknesses of case study research.
Second, the ‘‘thick descriptions’’ (Ryle, 1949; Geertz, 1973a) of processes in
specific context that increase the accuracy of findings in case study research reports
also reflect chaotic complexity to researchers using classical empirical positivistic
methods. Such thick descriptions also detract from the necessity to generalize beyond
the immediate case to support the validity of the propositions that the case study
researcher concludes from her observations.
Third, researchers criticizing case study methods view the variability in multiple-
person interpretations of verbal data in thick descriptions as a telling weakness. The
use of ‘‘member checks’’ (i.e., asking other case study researchers to read the original
case data and interpretations by the original researchers, see Hirschman, 1986) does
not sway the view of failure to achieve objectivity in interpretations of textural data
on opinions, beliefs, and interpretations and such soft data are integral to thick
descriptions.
Fourth, the practical relevance of a case study report to other contexts,
normative implications, or prescriptive implications is questionable both on grounds
of the absence of deductive theory and the collection of data to one or a very few
contexts — case study research provides inadequate replications to support either
generalization or practical relevance to aiding decision-making in other contexts.
This argument includes the view that the process in the case study report is not
necessarily relevant to any other contexts.
In summary, Figure 1 reflects the primary strength and weakness of classical
empirical positivistic and case study research methods in terms of Thorngate’s (1976)
first two objectives for advancing behavioral science: achieve generalization and
accuracy. The visual message of Figure 1 is that the strength of classical empirical
positivistic research and variable-based theory development (via fixed-point surveys
typically using 1–5 or 1–7 scaled items) is its ability to generalize via quantitative
modeling; however, Figure 1 depicts variable-based theory using classical empirical
positivism as failing to provide accurate, relevant views of real-life processes.
Figure 1 shows the opposite image for case study research methods and case-based
High
Fixed-point Fuzzy set
survey social science,
historical
research
General
Case study
Anecdote, research
SPI
Low
Low High
Accuracy
Figure 1: Research methods’ attainment of objectives.
theory development — highly accurate portrayals of real-word processes and little to

no generalization across other contexts.
Two additional sets of research methods appear in Figure 1. Anecdotes and
subjective personal introspections (SPI) (e.g., Sloan, 1964) appear as offering low
generalization and low accuracy. Fuzzy set social science (Ragin, 2000; Rihoux &
Ragin, 2009) and historical research (Golder, 2000; Golder & Tellis, 1993) methods
appear in Figure 1 as offering high generalization and high accuracy. The review that
follows includes all four of these methods.
Building Bridges across the Chasm
Members of both research camps appear to be guilty of the ‘‘law of the

instrument’’ — one uses the tool one has even if in context the tool’s use is highly
inappropriate. Dropping one’s tools is a proxy for unlearning, for adaptation, for
flexibility, in short, for many of the dramas that engage organizational scholars. It is
the very unwillingness of people to drop their tools that turns some of these dramas
into tragedies. These dramas, however, are not confined simply to the people in
organizations that scholars study. The scholars themselves are equally at risk.
Kaplan’s (1964, p. 28) ‘‘law of the instrument’’ portrays part of the risk: ‘‘Give
a small boy a hammer, and he will find that everything he encounters needs
pounding. It comes as no particular surprise to discover that a scientist formulates
problems in a way which requires for their solution just those techniques in which he
himself is especially skilled’’ (Weick, 1996). Learning to drop one’s tools and pick-up
tools more useful to the task-at-hand is difficult for several reasons that Weick (1996)
describes. Reasons include evidence supporting the assertions that experience
breeds comfort and same-group interactions support the status quo over change.
For example, ‘‘Campbell (1979) has shown how the learning theory developed in the
tight research group surrounding Kenneth Spence was less powerful than the
learning theory developed within the much looser group surrounding Edward
Tolman. These differences are explained in part by the relative ease people in
Tolman’s group had evaluating and changing ideas without regard for the effect of
these changes on their reputations and the relative difficulty people in Spence’s group
had when they tried to do the same thing’’ (Weick, 1996, p. 309).
Weick’s (1979, p. 261) battle cry is relevant for overcoming one camp thinking
in theory creation and research methods, ‘‘Complicate yourself ! Consider different
causes, other solutions, new situations, more complex alternatives, and take pleasure
in the process of doing so.’’ Several research methodologists provide advances
in theory and method that respond usefully to Weick’s battle cry and provide
useful tools for achieving three primary objectives relevant to Thorngate’s (1976)
‘‘postulate of commensurate complexity.’’ The postulate states that it is impossible
for a theory of social behavior to be simultaneously general, accurate, and simple and
as a result organizational theorists inevitably have to make tradeoffs in their theory
development. For example, the more general a simple theory is, the less accurate it
will be in predicting specific details. Karl Weick (1969) uses a clock face metaphor to
illustrate this postulate, with the word general inscribed at twelve o’clock, accurate at
four o’clock and simple at eight o’clock (Figure 2A). Weick uses this metaphor
to illustrate the inevitable dilemmas that are present in social scientific research:
one can only obtain the benefits of any two of the virtues of simplicity, generality and
accuracy, by sacrificing the third one’’ (Moss, 2002). Weick argues that theories
and research are classifiable as being either: ‘‘two o’clock research,’’ which is general
and accurate but difficult; ‘‘six o’clock research,’’ which is accurate and simple,
but weak on generality, or; ‘‘ten o’clock research,’’ which is general and simple, but
not accurate.
Figure 2B visualizes dropping the metaphorical tool of Weick’s clock and
Thorngate’s postulate of commensurate complexity and picking up a ‘‘property
space’’ (Lazarsfeld, 1937) view that all theoretical objectives are achievable for
behavioral scientists. These objectives include generality, accuracy, and complexity
(coverage). Property space analysis depicts a truth table view of all possibilities
involving possibility combinations of conditions or factors; with three objectives,
eight possible combinations are possible at least theoretically.
Figure 2b includes nine possibilities with the ninth indicating ‘‘triangulation’’ —
that is, the use of three sources/methods in classical case study research to confirm
observations (cf. Denzin, 1978). Note that in the three-dimensional box that the
objective of coverage (i.e., detail or complexity) replaces the objective of simple in
Weick’s clock. While the objectives of building a theory that is understandable and
useful support the call for simple models, advances in new research tools support the
7 8
3 4
High
9
General
5 6
High
1 2 Coverage/
Low Low Detail
Low Accuracy High
(A) (B)
Figure 2: Weick’s Clock and Woodside’s Box. (A) Weick’s (1969) Clock Metaphor
of Thorngate’s (1976) Postulate of Commensurate Complexity. (B) Woodside’s Box
Metaphor of Case and Multiple Case Study Research. Notes: Key to numbers in
Woodside’s box: (1) Anecdote, SPI; (2) Thick descriptions; role playing; FMET;
CPI; DSA; (3) Fixed-point surveys; (4) Fuzzy set social science; historical analysis;
simulation models of thinking and deciding; (5) Multiple anecdotes in different
contexts (e.g., Terkel, 1974; Tucker, 1967); (6) Multiple case studies in same contexts
(e.g., Nutt, 1984); (7) Naı̈ve observation; (8) Multiple-case system dynamics
modeling; (9) Triangulation: mixed-methods; decision systems analysis. SPI,
subjective personal introspection; CFI, confirmed personal introspection; FMET,
forced metaphor elicitation technique; DSA, decision systems analysis.
proposal that building useful and understandable behavior process theory that
includes coverage/complexity (e.g., delays in influence, feedback loops, tacit, and
deliberate knowledge) is achievable and has been occurring since the 1990s. The
intention is that the three-dimensional box includes a postulate of disproportionate
achievement: building and testing useful process models that are general, accurate,
and complex is possible.
A brief discussion follows for each of the nine locations in the three-dimensional
box with references to literature for each for further examination. The discussion on
fuzzy set social science (corner 4 in the box) lingers a bit because this set of tools is
new in organizational process research.
An Anecdote and Subjective Personal Introspections
An anecdote is a short story that tells an interesting tale often to support an argument.
Senator John McCain talked about the hardship a tax increase would place on
‘‘Joe the Plumber’’ in a 2008 U.S. Presidential debate with his opponent, Senator
Barack Obama. Because an anecdote is not necessarily typical of others or even the
protagonist in the story (e.g., the Joe in the ‘‘Joe the Plumber’’ story was not working
as a licensed plumber), anecdotal research is low in generality, accuracy, and coverage.
SPI (see Gould, 1991, 1995; Holbrook, 1986, 1995, 2005b) includes a family of
research methods that rely extensively or even exclusively on the researchers’ life
experiences as data (Wallendorf & Brucks, 1993). Woodside (2004a, 2006) describes
how SPI reports are representations of folk explanation theory of behavior
explanations (Malle, 1999, 2004) but that such explanations usually fail to include
implicit (unconscious) thinking of the informant–researcher and usually fail to
include sufficient coverage (i.e., complexity) to accurately explain real-life processes;
Woodside (2004a, 2004b, 2006) recommends and describes examples of using mixed-
methods strategies to complement and enrich SPI reports with confirmatory personal
introspections (CPIs).
Thick Descriptions, Role-Playing, and Forced Metaphor

Elicitation Techniques (FMET)
These methods provide highly accurate descriptions but usually are low in generality
and low in the amount of actual coverage/complexity of process behavior. Case study
reports using the long interview method (McCracken, 1988) focus primarily on
individual retrospective descriptions by a few informants of process experiences.
McCracken (1988) recommends completing five long interviews with informants
for each theoretical sample of interest. A theoretical sample includes selecting
representative cases (e.g., informants) for specific, or for all, cells in a property space.
Role-playing (also known as, ‘‘forecasting using structured analogies,’’ see
Green & Armstrong, 2007a, 2007b) provide high accuracy in process information but
low generality and low coverage. Role playing includes informants each adopting
specific roles found in real-life cases and interacting with each other with their
individual views of how each would contribute in the (inter)organizational process.
Role-playing studies are useful in accurately predicting process outcomes.
Advances are occurring in the 21st century in moving role-playing toward modest
contributions in generality and coverage (e.g., Green & Tashman, 2008). Forced
metaphor elicitation techniques (FMET) include the attempt to capture both tacit
and deliberate knowledge from informants about colleagues and themselves in
enacting (inter)organizational processes (Woodside, 2008a, 2008b, 2008c). To gain
tacit knowledge, von Wallpach and Woodside (2009) report asking informants at
three different organizational levels in a manufacturing firm to identify themselves
and others as particular animals using photographs and to describe behaviors each
person enacts as representative of their specific animals. Their study indicates that
such tacit zoomorphistic data complements, deepens, and sometimes conflicts with
explicitly thought descriptions of organizational processes.
Fixed-Point Surveys
The survey method with five- or seven-point scale items (e.g., strongly disagree
to strongly agree) likely is the dominant logic in use in Ph.D. dissertations.
Applications of such scaled items via postal and/or Internet data collection usually
have one respondent per firm; these applications typically have less than 25 percent
usable responses of the mailings or Internet contacts; the data analyses that follow in
these studies focus on performing structural equation modeling. Most structural
equation models of (inter)organizational processes use such structured responses to
test the models’ nomological and internal validities. Such SEM modeling reports
offer evidence of being general by large sample sizes (e.g., nW300), with the inherent
inaccuracy problems of self-reports only data, low correspondence of fixed-point
scales with actual processes and events, and shallow coverage inherent with data
collected retrospectively from the stance of one period and usually one informant
responding per organization.
Fuzzy Set Social Science (FSSS), Historical Method, and Simulation

Models of Thinking and Deciding
Corner 4 in the three-dimensional display of the postulate of disproportionate

achievement indicates high generality, high accuracy, and low complexity (or
coverage). Three research methods are representative of the achievements and
limitations of corner 4: FSSS, the historical method, and simulation models of
thinking and deciding. This section briefly describes each of these methods. Each
method includes in-depth case analysis of a few (e.g., 3 to 30) unique cases or a few to
several (e.g., 3 to 100) thinking and deciding processes.
FSSS
FSSS bridges quantitative and qualitative approaches because the methods in

this tool kit are simultaneously qualitative and quantitative (Ragin, 2008, p. 82).
FSSS focuses on analyzing alternative combinations of antecedent conditions that
represent causal complexity rather than the analysis of net effects; FSSS identifies
causal recipes (specific combinations of causally relevant ingredients relating to an
outcome) and thereby unravel causal complexity.
FSSS builds on set theory and fuzzy sets (Zadeh, 1965) analysis using Boolean
algebra rather than linear algebra. Examining all logically possible combinations
of causal conditions makes it possible to construct experiment design-like contrasts
(where only one causal condition at a time is allowed to vary) and thus offers a
thorough analysis of the effects of relevant causal conditions. In effect, the impact of
each cause is examined in all logically possible contexts (the 2k configurations
of conditions, k ¼ number of causal conditions) (Ragin, 2008, p. 125).
The following discussion is a brief introductory example of FSSS. A fuzzy set
scale allows for fine gradations of the degree of membership in a causal conditional
recipe. A fuzzy set is viewable as a purposively calibrated scale transformation of a
continuous variable. Such calibration is possible only through the use of theoretical
and substantive knowledge that is essential in the specification of the three qualitative
breakpoints (full membership ¼ 1.0; full non-membership ¼ 0.0; and maximum
ambiguity — the crossover point ¼ 0.5) (Ragin, 2008, p. 30).
Figure 3 illustrates the creation of three fuzzy set purposively calibrated scales
from case data in a business-to-business process study of marketing and purchasing
industrial chemicals (Woodside & Wilson, 2000). Less than one percent of all
customers for the large manufacturer of industrial chemicals were fully in the
membership of customers with large purchase requirements. Customers with purchase
requirements for the category of chemicals in this study are classifiable more out
than in the large customer requirements membership. The manufacturer classifies
customers willing to single source 90 percent plus of their purchase requirements for
the category as fully in the membership of customer willing to single source.
Customers buying 50 percent of their requirements from this manufacturer are
classifiable as more out than in membership of willing to single source. Customers
aggressively demanding price reductions plus additional benefits (e.g., the
manufacturer building storage facilities for the category on the customers’ site at
no charge) are classifiable as fully in the membership of customer aggressive with
respect to price setting. Customers demanding ‘‘cost avoidance’’ objectives (i.e., price
increases less than published inflation rates for the category) are classifiable at the
crossover point. Customers expressing willingness to pay ‘‘market prices’’ for the
category are classifiable as more out than in membership for customer aggressiveness
with respect to price.
Three common operations on fuzzy sets are set negation, set intersection, and set
union (logical or). Logical and: compound sets are formed by the combination of two
or more sets, an operation commonly known as set intersection. With fuzzy sets,
taking the minimal membership score of each case in the sets that are combined
accomplished a logical and a set intersection. Table 1 includes the fuzzy set scores
and the set intersection of their three-way combination (causal recipe) for eleven
customers in the study of marketing and buying of industrial chemicals. The midlevel
dots are used to indicate set intersection (combination of aspects) for the three causal
conditions (A B C). Note the intersection scores are equal to the lowest score from
the three prior columns in Table 1. The intersection value indicates the degree each
case is more in or out of the intersection membership. Negation: a fuzzy set can be
negated to indicate the degree that the case is not a member of the set. To calculate
the membership of a case in the negation of fuzzy set A, simply subtract its
membership in set A from 1.0 as follows:
(Membership in set BA) ¼ 1.0 (Membership in set A) or BA ¼ 1.0 A, where
‘‘B’’ indicates negation. Thus, for customer case number 1, its membership in (Not a
member of the large customer group) has a negative score of 0.1. Note that negation
membership, ‘‘Not a Large Customer,’’ is asymmetric to membership in the target
concept of small customer membership, that is, a customer can be more in the out
than in the large customer membership (Not a member of large customers) and still
not be full member of the small customer membership. This point holds for the other
two causal conditions (B and C) in Table 1. Dual coding of key causal conditions has
important theoretical benefits.
(A) (B) (C)
Full membership + 3.0 1.0 90%+ 1.0 “Cost reduction +” 1.0
+ 2.0 0.7 80% 0.7 “Cost reduction” 0.7
Cross -over point + 1.0 0.5 70% 0.5 “Cost avoidance” 0.5
0.0 0.2 50% 0.2 “Market price” 0.2
Full nonmembership < 0.0 0.0 < 50% 0.0 “List price” 0.0
Z-value Fuzzy Set Value SOBX Fuzzy Set Value Gloss Fuzzy Set Value
Figure 3: Fuzzy set scaling examples. (A) Customer has large annual purchase requirements. (B) Customer willingness to
single source requirements. (C) Customer objective (aggressiveness) with respect to price. Key: SOBX, share of business
awarded to firm X (our firm); Code, buyer statement indicating aggressive stance in price negotiation with firm X (our firm);
‘‘Cost reduction + ’’ indicates buyer wants lower price for next year in real terms and extras (e.g., free construction); ‘‘Cost
avoidance’’ indicates buyer wants price increase to be less than rate of inflation.
Bridging the Chasm between Survey and Case Study Research
27
Table 1: Fuzzy set scores for customer SOB awarded to X.
Customer A. Large B. Willingness C. Price A B C Y. Customer SOBx

case number customer single source objective annual agreement
1 .9 .7 .9 .7 1.0
2 .6 .7 .8 .6 .9
3 .9 .2 1.0 .2 .8
4 .5 .9 .3 .3 .5
5 .2 .9 .6 .2 .9
6 .2 .2 .2 .2 .5
7 .9 .2 .3 .2 .3
8 .7 .9 .1 .1 .6
9 .1 .3 .9 .1 .4
10 .1 .4 .1 .1 .3
11 .6 .9 1.0 .6 .2
Logical or: two or more sets also can be joined through the logical or: the union of
sets. The logical or directs the researcher’s attention to the maximum of each case’s
memberships in the component sets. A case’s membership in the set formed from the
union of two or more fuzzy sets is the maximum value of its memberships in
the component sets. The addition sign is used to indicate logical or, for example the
logical or membership for case number 1 for the combination of the three causal
conditions in Table 1 equals A + B + C ¼ 0.9.
With fuzzy sets, membership scores in one set (a causal condition or a
combination of causal conditions) that are less than or equal to their corresponding
membership scores in another set (e.g., the outcome) indicates a subset relationship.
Observe in Table 1 that the causal recipe membership score for A B C are
consistently less than or equal to their corresponding membership scores in customer
share of business awarded to firm X (the chemical manufacturer marketing the
category) in the study — with the exception of customer case number 11.
Figure 4 shows the plot of the causal recipe of the intersection representing
the conjunction of the causal conditions (A B C) and the outcome membership of
customer share of business awarded to firm X. The pattern of results is consistent
with an argument of sufficient causation — an upper-left triangular plot, with the
degree of membership in the causal combination of the horizontal axis and the
degree of membership in the outcome on the vertical axis, signals the fuzzy set
relation. The plot in Figure 4 shows sufficiency but not necessity for the conjunction
of A B C on the outcome membership. Other paths to high membership scores on
the outcome condition exist but this observation does not take away from the finding
of sufficiency in high membership scores in the causal recipe resulting in high
membership scores in the outcome condition — the argument of sufficiency but not
necessity permits multiple paths to high scores for the outcome condition.
1.0 1
Y= 5 2
New
SOBX 0.8 3
Award
0.6 8
4
0.4 9
7
10
0.2 6 11
0.0
0.0 0.2 0.4 0.6 0.8 1.0
X = Causal Condition A·B·C
Figure 4: Plot of Y (new SOBX award) by causal condition A B C (n ¼ customer

case number).
Note:
X X
ConsistencyðY i X i Þ ¼ ½minðX i ; Y i Þ= ðX i Þ
¼ 27=27 ¼ 1:00 without customer case number 11
¼ 29=33 ¼ :88 with customer case number 11
X X
CoverageðXi YÞ ¼ ½minðXi ; Y; Þ= ðYi Þ
¼ 27=62 ¼ :44 without customer case number 11
¼ 29=64 ¼ :45 with customer case number 11
Measures of Associations
Consistency, like significance, signals whether or not an empirical connection merits
the close attention of the investigator. If findings from the membership analysis
are inconsistent with the hypothesized relation, then the hypothesis or conjecture is
not supported (Ragin, 2008, p. 45). Coverage, like strength, indicates the empirical
relevance or importance of a set-theoretic connection. Coverage estimates assess
the proportion of cases following a path to high outcome scores; coverage is a
straightforward indicator of the empirical importance of a causal combination
(Ragin, 2008, p. 55).
The formulas and specific estimates for consistency and coverage for the causal
combination (conjunction A B C) appear at the bottom of Figure 4. The evaluation
of the set relationships between the causal recipe and the outcome condition indicates
high consistency and moderate amount of coverage.
The uniqueness of case number 11 in the findings in Table 1 and Figure 4 merits
further attention and its discussion permits extending Gibbert’s (2006) observations
about ‘‘generalizing about uniqueness.’’ Further discussion about customer 11 with
the manufacturer marketing the category resulted in confirmation of this customer’s
unique relationship with this marketer. The manufacturer sales manager reported
monthly complaints by this customer and continuing attempts to renegotiate prices
during the annual contract. While this sales manager did not use the expression,
customer 11 reflects the industrial customer equivalent to Van Maanen (1978)
‘‘The Asshole,’’ in his study of a distinct but familiar type of person to the police.
Thus, unique findings indicate a paradox worthy of further investigation. ‘‘General-
izability demands the research findings are not idiosyncratic to the firm or the sample
of firms studied’’ (Gibbert, 2006, p. 124). Thus, the researcher should look for the
presence of further assholes or other seemingly unique cases before concluding that
adding condition D is relevant for model building and testing.
Creating a fourth causal condition, BD ¼ ‘‘Not an Asshole’’ would place all
customers in Table 1 above the crossover point (0.5) except for customer 11.
Customer 11’s low score on this causal condition (BD ¼ 0.00 for customer 11)
would shift his A B C BD conjunction score to the left and result in high
consistency for this more complex causal recipe.
Note that Table 1 and Figure 4 examine only one causal condition’s relation to the
outcome condition. Additional causal conditions warrant examination and these
include A, B, C, A B, A C, B C, BA BB BC, as well as BA B C, and others.
Ragin, Drass, and Davey (2007) provide a software program to ease the calculations
involved in creating complex conditions and estimating their consistency and
coverage (www.fsqca.com).
Historical Method
The historical method fits well in the corner 4 in the box that includes displaying the
possibility of achieving disproportionate achievement; corner 4 indicates high
generality, high accuracy, and low coverage. The historical method includes theory
and analysis relating to archival data and dynamic influences on outcomes. Smith
and Lux (1993, p. 595) provide a useful exposition of historical analysis even though
their view is too restrictive, ‘‘Historical research provides a qualitative interpretive
method uniquely formulated to explain the causes of change through time.’’ Golder
(2000, p. 157) offers a broader definition than Smith and Lux’s view: historical
analysis is ‘‘the process of collecting, verifying, interpreting, and presenting evidence
from the past.’’ Golder (2000) is must reading because the exposition provides a solid
foundation of where the field went and is going — a thorough historical analysis of
historical analysis would indicate that the field offers well-formed models and tests of
these models from the perspective of variable-based research (versus FSSS’s reliance
on case data theory and conjunctive condition tests). Tellis (1988) and colleagues
(e.g., Tellis & Franses, 2006; Tellis & Golder, 1996; Tellis & Wernerfelt, 1987; Tellis,
Niraj, & Yin, 2009) are leading contributors to the literature advancing theory and
research using historical analysis.
A prime strength of historical analysis is its use of independent and dependent
variables very useful and appealing for theorists, executives, and both marketing
and financial analysts. Market share and market share leadership and switches are
sometimes dependent variables in historical analysis. Independent variables often
include transforming qualitative information into ordinal scales useful for model
testing (e.g., see the product quality scale in Appendix A, Tellis et al., 2009).
Historical analysis is often limited in the range of questions it can address; thus,
the placement in corner 4 rather than 8 in the three-dimensional box. However,
historical analysis is useful in addressing profoundly important questions and
providing useful answers to these questions. Clayton Christensen’s dissertation
report (Christensen & Bower, 1996) on the impact of disruptive technologies of firms’
successes and failures and subsequent book, The Innovator’s Dilemma (Christensen,
1997), exemplifies the use of historical analysis and the immense impact on theory
and practice that the method can provide.
Simulations of Thinking and Deciding
Also appropriate for placing in corner 4 of the three-dimensional box are simulations
of thinking and deciding. Such simulations are in the form of alternative paths in
binary flow diagrams (e.g., Morgenroth, 1964; Howard & Morgenroth, 1968) and
ethnographic decision tree models (Gladwin, 1989). These models provide rigorous
development of generalizable theory using accurate details of process data on very
focused issues. Such models are generalizable to multiple cases (individuals) facing
a given dilemma, for example, a given firm processing information deciding on a
response to a major competitor’s price change (Howard & Morgenroth, 1968) for
several periods — with each price change and response serving as a case study.
Reports on Multiple Anecdotes in Different Contexts
These studies are low in generalization and accuracy but provide extensive coverage.
Possibly, placing Studs Terkel’s (1974) Working, and W.T. Tucker’s (1967)
Foundations for a Theory of Consumer Behavior in corner 5 of the three-dimensional
box is a bit unfair because both reports include thick descriptions of how individuals
interact with others and go about completing activities. These reports build usually
from one interview in one setting per informant. Both reports do include general
statements (propositions) for explaining individual behavior. However, such reports
do not provide rigorous analysis of variable or case-based data or well-formed
falsifiable propositions. Consequently, formal analysis of the descriptions and
explanations of the case data in these reports is absent. Such reports are useful in
nurturing insight and nuanced thinking about the complexity of behavior and
implicit and explicit motives antecedent to action.
Multiple Case Studies in Similar Contexts
These studies offer highly accurate reports with substantial complexity/

coverage of organizational and individual processes. Nutt’s (1984) report on 78 case
studies on types of decision-making exemplifies corner 6 in the three-dimensional
box. Mintzberg, Raisinghani, and Theoret (1976) report on the ‘‘structure of
unstructured decisions’’ fits into corner 6 as well. Both studies provide useful
though modest amounts of description of processes and indicate that search and
evaluation processes occur in several different periods in decision processes and
even after decisions are made. The propositions these studies develop are insightful
but stop short in proposing testable hypotheses either using variable or FSSS
methods.
Naı¨ve Observation
Naı̈ve observation may offer a great amount of coverage as to what steps occur in a
process as well as sophisticated generalizations that are sometimes highly inaccurate.
An untrained and inexperienced American attempting to explain a Japanese tea
ceremony is illustrative of corner 7 in Woodside’s box. Famously, Freeman (1983)
attacked Margaret Mead’s (1928) work for presenting a naı̈ve observational
report on sexual adolescent behavior in Samoa. Unfortunately, Mead was unable
to respond as she had died before Freeman’s attack. Characteristics of naı̈ve
observation include ‘‘participant observation’’ with the researcher living and acting
in the social and work lives of informants for several weeks, months, or a year or
more. Street Corner Society (Whyte, 1943/1993) exemplifies such research in the
environment of an Italian–American ‘‘street gang’’ (more like an informal club of
single men than a gang). Whyte was a participant observer for 3.5 years in this study.
In an industrial marketing setting, Woodside and Samuel’s (1981) study of corporate
buying agreements is the result of Woodside living and working inside a corporate
purchasing office for six weeks with Samuel providing an informant’s interpretation–
inside ‘‘member check’’–of the researcher’s (Woodside’s) interpretations of what
was happening and the meaning of actions of individuals and groups. The presence
of, and reliance on, informant member checking is another characteristic of naı̈ve
observation (e.g., Whyte relied on ‘‘Doc’’ (the de facto leader of the group) to help
him interpret events and the meanings of interactions among street gang members
with each other and other persons in the local society).
Multiple-Case System-Dynamics Modeling
Research methods representative of corner 8 in the three-dimensional box are

very ambitious. Multiple-case system-dynamics models are representative of such
ambitious methods. Such models include tacit and deliberate knowledge of
individuals and subgroups in organizations and simulations of prior actions to
predict future actions. Such models include attempts to include feedback loops
(restarts of problem definitions and search for solutions, ‘‘hidden demons’’ of
seemingly minor conditions that have substantial influence on outcomes in causal
recipes with other conditions, and what-if analyses).
Figure 5 is a template showing many of the participants in a firm and feedback
loops in the decisions and actions among participants at different stages in
attempting to adopt superior new technologies with some firm members attempting
to thwart the adoption. While this template represents a summary of propositions
by Woodside and Biemans (2005), many of the participants, decisions, outcomes,
and feedback loops that Figure 5 shows are apparent in system-dynamic models of
firms attempting to change direction (e.g., Huff, Huff, & Barr, 2001).
Huff et al. (2001), Maani and Majaraj (2004), and Repenning (2001) exemplify
formal model building and testing of firms attempting to change direction (i.e.,
adopting superior new technologies). Their treatise includes the same propositions as
(and more complexity/coverage than) Christensen (1997) and Christensen and Bower
(1996) but formally models the impact of forces of inertia versus the forces of change in
the firm. Such modeling successfully maps thinking, decisions, actions, and outcomes
through several months and years to provide compelling insights into why some firms
are successful in adopting superior new technologies while other firms fail to do so.
Huff et al. (2001) is a remarkable treatise that illustrates how path dependencies lead to
specific outcomes rather than independent assessments via checklists of ‘‘key success
factors’’ or ‘‘key failure factors.’’ No one factor (i.e., antecedent condition) is necessary
or sufficient to result in adaptation or continuing stability (and rejection of superior
new technologies and strategic actions) by a firm, rather, different streams of thinking–
decisions–actions (i.e., complex causal recipes) are sufficient for one outcome or the
other (change versus stability). The empirical demonstration of this key insight using
system-dynamics modeling for different firms in one industry (the pharmaceutical
industry) is Huff et al.’s remarkable achievement.
Triangulation: Mixed-Methods and Decision Systems Analysis
The use of mixed or multiple methods in case study research usually contributes to
increasing accuracy and complexity/coverage in a study more so than generality.
A mixed-method approach is likely to provide confirmation and disconfirmation
of some beliefs and feelings of participants collected during interviews by examining
data collected using alternative methods within the same context (e.g., a given firm
and among stakeholders of the firm) during the same days, weeks, and/or months of
34
Key:
d 1. /d = delay
Continuing use of the installed-base technology C = Champion
or partial to complete transfer to new technology U = Lead user
15. Death M = Maven
of SBU / O = Opposer
Firm L = opinion leader
d
d d d d
2. 3. 4. 5. 6.
Innovation Manufacturing Diffusion Opposer Adoption
5a.
2b.
2a. 3b. 4a. 4b. Gatekeeping, 5b. 6a. 6b.
Creating
Idea 3a. Batch runs, Marketing, User delaying Advocating Learning: Local
working
generation, Engineering, publicity, publicity, learning: strategy, the dominant information, testing;
prototypes,
testing, designing social social skills denying logic; skills, paradigm
structure,
revising networking networking and insights counter- politicking insights shifting
procedures
attacking
U, M
C2 C3 O L
2c. 2d. 3c. 3d. 4c. 4d. 5c. 5d. 6c. 6d.
Identifying Overcoming Identifying Overcoming Identifying Overcoming Identifying Overcoming Identifying Overcoming
obstacles, obstacles, obstacles, obstacles, obstacles obstacles, obstacles, obstacles, obstacles, obstacles,
Case Study Research: Theory Methods Practice
failing succeeding failing failing and failing succeeding failing succeeding failing succeeding
7. 8. 9. 10. 11.
3rd 3rd 3rd 3rd 3rd
12. 13.
party A party A party A party A party A
Network Network
Cn1 Cn2
Figure 5: A dynamic theory of innovation, manufacturing, diffusion, adoption/rejection (IMDAR) of superior products/
services built on using new technologies. Adapted from Woodside and Biemans (2005, Figure 1, p. 382).
4 2 Interviewing
Participant 1
observation participants
7 Researcher creates a 14-page

Researcher (Woodside) attends open-ended interview form
corporate purchasing planning 5 6 and completes personal one-
meetings and informal daily on-one interviews with
meetings among participants manufacturing plant buyers
at GTE corporate headquarters at 9 GTE plant locations during
in Stamford, Connecticut; the six weeks of data collection.
GTE provides Woodside with an 3 Samuels (mid-level corporate
office at GTE corporate purchasing buyer) helps Woodside
office during the six weeks of interpret the open-ended data
observation and to revise early versions
of decision-system maps.
Document
Woodside and Samuel review
analysis
written CPAs.
Corporate purchasing agreements (CPAs) with major

suppliers provided to Woodside; these CPAs are not
contracts by serve as letters of understanding of material
requirement purchase amounts and prices that both supplier
and customer plants will adhere to within the next 12 months
Figure 6: Triangulation Methods Application in Case Study Research. Source:

Compiled from Woodside and Samuel (1981).
the study. Rich, deep, insights into what is happening and why it is happening
follow from such mixed-methods research studies. Woodside and Samuel’s (1981)
decision systems mapping (Capon & Hulbert, 1975; Howard, Hulbert, & Farley,
1975; Hulbert, 2003) of corporate purchasing processes is an application of a mixed-
methods strategy.
Figure 6 illustrates the main components of this mixed-methods application.
Note that Figure 6 includes the collection of seven categories of data. Categories 1–3
represent explicit and implicit information from each of three separate sources.
Categories 4–6 represent explicit and implicit information from each two-way
combination of sources.
Category 7 represents information from all three sources within the same context
(i.e., a location in time with the same people present). The following vignette includes
examples of case data collection using direct observation, interviews, and product
demonstrations individually and in combination. During a marketing performance
audit (Woodside, 1977), the researcher (R) traveled by car for seven hours with an
informant (I1) to several customer sites.
During the sixth hour of the trip I1 informed R, ‘‘Bobby [sales manager] never
rides with me. I never get any help from Bobby.’’ Three additional road trips
separately with three additional salespersons provided similar stories. One of the
four salespersons offered the following conclusion about his relationship with Bobby,
‘‘I hate Bobby — I really do.’’
The department’s administrative assistant (I5) reported the following story during
a second, two-hour, individual interview with R.
(I5) At lunch time if the secretary goes to lunch and I’m out to lunch, Bobby will
immediately leave. Do you know why?
(R) No, I don’t. Please tell me.

(I5) Because Bobby does not know where anything is. If a customer calls in
[to ask about the status of an order or to buy something], Bobby does not know
what to do.
One day during the study, R asked Bobby to demonstrate how to use the
equipment that the salespersons were selling. Bobby responded, ‘‘OK. Jim, show
Arch how to use this equipment.’’ Jim demonstrated operating the equipment as
Bobby and R watched. Bobby was observed staying in his office each day over a
three-week period. At the end of three weeks R asked Bobby about getting out and
riding with each of the salespersons to go to customer sites and meet with customers.
This question started the following exchange:
Bobby: ‘‘I’m not a sales manager, I’m a product manager.’’
R: Don’t the four salespersons report to you? Bobby: Show me in a

textbook where I’m supposed to ride with sales people. During this
discussion, Bobby offered the following summary statement. ‘‘I don’t
like the salesmen. Do you? I really don’t like marketing either. And I’m
not going to ride with the salesmen.’’
At the time of study, none of the four salespersons were meeting their monthly
sales quotas. Two months following the audit, the department was closed and Bobby
was placed in another department in the firm. (Bobby’s father was a member of
the firm’s board of directors and a former governor of the state.) Field case studies
employing a triangulation of methods is time consuming because such an approach
usually includes re-interviewing the same and new informants during the course
of data collection and attending key meetings and events that are scheduled for
specific days well after the start of the study. Re-interviewing the same person on
different days and often in different contexts/locations is valuable because of the
opportunity to clarify information collected from the first interview and, equally
important, because a second interview creates a sense of ‘‘us’’ between the informant
and the interview. During one of his stand-up presentations, Chris Rock (American
comedian and social commentator) expresses this proposition succinctly, ‘‘When you
meet someone for the first time, you are not meeting that person, you are meeting his
representative.’’
Mixed-methods designs usually result in compelling evidence that (1) deliberate
thinking strategies that informants describe in interviews differ substantially from
tacit (implicit) thinking strategies and from actual (implemented) strategies observed
and (2) emotions among participants usually run high for specific process steps that
are occurring in the firm. For example, reports of disagreement and anger relating to
specific personal relationships and events are relevant to the success and failure
of adopting superior new technologies and other processes in organizational
behavior; data collection via postal surveys rarely capture the nitty–gritty dramatic
dynamics of the emotions of process participants. Being there — observing in field
contexts — is necessary to achieve high coverage/complexity among emotional and

cognitive communications among participants.
Recommendations for Bridging the Chasm
Achieving accuracy is paramount. Generality and complexity are add-on objectives.

To some meaningful extent, process researchers can plan their research designs
to accomplish all three objectives to a meaningful degree. However, attempting to
insure generality at the cost of accuracy and complexity will only insure the
continuation of research reports of little importance and even less impact (see
Armstrong, 2003 for a review of the lack of impact of articles and studies in general
by marketing scholars). Achieving accuracy in describing, understanding, and
predicting organizational and individual processes includes applying research designs
located in the middle and top right of the three-dimensional box: locations 4, 8, and 9
and placing less reliance on research designs illustrated by locations 1, 2, 3, 6, and 7.
Research designs focusing nearly exclusively on applying corner 3 methods (e.g.,
fixed-point surveys) can be useful for studying such issues as customer’s assessment
of services delivered, satisfaction with the overall service–product experience, and
intention to buy–use the service in the future. But such post hoc focused studies are
not descriptions and explanations of on-going thinking, deciding, actions, and
outcomes processes. For studies focusing on describing, explaining, and/or predicting
management processes and process dynamics, fixed-point surveys coupled or
uncoupled with structural equation modeling (SEM) or other regression-based
statistical analysis, are insufficient and represent achieving generality without
accuracy or complexity — even though box-and-arrow depictions of SEM output
appear to be complex. Such survey-based SEM reports have fatal weaknesses —
especially given the widespread use in data collection in SEM applications of one-
shot surveys relying on explicit-only informant reports of 7-point scales that are
several steps removed from real-life thinking and actions, and result in useable
responses of less than 25% of the total surveys mailed.
This chapter closes with a brief discussion of Table 2. This figure summarizes
some specific steps to take that are helpful to achieving each of three research
objectives. Note that each box in the figure includes an exemplar source that further
illustrates the step. Box 1 in Table 2 urges a researcher to plan to examine members
in specific-to-all reference classes in a property space analysis (Lazarsfeld, 1937;
Woodside, 2008a, 2008b, 2008c) rather than planning to conduct a general sampling
of possible informants. For example, a few customers may be extremely important to
the current profitability of a firm and special attention is necessary in planning the
study to find and collect data on these customers (e.g., Woodside & Wilson, 2000).
Box 2 in Table 2 suggests achieving generality by collecting multiple applications of
thinking and deciding processes for a reference class of decisions (one property space
location). Morgenroth (1964) and Howard and Morgenroth (1968) are the useful
sources illustrating the achievement of generality in model building for information
Table 2: Recommendations for achieving generality, accuracy, and complexity.
Objective/ Step 1 Step 2 Step 3

steps
Generality 1. Apply property 2. Collect multiple 3. Create degrees of

space analysis to instances of process freedom (DOF)
uncover all process data within each tests to generalize
categories and use reference class (i.e., single case data to
theoretical for each property theory; exemplar:
sampling of spaces; space); exemplar: Wilson and
exemplar: Howard and Woodside (1999);
Woodside (2008a, Morgenroth (1968) conduct FSSS and
2008b, 2008c) and Montgomery test conditional
(1975) recipes; exemplar:
Ragin (2008)
Accuracy 4. Re-interviewing 5. Do multiple 6. Apply triangulation
the same informant individual using multiple
during and after interviews of methods that go
observing the different informants beyond interviewing
process; exemplar: during and after and observing;
Pattinson and observing the exemplar: Pettigrew
Woodside (2007) process; exemplars: (1975) and
Woodside, Woodside and
Pattinson, and Samuel (1981)
Miller (2005);
Pattinson and
Woodside (2007)
Complexity 7. Adopt systems 8. Build and test 9. Build and test
thinking, mapping simulation models system dynamics
strategic thought, of thinking and models; exemplars:
and decision deciding; exemplars: Hall (1976) and
systems analysis; Howard and Huff et al. (2001)
exemplars: Senge Morgenroth (1968)
(1990), Huff (1990) and Gladwin (1989)
and Howard et al.
(1975)
processing, thinking, and deciding price increases and decreases by one group of
decision-makers across multiple applications. These two articles include testing the
resulting thinking–deciding model created from one set of decision data for a hold-
out sample of decisions; a seminal contribution to process modeling that achieves
generality in one firm across many decisions in one reference class of actions.
Box 3 includes generalizing from data for a single case to theory using Campbell’s
(1975) recommendations for conducting degrees-of-freedom (DOF) analyses
(see Wilson and Woodside (1999) for such an (DOF) analysis in case study research).
Similar to a medical doctor looking to match a patient’s symptoms of an illness to
one more possible diseases, DOF analysis is a pattern matching process whereby
the examination attempts to match several (e.g., n ¼ 10) characteristics occurring in a
process in a case with the predictions regarding these same characteristics made by
competing theories (bounded rational model versus political model versus garbage can
model of thinking and deciding, see Wilson and Woodside for such a DOF analysis).
Box 4 recommends increasing accuracy by re-interviewing the same individual in
multiple settings. Such multiple rounds of emic (informant-view) data help to clarify
and deepen etic (researcher view) interpretations of what happened, why it happened,
and learning the consequences of what happened. Woodside, Pattinson, and Miller
(2005) provide thick descriptions of process data using this approach. Box 5 extends
the method to include conducting multiple rounds of interviews with multiple
participants in the process under study.
Whyte (1943) applies this procedure in his Street Corner Society as do Pattinson
and Woodside (2007) in their ‘‘Innovation and Diffusion of Software Technology.’’
Box 6 recommends achieving accuracy by including a triangulation of methods in
planning the design of process studies. Woodside and Samuel (1981) illustrate this
approach in business-to-business case study research. Box 7 focuses on enriching
studies by acquiring data showing the details and complexity in process data using
such tools as mapping strategic thought (Huff, 1990) and decision systems analysis
(Howard et al., 1975). Box 8 recommends achieving complexity by building and
formally testing simulation models of thinking and deciding; Howard and Morgenroth
(1968) illustrate the accomplishment of such modeling in case study research. Box 9
includes the most advanced steps involving description, understanding, and prediction
of complex processes for a reference class of decisions (e.g., innovate or maintain use of
current technology); Huff et al. (2001) provides details for taking this step successfully.
Limitations: the intention here is that the prior discussion and these recommenda-
tions are guideposts for (1) adopting the postulate for disproportionate achievement
regarding the principal objectives in planning research designs (i.e., planning research
designs to achieve all three principal objectives: generality, accuracy, and complexity/
coverage) and (2) an eclectic stance toward learning and applying new research
methods. The intention is not to present a full exposition of methodological theory or
applications of specific data analysis methods. The discussion is limited in scope and
by depth of coverage; hopefully, this brief discussion of the available references to
alternative research methods serves empirical positivists and case study researchers to
apply Weick’s (1979) wisdom to ‘‘complicate yourself ’’ by learning additional methods
and taking steps necessary for achieving all three principal research objectives.
Conclusions
This chapter advocates dropping belief in Thorngate’s (1976) postulate of ‘‘postulate

of commensurate complexity’’ — it is impossible for a theory of social behavior to be
simultaneously general, accurate, and simple and as a result organizational theorists

inevitably have to make tradeoffs in their theory development; and, picking a new
research paradigm: the postulate of disproportionate achievement: the building
and testing of useful process models that achieve all three principal research
objectives is possible now because of breakthroughs in software research tools
(e.g., system-dynamics software, see Sterman, 2000) and methods (e.g., FSSS, see
Ragin, 2008). Applying the postulate of disproportionate achievement requires
adding methods to the different tool boxes-in-use by both empirical positivists and
interpretive researchers.
Both empirical positivists and interpretive researchers need to recognize
the widespread tendency to fall into using Kaplan’s (1964, p. 28) ‘‘law of the
instrument’’ — one uses the tool one has even if in context the tool’s use is highly
inappropriate. Planning on using research designs that include taking modest-to-
substantial steps toward achieving accuracy, generality, and complexity is now possible
and appropriate.
Chapter 3
Storytelling Theory and Research
Synopsis
Case study research frequently includes collecting and interpreting stories

individuals tell about their lives and event that they believe that they know
about. Chapter 3 discusses storytelling theory and describes case study research
in consumer behavior of stories that consumers tell about buying and using
products and services. Storytelling is pervasive through life. Much information
is stored, indexed, and retrieved in the form of stories. Although lectures
tend to put people to sleep, stories move them to action. People relate to each
other in terms of stories — and products and brands often play both central
and peripheral roles in their stories. To aid storytelling research in consumer
psychology, this chapter develops a narrative theory that describes how
consumers use brands as props or anthropomorphic actors in stories they
report about themselves and others. Such drama enactments enable these
storytellers to experience powerful myths that reflect psychological archetypes.
The chapter includes findings from case study research that probes propositions
of the theory. Implications for consumer psychology and marketing practice
follow the discussion of the findings.
More often than not, in America at least, those who win in myth
markets are performing a myth of rebelling; the most successful
icons rely on an intimate and credible relationship with a rebel world:
Nike with the African-American ghetto, Harley with outlaw bikers,
Volkswagen with bohemian artists, Apple with cyberpunks, Mountain
Dew with slackers — protagonists who would rather pursue quixotic
activities than ‘‘grow up’’ and get serious about careers. (Holt, 2003)
Consumer Storytelling Theory
A myth is a traditional story about heroes or supernatural beings, often explaining

the origins of natural phenomena or aspects of human behavior. The myth of
rebellion builds from the outlaw archetype reflecting the motto, ‘‘[parental, follow-
the-leader, and societal] rules are meant to be broken’’ (cf., Mark & Pearson, 2001).
The central proposition in this chapter is that consumers often use products
and services as props or anthropomorphic identities to enact story productions that
reflect archetypal myths (cf., Holt & Thompson, 2004; Wertime, 2002). Storytelling
of such enactments includes conversations between consumers and brands on
both unconscious and conscious levels of thinking (cf., Wang, Baker, Wagner, &
Wakefield, 2007; Zaltman, 2003). The work of several other scholars in consumer
behavior (e.g., Adaval & Wyer, 1998; Arnould & Wallendorf, 1994; Hirschman,
1986; Holt & Thompson, 2004; Padgett & Allen, 1997) and in related fields of human
inquiry (Bruner, 1990a; Mitroff & Kilmann, 1976; Orr, 1990; Schank & Abelson,
1977; Zuikier, 1986) support the view that ‘‘y people think narratively rather
than argumentatively or paradigmatically’’ (cf., Weick, 1995, p. 127; Wells, 1989).
Stories and storytelling are central to achieving a deep understanding of
consumer psychology (cf., Escalas & Stern, 2003; Holt, 2004). The following
five propositions inform this conclusion. First, people naturally think narratively
rather than argumentatively or paradigmatically (Hiltunen, 2002; McKee, 2003;
Weick, 1995).
The movie director Steven Spielberg dramatically illustrates the point:

Once upon a time it was a small gathering of people around a fire
listening to the storyteller with his tales of magic and fantasy. And
now it’s the whole world. In Japan and in Finland, in the heartland
of America, in Italy and Spain, in Singapore and France y still
they gather to hear the stories. But now they gather in multiplexes
in Britain, Germany, Spain, Australia y or giant movie places in
Mexico. That’s what has thrilled me most about Jurassic Park
phenomenon. It’s not ‘‘domination’’ by American cinema. It’s just
the magic of storytelling, and it unites the world. And that is truly
gratifying. (1994, quoted in Hiltunen, 2002, p. xii)
A casual search of blogging (i.e., entries to personal online journals) expands on

Spielberg’s view. Weblogs are individuals’ own stories of their lived experiences,
beliefs, and attitudes that often include pictures (photoblogs) and video (vlogs).
Weblogs are expanding exponentially in the first decade of the 21st century — from
less than 1,000 in 2000 to 30 million plus in 2007 (Kluth, 2006). Blogging appears to
be a revolutionary away of communicating whereby individuals provide commentary
on their lives and the lives of others via such services as LiveJournal.com
and MySpace.com — to name only two of many such services. As a counterpart
to individual blogging Web sites, Wang et al. (2007) show that marketers attempt to
create social roles for brands in commercial Web sites increase consumers’ interest,
attention, and feeling of control (and indirectly their Web site purchase intentions
compared to informative [lecture format] Web sites).
Second, a substantial amount of information stored in and retrieved from memory
is episodic — stories that include inciting incidents, experiences, outcomes/evaluations,
Storytelling Theory and Research 43
and summaries/nuances of person-to-person and person-and-brand relationships

within specific contexts (see Fournier, 1998; Schank, 1990).
Third, retrieving, reliving, or repeat watching stories results in what Aristotle
(see Hiltunen, 2002) refers to as ‘‘proper pleasure’’ — a catharsis — that relates
usefully to the work of Holt (2003) and Jung (1916/1959): Watching, retrieving, and
telling stories enables the individual to experience one or more archetypal myths.
An archetype is an unconscious primary form, an original pattern or prototype in
the human mind; archetypes are not learned or acquired — they are with us from
birth and are as natural and embedded in us as our own DNA (Jung, 1916/1959;
Wertime, 2002).
Fourth, specific brands and products often play pivotal roles enabling consumers
to achieve the proper pleasure that results in a consumer mentally and/or physically
enacting a specific archetype — and reliving the experience by periodically retelling
a given story. The brand-consumer storytelling and pleasure outcome builds on
Bagozzi and Nataraajan’s idea (2000, p. 10) ‘‘that people need help in finding what
makes them happy, and this is where marketing comes in.’’ Happiness via brands
enabling consumers to enact archetypal stories is a micro complementary proposal to
Nataraajan and Bagozzi’s (1999) more macro explication of the role of marketing in
aiding consumers’ conscious quest for happiness.
Fifth, individuals seek clarity to make sense of prior conversations, events, and
outcomes from others and themselves by telling stories. ‘‘How do I know what
I think until I hear what I say?’’ (Weick, 1995) partly summarizes this proposition.
Story repetition is often a plea for clarity that may be achievable in part by
recognizing that the drama in the story is one illustration of one or more specific
archetypes (e.g., story of rebellion, mother-of-goodness, little trickster, ultimate
strength, the hero; see Wertime, 2002, for the storylines for these and other
archetypes). The above set of propositions builds from the proposals of Holt (2003),
Hiltunen (2002), Jung (1916/1959), Mark and Pearson (2001), and Wertime (2002).
The propositions help to explicitly describe how brands enable archetype outcomes.
The proposals here go deeper than Holt’s (2003, 2004) proposal that icons are
encapsulated myths; the proposals in the present chapter describe how consumers’
stories involving actions with brands and products provide a proper pleasure
(Aristotle’s Poetics, Butcher, 1961) that relates unconsciously to one or more
archetypes (Jung, 1916/1959) and helps consumers achieve deep satisfying levels
of sense making.
The observation that many consumers are motivated to report, via blogs and
other forms of personal journals, on their lived experiences involving buying and
using brands is one impetus for research on consumer storytelling. Such personal
online journals are unlike traditional diaries in two respects: Blog entries are
displayed in reverse chronological order and though traditional journals were private
or even secret affairs that were never linked to other journals, blogs are social in
nature, whether they are open to the public as a whole or only to a small select group
(Kluth, 2006).
Now, a new blog is created every second of every day, according to Technorati, a
search engine for blogs, and the ‘‘biosphere’’ is doubling in size every five months
(Kluth, 2006). A casual search of blogs via Technorati, Google, or Ask by brand
names and adjectives (e.g., ‘‘Honda’’ or ‘‘Hawaii’’ and ‘‘good’’ or ‘‘bad’’) brings
forth dozens of blog entries of seemingly mundane reports of buying and
experiencing a Honda or a visit to Hawaii.
Why would anyone care to share or read such stories? Three rationales help to
explain this behavior. First, telling stories is inherently pleasurable to the authors;
such storytelling allows authors to be both protagonist and audience and to vent
anger or report bliss about events and outcomes over and over again, and to enjoy
the nostalgia of reliving earlier experiences (see Schindler & Holbrook, 2003).
Second, to some extent consciously, but mostly unconsciously, storytelling permits
the teller to experience an archetype fulfillment; the plot line in the story told
provides evidence that the storyteller-as-protagonist represents a regular guy/gal,
lover, jester, creator, ruler, rebel, sage, hero, outlaw, magician, or some other
archetypal primal form. Third, telling stories revises and deepens sense making of
the meaning of events in the story and what the complete story implies about oneself
and others.
Schank (1990, p. 219) proposes that people think mostly in terms of stories.
‘‘They understand the world in terms of stories that they have already understood.
New events or problems are understood by reference to old previously understood
stores and explained to others by the use of stories.’’ Woodside and Uncles (2005)
empirically confirm the importance of behavioral primacy affecting not only what
consumers are able to understand but also what they prefer and do in subsequent
time periods.
The following stream of research provides a strong justification for storytelling
theory building in consumer psychology as well as for creating and testing the impact
of storytelling in marketing contexts (e.g., Adaval & Wyer, 1998). Consumers
often include products and brands in reporting their own (emic) lived experiences
(e.g., see Arnould & Wallendorf, 1994; Hirschman, 1986; Kozinets, 2002; Moore,
1985; Woodside & Chebat, 2001); consumers assign roles, actions, and relationships
to brands (Fournier, 1998) in the stories they tell to themselves and to others; brands
enable consumers to enact archetypal myths (e.g., drinking Mountain Dew enables
the slacker myth; see Holt, 2003).
Classical drama provides a definition of a ‘‘good’’ (McKee, 2003), or memorable,
story: A good story displays tension that includes one or more inciting incidents
preceded by conditions or settings that initiate the unconscious/conscious identifica-
tion of one or more goals, with actions by a protagonist and possibly additional
actors resulting in an outcome; the temporary occurrences of world blocks (e.g., an
antagonist temporarily preventing the protagonist from achieving the main goal)
and/or personal blocks (e.g., the protagonist lacks the skill to perform an act
necessary to reach the goal) serve to increase viewer and/or protagonist emotion and
involvement in a story. Taking steps to overcome blocks such as seeking and gaining
help from others occur frequently in stories.
This chapter expands on brands-as-icons storytelling research by developing a
narrative theory of consumer storytelling involving iconic brands. The theory’s core
proposition reflects the views of a leading screenwriting coach (McKee, 2003):
Consumers seek out or find themselves in unfamiliar and/or anxiety inducing

contexts; consumers consciously and/or unconsciously act in scenes that take them
through and out of these contexts; stories lead consumers to consciously and
unconsciously reflect and make sense of the story, their actions in the story, and the
outcomes of the story (cf., Weick, 1995).
Following this introduction, the next section of this chapter advances a theory
of consumer storytelling. The third section develops ten propositions that follow
from consumer storytelling theory. The fourth section presents a case study research
method that probes these propositions. The fifth section presents the findings of the
study. The final section includes a discussion relating the findings to the theory,
limitations, strategy implications, and future research suggestions.
Consumer Storytelling Research
Consumer storytelling theory builds on several related streams of theory and

research, including Holt and Thompson’s (2004) view that dramatic consumption
experiences must be scripted, either by experiential service providers or within the
institutional structure of a consumer subculture. Thus, the structure of a word-of-
mouth (WOM) communication is an important indicator of whether the message is a
story. A story’s structure includes two important elements: chronology and causality
(Delgadillo & Escalas, 2004). Regarding chronology, narrative thinking organizes
events in terms of a temporal dimension: Action occurs over time. This discussion of
story matches with classical drama versus vignettes or lectures; see Stern (1994) and
Escalas and Stern (2003) for details comparing these three forms of communications.
Wells (1989) provides thorough descriptions of drama and lecture communications
relevant for crafting advertising strategies.
Time figures in narrations as episodes (e.g., situations via scenes within acts in a
drama); each episode has a beginning, middle, and end, whereas time in reality is
an undifferentiated continuous flow (Bruner, 1990a, 1990b; Escalas, 1998). ‘‘Second,
narrative thought structures elements [scenes, action, talk, and acts] into an
organized framework that establishes relationships between the story’s elements
[e.g., actors including persons, products, and brands; see Fournier, 1998] and allows
for causal inferencing’’ (Delgadillo & Escalas, 2004, p. 187).
Escalas (1998) provides a narrative structure coding scale that reflects her
assessment of the literature (also see Bruner, 1990a, 1990b; Gergen & Gergen, 1988)
on what makes for a good, or well-crafted, story. Bruner (1990a, 1990b) proposes
two dimensions that relate to crafting a good story: the landscape of action and the
landscape of consciousness. The landscape of action consists of events that are visible
(by sight or imagination) to the casual observer: initiating event, resulting actions,
and outcomes. The landscape of consciousness allows the reader/viewer to get inside
the head of the story’s characters (e.g., protagonist and antagonist). As Delgadillo
and Escalas (2004, p. 187) emphasize, ‘‘According to Bruner, a story with both
a landscape of action and consciousness is a better story than one that contains only
a landscape of action.’’
Gergen and Gergen (1988) offer the concept of evaluative slope in theorizing
about storytelling: Events in a story are evaluated over time (as it occurs in the
narrative) for the degree these events improve or worsen the state of the protagonist.
Stories that have a steep incline or decline in evaluative slope and those that
alternate in sign (e.g., rising, falling, then rising again) evoke the most emotion
(see Delgadillo & Escalas, 2004).
Six propositions inform Escalas’ (1998) scale items (items appear in quotes) for
measuring the degree of good storytelling. A 5-point scale applies for each item
ranging from 1, not at all, to 5, very much. The story presents a protagonist engaging
in actions to achieve goals; (1) ‘‘To what extent do these thoughts consist of actors
engaged in actions to achieve goals?’’ The story informs about conscious and/or
unconscious thoughts of the protagonist and other actors; (2) ‘‘To what extent do
these thoughts let you know what the actors are thinking and feeling?’’ The story
informs about how personal evolution or change in the life of the protagonist occurs;
(3) ‘‘To what extent do these thoughts provide you with insight about the personal
evolution or change in the life of a character?’’
The story informs how events involving the protagonist, take place; (4) ‘‘To what
extent do these thoughts explain why things happen, that is, what caused things to
happen?’’ The story has an inciting event (a crisis or turning point) involving the
protagonist, along with a beginning and a resolution; (5) ‘‘To what extent do these
thoughts have a well delineated beginning (initial event), middle (crisis or turning
point), and ending (conclusion)?’’ The story presents the protagonist in clear-cut
situations; (6) ‘‘To what extent do these thoughts focus on specific, particular events
rather than generalizations or abstractions?’’ Consumer storytelling theory builds
from memory systems research as well (e.g., Bettman, 1979; Tulving, 1985). From
reviews of the theoretical and empirical literature, Tulving (1985) provides a ternary
classification scheme that consists of procedural, semantic, and episodic memory
mono-hierarchical subsystems.
The system at the lowest level of the hierarchy, procedural memory,

contains semantic memory as its single specialized subsystem, and
semantic memory, in turn, contains episodic memory as its single
specialized subsystem. In this scheme, each higher system depends on,
and is supported by, the lower system or systems, but it possesses
unique capabilities not possessed by the lower systems y Episodic
memory affords the additional capability of acquisition and retention
of knowledge about personally experienced events and their temporal
relations in subjective time and the ability to mentally ‘‘travel back’’ in
time. (Tulving 1985, p. 387)
Tulving (1985, p. 388) emphasizes that episodic memory associates with self-
knowing consciousness, ‘‘Autonoetic (self-knowing) consciousness is a necessary
correlate of episodic memory. It allows an individual to become aware of his or her
own identity and existence in subjective time that extends from the past through the
present to the future.’’ Tulving’s assessment fits well with Weick’s (1995) assessment
that people make sense of events, outcomes, and self by telling themselves and others
stories about what they have experienced. According to Weick (1995), all sense
making is retrospective and based on storytelling to self and others by the storyteller.
Consumer psychology and psychoanalytic research on brands as anthropo-
morphic identities, archetypes, and on brands as icons (see Fournier, 1998;
Hirschman, 2000a; Holt, 2003; Rapaille, 2004) informs consumer storytelling theory.
For example, a consumer and brand may be bound in a kinship relationship by
automatic (unconscious triggering) of inherited brand use from the consumer’s
mother. Fournier (1998) describes 15 consumer–brand relationship forms including
arranged marriages, kinships, flings, secret affairs, enslavements, courtships, and
others).
Jung (1916/1959, p. 101) defines archetypes as ‘‘forms or images of a collective
nature which occur practically all over the earth as constituents of myth and at the
same time as autochthonous (biologically based unconscious thinking) individual
products of unconscious origin.’’ Campbell (1968, 1974) argues that most archetypal
forms originated in Sumer and Akkad around 2500 BC. From this beginning, he
proposes, all other textual narratives represent a ‘‘provincial extension of the one
historic heritage and universal history of mankind’’ (Campbell, 1974, p. 133).
Archetypal images include such widely recognized symbols as the tree of life, the raven
and jackal as death images, owls as symbols of wisdom, ships as carriers of the dead,
birds as female figures, the earth as a womb, and so forth (Hirschman, 2000a, p. 60).
Hirschman (2000a, p. 60; also see Hirschman, 2000b) reports ample evidence of
archetypal thought among consumers in the stories they tell: ‘‘For example,
Campbell (1973, p. 71) writes of ‘the helpful crone and fairy godmother as a familiar
feature (supporting actors) of European fairy lore.’ This female figure serves as a
helper or guide to the novice in the story, providing special wisdom or knowledge to
help the novice complete his/her task.’’ Lisa, age 20, describes an exemplar of this
archetype in the character of Mrs. Garrett, the headmistress on the television series
The Facts of Life:
They lived in a boarding school. And from what I can remember about
the show, they were always doing something wrong and then being told
a lesson. Something happens, but it always ended by teaching a lesson
somehow. Mrs. Garrett, I remember her. She was like the keeper of the
school, or at least the girl’s dorm. And I guess she was like their mom,
their psychiatrist, because she always had the right answers and she
always had the right ways, and they were always asking her for advice.
And she was this old red-headed lady y So she was always directing
them in the right direction. (Hirschman, 2000a, p. 60)
Without referring to earlier work by Jung, Campbell, Hirschman, and Holt (2003)
interprets storytelling in television commercials as manifestations of primary
forms, that is, archetypes. For example, Holt (2003, p. 48) provides the following
interpretation for the ‘‘Do the Dew’’ ad campaign for Mountain Dew soft
drink brand, ‘‘With the ‘Do the Dew’ campaign, Mountain Dew reinvented the
wild man (prior campaign focus for the brand) as a slacker. In these spoofs of
extreme sports, all presented as do-it-yourself quests, the brand asserted that
the real men of America’s free-agent frontier weren’t the most buff or competitive
athletes, but the creative guys who pursued their stunts as whimsical art. Slackers
didn’t just face down dangerous situations that came their way. They sought
out insane life-threatening risks. The Dew guys upped the ante on masculine
risk taking to absurd levels, which, in the end, made fun of the idea that manhood
has anything to do with such feats. The people with real power, in Mountain
Dew’s worldview, were people with extreme — and very particular — tastes.
Slackers had no power as workers, but they could assert their will in the corporate
world by asserting their opinions. Companies and their managers would have to
take notice.’’
Holt and Thompson (2004, p. 425) advance such archetypal analysis in their
analysis of two consumers’ self stories. Holt and Thompson propose through their
analysis that ‘‘American mass culture idealizes the man-of-action hero — an
idealized model of manhood that resolves the inherent weaknesses in two other
prominent models (the breadwinner and the rebel) — what we call the ideology of
heroic masculinity — to construct themselves in dramatic fashion as man-of-action
heroes.’’ In a view the present chapter adopts, Hunt (1993) criticizes Holt’s (1991)
prior argument that such interpretative research should not be subject to audits,
triangulation, and purposive sampling but should be judged alone on its
insightfulness and ‘‘ability to convince the reader, no more.’’
Rapaille (2004) emphasizes that learning a core cultural archetype occurs early
in life through an imprinting epiphanic experience that establishes an unconscious
behavior pattern — and such imprinting occurs only during a critical period of
time (age 1–6), after which it is very difficult or impossible to imprint. Rapaille’s
interpretive research describes alternative collective unconscious archetypes that
drive different cultures; he describes how the American mind is very different from
the French mind, for example.
The early storytelling research by McClelland (1961, 1988) informs Rapaille’s
proposition that epiphanic story experiences imprint people with collective
unconscious archetypes. McClelland demonstrates this by analyzing unsolicited
stories written in everyday life (e.g., using multi-item closed end scales that
build in degrees-of-freedom for interpreting case data; see Campbell, 1975; Wilson &
Woodside, 1999), that cultures vary in their orientations toward needs for
achievement, power, and affiliation.
Holt (2003) proposes that some brands become icons — brands that permit
consumers to experience powerful myths consciously or unconsciously. Myths are
simple stories with compelling characters and resonant plots; myths help consumers
make sense of the world. ‘‘Myths provide ideals to live by, and they work to resolve
life’s most vexing questions. Icons are encapsulated myths. They [icons] are powerful
because they deliver myths to us in a tangible form, thereby making them more
accessible’’ (Holt 2003, p. 44).
Consumer self-awareness processing in making myths accessible is worthy of

attention by consumer researchers. For example, examining an unsolicited self-
reported (emic) story, Woodside and Chebat (2001) apply Heider’s balance and
change theories in examining how a Jewish couple consciously copes with automatic-
unconscious retrievals of Nazism and the Holocaust when considering the purchase
of a German car (cf., Holt, 2002). Research on storytelling (e.g., see Arnould &
Wallendorf, 1994; Fournier, 1998; McKee, 2003; Schank, 1990; Zaltman, 2003) is
useful because it helps clarify and deepen knowledge of how people resolve
paradoxes triggered in their minds by unbalanced states (ranging from a vague
conscious feeling of unease to awareness of a problem or opportunity arising from an
inciting incident). Learning stories enables the researcher to examine the complexity
often associated with initial balanced states (e.g., the personal prequel history of the
protagonist and supporting actors in a story that affects how consumers interpret
the situation in which an inciting incident occurs) that lead to imbalance and the
steps taken (consciously and unconsciously — see Wegner, 2002) to achieve old or
new balance states.
Essentially, a story expresses how and why life changes. It begins with a situation in
which life is relatively in balance: You come to work day after day, week after week,
and everything’s fine. You expect it will go on that way. But then there’s an event —
in screenwriting, we call it the ‘‘inciting incident’’ — that throws life out of balance.
You get a new job, or the boss dies of a heart attack, or a big customer threatens to
leave. The story goes on to describe how, in an effort to restore balance, the
protagonist’s subjective expectations crash into an uncooperative objective reality.
A good storyteller describes what it’s like to deal with these opposing forces, calling
on the protagonist to dig deeper, work with scarce resources, make difficult decisions,
take action despite risks, and ultimately discover the truth. (McKee, 2003, p. 52).
The drama versus lecture literature in consumer research offers propositions in
addition to McKee’s proposals on the crafting of stories by marketers to persuade
consumers (e.g., Stern, 1994; Wells, 1989). Wells (1989) provides guidance on how
dramas differ from lectures: Lectures are directed at an audience; dramas are
overheard. Television commercial dramas teach lessons about how products add to
the quality of life. Under certain circumstances, they also provide true samples of the
emotional rewards enabled by the advertiser’s brand. Stern (1994) transforms the
commercial drama–lecture dichotomy into three distinct communication forms:
classical drama, vignette drama, and lectures. Vignette dramas are constructed as
a series of stories in which unrelated characters replicate similar actions across a
variety of spatial settings, without necessarily showing change; the chronology has
neither beginning, middle, or end, nor any fixed order, with no loss in meaning
because each vignette stands alone (Stern, 1994, p. 609).
This chapter does not include examining the occurrence of vignette dramas in
consumer storytelling. The presence of all three forms of communications may occur
in varying degrees in the communications consumers provide to themselves and
others; Stern’s proposals are worth examining empirically in future research.
To probe the core and supporting propositions of the theory, the present chapter
includes an examination of case study data. To control against social desirability bias
(Fisher, 1993) and self-generated validity and other effects of measurement on belief,
attitude, and behavior (Feldman & Lynch, 1988), the case study research includes
searching for unsolicited weblog consumer self-reports of stories reflecting brands
causing inciting incidents (cf., Holt’s, 2002 report on why brands cause trouble) or
brands enabling consumers to enact myths (i.e., iconic brands).
The findings include strong evidence to support the theory; the findings and the
theory are useful for increasing an understanding of how consumers cope and solve
dilemmas found in the contexts of everyday and unique situations that they encounter.
Two marketing strategy implications that follow naturally from this research include
(1) creating new dramas that build on real-life self-reports of lived experiences with
brand icons and (2) practicing-to-improve brand icon stories to aid consumers in
unconsciously and consciously enacting archetypes via experiences with brands.
Consumer Storytelling Theory and Research Propositions

Several propositions follow from the discussion of consumer storytelling theory that
extends Escalas’ (2004) general hypothesis that the structure of word-of-mouth
communication is usually organized as a story. These additional propositions include
the following components of consumer storytelling theory.
P1. Narrative storytelling on purchasing/consumption requires a protagonist
consumer to experience an ‘‘inciting incident’’ (McKee, 2003) that focuses her
attention and results in action in response to this incident.
P2. Consumer storytelling theory extends beyond highly risky consumption acts to
the more mundane and improvisational presentations-of-self (to oneself and
others) in everyday life.
P3. Consumption stories protagonists tell about themselves consciously and/or
unconsciously often match the plot lines scripted by brand controllers (e.g., Nike’s
myth of individual achievement through perseverance) or by deep-seated,
‘‘hardwired’’ (see Hirschman, 2000b; Rapaille, 2004) cultural archetypal myths.
P4. Consumers’ presentations-of-self storytelling presents a protagonist engaging
in actions to achieve goals.
P5. These stories inform the listener or reader about conscious and/or unconscious
thoughts of the protagonist and other actors.
P6. These stories frequently describe epiphany personal realizations and changes
that occur in the life of the protagonist.
P7. The story informs how events involving the protagonist take place.
P8. The stories frequently include a second inciting event (a crisis or turning point)
involving the protagonist, along with a beginning inciting incident and a
resolution.
P9. The stories present the protagonist in clear-cut situations.
P10. The storytellers frequently provide a lesson learned — a gloss/gist of

wisdom — in a sense-making summary comment in the closing or opening emic
interpretation to the story.
Figure 1 shows some of these 10 propositions visually.
Figure 1 depicts elements in a story on two graphical dimensions: past, present,
and future on the x-axis; emotional intensity (see Delgadillo & Escalas, 2004) on the
y-axis. Figure 1 includes an ascending and descending evaluative slope with several
minor and one or more major emotional peaks. Figure 1 configures time in narrative
format as episodes (i.e., scenes within acts in a drama); each act has a beginning,
middle, and end, because the storyteller thinks of acts occurring in distinct units
of time even though in reality time is an undifferentiated continuous flow (Bruner,
1990a, 1990b; Escalas, 1998). A scene is equivalent to a situation; a situation
represents the conjunction of a physical place, specific time, physical objectives
(props) and atmospherics, and specific people (cf., Bearden & Woodside, 1976; Belk,
1975); any change in any one of these four factors represents a scene change within
an act. Interactions (e.g., talking, yelling) and actions and events (e.g., touching,
moving, eating, walking) occur within scenes.
Figure 1 shows several increasing emotional peaks. The peaks represent conscious
and/or unconscious minor/major inciting incidents (e.g., the protagonist recognizing
the possibility of going on a trip to Paris; deciding on going and making specific plans
for the trip; arriving in Paris; seeing the lights of Paris at night at the Eiffel Tower).
The expectation is that consumer storytelling reports include at least four emotional
peaks varying in height: (1) I Act 1, the protagonist has some suspicion or fully
conscious sense of awareness by the protagonist of the possibility of engaging in a
‘‘dramatic turn of events’’ (2) in Act 2, the protagonist engages in actions resulting in
recognizing that an inciting incident has occurred (e.g., seeing paintings in museums
in Paris that the protagonist has seen in textbooks as a child); (3) in Act 3, the
protagonist experiences a sense of resolution or fulfillment (e.g., she receives
travelers’ checks from American Express office in Paris replacing the checks stolen
along with her pocketbook); (4) in Act 4, telling the tale each time evokes a small
emotional peak again and again, and the protagonist engages in reflection — self-
interaction. Reflection is one dimension of sense making (Weick, 1995); Act 4
concludes the story often with a summary judgment with an implicit or explicit
statement of a lesson learned (P10).
Figure 1 shows one person, the protagonist, in all acts but not all scenes in the
story. Certainly, the protagonist is not always present in all acts that she may report
in telling the story. The intention is for Figure 1 to represent a rough template only of
the elements, flow, and content of stories that a protagonist may include in her
storytelling.
Most consumer storytelling likely occurs in narrated story form; the protagonist
writes or speaks in past tense in a looking-back reporting mode in describing
and interpreting scenes, acts, and outcomes — rather than reporting pure dramas.
52
S9
A2
Emotion S6
S1 S3 A2
A2 A2 V5
A5 A1
A1 A1
A4 A5 A6
S8 V8
A3 V7
V1 V3 V4 A1 S10 Evaluative
V2 A1 slope
V6 V9
S2
A7
S4 S7
Prequel Act 1 (Balance) Act 2 (Imbalance) Act 3 (Resolution) Act 4 (Sensemaking &
Storytelling)
Acts through Time (Seconds, Minutes, Hours, Days, Weeks, Months, Decades)
Figure 1: Consumer storytelling theory of structure and emotional slope. Notes: Ai, Actor; Vi, event/action; Si, scene; 2,
interaction/dialogue. Only A1, the protagonist, appears in all acts. In Act 4 the protagonist makes sense of the action when
telling the story to herself and/or others about the events occurring beforehand (prequel), the action leading up (Act 1) to the
inciting incident (Act 2) and how the story turns out (resolution, Act 3).
In pure dramas, the actions and the characters simply unfold in the storytelling,
and the expectation is that the readers, viewers, or listeners provide their own
interpretations (Boller, 1990; Booth, 1961; Iser, 1976; Martin, 1986). The distinction
between overt narration and pure drama matters. Here is why:
Overt narration tends to distance an audience from the characters.

Specifically, narration prompts an audience to view the depicted
experiences as ‘‘someone else’s.’’ In contrast, dramatized stories tend to
prompt an audience to view the depicted experiences as ‘‘their own.’’
Dramatic forms invite audience participation in the portrayed
experiences. Through their imaginary apprehension of the characters’
mental states and motives, audience members share in the enactment of
the story actions. (Boller, 1990, p. 622)
Figures 2 and 3 further extend and deepen understanding of relevant central

concepts in drama theory applied to marketing and consumer psychology. Figure 2
expands [433]Moisio and Arnould’s (2005) dramaturgical framework in marketing.
Building from Goffman’s (1959) drama contributions concerning presentation of self
Observable
drama structure:
the set of theatrical components—
Observable setting, actors/audience and
performance (set of activities
drama content:
that occurs before an audience) Observable
Observable cultural resources
that infuse activity with signs that drama interaction:
dramatically highlight and participants’ conversations and actions
portray confirmatory facts shaping, directing, structuring the
that might otherwise performance
remain obscure
Observable
Unobservable
Unobservable
drama structure:
how the participants and audience
interpret the emotional meaning,
Unobservable intensity, and slope via
drama content: observing the theatrical Unobservable
Unobservable archetype components during and drama interaction:
resources interpreted partly after the performance the meanings behind/
consciously but mostly underneath what the
unconsciously; during the drama participants say and do;
participants’struggles to make what the audience
sense of archetype hears/interprets the what the
relevancy participants say and do
Figure 2: Further extending the dramaturgical framework in marketing and

consumer psychology.
54
Archetype
Core archetype
myth affecting consumer
and brand enactment
mostly unconsciously
Consumer (t1)
2 1
4
Brand
3
Marketer
unconscious/
conscious story Consumer (t4)
portrayal 5 6
Ecological development stage 7
unconscious desire to enact
archetype
Story Enactment (t2), Reporting and Interpreted (t3)

Archetype story performed by consumer via brand use. Proper pleasure
Emic sense-making by consumer and marketer; (Aristotle, 294BC)
etic sense-making by researcher. Which archetype is enacted?
How was enactment accomplished? What complementary archetypes
also are enacted in the realized story?
8A
Key: 8A, B, and C = indicators of a specific archetype enactmen t. 8B
8C
Figure 3: Brand enabling archetype enactment by consumer.

in everyday life and the work of Grove and Fisk (1992), Moisio and Arnould (2005)
distinguish among drama structure, drama interaction, and drama content. Drama
structure refers to the set of theatrical components: setting, actors/audience, and
performance, or the formal components of drama (Grove & Fisk, 1992).
Drama interaction is the level of consumer involvement or activities, ranging from
active to passive, which can shape, redirect, and structure the unfolding of the drama
performance. Drama content refers to the cultural resources that ‘‘infuse y activity
with signs which dramatically highlight and portray confirmatory facts that might
otherwise remain unapparent or obscure’’ (Goffman, 1959, p. 30). Figure 2 extends
the above descriptions of drama structure, drama interaction, and drama content in
several ways. First, Figure 2 includes revisions of the descriptions of each concept;
for example, drama interactions explicitly include conversations and actions in
Figure 2. Such activity may include conversations in the drama between one or more
human participants as well as between a consumer and one or more brands. A brand
may express a viewpoint explicitly, for example, the ‘‘I AM A MAN’’ message
expressed in Burger King’s Double Whopper 2006 TV commercials.
Second, Figure 2 suggests a causal pattern from drama content directly
influencing drama interaction and indirectly influencing drama structure. However,
certainly drama structure and interaction affect drama content in some situations —
drama content is contingent on relevant drama structure, permitting such content to
be activated.
Third, the below-the-dotted-line dramaturgical framework in Figure 2 serves to
deepen sense making of each of the three concepts. The participants (actors) in the
drama as well the audience (again, the actors are also the audience members to their
own performances, as well as consumer psychologists and marketers) are both
consciously and unconsciously interpreting drama content, structure, and interac-
tions. A scene comes readily to mind here from Moonstruck, where Cher, playing a
widow, is explaining to Nicholas Cage why he lost his hand. The dialogue goes
something like this: Cher, conversing in character, tells the brother of her fiancé,
‘‘I know the real reason that you lost your hand [to an electric bread cutter] —
because you are a wolf trapped in a loveless engagement and you cut off your own
hand to escape from the marriage.’’ Cage, in character, responds to hearing this
revelation/interpretation by violently swiping food, plates, glasses, and a bottle of
liquor off his kitchen table, lifting Cher up into his arms, and announcing, ‘‘Son of a
bitch!’’ The profanity is not directed at Cher’s character, but is an assessment of his
own feeble prior attempts to make sense of how he came to lose his hand. Note here
that several different actor/audience members are interpreting the action and the
meaning of the action, including the two principal actors, the author of this chapter,
and you — the reader. The deeper archetypal interpretation revision by Cage’s
character shifts the blame of the loss of his hand from his brother as ‘‘the little
trickster’’ to himself as ‘‘the anti hero’’ (for a discussion of these two archetypes, see
Wertime, 2002). The little trickster frequently causes embarrassment and loss of
attention — what Cage’s character claims his brother did to him when he lost his
hand. The wolf in fairy tales and other myths represents the human potential for
destruction and evil — the archetypal profile of the anti hero (for elaborations of
archetypes applicable to marketing and consumer psychology, see Mark & Pearson,
2001; Wertime, 2002).
Figure 3 emphasizes the centrality of archetypes for deepening sense making of
classical dramas applied to marketing and consumer psychology. Arrow 1 in Figure 3
represents the proposal that mostly unconscious desire for archetype fulfillment
drives consumers to act. Arrow 2 proposes that the classical drama productions by
marketers reflect one or a few archetypes. An example of a marketer’s application of
the ultimate strength archetype is the drama Timex watches manifests when the
watch survives torture tests: ‘‘It takes a licking and still keeps ticking,’’ thereby
confirming the gist of this archetype. When an obstacle is there, it must be overcome;
strength must be proven in use. Wertime (2002) provides details of marketing
applications for 12 archetypes. Arrow 3 in Figure 3 reflects the proposal that the
archetype a marketer’s story produces in a brand drama should match with the
intended customer’s archetypal, mostly or entirely unconscious yearnings.
Table 1 provides gists of stories for 12 archetypes that Wertime (2002) describes,
along with examples of brands explicitly or implicitly informing consumers that
experiencing these brands demonstrates fulfilling one or more of these archetypes.
In Figure 3, arrows 4, 5, and 6 illustrate that the consumers’ enactment of a story
and their retrospective storytelling reflects one or more archetypes; the same proposal
applies to brand storytelling by the marketer. Consumer emic reports of own-lived
stories are unlikely to describe archetypal origins of stories because few consumers
have training in sense making using archetype theory. Etic interpreting by researchers
with such training is usually necessary to achieve such insights. In the case study that
this chapter later describes, the fairy godmother (i.e., mother-of-goodness archetype)
illustrates this point — the retrospective report by two American travelers on their
trip to Paris includes an older cousin paying all trip expenses of the younger cousin
and arranging events leading to transformational experiences for the younger cousin.
Considering arrow 5 in Figure 3, Mark and Pearson (2001) provide an etic
interpretation of the innocent archetype (i.e., natural living, childlike simplicity,
and experiencing paradise) of Volkswagen’s new Beetle car design. But looking at
the design through the archetypal lens, we see that the ‘‘face’’ of the new Beetle is
virtually identical to the face of an infant — with big eyes and a high, smooth
forehead. Research shows that throughout the animal and human kingdom, those
same baby-faced characteristics, the characteristics of the innocent, signal that there
is no threat and that the creature is in need of care. They are the facial characteristics
of the koala bear, the teddy bear, Mickey Mouse, and, most recently, the little
television creatures from the United Kingdom called the Teletubbies.
They are faces that win hearts the world over (Mark & Pearson, 2001).
A suggestion for future research is worth noting here. The reporting of such looking
through the archetypal lens by Mark and Pearson (2001), Rapaille (2004), and
Wertime (2002) — and Holt (2004) and Holt and Thompson (2004) even though this
work does not refer directly to archetypes — would be enriched by further interviews
with consumers and marketers during which the consumer researcher describes
the archetypal interpretations of the consumers’ earlier collected stories back to the
consumers. The question here is how do consumers (emic) interpret the researcher’s
Table 1: Archetypes, story gists, and brand examples.
Archetype Story gist Brand examples
Ultimate When an obstacle is there, it Timex — ‘‘It takes a licking

Strength must be overcome; strength and keeps ticking
must be proven in use
The Siren Power of attraction, linked Allure by Chanel; Envy
with the possibility of by Gucci
destruction
The Hero Fortitude, courage, and Michael Jordan and Nike
victory; a journey and shoes; Joe DiMaggio and
transformation Mr. Coffee; Power Puff Girls;
Forrest Gump
The Anti-Hero Universal message of Heavy metal icons; Howard
destruction and Stern; Jerry Springer;
attraction of evil; the Oakland Raiders; Che
bad dude Guevara; Harley-Davidson
The Creator Creative inspiration and the Coca Cola — the real thing;
potency of imagination; Walt Disney; Kleenex
originality; authentic
The Change Transformation, self- Curves — workout stores for
Master improvement and women; Gillette’s Mach 2
self-mastery Razor; Porsche 911
The Authority, influence and CNN; E.F. Hutton; Bill Gates;
Powerbroker domination — the world’s Microsoft
leading -y; the best y;
number one
The Wise Old Experience, advice and Levi’s; Obi-Wan Kenobi
Man heritage; staying the
test of time
The Loyalist Trust, loyalty and reassurance Coca Cola and ‘‘Mean’’ Joe
Green with boy of 12 TV
commercial; I Love Lucy;
Friends TV sitcom
The Mother of Purity, nourishment, Just Juice; Ivory Soap;
Goodness and motherly warmth Tropicana Orange Juice; Aunt
Jemima; Fairy Godmother;
Witch of the East; Snow White
The Little Humor, non-conformity, Dennis the Menace; Bart
Trickster and the element of surprise Simpson; Pee-Wee’s Big
Adventure; SpongeBob
SquarePants
The Enigma Mystery, suspense, and Zorro; Abercrombie and Fitch;
uncertainty Star Trek
Source: Developed in part from several chapters in Weretime (2002).
(etic) archetypal interpretations? Of course, such etic-enhancing–emic-reinterpreting

research would follow with another round of etic (i.e., researcher) interpretation of
the consumers’ reinterpretations of the archetype reports. Such research extends
Hirschman’s (1986) call for asking other researchers to provide ‘‘member checks’’ of
facts and interpretations collected in case studies by a field-investigating researcher.
Without referring to the literature on archetypes, Cox (1967) provides an early
example of multiple rounds of emic–etic–emic–etic interpretive case study research
that reflects this call for deepening sense making of archetype interpretations.
Let consumer researchers not be so arrogant and vain as to believe that their
interpretations of consumer storytelling reports are necessarily accurate, complete,
or the only ones applicable to interpreting the relevance of specific archetypes
(cf., Woodside, 2006).
Arrow 7 in Figure 3 emphasizes that ‘‘good’’ (see McKee, 2003), storytelling
results in ‘‘proper pleasure’’ (Aristotle, see Hiltunen, 2002). Living out a good story
and retelling such a story offers a captivating emotional experience, leading to
fulfilling one or more archetype outcomes (arrows 8 A, B, and C in Figure 3). Such
a theory increases understanding of why weblogs report seemingly mundane
experiences that include using product and service brands. The case studies that
follow illustrate this proposition.
Method
Based on the scarcity of research on consumer self-storytelling (relative to consumer

attitude research), our intention is that the propositions are conjectures for
developing a nascent stream of research work. Future work should adopt the use
of a competing multiple hypotheses probing approach, for example, examining
competing lecture versus drama theories of communications by consumers via
building and testing alternative sets of propositions (see Campbell, 1975; Wilson &
Wilson, 1988; Wilson & Woodside, 1999). This report probes two sets of consumer
storytelling versus advertising lecturing propositions.
Figure 4a is a conversion of the set of consumer storytelling propositions into
a degrees-of-freedom instrument for assessing how well a specific WOM/blog
communication matches with the storytelling paradigm. Figure 4b is a conversion of
advertising lecturing propositions (see Wells, 1989, for an elaboration of these
propositions).
Figure 4c illustrates the possibilities of advertising representing different
combinations of low/moderate/high levels of drama and lecture formats. Figure 4c
reflects the proposition that, although all combination levels do occur in advertising,
most advertising executions follow the lecture format; marketers seek to convince
consumers using attribute–benefit–satisfaction arguments for buying their brands.
Thus, Figure 4c depicts the most frequently occurring advertising structure, using the
large oval in quadrant 3.
Title of Blog Entry and Reference: ____________________________________________________________________________________________
For each proposition below, please circle one: N = No ? = Not sure Y = Yes
regarding whether or not the proposition matching storytelling format: No ? Yes
1.Protagonist experiences an inciting incident? N ? Y

If yes, describe the inciting incident: ________________________________________
2. Mundane presentation of self in everyday life of a protagonist is apparent in the story? N ? Y

Evidence: _____________________________________________________________
3. Unconscious and/or conscious attempt in blog to match a myth identified for brand? N ? Y
Evidence: _____________________________________________________________
4. Protagonist found in blog engages in actions to achieve goals? N ? Y

Evidence: _____________________________________________________________
5. The blog presents a story that informs about conscious and/or unconscious thoughts of the
protagonist and other actors? N ? Y
Evidence: ________________________________________________
6. The blog presents a story about how personal evolution or change in the life of the protagonist occurs? N ? Y
Evidence: __________________________________________________________________________
7. The blog describes a world or personal block that does/may prevent goal attainment? N ? Y
Evidence: _____________________________________________________________
8. The blog has a beginning, middle, and ending? N ? Y

Evidence: ______________________________________________________________
9. The blog presents a protagonist in clear-cut situations? N ? Y
Evidence: ________________________________________________________________
10. The blog has a protagonist who offers a lesson learned? N ? Y
Evidence: ________________________________________________________________
Figure 4a: Storytelling degrees-of-freedom instrument.

59
60
Title of Blog Entry and Reference: ____________________________________________________________________________________________
For each proposition below, please circle one regarding whether or not N = No ? = Not sure Y = Yes
the proposition matching lecture format: No ? Yes
1. Does the communication speak directly to third-party (e.g. reader) audience? N ? Y

Evidence: ______________________________________________________________________
2. Communication presents specific feature details of a product/service? N ? Y

Evidence: _____________________________________________________________________
3. Use of persuasion tools (e.g. selling words, “you should”, “limited time offer”)? N ? Y
Evidence: _____________________________________________________________________
4. Is a protagonist absent from the communication (e.g. no names given)? N ? Y

Evidence: _____________________________________________________________________
5. Does the communication present a source or spokesperson as an expert?

(highly credible, e.g. Michael Jordan spokesperson for Nike brand)? N ? Y
Evidence: It quotes AC Nielsen and the Center for Disease Control and Prevention.
6. Is a one or two-sided argument logic found in the communication? N ? Y

Evidence: _____________________________________________________________________
7. Does the communication include one or more conclusions? N ? Y

Evidence: ______________________________________________________________________
8. Does the communication include one or more recommendations for

the viewer/reader to follow? N ? Y
Evidence: _____________________________________________________________________
9. Does the communication specifically mention benefits from features

or from using the product/service? N ? Y
Evidence: _____________________________________________________________________
10. Does the communication include normative statements that signal what
conclusion is most appropriate and/or what actions are best? N ? Y

Evidence: ____________________________________________________________________
Figure 4b: Lecture degrees-of-freedom instrument.

High Drama
Quadrant 1: High drama, low lecture 10 Quadrant 2: High drama, high lecture Example: Brand-consumer
as heroes advertising; often
incorporates consumer
Example: 9 High drama/
High drama/ experiences into scenes
Apple’s famous High lecture
Low lecture in commercials along
“Big Brother” screen
with lots of reason-why
destruction TV Super Bowl 8
copy, see herohonda.com
commercial
High
5
Lecture
Example: Vignette television
Australian 2006 print & advertising drama
TV tourism vignette (see Stern 1994) 4
commercials, tagline,
“So where the bloody
hell are you? 3 Low drama/
High lecture
Examples: Showing water 2 Example:

flowing over rocks in car “Hard sell” car dealer TV
Low drama/ TV commercial; empty Corona Commercials run in local markets
Low lecture beer bottle rolling on table top at 1
beach; no voice over message
0 Quadrant 3: High lecture, low drama

Quadrant 4: Low drama, low lecture
Figure 4c: Drama/Lecture advertising grid. Note: Size of oval by grid location reflects hypothesis concerning relative
frequency of advertising drama–lecture combination, for example, most radio commercials reflect high lecture and low drama
structure (quadrant 3).
61
Stern’s (1994) description of vignette advertising represents a combination of low-

lecture and moderate-drama advertising (ads including several brief incomplete
stories) — appearing high in quadrant 4. Low-drama in combination with low-
lecture format advertising (bottom right of quadrant 4) likely rarely occurs; such
ads represent attempts to achieve high consumer message involvement or brand-
acceptance without-ad involvement via very unique, low-reactance advertising
executions. One example of a quadrant 4 TV commercial that comes to mind is a
Corona ad showing an empty bottle of its beer rolling sideways on a table with other
free-standing empty Corona beer bottles in front of a beach — a 30-second TV
commercial that includes sounds of seagulls. Some consumers have found this low-
drama and low-lecture Corona campaign to be so appealing that dozens of amateur
interpretive productions of the campaign are available for watching on Youtube.
com. Brand Web sites often represent information locations high in drama and
lecture formats — quadrant 2 in Figure 4c. The herohonda.com Web site showing
motorcycles fit for rebels in their twenties is one example of such a high-drama and
high-lecture communication execution. The herohonda.com sites includes lots of
reason-for-buying copy along with lots of details for ‘‘great escape’’ motorcycle trips,
for example, several days on a road trip from Manali to Leh, the capital city
of Ladakh, located in the Himalayas. Planning and going on such a road trip to
Leh — or just buying the motorcycle and dreaming of such a road trip — likely
unconsciously enables both the change master and the creator archetypes in the
minds of the planner/dreamer. The capability of designing brand Web sites high in
both drama and lecture formats is a distinctive strength of Internet brand marketing.
For this chapter, the method used for collecting consumer-generated commu-
nications includes doing Google searches for brands for which some consumers
experience high emotion during purchase or use — brands or experiences for which
archetypal myths are readily identifiable: Paris and travel, Versace and buying and
wearing a coat, and Tommy Hilfiger and wearing jeans. Paris, as personified in Greek
mythology, is known for extreme beauty and romance. Versace is the luxury brand of
clothing and other products and services (e.g., hotels) that often uses ads showing a
‘‘superhero corporate titan closing deals with her mental acumen and intoxicating
men with her overt eroticism,’’ for instance, Madonna wearing brightly colored,
hyper sexy daywear while playing corporate executive (Givhan, 2005). ‘‘Wearing
Hilfiger represents youthful success and the state of ‘cool’ being’’ (Rsingh.net, 2005).
The Google searches for consumer communications included entering the brand
name and blog for each of the three brands. For each brand, the first entry returned
from the search that was found not to include advertisements or obvious brand
sponsorship was selected for analysis.
Certainly this search method for brands and blogs conjunctions is not an
attempt to achieve representative sampling; the approach is an example of
theoretical sampling to learn whether at least some written communications
among some consumers in a naturally occurring environment matches with
one model of consumer storytelling theory. Future research is necessary to examine
how representative storytelling relating to products and brands is in consumer
blogs.
Appendices A–C include abstracts of the first blog communications passing the
selection screen described above for the three brands. The complete communications
and the blog writers’ Web sites are available for scholarly research from the authors,
but the findings exclude mention of Web site names because of the personal
information in the stories. Because the study focuses on interpreting consumer
storytelling, the sampling procedure includes seeking to find reports in quadrant 1,
high drama and low lecture.
This study does not include the attempt to estimate the shares of drama and
lecture formats; the aim is to examine whether consumer blogs do exist that fit into
the storytelling paradigm and indicate consumer experiences with brands enabling
archetype outcomes. Consequently, the blogs this chapter interprets score very low
on the lecture degrees-of-freedom instrument and very high on the drama degrees-of
freedom instrument. Additional research is necessary to provide information on the
relative use by consumers of drama and lecture formats in the blogs when they
mention experiences with products and brands.
Analyzing the Stories

The analyses of the stories include the following steps: (1) building events and emic
narrative interpretation maps within applications of [686]Woodside and Chebat’s
(2001) approach for updating Heider’s balance theory (Figures 5–7); (2) applying the
consumer storytelling degrees-of-freedom instruments (see Figure 4a) to each blog
communication; (3) applying the Leximancer computer software program (see
Smith, 2000) and comparing the results with the findings from the other two analyses.
Leximancer, a data mining tool useful for extracting the main concepts contained
within electronic documents, generates an automated analysis. The evidence for a
concept will be found in the texts under consideration.
These concepts are displayed on an interactive map that provides a single
holistic view of the information as well as allowing one to automatically search for
instances of the text that contain the given concepts. For the analyses in this chapter,
replacements are first made of ‘‘I’’ in the communications, with ‘‘emic’’ and ‘‘etic’’
replacing the use of ‘‘me’’ in the story to pick up these concepts (cf., Martin, 2005).
This step was done because Leximancer output does not report personal pronouns as
concepts, and the conclusion was drawn that the output should include actors as
important concepts.
Findings
The findings include a balance/imbalance mapping and flow diagram of relationships

for each blog communication, a consumer storytelling degrees-of-freedom analyses,
and a Leximancer output.
64
18. “We went on Fat

17. “I wanted Paige to get a feel Tire's day trip to 19.....“I know Paige will
for shopping experiences that she Monet's gardens and treasure the memory of
would not have at home (aka the house in Giverny, about this girl's trip for many
ubiquitous mall).” 16. “On our trip to Giverny, we met a young an hour outside Paris.” years to come.”
woman from Brisbane, Australia who was
traveling on her own and we invited her to join
us. Three of us enjoyed delicious and
innovative soufflés, while Paige had the rack of
3. Paris
+ lamb. We shared two dessert soufflés, one 11. Sites
chocolate and the other cherry/almond. Yum” • The Marais
• Notre Dame
• L'Arc de Triomphe - 248 steps up and 248 steps
+ down...
15. “Michael Osman is an American artists • Champs Elysee
1. Gayle living in Paris.” • Jacquemart Museum
“He supplements his income by being a • Louvre Lite
+ tour guide. “I” found out about him on • Musee D'Orsay
Fodors” • Les Invalides, Napoleon's Tomb and the
2. Paige “So I engaged Michael for two days.” Napoleon Museum
• Sacre Coeur
• Monmartre
+ 14. “They had decide to come to Paris • Rodin Museum
to find the Harley Davidson store so • Pompidou Museum
they could buy Harley Paris t-shirts.” • Train to Vernon, bike to Giverny with Fat Tire
4. “The occasion Bike Tours
5. “I am a
was my cousin • http://www.fattirebiketoursparis.com/
Canadian and get
Paige’s 16th” 13. “The father stretched out his cupped hands • Eiffel Tower
by in French.” which held all of the pieces they were able to
recover, including the memory stick and he
very solemnly said, “El muerto....”
9. “I bought a Paris Pratique pocket-sized book at a

6. “All I can say is WOW! We rented a 2 Metro station. This handy guide has detailed maps
bedroom, 1½ bath apartment (two of each arrondisement, as well as the metro lines,
12. Unforgettable Memories the bus lines, the RER and the SCNF (trains). I'll
showers), “Merlot” from ParisPerfect
“This trip had so many memories, but here are a few choice never be without this again.”
http://www.parisperfect.com/ and boy was it
ever perfect!” highlights........On our very first night, knowing that the Eiffel
Tower light show started at 10:00 p.m.... she [Paige] dropped
her camera…down 6 flights…we were stunned…Spanish
Family below standing below [with pieces of the camera]”
10. “Six months before our trip, I gave
7. “We had a full view of the Eiffel from 8. “We were walkable to many good Paige a couple of good guide books on
our charming little terrace. ....We were bistros, cafes and bakeries and only a Paris and suggested she let me know
within walking distance to two metro stops few blocks from the wonderful market what her interests were since after all,
(Pont d'Alma or Ecole Militaire).” street Rue Cler.” this was to be her trip.”
Figure 5: Elaboration of trip to Paris blog communication.

16. “Most of all, I wanted to go
15. “I reached the car (out of breath!).........slid back inside and talk to
the jeans up to my waist, zipped them up, and + Valentine. I wanted to tell her
2. Aching back fastened the button. Rapture! I was never so glad about the whole silly situation
− to be wearing pants in all my life. Happiness But life, as I well know, is not
flooded over me. I wanted to sing a song. I the movies, and I was not about
1. Scott wanted to hug someone. I wanted to apologize to to risk humiliation twice in one
Tommy Hilfiger and take back the nasty things evening—my hips snugly
− I'd said about his questionable parentage. “I was ensconced in denim.”
+
emotional! No hard feelings?”
3. Wheeled garment bag for −

Christmas trip home
+
14. “It was with horror that I realized my
next predicament: getting to the counter.
+ 5. Attractive sales My pants were looser than ever, and the
4. Store women motion of walking would surely just
speed their escape.”
−
12. “I did not have $700 to spend on
Italian luggage, so my choice was
6. Pants unzipped 13. “You know, I think I'm going
pretty clear after about five
go ahead and get the Travelpro.”
minutes.........choose the Travelpro .....
Her face lit up like a child's on because I had the Financial Acumen to
Christmas morning. recognize that it was a great buy.”
7. “Not only that, but I happened to be wearing a
pair of loose-fitting jeans that day, purchased in a
foolish attempt to look with it. (Damn you and your
marketing, Tommy Hilfiger!) These pants were
almost loose enough to fall off when they were
fastened, and in their current state I feared that they
11. “Valentine: Zis bag has
would slip to the ground at the slightest
many expandable pockets.
encouragement.”
Me: Ah. (Left-hand Hike.
Right-hand Cover.)
9. “Part of the problem was that Valentine: Zis makes it very
10. “I adopted a two-pronged
8. “There’s nothing more dreadful than Valentine, which I later learned useful for short trips or long
strategy. The first phase, the
being embarrassed in front of a person you was her name, was an excellent trips.
Hike.......The second phase of
find attractive, and this seemed like the saleswoman. I have never met Me: I see. (Right-hand
my plan, the Cover.....”
inevitable outcome of my current anyone so knowledgeable about Hike.)”
situation.” luggage.”
Figure 6: Elaboration of unzipped — Tommy Hilfiger jean wearer communication.

65
66
20. “I resorted to
2. Fur Coat 19. “I arrived at my mans standing in front of the + 21. Would I buy
− house, 'nice coat' he said....” TV and slowly Versace again? Best
unbuttoning my two coat I ever bought!
buttons”
1. Pollee
+
18. “Quick as a flash I told them I was going to a fancy
+ dress ‘vicars and tarts party’.“.....”I pulled up my coat
3. Designer Shop to show the top of a leg and told them that was all they
were getting!” 13. “It didn't take long to
change into the lingerie and
on a whim I didn't redress
but just put on my coat.”
17. “The problem I had was that the evening was very
4. “I have discovered a + warm and they asked why I was wearing a coat when it
lovely second hand ladies
was so hot.”
wear shop in Beachamp
Place near Harrods in
Knightsbridge which has 12. “On speaking to the man of
designer wear at vastly 16. “Versace did not let me down. The coat, all
the moment I found he was at
reduced prices.” encompassing, only using two of the buttons
home .... We agreed I would get
provided kept my chastity covered. This coat has
a takeaway on the way back.”
quality and after several dry cleanings because of
the colour still looks as good as new!”
5. I bought a Valentino short leather skirt

(and the leather is baby soft, ok I know I
said I wouldn't wear fur-so I'm 15. “The local constabulary [two police
inconsistent! ) which would cost £600 11. “Proud of my purchase and
officers] wanted to discuss me using a mobile
new for £110 and it really as new-you having just been into Rigby and
phone while driving.”
would never have guessed it's second Pellar for my six month refit of
hand. lingerie I started my journey ....”
14. “It didn't take long to change into

the lingerie and on a whim I didn't
redress but just put on my coat.”
6. “I spoed [English for spied]
a beautiful cream coloured
cashmere coat .-Versace no 7. “The fit of the coat 10. “Having worn Versace I
less for 150 pounds......... was wonderful and I 9. “This coat was less than 6 would definitely look to buy
No way would I be able to felt like I was one of 8. “Versace is available at so something else from their
months old and had been
afford one of these at the real those film stars many stores now, almost always 'stable' as the finish on this
sent in by a lady who gets
price of nearly £1,000.” sweeping into the room tied up by anti theft devices so coat is the best. It has been
her clothers for nothing and
wearing the most this was the first time I had ever hand stitched and reeks of
wears them once or twice
wonderful outfit.” been able to try Versace on.” quality.”
and then sends them in for
resale.”
Figure 7: Elaboration of Pollee buying and wearing a Versace coat.

Gayle Takes Cousin Paige to Paris
The first story (Figure 5 and Appendix A) reports on a trip to Paris by a Canadian,
Gayle, and her cousin, Paige. Early concepts in the report include mention of three
principals within a clearly storytelling communication: Gayle, Paris, and Paige.
The relationships among these three principals are all positive and in balance. Gayle
reports previous trips to Paris, and this particular trip represents a coming-of-age
transformation for Paige, with the implication that Gayle and Paige’s mother are
paying for the Paris experience of Paige changing from child to woman. Although
this is not consciously stated, Gayle’s actions and her self-interpretations reflect
a fairy godmother myth enactment.
Gayle’s storytelling includes the possibility of romance, with Gayle providing
a likely Prince Charming for Paige.
But my cousin is a budding artist and I thought it might be enjoyable

for her to experience some of Paris’ best art offerings through the eyes
of another artist. So I engaged Michael for two days. His warmth,
charm, wit and quirkiness were infectious and by the end of two days
we cried when we said goodbye and feel like he’s part of our family.
He brought insight and intelligence, historical perspective and
tailored the itinerary to Paige’s interests. Paige will always remember
him fondly y after all y he taught her the meaning of allegory.
(Appendix A)
Gayle does insure that Paige experiences iconic sites and scenes — actions
uniquely available in Paris: for example, museums, the Eiffel Tower, and the
Metro. Gayle offers a chronological report with an increasing emotional slope
that includes both Parisian sites and the breakage of an expensive camera as
high points. The degrees-of-freedom (DOF) analysis for the trip to Paris case
study includes supporting evidence confirming each of the 10 degrees-of-
freedom issues, leading to the conclusion that Gayle’s communication is a narrative
story.
Examining the Leximancer output for Gayle’s story, the rather primitive stage
of the output from the software program is the first conclusion that might come.
As the description mentions for Leximancer, this structured analysis complements
rather than substitutes for other interpretation methods (e.g., balance and
flow diagramming and applying a storytelling degrees-of-freedom instrument).
(Leximancer diagrams for the cased studies are available from any of the authors)
The close association of Paige and Gayle (emic ‘‘I’’) is the main observable finding
from the Leximancer output for this story.
Note in the closing lines of the story that Gale connects the concepts of Paige,
treasure, and memory for their trip to Paris. The metaphor of their trip experiences,
representing acquisition of a forevermore treasure, expresses achieving an important
goal — one of the elements in good storytelling.
Scott Wears Tommy Hilfiger Jeans Unzipped While Shopping

for a Wheeled Garment Bag
The start of Figure 6 includes two negative associations in a balanced three-way

relationship that includes Scott (clearly the protagonist in this report), an aching
back, and a wheeled garment bag. The positive relationship between Scott and the
bag creates balance that overcomes the negative relationship between Scott and an
aching back. Scott’s first mention of Tommy Hilfiger jeans reflects one example of
‘‘brands causing trouble’’ (Holt, 2002).
Not only that, but I happened to be wearing a pair of loose-fitting jeans

that day, purchased in a foolish attempt to look with it. (Damn
you and your marketing, Tommy Hilfiger!) These pants were almost
loose enough to fall off when they were fastened, and in their current
state I feared that they would slip to the ground at the slightest
encouragement. (Appendix B)
This quote indicates Scott’s meta conscious (thinking about thinking) awareness
of his attempt to live the myth that he perceives Tommy Hilfiger pitches — what
Rsingh (2005) summarizes as the ‘‘state of cool being.’’ Scott’s parenthetical editorial
reflects a nascent view of Holt’s treatise that some consumers become inflamed
(in Heider’s change paradigm) to unfreeze and refreeze in adopting an anti-branding
sentiment and lifestyle in rebellion to marketers, cultural authority. Note in Figure 6
that two story streams occur in a parallel format: Scott’s concern about his jeans
being unzipped while talking with a sales woman and his buying a wheeled garment
bag. Box 15 includes some indication of an emotional high from finally getting to zip
up his jeans: ‘‘Rapture! y Happiness flooded over me.’’ Box 15 also indicates Scott’s
return to a believer in the myth intended by the marketer for Tommy Hilfiger:
‘‘I wanted to apologize to Tommy Hilfiger and take back the nasty things I’d said
about his questionable parentage.’’
The degrees-of-freedom analysis includes substantial evidence that Scott’s report
is narrative storytelling. The use of quotes by Scott in Box 11 hints at his shifting
to a pure drama form of turn-taking conversation (see Schenkein, 1978) between
Valentine (the sales woman) and himself.
The Leximancer output interprets a dominance of the wheeled bag over the pants
in Scott’s story. However, Scott as a self-concept ‘‘I’’ overlaps more with the pants
than the bag.
Appendix B includes a unique post-story epilogue offering by Scott.
‘‘EPILOGUE: It turned out that the garment bag was too small, so I couldn’t
even take it on my trip. I wanted to return it, but I couldn’t bring myself to go back.
The bag is still sitting in my closet. I haven’t worn those jeans since that night, and
I haven’t been back to the store.’’ The ironies of neither using nor returning the bag
or the jeans serves to illustrate Scott’s story as a satire of his life with brands and
serves as a self exposé of a ‘‘secret affair’’ with Tommy Hilfiger, that is, a privately
held relationship considered risky if exposed to others (see Fournier, 1998).
Pollee Buys and then Wears a Versace Coat sans Knickers

(i.e., No Underpants)
Appendix C is an abstract of the third blog communication. Figure 7 offers balance/

imbalance mapping results. The author of this communication, Pollee (an alias),
entitles the report, ‘‘Long coat, no knickers.’’ Given that the substantial evidence
found in the entry supports all 10 dimensions of storytelling, the report reflects a
narrative story.
Note in Pollee’s story that she also buys lingerie at Rigby and Pellar (a retail
store), and she decides to pull ‘‘into a garage to use the loo.’’ She decides to surprise
her ‘‘man of the moment’’ by removing her dress and knickers in the loo, to wear
only her coat and lingerie (likely a slip and stockings) home to ‘‘my man.’’ She is
stopped by two police officers after the garage stop for using a cell phone while
driving (illegal in Britain), and the officers also ask why she is wearing a coat on such
a warm (August) evening. ‘‘Quick as a flash I told them I was going to a fancy dress
vicars and tarts party.’’ Such a party requires guests to arrive dressed as a vicar or a
prostitute and are an English tradition.
This story expresses characteristics of the Versace coat assisting in a courtship
relationship and participating directly in a secret affair (see Fournier, 1998).
Courtship refers to an interim relationship state, possibly on the road to a
committed partnership contract; the man of the moment morphs into ‘‘my man’’
with no hindrances apparent to a possible committed relationship. Versace is a
co-conspirator with Pollee in a secret affair when she wears the coat without
knickers — a highly emotive, private bonding that leads to a risky exposure to police
officers. Note in the story’s ending that Pollee includes a happy face symbol ( :) ) in
suggesting her secret affair drama with Versace leads to a final emotional high point;
her man achieves an erection (‘‘only the Nan bread went soft’’), and the happy face
follows this development.
The opening relationship in Figure 7 includes a brief imbalanced relation-
ship reference for Pollee, fur coats, and design shops. Pollee twice refers to the
point that she ‘‘wouldn’t wear fur.’’ But her addition, ‘‘I’m inconsistent,’’
indicates an imbalanced state that is not resolved in the next scenes and acts in
the story.
The output from the Leximancer analysis indicates that the concept relating
most closely to Pollee (emic) is price. Such a relationship might be overlooked from
studying the interpretive mapping and the results from applying the consumer
storytelling degrees-of-freedom instrument. The second hand shop enabling Pollee to
buy a Versace coat at a price about 15 percent of the new coat retail price is a scene
critical in developing the plotline of the story.
Pollee’s story includes several touch points matching the Versace myth of
enabling a woman to intoxicate men with her overt eroticism. Note that Pollee’s
attempting to kiss her man results in his shifting out of the way to continue to
watch football on the TV. Only by ‘‘slowly unbuttoning my two buttons [of
my Versace]’’ does she achieve the intoxicating affect — the ultimate goal intended in
the story.
Discussion, Limitations, Strategy Implications,

and Future Research Suggestions
The findings support the view that at least some of the time individuals write stories
in naturally occurring blog communications that include enactments of brands
as archetypical icons. This conclusion complements and extends the work of Holt
(2003, 2004), Holt and Thompson (2004), and related literature (e.g., Adaval &
Wyer, 1998; Arnould & Wallendorf, 1994; Boller, 1990; Delgadillo & Escalas, 2004;
Fournier, 1998; Hirschman, 1986; Padgett & Allen, 1997; Woodside & Chabet, 2001).
Several previous studies may introduce researcher/experimenter biases (Fisher,
1993; Shimp, Hyatt, & Snyder, 1991) in informant reports; for example, Delgadillo
and Escalas’ (2004) procedure in their Study 1 reads as follows: ‘‘Please tell us about a
recent experience you’ve had with a product or brand y.’’ The use of the word ‘‘tell’’
would likely suggest storytelling to some informants. Similarly, Holt and Thompson’s
probing follow-up questions in the face-to-face interviews with the two informants,
the focus of most their article, are likely to stimulate self-generated validity problems
(see Feldman & Lynch, 1988). However, a content analysis (Kassarjian, 1977) method
avoids such issues. At the same time, the analysis and findings provide confirmatory
evidence supporting Holt’s views of how and why brands become icons.
The use of content analysis is not devoid of its own validity and reliability issues.
The authors of this report worked together in studying the three blogs in search of
evidence in support of the consumer storytelling theory. Future research should
include the use of independent member checks (Hirschman, 1986; Woodside, 2006)
and comparisons of multiple informants separately completing the consumer
storytelling and lecture degrees-of-freedom instruments. The use of different data
analysis tools for interpreting the blog communications confirms the instructions
Smith (2000) provides for using Leximancer: The appropriate view is that the
software program complements independent interpretive assessment of text and oral
communications rather than use as a stand-alone tool. Examining natural reports by
consumers-as-authors using Woodside and Chebat’s (2001) application of Heider’s
balance theory usefully complements Leximancer applications.
Marketing and advertising strategy implications include the following points.
First, crafting a story whereby the brand is a supporting actor enabling the
protagonist to achieve conscious and/or unconscious goals likely helps build very
favorable consumer–brand relationships (e.g., committed partnerships, best friend-
ship, flings, or a secret affair). The storytelling analysis in this chapter includes
identifying self-oriented thinking by the storyteller with near-conversational
interactions with the primary brands appearing in the stories (e.g., Gayle’s ‘‘I love
Paris’’; Scott’s, ‘‘Damn you and your marketing, Tommy Hilfiger!’’ Pollee’s ‘‘Versace
did not let me down’’). Consequently, learning — not only thinking about — what
buyers and users say to the brand and what the brand says first and back-and-forth
in such conversations is likely to provide valuable clues for designing highly effective
marketing and advertising strategies.
Narrative reports and drama enactments are more likely to encourage vicarious
participation (see Boller, Babakus, & Olson, 1989) whereas lecture forms of
advertising tend to evoke argumentative forms of thinking. Given that learning via
storytelling is more memorable and retrievable than lecture-based learning (Bruner,
1990a, 1990b; Schank, 1990), learning the stories consumers tell in natural settings
represents a useful grounding for the art of crafting naturalistic stories or acceptable
fantasies involving brand experiences — stories and fantasies that are acceptable and
enjoyable for the intended audience. Consumer storytelling research provides
evidence as to how and why brands become archetypal icons in the informants’
own words. Informant reported enactments of the iconic roles played by brands
likely include symbols and expressions that match with imprinted unconsciously
driving myths that affect the informants’ behavior. Such originating core myths may
be uncovered via word searches using blog search engines (e.g., Technorati, Feester,
or Blogdigger). Consequently, becoming aware of the consumer enactments of
imprinted myths via brand icons provides direction for story genre and consumer–
brand relationships (e.g., the fairy godmother myth is one plotline worthy of
producing alternative enactments to encourage one segment of travelers to visit Paris;
showing alternative scenes of a protagonist wearing a Versace coat sans knickers is
a story gist useful for rejuvenating romance in a tired relationship).
A stream of studies supports the hypothesis that drama- versus lecture based
advertising is more effective in situations reflecting everyday life (Adaval & Wyer,
1998; Delgadillo & Escalas, 2004; Chebat, Vercollier, & Gelinas-Chebat, 2003).
The present chapter verifies that not only Goffman (1959) and other behavioral
scientists have a compelling need to interpret the presentation-of-self in everyday
life, but also some consumers are driven to offer their own interpretations
(i.e., narratives) of presentations (i.e., dramas) of their everyday lives — lives that
enact myths via the use of iconic brands.
Appendix A
Appendix ‘‘I Love Paris y Anytime at All y,’’ posted by Gayle on November 4,
2004.
Here is my very lengthy and belated trip report from my June trip. The occasion
was my cousin Paige’s 16th birthday. Her mom and I are like sisters and ever since
Paige was born we planned to take her to Paris to celebrate this milestone. I have
been to Paris many times, but my cousins have not. I am a Canadian and can get by
in French. They are Californians and though adept at Spanish, were intimidated
to be in a cultural environment they didn’t know. I am so proud of them both. They
rose to the occasion and we had a blast!
Getting There
I flew direct from Toronto on Air France and they flew LAX-JFK-CDG on
Air France, all without incident. I had arranged for a taxi service to meet us at the
airport (I arrived earlier than they did). As is usual in France the meter had been
started while they were waiting so the originally quoted fare ended up being
substantially higher. I had forgotten about this possibility, which of course resulted
in the fare adding up to substantially more than we’d been quoted when I reserved
via email.
The Apartment
All I can say is WOW! We rented a 2 bedroom, 11/2 bath apartment (two showers),
‘‘Merlot’’ from ParisPerfect http://www.parisperfect.com/ and boy was it ever
perfect! Located on avenue de la Bourdonnais at the intersection of rue Rapp and
rue St. Dominique in the 7th, we had a full view of the Eiffel from our charming
little terrace. The apartment was lovely, with a modern kitchen and baths, very nice
furnishings and accessories.
The second bedroom was very tiny and I would not recommend it for an adult
couple, but for just me it was just fine. My cousins shared the master. The location
of this gem was perfect. We were within walking distance to two metro stops
(Pont d’Alma or Ecole Militaire) and several buses stopped outside our door. We
were walkable to many good bistros, cafes and bakeries and only a few blocks from
the wonderful market street Rue Cler.
The apartment manager was a lovely woman, formerly of Arizona, and a welcome
basket filled with practical goodies was waiting when we arrived, including a baguette
bag for us to keep, a lovely bottle of wine, laundry detergent tabs, some milk, snacks
and a wonderful binder filled with practical information as well as lists and
descriptions of restaurants and attractions. We will be renting from ParisPerfect
again, for sure. There was only one issue. ParisPerfect quotes their rental rates in
U.S. dollars. If you wish to put the rental on a credit card, then the card is actually
charged in Euros. This worked heavily against us. In future, I would ask to be quoted
a rental price in Euros to avoid the fluctuation which I was the victim of since
I booked the apartment one year prior. Or I would send them a U.S. money order,
especially now since the Canadian dollar is so much stronger against the U.S. dollar
than against the Euro.
Getting Around
I told my cousins to bring with them small 1 1/8 headshots and on Saturday we
procured our Cartes Oranges (zones 1–3) and our Museum Passes at our nearest
Metro station, in addition to a carnet of 10 single tickets for getting around until the
Carte Orange ticket became active on Monday morning.
For those of you who don’t speak French, I saw several folks who only spoke
English slip a piece of paper through the ticket wicket indicating what they wanted,
making their transactions very simple. If memory serves 3 Cartes Orange and 3
Museum Passes plus our carnet of 10 tickets cost 219 Euros or 73 Euros each I also
found out later that kids/students (I think under 18) don’t need a Museum pass as
they get in free. However, they cannot always access certain Figureions [sic], so I did
not regret that we got Paige a pass.
On previous trips I have comfortably and adeptly used the metro, but on this trip
I decided it was time to learn to use the more convenient and much more scenic bus
routes which crosses [sic] the City of Light. I am so glad I did. We got to see so much
more of the neighborhoods, travel with locals and actually found it more convenient
to get to the sites we were going to with a bit less walking and a lot less stairs.
I bought a Paris Pratique pocket-sized book at a Metro station. This handy guide
has detailed maps of each arrondisement, as well as the metro lines, the bus lines, the
RER and the SCNF (trains). I’ll never be without this again.
The Sites
Six months before our trip, I gave Paige a couple of good guide books on Paris
and suggested she let me know what her interests were since after all, this was to
be her trip. So with her list, I had prepared a basic itinerary as a guideline but
I wanted to also be flexible and spontaneous. We got through about half of what we
intended to but we experienced so much more we hadn’t planned. We walked 8–10
hours a day and thank goodness for my Puma’s (black, not tourist white) or I would
have died!
As this was their first trip we had to cover some of the basics. Here, roughly was
where we went.
Sites
The Marais
Notre Dame
L’Arc de Triomphe — 248 steps up and 248 steps downy
Champs Elysée
Jacquemart Museum
Louvre Lite
Musée D’Orsay
Les Invalides, Napoleon’s Tomb and the Napoleon Museum
Sacre Coeur
Monmartre
Rodin Museum
Pompidou Museum
Train to Vernon, bike to Giverny with Fat Tire Bike Tours
http://www.fattirebiketoursparis.com/
Eiffel Tower.
Unforgettable Memories
This trip had so many memories, but here are a few choice highlights: A Picture is
Worth y On our very first night, knowing that the Eiffel Tower light show started at
10:00 p.m., I steered my cousin out onto our tiny terrace to see the glittering
spectacle. She was so thrilled and excited and was snapping pictures with her brand-
new very expensive digital camera, but did not have the strap around her wrist.
Yup, you guessed it, in her excitement, she dropped the camera and we both watched
in shock as it bounced off our little marble-topped bistro table, through the ornate
iron railing and down the six flights to the pavement below. We were stunned. It was
not funny, yet we laughed so hard we were crying. I went racing down to the street to
get it and as I ran out the door, heard her conversing with the people down below in
Spanish! When I opened the door to the street there was an entire Spanish family:
grandma, grandpa, mom, dad and a few kids standing waiting for me. The father
stretched out his cupped hands which held all of the pieces they were able to recover,
including the memory stick and he very solemnly said, ‘‘El muertoy.’’ I thanked him
and returned upstairs with the remains. Lesson learned y always use the strap!
D-Day Lessons
On our second evening we wanted an easy, simple, and inexpensive dinner so we

grabbed a table right outside our apartment door at Le Dome, a neighborhood cafe.
A few tables over there were four guys of varying ages, clearly American and from
their haircuts and build, I guessed military. They asked if we could take their picture
and we struck up conversation, only to learn that they were indeed military, guests of
the French government, invited to participate in the D-Day commemoration in
Normandy because they were from the same battalion that liberated Omaha Beach
and they were invited to meet the veterans and represent their battalion today.
They were so honored to be there and had been very moved by the ceremonies.
A couple of them had never been to Europe and this was their great adventure. They
had decided to come to Paris to find the Harley Davidson store so they could buy
Harley Paris t-shirts. But alas, it was not meant to be as the store was closed and they
had to leave the next morning at the crack of dawn. I told them we’d go get them the
shirts and send them to them from California. Needless to say they were thrilled.
We ended up walking to the Bateaux Mouches with them and then along the Seine
to the Eiffel, talking about their jobs and their lives as infantry-men. We heard how
hard it is on families, that most of them don’t care about the politics of the day,
or the current administration, that they have a job to do, whether they agree or not,
and that many of the long-term benefits outweigh the short term sacrifices.
My 16-year-old cousin was simply awestruck by the entire evening’s events. When we
said goodnight and wished them well, Paige turned to us and told us that she had
learned so much and that it had been a night she’d never forget. Of course, we sought
out the Harley store and got them their T-shirts.
Artist in Residence
Michael Osman is an American artist living in Paris. He supplements his income by

being a tour guide. I found out about him on Fodors. Much has been written there
about him y good and bad as well as simply stupid stuff. But my cousin is a budding
artist and I thought it might be enjoyable for her to experience some of Paris’ best art
offerings through the eyes of another artist. So I engaged Michael for two days.
His warmth, charm, wit, and quirkiness were infectious and by the end of two days
we cried when we said goodbye and feel like he’s part of our family. He brought
insight and intelligence, historical perspective and tailored the itinerary to Paige’s
interests. Paige will always remember him fondly y after all y he taught her the
meaning of allegory.
Worth a Try
From the beginning, Paige was a French enthusiast. She asked for a French lesson
every day and would then actually use the new words and phrases wherever we went.
On several occasions, people commented on how polite she was and it served her well
when she was rewarded with additional pain au chocolat or extra dessert treats.
The Men, the Women, the Children
Everywhere we went people were charming, engaging, and welcoming. We especially

noticed how beautiful the women were and their style and flair for dressing, even in
the hottest weather. But the men were equally charming and flirtatious. I got my
come-uppance when I hastily, and without thinking, commented in English to what
I thought was a non-English speaking waiter that he had the most exquisite
eyes I’d ever seen and without skipping a beat, he replied to me in French, ‘‘Only
when I am in your presence!’’ Swoooon!
A Manger [French for ‘‘a diner’’]
We were determined to eat casually and simply. Breakfasts were at our cafe outside
our apartment where for 6 Euros we got orange juice, a buttered baguette,
a croissant, and a cafe crème. Lunches were often on the run, crepes, salads,
sandwiches, etc., usually in close proximity to wherever we were sightseeing. Dinners
were always in our neighborhood and came from either the detailed and lengthy
list of restaurants we were provided by Paris Perfect, or from information from my
various guidebooks. Thanks to so many Fodor’s Posters who provided excellent
restaurant recommendations, as well as to previous guests of ParisPerfect who left
copious notes on their favorite haunts.
A few we enjoyed were Cafe de Mars, La Terrace de 7eme, Bistro du Papa.

After reading a previous Fodorite’s trip report, we headed off to find La Cigale, for
souffles y La Cigale 4 rue Recamier 75007 in the 7th, in a little alley, hard to find.
On our trip to Giverny, we met a young woman from Brisbane, Australia, who was
traveling on her own and we invited her to join us. Three of us enjoyed delicious and
innovative souffles, while Paige had the rack of lamb. We shared two dessert souffles,
one chocolate and the other cherry/almond. Yum!
One of our favorite meals was at Relais de L’Entrecote. One of my all time
favorites, this is a steakhouse like no other you’ve ever been to. No menus. No
reservations. Cash only. For a very low price you get gorgeous a salad, steak sliced
in the best sauce I’ve ever had, unending hot, crispy frites, wine and dessert. A real
neighborhood joint. The review in Zagat says it like this and I concur, ‘‘When
ordering from the friendly but no-nonsense servers, rare, medium or well done
are the only words you need to know because the reasonably priced beef’s the only
game in town here. 2 locations. 6th arrondissement, 8th arrondissement y 20 bis,
rue St-Benoit (St-Germain-des-Prés) Paris 014 5491600, or 15, rue Marbeuf
(Franklin D. Roosevelt) Paris 014 9520717.’’
Shop ’til We Drop
I wanted Paige to get a feel for shopping experiences that she would not have at home
(aka y the ubiquitous Mall).We strolled the designer rich (and rich designer)
Avenue de la Montaigne, the boutiques of St. Germain, the overwhelming Galeries
Lafayette (where I discovered the champagne bar and parked myself to imbibe while
my cousin shopped), the trendy stores of the Marais, the markets, the museum shops,
an art fair, and the bookanistes. She became very adept at scoping out the hundreds
of shops that carried teensy, weensy (size 0) clothes that would set her apart at home,
focusing on the trendiest most fun clothing and accessories.
Her initial shyness soon evaporated and she would stride into a store issuing the
requisite ‘‘Bonjour madame/mademoiselle’’ upon entry and ‘‘Merci, bonne journee’’
upon exiting. One of my favorite experiences was when she asked if we could go
into Chanel on Avenue de la Montaigne y just to look around. She politely asked to
see the earrings and picked up several of the small, plastic button earrings with the
ubiquitous Chanel ‘‘C’s.’’ When told they were 230 Euros, without a beat, and with
a polite smile she said, ‘‘Thank you, they are a bit more than I was hoping to spend.’’
When we left the shop, she commented on the fact they were just a pair of plastic
earrings :)
Monet’s Gardens
We went on Fat Tire’s day trip to Monet’s gardens and house in Giverny, about an
hour outside Paris. We took the train with the group and then picked up the bikes at
the train station in Vernon, biked to the market to pick up lunch items, had a picnic
lunch then biked to the house for the afternoon. It was lovely and I highly
recommend it. We were there at peak season and it was mobbed, so I’d suggest it may
be more enjoyable in the off season, though the lily ponds may not be in bloom.
This company also has several other tours, including a night tour, a day tour, a tour
of Versailles, and segue tours y see their Web site http://www.fattirebiketoursparis.
com/. I reserved on line ahead of time.
The End
I could go on and on y but look, I already have. Obviously we had a wonderful time
and I know Paige will treasure the memory of this girl’s trip for many years to come.
Appendix B
‘‘Unzipped,’’ by Scott, posted on his blog site, December 1999.

It was the evening of December 20th, three days before I was to depart for Indiana
on my annual pilgrimage home. I needed a wheeled garment bag so that I could give
my aching back a rest during the long journey.
As usual, the week before Christmas turned out to be a frenzy, so I had put off my
luggage purchase until the last minute. If I wanted a new garment bag, I had to get it
that night. Not wanting to run the gauntlet of holiday-crazed suburb dwellers that
were crowding the large shopping mall, I opted to visit a luggage store in a small strip
mall near my office. I parked my car and strode for the store entrance, flush with the
happy willingness to buy that seems to possess Americans in the busy days before
Christmas. I pulled the door open, its bell jingling merrily.
At the sound the saleswoman moved from behind the counter to greet me at the
front of the store. As my foot crossed the threshold I immediately realized two
things:
1. The saleswoman was extremely attractive.

2. My pants were completely unzipped.
To say that this woman was extremely attractive would actually be an

understatement. She was gorgeous. She stood 5_7_ and had shoulder length
black hair, high cheekbones, and a perfect button nose. Her well-proportioned
mouth spoke English with a vaguely European accent, which added to the mystery.
Her body was nothing but curves. She was a statue wrapped in a blue-ribbed
turtleneck sweater and stylish black pants, both garments tight enough to observe the
sculptor’s fine craftsmanship from across the room. Her slender legs looked even
longer when punctuated by square-toed high-heeled boots, their shiny black leather
winking from beneath the hem of her slacks.
To say that my pants were unzipped would likewise be an understatement.

In point of fact, they were about to fall off. The button of my jeans had apparently
come undone in the car, and the zipper had fallen completely open under the
inevitable tug of gravity. The only thing keeping the pants attached to their owner at
all was a brown belt, and the 1-inch strip of worn leather did not look like it was up
to the task. Not only that, but I happened to be wearing a pair of loose-fitting jeans
that day, purchased in a foolish attempt to look with it. (Damn you and your
marketing, Tommy Hilfiger!)
These pants were almost loose enough to fall off when they were fastened, and in
their current state I feared that they would slip to the ground at the slightest
encouragement. I told this story to a friend who cheekily applauded my bravado,
suggesting that approaching such a woman with an open zipper constitutes the
sincerest form of honesty. To him let me just say that it’s easy to laugh when your
pants are well secured. I, on the other hand, was mortified. There’s nothing more
dreadful than being embarrassed in front of a person you find attractive, and this
seemed like the inevitable outcome of my current situation. I was instantly
transported back to eighth grade, when my friend Kyle had yanked down my pants
in front of a girl that I really liked at the time. I was devastated. I was so embarrassed
that I couldn’t even look at her after that day. My ears burned for months afterward.
Fourteen years later, I shuddered at the memory.
‘‘May I help you?’’ She asked again. I had apparently missed her original query,
lost in thought as I was. I quickly realized that there were only two choices: make my
purchase and leave as quickly as possible, or walk out the door and never, ever come
back. If I gambled and won, I’d have my luggage and no one would be the wiser.
If I lost y well, it would be eighth grade all over again. As I looked at this radiant
beauty, her smile lighting up the store, I felt that familiar sickly dread of junior high
welling up inside me. The only thing that kept me from bolting was my poor, aching
back. ‘‘Yes,’’ I replied, ‘‘I’m looking for a wheeled garment bag.’’
The saleswoman smiled at me. ‘‘We have several excellent models, sir,’’ she replied
warmly as she strode with feline grace to a rack of luggage, leaving a trail of perfume
and clipped vowels in her wake. This was going to be harder than I thought. Part of
the problem was that Valentine, which I later learned was her name, was an excellent
saleswoman. I have never met anyone so knowledgeable about luggage.
She knew every manufacturer, she knew every bag and, most devastating to my
cause, she knew where everything was in the store. I’d hoped that if I asked enough
questions she might be forced to go in the back room for a minute and leave me alone
to put my pants back on. Alas, this was not to be. She moved with the cool assurance
of a master, ready to handle any question I threw at her. To make matters worse,
I couldn’t take my eyes off of her. She was beautiful. She made luggage sales look like
the sexiest thing in the world. I watched, mesmerized, as her shapely fingers caressed
glistening black nylon. Her silky voice enchanted me with a siren song of aluminum
frame construction and lifetime warranties.
My reverie was snapped only by the sensation of my jeans sliding down my
legs. I realized that I had to regroup and concentrate if I was going to make it
out with my dignity intact. I took a deep breath, cleared my mind, and got to work.
I adopted a two-pronged strategy. The first phase, the Hike, was an attempt to hike
the pants up to temporarily prevent them from slipping off of my hips and onto the
floor. To execute the Hike, I casually hooked a thumb through one of my belt
loops and tugged the waist as high as I could get it. I imagined myself doing this
contemplatively, as an aristocrat might pull on his sleeve to signal a bid at an auction.
The second phase of my plan, the cover, was intended to prevent anyone from
discovering the hiking activities that were then taking place. I thrust my non-hiking
hand into the pocket of my long black coat and casually pulled that side of the coat
around me. At the same time I pulled my shoulders into my coat to cover more of my
front, as if a chill wind had just blown through the store.
I put my plan into action immediately. The resulting exchange would not have
looked out of place in a Charlie Chaplin movie. Although I attempted to make it
look natural by switching my hiking and covering hands, I had to adjust so often that
my clothing was constantly in motion. I was a blur of blue denim and black wool.
Valentine: Zis bag has many expandable pockets.

Me: Ah. (Left-hand Hike. Right-hand Cover.)
Valentine: Zis makes eet very useful for short trips or long trips.
Me: I see. (Right-hand Hike.)
Valentine: Zis bag also uses rip-stop ballistic nylon and durable polyurezane veels
[sic].
Me: (Left-hand Hike.) Polyurethane wheels, huh? (Right-hand Cover. Left-hand
Hike.) That’s good. (Right-hand Hike.)
We continued in this manner for several minutes as I tried to focus on the

important luggage facts being shot at me, machine gun style, from Valentine’s perfect
mouth. She gushed over a $300 suit bag made by Travelpro, enumerating its many
features while frequently stressing its high-quality American manufacture. She
claimed it was the very best bag in the store, second only to a more expensive Italian
bag. I nodded appreciatively, not bothering to work out the math. (I’m telling you,
she was good.)
I did not have $700 to spend on Italian luggage, so my choice was pretty clear after
about five minutes. I couldn’t appear too anxious, though. If I my retreat were too
hasty, Valentine might get suspicious and give me the once-over. One well timed
glance at my nether regions and I was done for. I was also worried about looking like
a cheapskate, even in this time of crisis. I had to examine a few other options and
then, after careful consideration, choose the Travelpro not because I couldn’t afford
the other bags, which of course I couldn’t, but because I had the Financial Acumen
to recognize that it was a great buy.
I poked and prodded at the other bags, rubbing my chin thoughtfully whenever
I had a free hand that wasn’t busy Hiking or Covering. Valentine helpfully
demonstrated every zipper on every bag, and each zzzzzzzzzip! ratcheted up my
anxiety level. My pants kept slipping. The store, previously empty, was now nearly
full, as if a tour bus of luggage hungry sightseers had just pulled up. The time had
come to make my move. I waited for her to finish talking about the quality of
American ball bearings and then, with a thoughtful pause, I said, ‘‘You know, I think
I’m going [to] go ahead and get the Travelpro.’’ Her face lit up like a child’s on
Christmas morning. ‘‘Oh, zat is an excellent choice, sir!’’ she glowed, putting her
delicate hand on my arm. The feeling of warmth at her touch was tempered only
slightly by the thought of what might happen if she jostled me too forcefully.
‘‘Right zis vay sir, and ve’ll get you all taken care of.’’ She pulled my new bag up
onto its wheels, thrust the handle at me, and slinked over to the cash register, her
pants swishing as she walked. It was with horror that I realized my next predicament:
getting to the counter. My pants were looser than ever, and the motion of walking
would surely just speed their escape. My only chance was to walk quickly and use
inertia to keep them up. I made it in two fast strides, but I nearly tripped on the way.
Safely across the store, I could now lean on the counter to halt the downward
progress of my trousers. The trip had frightened me, though; I was clearly losing my
concentration. I had to get out of there fast.
Valentine gathered the paperwork needed to ring up my purchase. For some
reason, it seemed that everything she needed to complete our transaction was in
an extremely low drawer behind the register, which required her to keep bending
over suggestively. I tried not to look, but my leaning posture put me in a perfect
ogling position. Not wanting to move my legs and risk disaster, I turned my
head awkwardly to the right and gazed at a display of backpacks to reassure the
others in the store that I wasn’t staring. The smirk on the face of the other store
employee told me that I wasn’t going a very good job. (Why couldn’t he have been
my salesman?)
Finally, Valentine stood up and began to process my credit card. It was taking
forever. I prayed that my credit card wouldn’t be rejected. I began to sweat. Finally
the machine spit out a receipt, which Valentine offered to me, smiling. I signed it
quickly and turned to go, but she wasn’t through. She handed me a passel of items
and explained each one at length: my credit card, my receipts, the owner’s manual,
the warranty card. I patiently waited until she was through and then, with as
attractive a smile as I could muster, yanked everything out of her hand. Not even
bothering to put my credit card back into my wallet, I turned on my heel and hurried
out. ‘‘Zank you sir, and have vonderful holiday!’’ Her accent played in my ears as
I wheeled my trophy out of the store.
I stood on the sidewalk outside. It was 150 feet to the car. My hands would be full
during the journey, so I wouldn’t be able to make any pants adjustments on the way.
If they started to slip at all, they were going straight to the ground. I had a horrible
vision of myself standing in the middle of the brightly lit parking lot with my pants
around my ankles, surrounded by Valentine and the people from the tour bus, all of
them cackling and pointing. I shuddered. This was it. I performed one final Hike,
grabbed the suitcase handle, and made a break for it.
I took long strides, keeping my legs perpendicular to the ground for as long as
long as possible. I looked like a football player high-stepping through a series of tires.
I was sure Valentine was watching me; I could feel her cool blue eyes burrowing into
my back. The parking lot seemed to get longer with every step I took, like that
hallway in ‘‘The Shining.’’
I reached the car (out of breath!) and frantically pressed the unlock button on my
car remote (hurry up hurry up!) until the lights flashed and I heard the metallic snick!
of the lock. I tore open the trunk (my God, they’re slipping!), threw the suitcase
inside (faster, faster!) and tossed in the handful of papers Valentine had given me
(no time, just go!). My credit card clattered against the garment bag’s wheels.
I slammed the trunk closed (almost there!), whirled around, whipped open the
driver’s door, and dove for the safety of the cockpit. The darkness was like a tall,
cool glass of water. After a long drink I pressed my back against the seat, slid the
jeans up to my waist, zipped them up, and fastened the button. Rapture! I was never
so glad to be wearing pants in all my life. Happiness flooded over me. I wanted
to sing a song. I wanted to hug someone. I wanted to apologize to Tommy Hilfiger
and take back the nasty things I’d said about his questionable parentage. ‘‘I was
emotional! No hard feelings?’’
Most of all, I wanted to go back inside and talk to Valentine. I wanted to tell her
about the whole silly situation and explain why I had been acting so strangely. In the
end, she’d probably think it was cute. She’d laugh and put her hand on my arm
again. She’d find the story charming, like in the movies, and insist on buying me
a cup of coffee. But life, as I well know, is not the movies, and I was not about to risk
humiliation twice in one evening. I sped out of the parking lot and into the night, my
hips snugly ensconced in denim.
Epilogue: It turned out that the garment bag was too small, so I couldn’t even take
it on my trip. I wanted to return it, but I couldn’t bring myself to go back. The bag is
still sitting in my closet. I haven’t worn those jeans since that night, and I haven’t
been back to the store.
Appendix C
‘‘Long coat, no knickers y,’’ by Pollee, posted August 24, 2001.
Well I wouldn’t wear a fur coat — not a real one anyway — so the title had to
reflect that!!
I have discovered a lovely second hand ladies wear shop in Beachamp [sic] Place
near Harrods in Knightsbrisge [sic] which has designer wear at vastly reduced prices.
I bought a Valentino short leather skirt (and the leather is baby soft, ok I know I said
I wouldn’t wear fur — so I’m inconsistent!) which would cost d600 new for d110 and
it looks really as new — you would never have guessed it’s second hand.
Anyway — onto the op — at this lovely shop I spoed [English for spied] a
beautiful cream coloured cashmere coat — Versace no less for 150 pounds. Sold to
the lady slobbering in the corner! The feel of the cashmere was sooo soft and it was
lined with a beautiful silver embossed lining that felt beautiful against the skin and
epitomized luxury all the way. No way would I be able to afford one of these at the
real price of nearly d1,000.
The fit of the coat was wonderful and I felt like I was one of those film stars
sweeping into the room wearing the most wonderful outfit. Versace is available at so
many stores now, almost always tied up by anti theft devices so this was the first time
I had ever been able to try Versace on.
There is a great Versace store at Biscester which does have cheaper prices —
worth a look! This coat was less than 6 months old and had been sent in by a lady
who gets her clothes for nothing and wears them once or twice and then sends them
in for resale.
Having worn Versace I would definitely look to buy something else from their
‘‘stable’’ as the finish on this coat is the best. It has been hand stitched and reeks of
quality. Proud of my purchase and having just been into Rigby and Pellar for my six
month refit of lingerie I started my journey home wanting to show off my purchases.
On speaking to the man of the moment I found he was at home not doing a great
deal and, as he was on his own, I thought I could pay him a visit. We agreed I would
get a takeaway on the way back.
My mind wandered as I negotiated the motorway and I was so eager to show off
my new ‘‘buys’’ that I cooked up a surprise. I have always wanted to do this and this
was the perfect opportunity. I picked up the take away and then pulled into the local
garage to use the loo.
It didn’t take long to change into the lingerie and on a whim I didn’t redress but
just put on my coat. Luckily I had been wearing a skirt in any case so had the
required stockings to complete the look :-)
I stuffed my bags with my remaining clothes and left the garage trying not to look
too conspicuous! No one, of course, paid any attention to me at all but I really
thought they knew what I was up to.
Now the food was getting colder the hotter I became and I really wasn’t
concentrating too well on my driving. As a precaution (clever girl — I’m learning)
I thought I would ring my man to check he was still alone, after all I didn’t want to
make a fool of myself. Yes he was alone and where was his dinner! Well I thought y
I hope to be giving you some food for thought in a few minutes :-)
As I rung off down I caught sight of a blue light behind me y the local
constabulary wanted to discuss me using a mobile phone while driving y well they
don’t have a great deal to do in this rural area! I duly pulled over and sat in the car
having wound the window a little — after all you never know if they’re real. They
were real and asked me to get out of the car.
Now I don’t know if you have ever stolen a frozen chicken by stuffing it up your
skirt and trying to walk away casually, but if you have you will be able to visualize
how I got out of the car y trying the hardest not to reveal what was (or wasn’t)
underneath my coat.
Versace did not let me down. The coat, all encompassing, only using two of the
buttons provided kept my chastity covered. This coat has quality and after several
dry cleanings because of the colour still looks as good as new!
The problem I had was that the evening was very warm and they asked why I was
wearing a coat when it was so hot. The fear on my face convinced them that I was
smuggling drugs or something and their manner was less convivial. Quick as a flash
I told them I was going to a fancy dress vicars and tarts party. [Refers to a party
where the theme requires guests to arrive dressed up as either a vicar or a prostitute.
It is a very English tradition.]
‘‘Let’s have a look’’ said the youngest one, who was the one if I had a choice
I would have accepted a drink from! I pulled up my coat to show the top of a leg and
told them that was all they were getting! They laughed and we all went our separate
ways y thank goodness.
Five or six minutes later I arrived at my man’s house, ‘‘nice coat’’ he said as
I wafted in like a film star, ‘‘did you remember the poppadoms?’’ [Indian crisp
bread served at curry houses in England often as an accompaniment to the main
course.]
As I tried to kiss him he shifted me out of the way as the football was on.
I resorted to standing in front of the TV and slowly unbuttoning my two buttons y
Luckily he has a microwave, and only the Nan bread went soft :-)
Would I buy Versace again? Best coat I ever bought!
Chapter 4
Creating Visual Narrative Art

for Decoding Stories$
Synopsis
Chapter 4 shows and tells how to create visual art to achieve deep
understanding about stories that individuals tell. Creating visual narrative art
(VNA) of stories achieves several objectives. First, creating VNA revises and
deepens sense-making of the meaning of events in the story and what the
complete story implies about oneself and others. Second, creating VNA
surfaces unconscious thinking of the protagonist and other actors in the story
as well as the storyteller (recognizing that in many presentations of stories an
actor in the story is also the storyteller); unconscious thinking in stories relating
to consumer and brand experiences reflect one or more archetype (Jung 1916/
1959) fulfillments by the protagonist and the storyteller; given that almost all
authors agree on a distinction between processes that are unconscious, rapid,
automatic, and high capacity, (System 1 processing) and those that are
conscious, slow, and deliberative (System 2 processing, see Evans, 2008), VNA
enables and enriches processing particularly relating to system 1 processing–
enabling more emotional versus rational processing. Third, creating VNA of
stories is inherently and uniquely fulfilling/ pleasurable/healing for the artist;
using visual media allows artists to express emotions of the protagonist and/or
audience member, to vent anger, or report bliss about events and outcomes that
words alone cannot communicate; VNA provides a tangible, emotional, and
holistic (gestalt) experience that is uniquely satisfying and does so in a form
that many audience members enjoy over and over again. Chapter 4 elaborates
on the rationales for its central proposition, briefly reviews relevant literature
on VNA, and illustrates one mode of VNA for the complementary stories told
by a consumer and brand.
$
This chapter is co-authored with Arch G. Woodside, Boston College, and Carol Megehee, Coastal
Carolina University.
Introduction to the Literature and Theory of Visual Narrative Art
Visual narrative art (VNA) includes scenes and/or all the acts/episodes in a story
using one or more illustrations via paintings, sculpture, photographs, physical
movements, film, or other media beyond verbal reporting that creates a picture in the
mind relating to events involving symbols, people, animals, and other objects within
contexts relevant to action in the story. Including physical displays of affection and
aggression, VNA is the oldest storytelling medium with cave paintings of animals and
depictions of humans hunting animals in Europe dating back more than 30,000
years. In more recent times (e.g., since the birth of Siddhartha Gautama (Buddha) in
485 BCE to the present), VNA remains the major medium for storytelling especially
in societies with high levels of illiteracy.
This chapter responds to Felman and Laub (1992, p. xv) research call for
‘‘textualization of the context.’’ ‘‘The basic and legitimate demand for contextualiza-
tion of the text [e.g., the place, time, and prequel events in a story] itself needs to be
complemented, simultaneously, by the less familiar and yet necessary work of
textualization of the text.’’ Research on VNA builds from dual processing accounts
of reasoning, emotions, judgment, and social recognition. Dual processing research
recognizes two different modes of mental processing. The following labels summarize
these different modes: System 1 and System 2 processes (Kahneman & Frederick,
2002; Stanovich & West, 1998). Dual processing theory includes theory and empirical
evidence supporting the following attributes of System 1 processes: unconscious
thinking, holistic, evolutionary old, associative and parallel, shared with animals,
domain-specific, independent of general intelligence, and independent of working
memory (Evans, 2008). Theory and empirical evidence support the following
attributes of System 2 processes: conscious thinking, analytic, evolutionary new, rule-
based, uniquely human, domain-general, linked to general intelligence, and limited
by working memory capacity (Evans, 2008). The present chapter suggests and
demonstrates the usefulness for researchers to work (create and collect dual
processing data) and report in dual media to reflect dual processing: creating VNA as
well as verbal reports of concepts, processes, outcomes, and interpretations.
Dual processing theory and research, the literature on visual narrative art, and
storytelling theory and practice literature support the propositions and case study that
this chapter reports. However, while this chapter connects these three literature streams,
the presentation here does not include extensive reviews of the three literature streams.
VNA was quite in vogue during the Buddha’s lifetime especially in the form of
pata-chitra — a cloth-banner serializing a story and revealing a moral — now a
traditional tool of knowing and stimulating a mind to know (Jain & Daljeet, 2006).
Nigel Saul and colleagues’ (1997) The Oxford Illustrated History of Medieval England
provides dozens of photographs of sculptures, paintings, and architecture that
confirm their point, ‘‘History has often taken shape in the telling of tales’’ (Nelson,
1977, p. 25). Spiegleman’s Maus I (1986) and Maus II (1991) present a compelling,
highly emotional, VNA portrayal of the Holocaust that portrays Jews as mice and
Germans as cats. ‘‘It tells an absorbing story, a moving portrait of a flawed family.
It is also complex aesthetically and politically in ways specific to comics’’ (Chute,
Creating Visual Narrative Art for Decoding Stories 87
2008, p. 456). (In his review of graphic narratives and nonfiction comics, Chute
(2008) reports on Spiegleman’s public and successful fight with the New York Times
to get his Maus books moved from the fiction to the nonfiction best-seller list.)
Chapter 4’s focus is to distinct from earlier reports relating to VNA in consumer
culture research that report on decoding themes appearing in fiction and nonfiction
comics (e.g., Barcus, 1961; Berger, 1971; Belk, 1987; Chute, 2008; Ewert, 2000;
Kassarjian, 1983, 1984; Spiggle, 1985, 1986). The focus here is on advancing a VNA
template for the case study researcher to create visualizations of stories relating to
text and non-text materials as a tool to assist in decoding stories into both
unconscious and conscious meanings. The use of ‘‘researcher’’ here is inclusive of
professional social scientists as well as the human protagonist and other actors
engaging in self-reflection (e.g., Spiegleman, 1991; and subjective and confirmatory
personal introspections, see Gould, 1991; Woodside, 2004a, 2006) using media that
include going beyond verbal reporting.
Following this introduction to VNA and VNA research, the second section
describes the central objective and presents rationales for creating VNA in consumer
research. The third section three presents a general template for creating VNA. The
fourth section four applies the creative process and shows a VNA for a consumer
self-report on shopping, buying, and using a luxury brand of clothing (a Versace
woman’s coat); building on Fournier (1998) and Chapter 3, the resulting VNA also
depicts what a seemingly innate object, a brand, says and does in interacting with the
consumer. The fifth section five decodes the VNA of the story involving the human
and brand protagonists to enable the VNA viewer to achieve understanding of both
conscious and unconscious meanings of the whole story from its parts. The sixth
section six includes conclusions, implications for learning in consumer research using
VNA, limitations, and suggestions for further research.
Central Objective and Rationales for Creating

VNA in Consumer Research
The central objective here is to illustrate the creation of VNA in contextualizing text so
as to explicate the dual processing that occurs in the meanings in stories. The chapter
probes the proposition that incorporating/creating visual media parallel with analyzing
written text assists particularly in surfacing System 1 (unconscious) processing and
consequently serves to deepen and clarify meaning in a story’s scenes and its gestalt.
Rationales for creating VNA include the general finding that unconscious thinking
is the dominant mode for most mental processing and such processing is unavailable
or only available within narrow limits of meaning and detail for conscious thinking
and communicating verbally (for reviews, see Wegner, 2002; Wilson, 2002). Creating
VNA to parallel verbal (oral and written) narrative nurtures dual processing to a
greater extent and in ways that verbal narratives alone do not. Research using true
experiments supports the proposition that nonverbal (e.g., image, music) processing
involves different psychological mechanisms than verbal processing and that
nonverbal versus verbal cues elicit more imagery (visual associations with action and
motion), thereby suggesting that such cues may be particularly useful for
investigations of feeling/emotional advertising (e.g., & Punj, 1998).
Second, relationships involving communication exchanges between consumers
and brands (Fournier, 1998) provide the means for awakening and fulfilling
collective unconscious archetypal forms that VNA assists in surfacing. VNA in
conjunction with verbal narrative art permits, recognizes, and insists on uncon-
scious–conscious dual processing interpreting during the VNA creation process and
by most audience members when art uses both media (e.g., film and comics).
Third, ‘‘Thoughts typically occur as nonverbal images even though they are often
expressed verbally. Thus the way in which thoughts occur may be very different from
the way in which they are communicated’’ (Zaltman & Coulter, 1995, p. 37). Using
historical analysis of research in the natural sciences, learning research, and
technology, Ferguson (1977, p. 827) provides substantial evidence confirming that,
‘‘‘Thinking with pictures’ is an essential strand in the intellectual history of
technological development.’’
Fourth, ‘‘Human memory is story-based’’ (Shank 1999, p. 12). Story-based,
episodic memories provide multiple indices for retrieval and interpretation and
therefore facilitate learning (Shank, 1990, 1999). Telling stories and repeating these
stories often brings clarity (Weick, 1995). Retrieving and repeating stories provides a
catharsis — an emotional cleansing that relieves tension by bringing unconscious or
repressed memories to consciousness (Woodside, Sood, & Miller, 2008). Specifically,
storytelling allows the individual to experience one or more archetypal myths.
An archetype is an unconscious primary form in the collective human subconscious
that represents an idea. An archetype is a prototype that is not learned but is with us
from birth (Jung, 1916/1959; Wertime, 2002).
The title and movie, Rebel without a Cause (1955), reflect a central proposition of
archetypal theory: an archetypal form stimulates and directs a collective unconscious
form of processing that a story encapsulates. Reflecting such unconscious processing,
in the movie the protagonist, Jim Stark (a teenager played by James Dean) is asked,
‘‘What’s wrong?’’ He responds, ‘‘I don’t know. Leave me alone.’’ The movie tells the
tale of a teenager feeling betrayed by his bickering parents and parental control of his
actions.
Without referring to Jung’s archetypal theory, Holt (2003) proposes that
marketers can be highly successful when their brands enact rebellion.
More often than not, in America at least, those who win in myth markets
are performing a myth of rebelling; the most successful icons rely on an
intimate and credible relationship with a rebel world: Nike with the
African-American ghetto, Harley with outlaw bikers, Volkswagen
with bohemian artists, Apple with cyberpunks, Mountain Dew with
slackers — protagonists who would rather pursue quixotic activities than
‘‘grow up’’ and get serious about careers. (Holt, 2003, p. 43)
Table 1 summarizes the rebel (anti-hero) and eleven additional archetypal forms,
their stories, and examples of brands showing and telling productions of these stories.
Table 1: Archetypes, story gists, and brand examples.
Ultimate strength When an obstacle is there, it must be Timex — ‘‘It takes a licking and
overcome, strength must be keeps ticking’’
proven in use
The Siren Power of attraction, linked with the Allure by Chanel; Envy by Gucci
possibility of destruction
The hero Fortitude, courage, and victory; a Michael Jordan and Nike shoes; Joe
journey and transformation DiMaggio and Mr. Coffee; Power
Puff Girls; Forrest Gump
The anti-hero Universal message of destruction Heavy metal icons; Howard Stern;
and attraction of evil; the bad Jerry Springer; Oakland Raiders;
dude Che Guevara; Harley-Davidson
The creator Creative inspiration and the potency Coca Cola — the real thing; Walt
of imagination; originality; Disney; Kleenex
authentic
Creating Visual Narrative Art for Decoding Stories
The change master Transformation, self-improvement, Curves — workout stores for

and self-mastery women; Gillette’s Mach 2 Razor;
89
Porsche 911
90
Table 1: (Continued )
The powerbroker Authority, influence and CNN; E.F. Hutton; Bill Gates;
domination — the world’s Microsoft
leading y; the best y; number
one
The wise old man Experience, advice, and heritage; Levi’s; Obi-Wan Kenobi
staying the test of time
The loyalist Trust, loyalty, and reassurance CocaCola and ‘‘Mean’’ Joe Green
with boy of 12 TV commercial;
I Love Lucy; Friends TV sitcom
The mother of Purity, nourishment, and motherly Just Juice; Ivory Soap; Tropicana
goodness warmth Orange Juice; Aunt Jemima; Fairy
Godmother; Witch of the East;
Snow White
The little trickster Humor, non-conformity, and the Dennis the Menace; Bart Simpson;
element of surprise Pee-Wee’s Big Adventure;
SpongeBob SquarePants
The enigma Mystery, suspense, and uncertainty Zorro; Abercrombie and Fitch;
Star Trek
Source: Developed in part from several chapters in Wertime (2002).

Chapter 3 develops the central proposition that Table 1 implies: a mostly

unconscious process creates the kernel of a consumer–brand relationship and the
consumer’s desire for archetype fulfillment drives her or him to act — to undertake a
storytelling journey.
Today, people often share their stories by way of blogs on the Internet. Blogs are a
rapidly growing form of information that are open to the public or to small select
groups (Kluth, 2006). Sharing stories via blog entries is pleasurable because both
authors and readers can relive experiences, vent anger, share happy times, and
vicariously experience the role of protagonist in the story (Shindler & Holbrook,
2003). In addition, storytelling permits archetype fulfillment and ‘‘revises and
deepens sense making of the meaning of events in the story and what the complete
story implies about oneself and others’’ (Woodside et al., 2008, p. 100).
All great storytellers deal with the fundamental conflict between subjective
expectations and cruel reality (McKee, 2003). Both storyteller and listener should
approach the story with skepticism. The skeptic knows that the real truth lies below
the surface story in the unconscious and unexpressed (Wilson, 2002). A story
typically begins when the protagonist’s life is in balance but some triggering event
comes along and throws him out of balance. The heart and soul of the story, the
journey, is the protagonist’s struggle to restore balance or achieve a new balance,
describing what it is like to deal with opposing forces, digging deeply, taking risks,
facing ugly truths, but finally achieving a catharsis — emotional fulfillment and
cognitive understanding about himself and other important people in his life.
Consumer–Brand Storytelling Master Template
Storytelling theory and practice literature just reviewed forms the basis of the phase
dynamics theory (PDT) of epiphany travel which can be used as a master template
for analyzing consumer self-narratives (Woodside & Megehee, 2009). PDT consists
of five stages or phases in the protagonist’s cognitive and emotional preparedness to
more fully answer the question, ‘‘Who and I?’’ Triggers, or motivating states and
experiences, provide segues between phases in preparing for starting a story (Prequel
and Awakening), engaging in the story’s action (Journey and Catharsis), and later re-
experiencing (Post-Journey Storytelling and (Re) Interpretations) a journey that
leads to personal enlightenment. Throughout the process and in the retelling of the
experience, the protagonist fulfills one or more archetypes. Figure 1 illustrates the
basic PDT model.
Figure 1 shows a sequential polyscenic mode of visual narration whereby the
narrator opts to separate each of several scenes in a story using some dividing
motif — an architectural structure such as walls or trees or the use of distinct panels
such as a sequence of four to eight panel episodes by creators of many comics. The
sequential polyscenic mode is one of several possible modes of VNA. Jain and
Daljeet (2006) present examples of Buddhist art (sculpture and pata-chitra) to
illustrate sequential polyscenic and three additional modes of VNA. A continuous
mode is the presentation of a whole chain of events without a frame separating one
92
Prequel Journey
Awakening Catharsis Post-Journey
Storytelling
T1 and (Re)
T2 6 Interpretations
1 T3 T4
8
Emotional and
On-going life Recognizing cognitive
experiences that a journey is understanding/
may include unlikely to be fulfillment and
pre-, or semi- necessary; understanding
conscious feelings 2 5 about oneself
completing
that something is planning and possibly
unknown or steps to start 4 other persons Telling of
missing about the journey; important parables and
oneself; consciously in one’s life; experiencing
unconsciously asking, “Who 3 deeper knowledge “proper
asking, “Who am am I?” 7 of who I am. pleasure”
I?” by reliving
events
Key: Ti = trigger—for example, T1 is a communication, event, that provides/surfaces conscious thinking about and outcomes
(1) problems/opportunities and (2) people and places necessary to reach to enable the achievement of desirable end states in the journey;
storytelling
dynamics
= a world (e.g., landslide) or personal (e.g., lack of knowledge or skills) block often that is unexpected that occurs during the journey
= help from a person—often a person on the scene that provides advice and assistance; serves to help craft solutions and means around world blocks
= delays, breakdowns
= physical locations = feelings
of failure
= weather or environmental conditions causing a delay or setback in the journey = antagonist
= good times/moments occurring during the journey

= feelings
= rope signifies help in the form of an animal, product or natural object of success
or = protagonist
(other archetype
May be a protagonist;
see Table 1
Figure 1: Phase dynamics theory of epiphany travel.

Figure 2: Example of monoscenic storytelling. Note: How many persons are in this
scene? Who are they? Which person is talking? What is the person saying? Are all
persons happy, sad, scared, good, wicked? What is going to happen next? Can you
create a story with a climax and a conclusion from this scene.
event from another. A synoptic mode presents multiple interpretations of the same or
similar stories with the various episodes representing synopses of a theme without
regard to temporal sequence. A conflated mode of a polyscenic narration portrays a
long series of events stretching over an expanded and well defined geography though
lacking in chronological order (Jain & Daljeet, 2006).
A monoscenic mode presents either the ‘‘seed’’ out of which grows an entire story
(Jain & Daljeet, 2006) or a vivid scene that captures the climax or gist of the story.
Figure 2 presents a story in the monoscenic mode that might remind you of the
Grimm’s fairly tale of Hansel and Gretel meeting an evil old witch. The illustration
depicts two protagonists and a tall bent-over person in a ragged dress and carrying a
crooked walking stick; the scene might automatically bring to mind the story of an
evil witch who plans on eating the two children. VNA via literary illustrations,
paintings, and other media (e.g., seeing–reading comics, attending theatre produc-
tions, watching professional wrestling, and other sporting events) provides a tangible,
emotional, and holistic (gestalt) experience that is uniquely satisfying and does so in a
form that many audience members enjoy over and over again (see Schindler &
Holbrook, 2003).
Applying the Master Template in Creating Visual Narrative Art
This section presents a VNA without (Figure 3) and with contextual text notes
(Figure 4) for Pollee’s self-reflective account of her shopping for clothing where she
94
Medusa
£ 1000 £ 150
BRAND:
VERSACE
Knightsbridge
Figure 3: Pollee shopping, buying, and using Versace cashmere coat and lingerie.
3
Medusa
6
BRAND: A
£ 1000 £ 150
VERSACE 2
4 5
Textualizing the Contexts

1. Pollee shopping in store at Beauchamp Place and
1 spots a Versace coat on sale for very low price that
she is able to try-on because no anti-theft device 8
is attached to the coat.
2. After buying coat, Pollee goes to Rigby & Peller
and buys luxury-sexy lingerie.
3. Pollee calls boyfriend and buys takeaway food to
take to his place. 9
4. Pollee stops at loo and transforms in to a Siren by…
5. Taking-off dress and wearing only coat with lingerie
to surprise boyfriend.
6. Pollee talks to boyfriend on cell phone.
7. Police stop Pollee for talking on cell while driving.
8. Pollee explains wearing coat in summer by telling
Knightsbridge officer that she is traveling to a vicars and tarts party.
9. Pollee arrives home to see her boyfriend watching
football on TV; takes off coat, shows him her
transformation; they embrace and have sex.
Figure 4: Textualizing the visual contexts of Pollee’s shopping, buying, and using Versace cashmere coat/lingerie.
95
spies, buys, and wears a recycled Versace coat. The section first reports a PDT textual
decoding of Pollee’s story. The section also shows a VNA creation of the story along
with a decoding of this VNA.
Analyzing a Consumer Self-Narrative from a Blog
‘‘Long coat, no knickers y,’’ a blog by Pollee, posted August 24, 2001, illustrates the
master template for analyzing consumer self-narratives. Pollee’s full self-report
appears in Appendix C in Chapter 3. As the following text and visual decoding of the
story reveal, Pollee’s self-report exemplifies a story in which the two protagonists —
Pollee and Versace interact with each other and supporting actors in the story to
fulfill the Siren archetype.
Prequel
In the prequel phase, the protagonist, in her on-going life, may experience
unconscious, preconscious, or semiconscious feelings that something is unknown
to him or missing in her life. From time to time the protagonist may ask herself,
‘‘Who am I? What is my purpose? Why am I here?’’ But until the meta-conscience is
aware that these questions be answerable, they remain below the level of active
conscience. This void may lie dormant in the person for years unless and until a
tipping point (Gladwell, 2002) — a small change with a big, immediate effect — is
reached. In this model, the tipping point is represented as a trigger, or the catalyst for
change.
Pollee, the consumer protagonist in ‘‘Long coat, no knickers y,’’ reveals at the
beginning of the story that she likes shopping at a second-hand ladies store in
Knightsbridge. Even though she would never wear real fur, she confesses to buying a
short leather Valentino skirt. Pollee even admits that she is inconsistent in the fur/
leather principle. Although she tells us that she cannot afford designer clothes at new
retail prices, what she really reveals about herself without actually saying so is that
designer labels are important to her.
Prequel-Awakening Trigger (T1)
The Prequel-Awakening Trigger (T1) is a communication or event that surfaces to

provide conscious thinking about (1) problems and/or opportunities and (2) people and
places necessary to reach to enable the achievement of desirable end states. This trigger
moves the consumer protagonist from a low-level, free-floating idea of a yet undefined
need to be fulfilled to recognition that the need is one of self-discovery or self-definition.
Information search — at least internal search — is triggered at this point.
‘‘At this lovely shop I spoed a beautiful cream coloured cashmere coat — Versace
no less for 150 pounds. Sold to the lady slobbering in the corner!’ And so begins
Pollee’s great adventure! In the early scenes in Pollee’s self-report, she says nothing
explicitly that responds to communications by Versace, the brand protagonist.
Versace says very little verbally but its visual communications to Pollee and others
appears clear. Figure 3 includes visuals of the irresistible beauty and charms of Sirens
as the archetypal myth capturing the (initially) unconscious thinking of Pollee and
the visual communications of Versace. See the Versace photograph (monoscenic
story gist) appearing on top of the head of the woman portraying Pollee in Figure 3
(top left) and the following interpretations relating to Versace and Versace’s logo.
We’ve been conditioned to be at least titillated if not outright shocked

by Versace. Garish, a little trashy, but always excellently constructed,
Versace has been the label to go to for purely sexy luxury. It’s naughty:
Medusa is the brand’s logo, after all. (WSJ, 2008, p. 1)
Medusa represents a dark force in the brand protagonist’s story. McKee (2003)
emphasizes that creating a dark force makes for good stories. Inciting incidents and
dark forces ‘‘throw life out of balance.’’ The story goes on to describe how, in an
effort to restore balance or to achieve a new balance, the protagonist’s subjective
expectations crash into an uncooperative objective reality. ‘‘A good storyteller
describes what it’s like to deal with these opposing forces, calling on the protagonist
to dig deeper, work with scarce resources, make difficult decisions, take action
despite risks, and ultimately discover the truth.’’ (McKee, 2003, p. 52)
Medusa [in Greek mythology] was one of the Gorgons [three sisters],
and the only one who was mortal. Her gaze could turn whoever she
looked upon to stone. There is a particular myth in which Medusa was
originally a beautiful maiden. She desecrated Athena’s temple by lying
there with Poseidon. Outraged, Athena turned Medusa’s hair into
living snakes. (Lindemans, 2009)
Awakening
Upon further evaluation, the consumer protagonist recognizes that a journey might
be necessary to complete the missing or unknown aspects of his life or person. In the
awakening phase, the protagonist plans steps to begin the journey and consciously
asks, ‘‘Who am I?’’ Travel options (e.g., where to go and what to see) are examined
and begin to gel in this phase.
Pollee reveals all kinds of wonderful things about this coat as if she is trying to
justify the purchase and reduce any dissonance she has about spending her 150
pounds. The verbal motives for purchasing the coat include relative price (i.e., getting
a 1000 pound coat for 150), condition (i.e., less than 6 months old, worn only once or
twice), and quality (i.e., silver embossed lining, hand stitching). Her possibly initially
unconscious motives are to own a Versace (luxury designer) item and to transform
herself into a Siren.
Clothes with the Versace label are usually locked up with anti-theft devices
because they are so valuable. Such security devices have prevented Pollee from trying
on Versace, until now. Pollee speaks of the feel of the coat against her skin (‘‘sooo
soft’’), the reek of quality, and of feeling like a film star. In this cashmere Versace
coat, Pollee becomes the Siren.
Awakening-Journey Trigger (T2)
The Awakening-Journey Trigger (T2) is a communication or event that compels the

protagonist to finalize specific travel plans. She will start a quest and her travels will
be to a specific location at a specific time in the not-too-distant future. T2 leads the
protagonist to believe that the discrepancy between the actual state and the desired
state is large enough and important enough to seek additional information. External
information search, evaluation of alternatives, and selection and purchase of
products take place between awakening to the need for a journey and actually taking
the journey.
Pollee puts together the purchase of this sensual cashmere coat with her ‘‘six
month refit of lingerie.’’ The lingerie purchase occurs at Rigby & Peller — a luxury
store brand specializing in sexy lingerie. Now the real journey begins! Unconscious
thoughts surface in Pollee’s mind and she asks herself, ‘‘What is the best way to show
off my purchases? How can these products bring out my inner-Siren?’’
Journey
During the journey to enact the archetypal storyline, the consumer protagonist
experiences the ups and downs/perceived positives and negatives in various physical
locations punctuated by — often unexpected — world (e.g., landslide) or personal
(e.g., feelings-of-failure) blocks that occur during the journey, delays (including
weather or environmental conditions causing setbacks in the journey), help (often a
person on the scene that provides advice and assistance that serves to facilitate the
crafting of solutions and means around world blocks), and, of course memorable
good times/moments that occur during the journey. Although the story of an
epiphany journey can include aspects of pre- and post-journey experiences and
interpretations, the journey provides the preponderance of the details of the story.
Pollee calls her ‘‘man-of-the-moment’’ to make sure he was home alone and not
busy. She will pay him a visit and bring some take-out food. Once she is in her car and
on the highway she comes up with an idea to surprise and show off her purchases to
her man. Pollee picks up the food and then pulls into ‘‘the local garage to use the loo.’’
In no time, Pollee changes into her new lingerie, and ‘‘on a whim,’’ puts the coat
over the lingerie without getting redressed. Her stockings complete the look.
‘‘I stuffed my bags with my remaining clothes and left the garage trying not to look
too conspicuous! No one, of course, paid any attention to me at all but I really
thought they knew what I was up to.’’
As she drove, ‘‘the food was getting colder the hotter I became.’’ She wasn’t
concentrating on her driving, and as a precaution, she checked on her man again to
make sure he was still home alone. ‘‘Yes he was alone and where was his dinner! Well
I thought y I hope to be giving you some food for thought in a few minutes ;-)’’ No,
Pollee was concentrating on her destination and not on her driving!
Just as she hung up, Pollee ‘‘caught sight of a blue light behind me y the local
constabulary wanted to discuss me using a mobile phone while driving [illegal to do
in the United Kingdom] y well they don’t have a great deal to do in this rural area!’’
She pulled over and rolled down her window to be certain they were real policemen.
‘‘They were real and asked me to get out of the car.’’
‘‘Now I don’t know if you have ever stolen a frozen chicken by stuffing it up your
skirt and trying to walk away casually, but if you have you will be able to visualise
how I got out of the car y trying the hardest not to reveal what was (or wasn’t)
underneath my coat.’’ But the Versace coat did not let her down. The coat’s quality
was as good as new even after several dry cleanings. ‘‘Only using two of the buttons
provided kept my chastity covered.’’ Pollee’s going to have some fun with the
policemen!
How can she explain wearing the coat on such a warm evening? She hopes the
cops don’t read the fear on her face as her being up to something worse than just
dialing and driving. But staying in character, and ‘‘quick as a flash I told them I was
going to a fancy dress vicars and tarts party.’’ ‘‘‘Let’s have a look’ said the youngest
one, who was the one if I had a choice I would have accepted a drink from! [Note
that Pollee reveals that she would consider a new man-of-the-moment depending on
the context.] I pulled up my coat to show the top of a leg and told them that was all
they were getting! They laughed and we all went our separate ways y thank
goodness.’’ Not only did she not get a ticket, but she and the two policemen got a
good laugh out of the situation.
Journey-Catharsis Trigger (T3)
The Journey-Catharsis Trigger (T3) is the eureka! Ah-ha! Or, moment-of-truth.

A few minutes later, Pollee wafts into her man’s house like a film star and he says,
‘‘Nice coat y where’s the food?’’ She tries to kiss him, but he pushes her out of the
way, ‘‘As the football was on.’’ Pollee starts to engage in a striptease. ‘‘I resorted to
standing in front of the TV and slowly unbuttoning my two buttons y Luckily he
has a microwave, and only the Nan bread went soft ;-)’’
Catharsis
Sometime during, but also at the end of and following the journey, the protagonist
experiences an epiphany — a sudden intuitive realization of ‘‘the meaning of it all.’’
This follows not only a cognitive understanding of the experience, but a feeling of
‘‘cleansing’’ or renewal of self (catharsis). In this phase, emotional and cognitive
understanding of self — and possibly other people important in life — leads to

fulfillment and a deeper knowledge of, ‘‘Who I am.’’
The ‘‘new’’ Versace coat, or what’s underneath it, prevails over football on TV!
Proof to Pollee that she is a Siren! ‘‘Would I buy Versace again? Best coat I ever
bought!’’
Catharsis-Storytelling Trigger (T4)
The Catharsis-Storytelling Trigger (T4) provides the bridge between emotional

fulfillment and cognitive understanding and the retelling and reinterpretation of the
journey. What made Pollee, the protagonist, think of posting this to her blog and
sharing it with the rest of the world? Whatever her motivation, the blog provides
positive word-of-mouth for the Versace brand. (This is four years after Versace was
killed, so was it a scarcer item? Was this Donatella’s or Gianni’s design? Does it
matter?)
Post-Journey Storytelling and (Re)Interpretations
Post-journey storytelling allows the epiphany traveler to relive and reinterpret events
and outcomes in the journey. The story evolves to become the experience itself.
Through interpretation, the story may evolve to the form of parable — metaphoric,
symbolic stories with a lesson, moral, or message.
Why would the consumer protagonist want to share his story over and over again?
First, storytelling allows the consumer protagonist to experience ‘‘proper pleasure’’
(Aristotle in Hiltunen, 2002) by reliving events and outcomes in the journey (Butcher,
1961). Storytelling is inherently pleasurable; it allows the teller to be both protagonist
and audience, to vent anger or restore bliss, to relive the experience over and over
again. Second, storytelling allows the protagonist to fulfill a primal archetypal form.
Finally, storytelling revises and deepens sense making; repetitive storytelling clarifies
and sharpens the meaning of events and contributes to the definition of self and
others (Woodside et al., 2008).
How many of Pollee’s creative and humorous thoughts actually occur in their
correct place in the story, and how many did she create after the fact in the process of
storytelling? For example, did she actually think of what it would be like to steal a
frozen chicken when she was getting out of the car to face the policemen, or did she
think of it as an embellishment to her story while writing her blog? Perhaps, she
thought of it after the fact but before she wrote the blog and these funny ideas were
what stimulated her to post the story.
How much of the story was retroactive rationalization? The Siren archetype is a
thread that runs throughout the story as if it were scripted. Pollee stays in character
with the sensual designer clothes, the naughty lingerie, the flirtation with the cops
(including the vicars and tarts cover story), the food, and the strip show in front of
the TV. The ‘‘man of the moment’’ fills a stereotype of the boyfriend who is only
interested in food and football until the Siren successfully convinces him to be
interested in her (sex).
This blog has a life beyond the first posting or retelling. Blogs are highly social in
nature in that they invite commentary from others. Once information has been posted
to cyberspace it takes on a life of its own and it is virtually impossible to retract once
posted. Others will clarify, alter, and transfer information posted. Unbeknown to the
original author of this blog, her story lives on in academic research!
Archetypal Identification as Outcome
By purchasing the cashmere Versace coat, undergoing her adventure in her new coat,
and later retelling her story on in a blog, Pollee fulfilled her Siren archetype. In Greek
mythology, the Siren was one of a group of sea nymphs that would lure sailors to
their deaths on the rocks surrounding their island by sweetly singing their siren song.
In general, a siren is a beautiful, seductive woman — a temptress — who by power of
attraction, can lure men with her allure to their possible destruction. Not only was
Pollee able to avoid a traffic ticket by flirting with the policemen, but she was able to
lure her ‘‘man of the moment’’ away from the football game on TV!
Figure 3 is a VNA creation of the main and secondary story lines in ‘‘Long coat,
no knickers’’ — a blog posting by ‘‘Pollee.’’ The main point depicted in Figure 3 is
that Pollee, the consumer protagonist, fulfills a Siren archetype. Finding the deeply
discounted but beautiful and luxurious Versace coat and pairing it with (just) her new
lingerie allows Pollee to enact her inner temptress. Even the close encounter with the
cops illustrates the principle that all great storytellers deal with the fundamental
conflict between subjective expectation and (sometimes cruel) reality.
Much like Clark Kent using a telephone booth to transform from his cheap
business suit and thick horn-rimmed eyeglasses into the blue bodysuit with red cape
of Superman, Pollee uses a loo to transform from middle-aged woman shopper into
the Versace cashmere coat and Rigby & Peller lingerie of a Siren. She overcomes
world blocks (police stopping her journey and asking for an explanation for wearing
a winter coat in the summer and her boyfriend’s focus on the football game on TV)
with humor and creativity.
Textual Decoding via a Sequential Mode of VNA
Figure 4 includes textual decoding of the VNA for the Pollee–Versace story. Note
that Figure 4 replaces the continuous mode of narration appearing in Figure 3 with
a sequential mode of narration by using dotted triangles as a dividing motif.
The dotted triangles show the six physical locations that appear in the story.
The use of a dividing motif in Figure 4 helps to clarify the possible balances and
imbalances in relationships among stories that the brand, consumer, archetype, and
reported action tell. Figure 5 reconfigures and simplifies Figure 4 in consideration of

Heider’s balance theory (Heider, 1958; Woodside & Chebat, 2001). Woodside
and Chebat’s (2001) application of Heider’s balance theory to a Jewish couple buying
a German car illustrates a transformation from a psychologically balanced state to
an imbalanced state to a new balanced state. The unconscious automatic associa-
tion of Volkswagen with Nazi Germany prevented the couple from initially
considering the purchase of a Volkswagen Super Beatle. The couple went through
several iterations of conscious thinking to give themselves conscious permission to
buy the car.
Figure 5 shows a conjunction of four triangles: ABC, BCD, ABD, and ACD. Note
that Figure 5 shows balanced positive relationships for all four triangles. Considering
Heider’s balance theory and a close reading of Pollee’s self-report indicates that the
imbalance occurring is Pollee’s prior inability to enact her unconscious desire to
transform into a Siren. Building from Jung’s theory of archetypal unconscious
processing, double-headed arrow 1 in Figure 5 is the initiating driving force awaiting
activation to permit Pollee to experience triangle BCD.
Versace uses a somewhat shocking, monoscenic, visual story at node A in Figure 5
to associate with the Siren archetype. Versace’s visual communication to Pollee is
visceral, emotional, and instinctual — a factual communication would be counter-
productive. Versace’s visual association in triangle ABC suggests that Versace and
Pollee attach kindred spirit forces representing a combined undeniable force. Putting
on the Versace coat has a dramatic impact similar to Clark Kent’s donning his
Superman costume. Pollee and Versace combine to enact the story that Pollee reports:
ABD in Pollee’s mind becomes Pollee’s ACD lived experience. Such an analysis is
extendable to other archetypes and customer–brand stories (cf., Holt, 2003).
A visual imbalance does appear in Figure 5. Note that in the illustration at node
A the central model is wearing clothes while the other four models appear without
clothes on a beach — an imbalance (paradox, riddle, conflict). Only the central
model is looking at the viewer and she appears ready to say, ‘‘Well, are you coming
or not?’’ One interpretation is that the central model represents the consumer
transported into Medusa to a place where beautiful immortals gather. What serves as
the central model’s transporter? A Versace black, leather, bare-back pants-suit serves
to transport the central model into Medusa to this place where she can associate with
beautiful, naked, young immortals. The central model’s facial expression, facial
makeup, and hair all express her archetypal transformation.
Conclusions, Implications, Limitations, and Suggestions

for Further Research
Alison Krauss and Union Station’s song lyrics (2008), ‘‘When You Say Nothing at
All’’ illustrate the wisdom of the title to Zaltman and Coulter’s (1995) article, ‘‘Seeing
B
Archetype:
Siren
2 1
Consumer, Pollee, with unconscious/
Brand, visual + C conscious desire to enact archetype
A
+
message, and
monoscenic story
portrayal
3 +
BRAND:
Medusa VERSACE
+ 4 + +
5 6
D
Story enactment, gist, that
follows Siren plot and consumer-
brand unconscious/conscious
interactions
Figure 5: Brand and consumer interacting in storytelling production of Siren archetype. Source: Original visual structure that
follows from Figure 3 template in Woodside et al. (2008).
103
the Voice of the Customer: Metaphor-Based Advertising Research.’’ Here is the

refrain in Krauss’ song:
The smile on your face lets me know that you need me
There’s a truth in your eyes sayin’ you’ll never leave me
The touch of your hand says you’ll catch me if ever I fall
You say it best when you say nothing at all. (Krauss & Union Station,
2008)
The present chapter reviews relevant literature that supports the proposition that
most communication is nonverbal, consumers think in stories via pictures mostly —
not in words, and dual processing theory and research indicate that the oldest
thinking process (System 1) has attributes that associate with the archetypal
collective unconscious: automatic, implicit, holistic, evolutionary old, emotional,
low-effort, visual-contextualized, and primal (shared with animals). Nonverbal
messages appear to enhance effectiveness in ways that are not achievable with words.
Consumers and brands often talk to each other because such a form of expression
represents logical, rationale, sane behavior — System 2 processing. System 2
processing is explicit, controlled, evolutionarily recent, textual, abstract (low
context), uniquely human, rule-based, analytical, and often down-right boring,
ignorable, and unconvincing.
The theory and research in this chapter supports a complementary proposition to
Weick’s (1995, p. 12) famous System 2-related, sense-making suggestion that we do
not know what we think until we hear what we say. The System 1-related
complement is that we do not know what we feel (desire, love, hate) until we visualize
what we (unconsciously) see. Because most thinking is unconscious, to learn what we
are really thinking Wilson (2002) informs that we (researcher and informant) need to
focus on our observable behavior — what we actually do and not what we say we
believe in. Creating and interpreting VNA surfaces substantially important
unconscious thinking/feeling-behavior associations.
Creating VNA requires the researcher and/or the informant to work in an
evolutionarily old and primal mode of communication. Such a medium is useful
because ‘‘all consumers have relevant hidden thoughts: ideas they are not aware of
possessing but are willing to share once discovered’’ (Zaltman & Coulter, 1995,
p. 39). Zaltman and Coulter (1995) emphasize the possibility of accessing these
hidden or deep structures using a variety of techniques such as those used in art
therapy, and especially phototherapy, can be very effective in surfacing such
thoughts (Weiser, 1988; Ziller, 1990). What a person notices in a picture will always
mirror the inner map that she or he is unconsciously using to organize and
understand what the senses perceive (Weiser, 1988).
Implications that follow from VNA-related theory and research include the view
that creating picture collages of stories consumers and brands tell are worthwhile
because they build from theory and empirical evidence that dual processing accounts
for human behavior. VNA relates directly to the dominating influence of
unconscious (System 1) processing while the dominant research method of asking
questions does not and assumes that only conscious (System 2) processing is relevant.
Another implication is that creating VNA may be useful in combination with
consumer verbal storytelling. The consumer acting as a storytelling informant can be
assisted in interviews in using the Internet to collect drawings and photographs,
organize the collection into a story, and describe the story verbally that unfolds
visually in the resulting VNA. Such a story-visualizing and storytelling conjunctive
approach represents one type of forced metaphor elicitation technique (for another
FMET approach, see Woodside, 2008a, 2008b, 2008c).
One of several limitations to creating VNA as a research method is that such
creations usually require more than one session to create a complete VNA
representation. Creating an insightful VNA likely requires a few days to a week or
longer to permit surfacing of unconscious thoughts as the collage takes-on forms and
shape. Using naturally occurring individual self-reports via storytelling with or
without photographs (e.g., in blogs), researchers can expect to review such case data
several times over several days in creating a VNA to represent the story. Issues of
validity, reliability, and generalizability are relevant for research using VNA.
Following some amount of training in creating VNA, validity and

reliability can be probed by asking an informant to create VNA of
their own verbal story separate from a researcher’s own VNA creation
of the same story. The resulting emic and etic visualizations can be
compared for similarity, differences, completeness, and accuracy.
Some meaningful degree of generalizability is achievable by selecting
samples of informants based on theoretical profiles (e.g., middle-aged,
heterosexual females currently involved with a man-of-the-moment)
for VNA research. While not referring directly to VNA, McCracken
(1988) suggests that five interviews per profile provides practical
generalizability if the results from such anthropological data are highly
similar. The VNA in this report may be relevant only for Pollee and
Versace; the study does not include independent confirmation of the
accuracy and usefulness of the VNA or its generalizability — certainly
another limitation that future research should address.
Chapter 5
Subjective and Confirmatory Personal

Introspection
Synopsis
Research findings support the view that a multiple methods approach is

necessary to surface the substantial amount of relevant thinking processes that
occur both consciously and unconsciously within different phases of consumer
decision making. Chapter 5 advocates viewing all studies that ask informants
questions as representative of researcher–informant introspections. Because
answers to questions differ substantially depending on how the questions are
framed, applying multiple, explicit, question frames to acquire conscious and
unconscious thoughts in researcher–informant introspections is helpful. This
chapter reviews multiple methods, including metaphor elicitation of uncon-
scious thinking, useful for achieving and confirming thick descriptions of
conscious and unconscious thinking associated with informants’ deep-seated
beliefs and observable actions.
Introduction
A substantial body of social and case study research now supports a ‘‘creative
destructive’’ (Loewenstein, 2001) view of consumer decision making (see Bargh,
2002; Zaltman, 2003). Such a view proposes that in actuality individuals have far less
access to their own mental activities than survey researchers give them credit for.
A stream of studies supports the conclusion that informants are able to only partially
retrieve and report the reasons for their actions (Ericsson & Simon, 1993; Woodside,
2004b; Woodside & Wilson, 2003). ‘‘Ninety-five percent of thinking takes place in
our unconscious minds — that wonderful, if messy, stew of memories, emotions,
thoughts, and other cognitive processes we’re not aware of or that we can’t
articulate’’ (Zaltman, 2003, p. 9).
Consequently, this chapter offers a workbench model of informant’s thinking
related to interpreting and answering questions that the informant asks him- or
herself or is asked by another researcher (e.g., another person addressing a question
to an informant). The chapter suggests the use of a combination of introspective
conscious and unconscious thought retrieval elicitation techniques to achieve three

objectives: (a) confirming both the belief and evaluations held consciously and
unconsciously by the informant, (b) confirming the existence of experiences and
related outcomes as described by the informant, and (c) achieving a deep under-
standing of how consumers become aware of their own desires (see Belk, Güliz, &
Askegaard, 2003) that affect their search behavior, purchases, and use of products
and services.
Figure 1 portrays multiple issues related to thinking about an issue raised by a
researcher — regardless of whether or not the researcher is the same or a person
different from an informant. Table 1 summarizes some of the research concerns and
findings that relate to the first nine issues in Figure 1. Given that the dominant logic
in consumer research continues to rely on asking questions and that attempts at
answering questions require introspection (i.e., interpreting the questions, retrieving
information from memory, editing, and reporting), Levy’s (1996, p. 172) views
represent a sound defense for advancing introspection research methods:
In a casual sense, introspection is an inevitable part of consumer

research, used by all research workers, as it means looking within one’s
self to know one’s ideas and feelings. That is, introspection is another
word for being self-conscious, aware, thoughtful, having ideas, and
knowing what they are.
Relevant to the researcher asking a question (the starting oval in Figure 2), some
amount of national and micro-culture-based unconsciousness (see Hofstede, 2001)
occurs in framing and verbalizing questions. For example, how an Australian, a Brit,
and an American frames and asks a respondent about beverages that the respondent
consumed for breakfast will vary unconsciously by word content, dialect, and
speaking pattern — even though the three researchers are using the same language.
Hofstede’s (2001, p. 4) view about provoking behavior applies to the researcher
asking questions: ‘‘Strategies using provoking behavior inevitably contain a
Heisenberg effect, in that the researcher interferes with the behavior observed. This
means that such behavior [as reported by the informant] cannot always be extrapolated
to circumstances in which the researcher is not present.’’ Also, the events and thinking
process reported retrospectively by the informant may be viewed accurately as only
approximations and highlights of the events and processes that actually occurred.
Researcher interference with the behavior observed also applies to the researchers’
recordings and interpretations of nonreactive measures (for a discussion of this point,
see Chapter 1 in Webb, Campbell, Schwartz, & Sechrest, 2000).
Subjective Personal Introspection
Subjective personal introspection (SPI; see Holbrook, 1986) includes a family of

research methods that rely extensively or even exclusively on the researchers’ life
experiences as data (Wallendorf & Brucks, 1993). Holbrook (1986, 1995, 1999, 2003,
1. Researcher (R) frames and asks a question; 3. Meta-cognition: 5. I searches memory
the question may be in a survey, asked orally by a I asks herself automatically for
person other than the informant (I), or why question relevant experiences
generated reflectively by I. is being asked and knowledge that
(see relate to the
Schwartz 1999). topic(s) in the question
as interpreted and
framed by I.
4. I crafts meaning
10. R aware, for the question:
understands, 2. I aware
reframes question 6. Unconscious
and interprets question is and topics (see
being asked? thinking activated
answer Hilton 1995: by attending to
being given? Ericsson and attending to the
Simon 1993) question and
memory search.
8. (a) I edits and crafts answers that 7. (a) I retrieves experiences and concepts
she believes relevant to question; from memory;
9. I verbalizes
(b) I edits answer to appear sane, wise (b) Interprets and assigns meaning;
an answer. (c) I edit answer based on context, status, (c ) collects additional meanings
(d) I automatically selects words to express automatically
herself
Figure 1: Introspective thinking process related to answering a question.

Subjective and Confirmatory Personal Introspection
109
Table 1: Research concerns and findings for 10 issues related to introspection.
Issue Research concerns and findings
Researcher (R) frames and asks Alternative question frames have larger
question influence on how informant (I) interprets
and answers the question
Informant (I) aware of question I may not attend or interpret R’s statement
asked? as a question
Meta-cognition: I asks herself Meta-cognition likely to be an implicit step
why question asked that I does not verbalize unless unsure, not
confident, about rationale
I crafts meaning; reframes Meaning/interpretation I assigns may poorly
question and topic match with R’s framing of the issue and
general focus of the inquiry
I’s search of long term for I’s initial search is done automatically, a
experiences and knowledge spreading activation relevant to question
as I has framed it. Occurs quickly among
easily retrieves concepts; evaluations are
assigned automatically among concepts
retrieved based on unconscious thinking
Unconscious thinking activated Most thinking occurs unconsciously;
I partially able to uncover unconscious
thinking by reflection and use of thinking
probes (e.g., metaphor elicitation
techniques)
I retrieves experiences and I automatically and unconsciously decides
knowledge from memory when to stop retrieving interprets own
thoughts; assigns meaning. experiences,
interpretations, and assigning meanings
I crafts answers; does editing Crafting and editing done automatically
of thoughts partially and by controlled thinking (I may
ask, ‘‘Will I get in trouble if I say what
I really think?’’)
I verbalizes answer I unlikely to provide a rationale for beliefs
without probing by R in order to limit
cognitive effort and limit interviewing time
2005a) provides a stream of introspective empirical reports focusing on one method

of introspection: ‘‘impressionistic narrative accounts of the writer’s own private
consumption experiences’’ (Holbrook, 2005a, 2005b, p. 3). Holbrook’s contributions
stimulated the work of other consumer researchers (e.g., Gould, 1991, 1995;
Hirschman, 1992; Williams, 1992) and work with a colleague (Holbrook &
Kuwahara, 1998).
Your Name: _________________________________ Today’s Date: ______________________________
Instructions: Please read the instructions for all six rows in Figures 2a and 2b before answering.
In the hexagon below, In some way relating to where to

1. In the square below, draw and In the oval below, write down go for a university or college degrees
write down 3 words about
name the beverage that represents 3 words that describe that tells about please tell as story or true event that
yourself that follow from the
who you are on a Friday night at a party. the object in the square to the left illustrates the words in the hexagon
words in the oval to the left
to the left.
2. In the square below, draw and name

the motor vehicle that best represents
who you really are:
3. In the square below, draw and name the

motor vehicle that best represents
your carefree side, the vehicle that you
would most like to be:
Figure 2a: Metaphor and storytelling exercise.

111
112
In the hexagon below, In some way relating to where to

4. In the square below, draw and In the oval below, write down
write down 3 words about go for a university or college degrees
name the animal that best represents 3 words that describe that tells about
yourself that follow from the please tell as story or true event that
who you really are if you were an animal the object in the square to the left.
words in the oval to the left. illustrates the words in the hexagon
other than a human:
to the left.
2. In the square below, draw and name

the animal that you would most like to be
if you were an animal other than a human:
3. In the square below, draw and name the

beverage that best represents who you
really are:
Figure 2b: Metaphor and storytelling exercise.

Subjective and Confirmatory Personal Introspection 113
Wallendorf and Brucks (1993) discern five categories of introspection on the basis
of the level of closeness or intimacy between the researcher and introspector:
Researcher introspection. The ultimate level of closeness in which the researcher is

the sole introspector in the study; while several studies are available using this
method, ‘‘our review of the social science literatures indicates that consumer
research may be alone in this regard’’ (Wallendorf & Brucks, 1993, p. 141).
Guided introspection. People other than the researcher are asked to introspect or
think aloud about themselves and their actions; answering a written questionnaire
is one form of guided introspection.
Interactive introspection. The researcher assists others in their introspections
but the object of the study is the ‘‘emergent experiences of both parties’’ (Ellis,
1991, p. 30). Wallendorf and Brucks comment that ‘‘beyond Ellis’s initial
discussion, there does not appear to be a social science literature that has used and
refined this method.’’ The present study does advance Ellis’s perspective and
includes an interactive introspection among the triangulation of methods
employed.
Syncretic combinations. The most common form expands the sample beyond the
researcher but also incorporates details of the researcher’s life experience that she
or he is willing to document; unlike interactive introspection, this combination
does not have the researcher share his/her introspections with informants (e.g.,
Freud, 1908/1965).
Reflexivity within research. Ethnographic studies using participant observation and
relying on two sources of data: (a) observational and interview material from
people in a cultural group being studied and (b) reflexive material that emerges
from being a participant studying that cultural group.
Advancing Toward Confirmatory Personal Introspection
Employing a syncretic combination of methods likely leads to deeper sense-making

than SPI, as developed and advocated by Holbrook (1986, 2005a, 2005b). Such
syncretic combination includes two or more of the following methods: (a) researcher
self-introspection and written description of an observable stream of behaviors and
thoughts occurring while the behaviors occurred; (b) the researcher interviewing any
mentors who participated in the behavioral process under study; (c) the researcher
interviewing a cohort to learn the cohort’s interpretations about the researcher’s own
understanding of the occurrence of specific acts in the stream of behavior, and the
antecedents that lead to each act; (d) the reflective use of decision and conscious
processing tools by the researcher to help surface prior conscious thought (see Park,
Hughes, Thurkal, & Friedman, 1981); (e) metaphor-generating tools (e.g.,
Christensen & Olson, 2002; Zaltman, 2003) and the use of experimental research
designs (e.g., see Bargh, 2002) to uncover unconscious thinking relevant to the
behavioral process being investigated; and (f ) the use of an independently developed

survey schedules to assist the researcher–informant in his or her introspection.
The objectives of this combination of methods include:
Confirming the occurrence of milestone behavioral events and decision heuristics

used in making choices in specific phases of the behavioral process;
Uncovering unrecognized paradoxes and resolutions in differences in opinions
held by multiple persons participating in several phases of the behavioral process
under study;
Helping to make explicit the ‘‘implicit mental model’’ (Senge, 1990) relevant to
decision making within the original time frame of the introspection study; and
Surfacing thoughts held unconsciously that are applicable to choices made in the
process under study.
In his report of the following incident, Holbrook (2005a, 2005b, p. 48) offers a
backhand (i.e., he is not an advocate of the need for independent assessments of
researcher introspections) application of syncretic combinations in his inter-
pretive analysis of photographs from his family’s archive: ‘‘I often find myself
musing over what rampant lack of self-confidence would encourage a mechanical
reliance on such self-imprisoning safeguards and such vision-restricting formulas.
The closest I have come to a member check [an independent assessment of an SPI
for accuracy and completeness] has been inviting my ninety-one-year-old mother to
attend a conference where I presented some of this material and dutifully
making revisions in my comments as she called out occasional corrections from
the audience.’’ Note that he reports that his mother’s comments helped to revise
his incorrect remembrances and her silences imply confirmation of other parts of
his narration. Thus, the attempt here is to illustrate how unique forms of
introspective syncretic combinations can be planned that help to (at least partially)
overcome the Wallendorf and Brucks (1993) conclusion that researcher self-
introspection offers severely limited potential in contributing useful research in
consumer behavior.
The application of multiple research methods transforms Holbrook’s SPI to
achieve a confirmatory personal introspection (CPI) that includes many of the
criteria of scientific approaches to research without the loss of the criteria
representing the artistic approach (see Brown, 1998, for a review of both
approaches). CPI is likely to be found useful for designing products and marketing
communications that consumers find desirable and that motivate their purchase
behavior. These two views respond to Andreasen’s (1985) and Wells’ (1993)
‘‘research backward’’ guideline to answer the ‘‘so what?’’ question before
implementing an empirical investigation.
Given the substantial scientific evidence that most thinking is unconscious (for
reviews, see Bargh, 2002; Zaltman, 2003), the research tools to surface unconscious
thoughts described subsequently are worthy of attention. Field studies applying CPI
may demonstrate the usefulness for working from several complementary literature
streams to extend Zaltman’s (2003) treatise on how consumers think — including
strategies for theorizing from process data (e.g., Eisenhardt, 1989; Langley, 1999);
decision-plan net theory of individual-choice models (see Park et al., 1981);
‘‘autoethnography’’ and other personal introspection methods (Holbrook, 2005a,
2005b; Wallendorf & Brucks, 1993), as well as related views on sense-making (Weick,
1995); and unconscious and automatic influences on consumer judgment, behavior,
and motivation (see Bargh, 2002).
In bare-knuckle terms, the view here attacks the current dominant logic in
consumer research on learning the reasons for consumer choice of brands or store
sites by relying principally on a written, self-completed survey using mostly closed-
end (fixed-point) questions. Although most consumers are able to complete such
surveys, such a highly cognitive method excludes data collection of most thoughts —
embracing the finding that most thinking occurs unconsciously, consumers have only
limited accessibility to the unconscious, and ‘‘people generally do not think in
words’’ (Zaltman, 2003, p. 13). Behavioral research methods that enable consumers
to access their unconscious thoughts need widespread adoption in studies on product
and brand knowledge held by consumers. CPI research includes methods designed to
reduce the inherent attempts to self-edit and block unwelcomed, or socially
unacceptable (see Fisher, 1993), thoughts and to stimulate informants to report
reasons for their actions seemingly ‘‘too minor to mention’’ in open-ended written
responses.
Subsequent to this introduction, the literature review suggests placing introspec-
tion in consumer research within grounded theory (Glaser & Strauss, 1967) and
building theories from case study research (see Eisenhardt, 1989; Langley, 1999;
Woodside & Wilson, 2003). The third section urges the adoption of Hirschman’s
(1986) humanistic inquiry philosophy and method recommendations as much as
possible for researcher introspection. The fourth section describes the application of
multiple methods focusing on the same empirical ground covered by a researcher’s
introspection — multiple methods that are useful for surfacing unconscious thoughts
and aiding retrieval of conscious thoughts generated during the process being
examined. The fifth section offers implications for theory construction that follow
from CPI. The final section covers limitations, conclusions, and suggestions for
further research.
Introspection, Case Study Research, and Constructing

Grounded Theories
Building theories from case study research (Eisenhardt, 1989; Woodside & Wilson,
2003), the organizational decision-making literature (see Langley, Mintzberg,
Pitcher, Posada, & Saint-Macary, 1995), and the historical method in consumer
research (Smith & Lux, 1993) provides useful theoretical grounding for researcher
self-introspection. Eisenhardt (1989) informs that, ‘‘The case study is a research
study which focuses on understanding the dynamics present within single settings
y Moreover, cases studies can employ an embedded design, that is, multiple levels of
analysis within a single study (Yin, 1994).’’
Langley et al. (1995) review organizational literature that recognizes phases in
some decision making that often involve an iterative feedback process, but
nonetheless often lead to some form of choice agreement among multiple
participants in the process, followed by action. These authors emphasize that even
when a decision can be isolated, rarely can the processes leading up to it.
Thus, most of the literature notwithstanding, we believe that no decision can be
understood de novo or in vitro, apart from the perceptions of the actors and the
mindsets and cultures of the contexts in which they are embedded. On the contrary,
we shall argue that decision making must be studied in toto and inspiration, emotion,
and memory, and at the collective level to include history, culture, and context in the
vast network of decision making that makes up every organization (Langley et al.,
1995, p. 261).
Several authors in separate but complementary research streams emphasize that
subconscious processes play may major roles in decision making. Thus, Langley et al.
(1995, p. 268) wish to add to Simon’s administrative man, an ‘‘insightful man, who
listens to the voices emanating from his own unconscious, or perhaps, better
expressed, who sights the images that well up in his own imagination.’’ Smith and
Lux (1993) call for the study of unconscious motives (Mu) in their historical method
exposition in consumer research, ‘‘Unlike transparent motives that ‘‘just are,’’
unconscious motives arise from social circumstances that do not necessarily have to
stand as they are, but of which the individual may not be consciously aware at the
time the act occurs. Such unconscious motives might be likened to unreflected
experience (Thompson, Locander, & Pollio, 1989) in that they constitute a ground
against which behavior is the figure.’’
Unconscious processing is an additional observation found in the literature on
decision making and thinking research. ‘‘It is often difficult to say who decided
something and when — or even who originated a decision’’ (quoted in Quinn, 1980,
p. 134, and Langley et al., 1995, p. 265).
Core Propositions for Examining the Exchange Process for Major Services
Consequently, the core proposition (CP1) offered here is that consumer choice
processes include recognizable phases involving several persons participating in one
or more phases with conscious and unconscious thoughts/motives affecting the
buyers’ beliefs and actions, and that one or all of the participants may be unable to
consciously explain the causes of specific milestone subdecisions occurring in the
process. However, CP2, relating to the purchase process for a major consumer
durable (e.g., motor vehicle) or service (e.g., the selection of a university to attend for
a 3- or 4-year degree), states that one or more phases of the process are likely to
include substantial conscious effort, and some of the thinking involved (not all) can
be retrieved within a researcher introspective study.
Adopting the Humanistic-Inquiry Paradigm for Researcher Introspection
Hirschman (1986) provides four criteria appropriate for humanistic inquiry that, if
applied, are useful for increasing the usefulness of researcher introspections —
including subjective personal introspections:
Credibility. Representing the multiple realities of the process examined adequately,

possibly by the researcher submitting the interpretation to the scrutiny of the
participant in the process who provided some of the original data and seeking
responses as to the report’s authenticity.
Transferability. Transferring the resulting interpretation of the process to other
contexts, such as the purchase of a major service different from the one being
examined (e.g., processes involved in buying a medical procedure versus processes
for a university degree).
Dependability. Use of multiple human investigators enhances internal stability of
measures taken.
Confirmability. Seeking neutrality and objectivity, possibly by asking outside
auditors to confirm or disconfirm the interpretations by the researcher.
Crafting Formal Survey Protocols for Self-Interviews
One step toward increasing credibility in researcher self-introspection is for the

researcher to formally interview him- or herself on two or more separate occasions.
Although such a suggestion may seem absurd, calls for further reflection by decision
makers for more useful sense-making of events and causes of these events are made
elsewhere (e.g., Weick, 1993).
The use of a formal survey protocol for completing by the researcher, with him or
her adding and answering additional questions, is another suggestion that may
improve credibility — thus, the use of a written interview schedule of questions
crafted earlier by an informant and others serves to guide, broaden, and deepen emic
interpretations during subjective personal introspections.
Learning While Talking
One step toward accomplishing confirmability occurs by having two researchers

interview one another while interviewing themselves. This method helps operatio-
nalize Weick’s (1993) famous proposition, paraphrased here as ‘‘How do I know
what I think until I hear what I have to say? How do I know what I’ve done until
I tell aloud what I did?’’
Another step toward confirmability, as well as dependability, is having an ‘‘inside
auditor’’ answer questions that confirm or disconfirm as well as deepen the
interpretations reported by the researcher self-introspection (e.g., the corrections

offered by Holbrook’s mother on his interpretations of her father’s behavior).
The Inside Auditor
The use of such an inside auditor, as one’s mother or another person involved
directly in the focus of the introspection, is almost certain to be helpful for clarifying
and deepening researcher introspective reports. Consequently, an advance toward
CPI is made by using inside auditors as well as agreeing with Hirschman’s (1986,
p. 246) advocacy of an outside auditor:
To assess whether or not the interpretation is drawn in a logical and

unprejudiced manner from the data gathered and the rationale
employed, humanistic inquiry relies on the judgment of an outside
auditor or auditors. These individuals should be researchers them-
selves, familiar with the phenomena under study.
The Cohort Auditor
To stimulate the mental surfacing of observations about the self and the process
being examined by the introspection, an additional category of auditor is possible:
the cohort auditor (CA). CA is a person living in the same current environment as the
researcher completing a self-introspection that the researcher–introspector (RI) asks
to comment on the process and outcome under study. Thus, the RI both tells his/her
story to the CA and asks for questions and comments from the CA as the story is
being told. Besides being a sounding board, the CA is likely to provide information
that triggers retrievals relevant to the focus of the study by the RI that might
otherwise not occur.
The Forced Metaphor-Elicitation Technique (FMET)
The forced metaphor-elicitation technique (FMET) has the goals similar to the
Zaltman metaphor-elicitation technique (ZMET) and the Doyle and Sims (2002)
‘‘cognitive sculpting’’ technique. Using metaphor analysis as a research tool to
understand more deeply the unconscious linkages associating with a behavior; Sims
and Doyle (1995) illustrate cognitive sculpting research with informants’ use of
tabletop objects as metaphors of what they are saying and have done — resulting in
what Sims and Doyle refer to as ‘‘explicating knowledge.’’
Christensen and Olson (2002) provide an application of the ZMET for a study of
15 very highly involved mountain bikers: ‘‘Approximately 1 week prior to the
interview, each recruited participant was contacted and given a set of instructions.
First, they were asked to think about mountain biking. Then they were told to select
8–10 pictures that represent their thoughts and feelings about mountain biking and
bring the pictures to the interview. Each picture is a metaphor that expresses one or
more important meanings about mountain biking y Respondents participated in in-
depth interviews conducted by three interviewers trained in the ZMET methodology
and experienced in conducting ZMET interviews.’’ See Zaltman (2003) for further
details.
The FMET is a tool for surfacing metaphors for use in researcher introspection.
FMET includes four distinct steps. First, the respondent is asked to draw or select
pictures for three sets of two objects each (Figure 2 shows the FMET design):
The animal that first comes to mind that the RI believe that ‘‘represents some
aspects of who you are, what you are like.’’
The animal that first comes to mind ‘‘that you admire, might select to be if you
were an animal other than a human.’’
The beverage first comes to mind that ‘‘that best represents you most of the time
during the daytime.’’
The beverage that first comes to mind that ‘‘best represents you at home or a party
on a Friday night.’’
The motor vehicle that first comes to mind that ‘‘best represents the vehicle you
really would most likely be if you were, in fact, a motor vehicle.’’
The vehicle that first comes to mind that ‘‘best represents the fun, carefree side
of you.’’
Nearly all individuals can identify themselves as more than animal, beverage, and
vehicle, depending upon the situation being framed by the question (cf. Dichter,
1985; Woodside, Floyd-Finch, & Wilson, 1986). Thus, the FMET attempts to
capture the several unconscious beliefs about the RI. Unlike the ZMET, the FMET
does not focus on selecting items in pictures related directly to the behavior being
examined — the ZMET is more likely to cause greater cognitive effort and strain
caused by attempts to find pictures that associate with the behavior being examined
than the FMET. The ZMET appears to require great cognitive effort and substantial
interviewer training and skill in order to interpret the pictures selected by the
informant; as discussed subsequently, the FMET is designed for the RI to self-
interpret how the specific objects selected associate with the choice behavior under
investigation.
Second, the FMET asks the RI to say or write the two to four features that first
come to mind about each of the six objects in the pictures. Third, the FMET asks the
RI to mention the first thoughts that come to mind — ‘‘what each of these features
tells you about yourself.’’ Finally the FMET asks the RI ‘‘to tell a story or true event
that illustrates the concepts and description’’ of the features just mentioned about
you related to the choice behavior under investigation. Figure 3 illustrates the fourth
step for the choice of buying a major consumer service — the choice of a university
for an undergraduate degree. The RI then has the opportunity to include the results
from using the FMET into her interpretation of her service choice.
Coffee The university I finally picked was known as one

Strong taste Self-Confident of the hardest in the country. This has increased
Warm Strong Personality my desire to prove to myself I could go to such a
Energetic Outgoing Attitude difficult university and made me see it as a
Dark Colour challenge I wanted to win.
Most of my peers from high-school were going to

Motorbike Independent study in the university of our city. This made me
Fast Willing to want even more to go somewhere else because I
One-Person Succeed wanted to leave the old life behind, start new
Vehicle Free friendships and be far enough away from home to
feel independent.
Parma is a relatively large productive city in the

Sports Car
Willing to be rich North of Italy, wealthy and full of university
Red
Smooth students. Offers a lot of jobs with good careers
Expensive
Elegant perspective, and at the same time many ways to
Fast
have fun and enjoy the university years.
I haven’t considered being far from

No relationships my family to go to university to be
Cat
Moody a problem.
Small
Selfish I’m really focused on my self and my
Clean
Independent Choices depend more on what I think
and who I want to become.
Panther •Brave
None of my peers was going to
Wild •Grumpy
study in Parma so I felt I was
Dangerous •Hard-headed
braver than them in going there
Fast •“Hunts” for
by myself.
Slim myself
Parma University offers lot of sports—

Water I want to be clean activities that I am involved in.
Fresh Healthy I love sport and keeping my body in shape
Natural Diet beverage and I considered all the universities under
this point of view.
Figure 3: Metaphor and storytelling exercise.
Figure 3 is an example application of one informant’s use of the FMET related to

the topic of selecting a university or college for an undergraduate degree. Note how
the informant’s responses in the fourth step help to uncover the matching of personal
features within the informant’s desires to the features provided by specific brand (in
this case, Parma University, Italy).
Achieving personal independence relates to the pictures that include animals that
hunt alone, a motorcycle — a vehicle often with one rider, and a two-seat sports car.
The RI refers to being ‘‘independent’’ to describe the cat and the motorcycle.
‘‘Independent’’ is used by the RI to describe herself as well. Such information is
helpful to achieve a deep understanding of the motivation to attend a university as
far away from her high school friends and her parents as possible — within the felt
limits of the reported inadequate transportation system. High need for achievement,
independence, freedom, and health via sports are core themes that connect the
animal, beverage, and motor-vehicle metaphors with the RI’s university decision.
The FMET findings increase understanding of the reasons for the RI’s choice of
product–service features for evaluating and ranking alternative universities.
A Confirmatory-Introspection Research Example
Figure 4 shows the phase theory used to consider possible subdecisions that may be
involved in the purchase of a major consumer service; that is, the selection of a
university in order to attend for a 3- or 4-year Bachelor of Arts degree. The rejection-
inducing dimensions (RID), tradeoff dimensions (TD), and relative-preference
dimensions (RPD) (cf. Park et al., 1981) shown for each decision phase in Figure 4 is
to indicate the possibility of such feature dimensions and not that they always occur
for each phase.
A RID is an aspect that an alternative must offer for the alternative to receive
further consideration. For example, a student wants to play college football and will
only consider colleges and universities that have a football team. Not having a
football team is an RID for this student.
A TD is an aspect that a student is willing to give up only if an alternative offers a
different aspect of equal value. For example, a student wants to live in a private room
on campus but is willing to consider a university that offers private rooms nearby but
off campus.
An RPD dimension is an aspect that the student finds attractive for an alternative
to offer but not having the dimension does not cause the student to reject the
alternative. For example, a student may prefer to go to a college or university that
has sororities and fraternities but is willing to consider colleges and universities that
do not have this feature.
This university-choice topic is a useful focus for several reasons. First, the
selection of a university for such a degree represents the purchase of a major service,
due to the time and often the financial expenditure for the student and her or his
parents — an important decision made relatively early in life for many persons in
many developed nations. Second, based on pretest interviews, most college students
are likely to be able to identify distinct phases in their choice process, which include
becoming aware of alternative universities; collecting information from family
members, friends, and teachers on what attributes to consider in making the decision;
selecting universities and colleges to visit; and making the final choice. Thus, the
decision is complex and time consuming, and one or more phases are likely to include
the combination of substantial amounts of conscious and unconscious thinking.
Third, for some phases of such an important service purchase, RIDs, TDs, and
RPDs are likely to be used. Fourth, this choice decision is relevant to the RI from
whom the data were collected: The RI was still attending her chosen university and
felt capable of reporting the details occurring in most of the phases of the process.
Figure 5 serves to demonstrate the value in using multiple methods in collecting
data to confirm and deepen the process under investigation. Note that Figure 5
depicts each method as confirming one or more pieces of data learned by one or more
other research methods as well as certain amounts of information found unique to a
particular method. Also, note that not all the information relevant to the process is
shown to be captured even when multiple methods are used.
Figure 5 includes cylinders to indicate that a certain amount of information not
relevant directly to the process is recorded. Seemingly non-relevant information may
122
P1 = the buying process for a major product-service include several identifiable phases (e.g., A, B, C, D, E)
P2 = feedback loops occur (e.g., revisions in thinking due to new information)
P3 = a few attributes are critical (i.e., rejection inducing dimensions) for each phase and other attributes are“nice to have features” (i.e., relative preference dimensions)
P4 = different influence sources affect the use of different attributes for different phases of the process, that is, the I’s shown in this figures may refer to different influence
sources (e.g., friends, parents, teachers, college tour guides)
E
B C D
A
a-c
a-c a-c a-c
a-c RID
RID RID RID
RID
d–f
d–f d–f d–f d–f
TD
TD TD TD TD
g–h
g–h g–h
g–h g–h RFP
RFP RFP
RFP RFP
a–c
a–c a–c a–c
I a–c
I I I
I
d-f
d-f d-f d-f
I g -h d-f
I g -h I g -h I g -h
I I g -h
I I I
I
Figure 4: Propositions (Pi) for the unconscious and conscious thinking–doing process for buying major product–service. RID,
rejection inducing dimension; TD, tradeoff dimension; RFP, relative preference dimension. The RID, TD, and RFP and each
of the phases may or may not be identified consciously by the informant. A–E are conscious and/or unconscious phases in the
decision-making process (e.g., A, I will attend a university; B, I become aware of alternative universities to attend (awareness
set); C, I select universities to visit; D, I select a university to attend; E, I attend the university I select; F (not shown), I stay or
transfer out of the university that I am attending; and G, I complete the university degree program. Lower case letters indicate
the attributes relevant to each decision heuristic.
Data collected during second interview with informant Data collected during interview
Data collected during first interview
with informant’s mentor
with informant
E
B C D
A
a-c
a-c a-c a-c
a-c RID
RID RID RID
RID
d–f
d–f d–f d–f d–f
TD
TD TD TD TD
g–h
g–h g–h
g–h g–h RFP
RFP RFP
RFP RFP
a–c
a–c a–c a–c
I a–c
I I I
I
d-f
d-f d-f d-f
I g -h d-f
I g -h I g -h I g -h
I I g -h
I I I
I
Data collected from forced

metaphor elicitation
technique
= data collected not directly relevant to study
Figure 5: Research toolkit for surfacing relevant unconscious and conscious thinking–doing processes.
123
enable the uncovering of information directly relevant to the process under study, for
example, an informant may need to talk and say little that is directly relevant to the
specific study to get to the point of being comfortable about what she is saying, as
well as to learn enough about what she thinks to elaborate deeply on her motivations.
Thus, although some data are identified as not directly relevant to the phases in the
process, it is preferable not to label any part of the data collected as useless
information.
Implications for Theory Construction
In some substantial sense, an informant always engages in researcher introspection

whether he or she is asking questions implicitly or explicitly to him- or herself, as well
as whether the questions are written or presented verbally. The informant creates an
interpretation of the question posed from whatever the source — thus, the informant
engages in researcher introspection. Also, metaphorically speaking, the informant
has to search the file drawers of his or her mind to find the drawer somehow labeled
by one or more concepts interpreted to be in the question. Then the informant has to
find the folder in the file drawer; open the folder; read and interpret the folder’s
contents; and select and use words, sentences, and other tools that he or she feels can
be interpreted by him or her and possibly by others in a manner that he or she
desires. Thus, respondents must make sense of the questions asked of them, and
autobiographical memory in retrospective thinking is involved in all survey research.
The work of Payne (1951), Grice (1975), Hilton (1995), Ericsson and Simon
(1993), and Schwarz (1999) supports the view that ‘‘question comprehension involves
extensive inferences about the speaker’s intentions to determine the pragmatic
meaning of the question’’ (Schwarz, 1999, p. 96). Such inferences apply to whoever
asks the question, no matter if the question is asked implicitly or explicitly.
Consequently, the informant is always an active participant in framing a question, as
well as in answering it.
From this perspective, the severe problems that Wallendorf and Brucks (1993)
note for researcher introspection does not lead to the conclusion that the method
should be abandoned in consumer research. Rather, the more useful conclusion is
that researcher introspection needs to innovate to include carefully executed
(rigorous) research procedures and the combined use of multiple data-collection
instruments. Such data-collection instruments should include tools designed
explicitly to surface unconscious thinking processes, especially because substantial
scientific knowledge now exists that unconscious thinking processes represent most
mental activities by individuals (for a literature review, see Zaltman, 2003).
Researcher introspective case studies aid in ‘‘opening up decision making’’
(Langley et al., 1995) by demonstrating that different models of decision making are
likely to be relevant for different phases occurring in the process. For example,
decision making as convergence without consciously considering alternatives may
best describe the process whereby the student RI became committed to attending
a university for the purpose of acquiring a degree. After such a convergence has been
completed, the phase involving visiting alternative universities reflects decision
making following Simon’s (1960) three-step sequence: first intelligence (i.e.,
diagnosing the problem), then design (i.e., finding alternatives to evaluate), and
finally choice. Langley et al. (1995) refer to this view as Model 1, organizational
decision making as sequential.
Regarding the selection and use of criteria to select universities to visit and reach
the choice of university to attend appears to represent ‘‘insightful man,’’ that is, the
RI may listen ‘‘to the voices emanating from his [her] own subconscious, or perhaps
better expressed, who sights and images that well up in his [her] own imagination’’
(Langley et al., 1995, p. 268). However, the relevancy of these features is questionable
because the data from the FMET were collected after, rather than during, the choice
process. The FMET data do appear to provide useful clues into the deep meanings of
RIDs in the RI’s decision plan net. Thus, decision making as insightful, identified by
Langley et al. (1995, p. 259) as Model 5, appears relevant especially for the selection
of dimensions to use for evaluating university alternatives.
Generalizing the results of the RI report to theory suggests that different models
of decision making are at times more or less relevant to the decision process,
depending on the phase in the process being examined. Thus, asking whether or not
the purchase of a major durable or service is made consciously or unconsciously is
less useful than asking where and how both thinking processes contribute to the
decision process.
Conclusions, Limitations, and Suggestions for Further Research
The use of multiple methods in research introspections does appear helpful in

confirming the occurrence of specific phases in the decision process and in achieving a
deep understanding of primary motivations within the individual that help to explain
the presence of attribute dimensions used in his or her choice of a university. The
combined use of decision-plan net analysis, the FMET, schedules of survey questions
completed by mentors, an inside auditor, and the RI him- or herself represents a
blending of emic (conscious and unconscious views of the individual native
informant) and etic (interpretation of the researcher after acquiring some distance
from the research site) perspectives. Though at first blush the view may appear to be
an oxymoron that an etic perspective can be acquired in researcher introspection,
asking others to confirm the occurrence of specific actions that relate to the behavior
of the RI that the RI is examining, as well as seeking views from an inside auditor,
serves the RI in viewing herself in the third person.
To increase sense-making about what one has done and to deepen understanding
of why it was done, Weick (1995) advises that one reflect (e.g., sleep on it) on the
meaning of what one does and says, as well as seek the views of mentors about the
meaning of what has happened — steps to achieve distance, an etic view of what
happened and why it happened. The use of explicit tools, including survey forms
designed and used explicitly by the RI to interview herself and others, as well as the
additional tools described in this chapter, is more likely to result in achieving a useful
etic view than is adopting Holbrook (2005b, p. 48) emic-is-enough assumption (e.g.,
‘‘I assume that my own introspections resonate so strongly with the photos taken by
my grandfather because ATH [Holbrook’s grandfather] has, in effect, captured the
essence of my own subjective personal introspection-based recollections’’).
Along with the RI asking herself multiple and the same sets of questions on
different days (e.g., for an example of same topic, consecutive weekly interviews and
informant reflections over 18 weeks, see Cox, 1967) regarding the process being
examined in the study, the interviewing of other persons involved directly in the
process as well as insider auditors is likely to always improve researcher
introspections. The following views by Weick (1995) and Allport (1985) as well as
Hirschman’s principles for humanistic inquiry are rationales for this suggestion.
‘‘Those who forget that sense-making is a social process miss a constant substrate
that shapes interpretations and interpreting. Conduct is contingent on the conduct of
others, whether those others are imagined or physically present’’ (Weick, 1995, p. 39).
Social psychology is ‘‘an attempt to understand and explain how the thought, feeling,
and behavior of individuals are influenced by the actual, imagined, or implied
presence of others’’ (Allport, 1985, p. 3).
Of course, the intention is not to generalize from one RI case study to many
consumers. The specific findings are applicable only to the RI. However, the two core
propositions are confirmed by the findings that CP1, the decision process examined,
includes recognizable phases involving several persons participating in one or more
phases with conscious and unconscious thoughts/motives affecting their beliefs and
actions, and that one or all of the participants are unable to consciously explain the
causes of specific milestone subdecisions occurring in the process.
CP2 relates to a purchase process such as the selection of a university to attend for
a 3- or 4-year degree. One or more phases of the process likely includes substantial
conscious effort, and much of the thinking involved (not all) can be retrieved within a
researcher introspective study. Such seemingly intuitive propositions need confirma-
tion and extension by additional research before concluding that they are obviously
valid. Heretofore, the view that whether or not to buy a major retail service, such as
whether or not to attend university, can be a convergence process instead of being a
decision (organizational decision-making Model 4 of Langley et al., 1995) receives
scant attention in the service marketing literature.
Unfortunately, the Webb et al. (1966, 2000) observations are still accurate early
in the 21st century.
Today, the dominant mass of social science research is based on

interviews and questionnaires. We lament this overdependence upon a
single, fallible method. Interviews and questionnaires intrude as a
foreign element into the social setting they would describe, they create
as well as measure attitudes, they elicit atypical roles and responses,
they are limited to those who are accessible and will cooperate, and the
responses obtained are produced in part by dimensions of individual
differences irrelevant to the topic at hand. But the principal objection is

that they are used alone. No research method is without bias.
Interviews and questionnaires must be supplemented by methods
testing the same social science variables but having different metho-
dological weaknesses. (Webb et al., 2000, pp. 1–2)
Ironically, although Holbrook (2005a, 2005b) criticizes using a ‘‘member check,’’

his use of family photographs and incorporating of his mother’s corrections to his
subjective personal introspection are two examples of applying supplementary
methods for transforming SPI into CPI. The irony extends beyond illustrating the
Webb et al. (2000) point that no research method is without bias to the observation
that subjective personal introspections alone are unlikely to detect substantial biases
in method and in the accuracy of reported findings.
Chapter 6
Overcoming the Illusion of Conscious

Will and Self-Fabrication
Synopsis
Naı̈ve subjective personal introspection includes the failure to recognize the

confirmability of one’s own attitudes and personal meanings learned explicitly
from self-examining such topics and explaining one’s own behavior.
Unconscious/conscious theory of behavior explanation follows from unifying
the research on unintended thought–behavior with folk explanations of
behavior. Chapter 6 describes advances in research confirming own attitudes
and personal meaning and suggests the need for applying multiple methods to
overcome the fundamental attribution error, inherent cultural prejudices, and
the general bias toward self-fabrication. The discussion is valuable for
achieving a deep understanding of how customers think, advancing from
subjective to confirmatory personal introspection, and understanding the need
to apply research tools useful for enlightening knowledge and overcoming the
inherent bias within subjective personal introspection.
Introduction
In reviewing relevant literature, Chapter 5 provides a series of telling propositions

regarding subjective and confirmatory personal introspection. The following points
reflect and extend these propositions.
1. The dominant logic in consumer research includes asking questions that require
some amount of subjective personal introspection by a respondent — whether or
not a separate individual is asking the question or the informant both asks him or
herself and answers the question and whether or not the informant answers the
question face-to-face with a researcher, reads the question in a survey, or ponders
the issue alone.
2. Because most thinking occurs unconsciously (Bargh, 2002; Wegner, 2002;
Zaltman, 2003) and the informant has limited access to his or her own unconscious
thinking, the informant is able to retrieve, interpret, and report (to herself and
others) only a limited amount of relevant knowledge and insight when answering
questions and pondering a specific topic.
3. The use of additional research tools (beyond self-interviewing and meditation)
aids in surfacing and confirming/refuting both events and personal meanings that
the informant otherwise concludes to be accurate answers reflecting his or her own
prior and/or currently held beliefs and attitudes.
The next section elaborates on advanced attribution errors, that is, (a) personal
denial of committing the fundamental attribution error, the illusion of will, including
unawareness that implicit (automatic, unconscious) meaning is relevant for subjective
personal introspection (SPI) and (b) disregarding and denying the usefulness of
member checks (independent assessments of an SPI for accuracy and completeness)
and other tools (e.g., implicit association tests; see Brunel, Tietje, & Greenwald, 2004;
Masion, Greenwald, & Bruin, 2004), the forced-metaphor elicitation technique for
uncovering unconscious meaning (see Chapter 6) — tools permitting access to
unconscious personal meaning that otherwise remain inaccessible to the researcher–
informant.
This chapter describes folk conceptual theory of behavior explanation (see Hilton,
1990; McClure, 2002; Malle, 1999, 2004; Malle & Knobe, 1997) and discusses the
theory’s relevancy to SPI research and theory. Highlighted topics also include the
critical importance of SPI and the use of mixed research designs in SPI research for (a)
theory building in consumer research; (b) deepening individual and group sense-
making; and (c) aiding in preventing unfair, bad, and downright dangerous decision-
making (e.g., Gaither, 2002; Kozak, 1996) — reasons that substantially extend Gould’s
(1995) defense of SPI in responding to the Wallendorf and Brucks (1993) criticism of
the method. Chapter 6 closes with conclusions and implications for theory and
research.
Deepening Subjective Personal Introspection
Gould (2005) proposes that research using methods to confirm personal introspec-
tions are irrelevant in SPI research — reporting insights through meditation is
sufficient and alone reflects ones thoughts and feelings. This view is incorrect in
several aspects, but the more important flaw is that it ignores the following main
point. Because subjective personal introspection pervades social research, and both
the researcher and informant (whether or not both are the same individual or
different persons) are unable to examine relevant unconscious and conscious data
using only one method (e.g., Buddhist meditation), applying mixed-methods designs
(Tashakkori & Teddlie, 1998) is essential for correcting event memory failures and
overcoming fundamental and advanced attribution errors.
Here is the essence of Gould’s (2005) position. Thus, when people tell
stories about their lives, confirmation could mean that the researcher
would seek a member check on the accuracy of accounting for certain
Overcoming the Illusion of Conscious Will and Self-Fabrication 131
events. For example, did the individual engage in a certain behavior or

not? But there is no member checking the meaning a person assigns in
the story. This is her own introspective story. People can compare
stories with one another but one does not generally ‘‘confirm’’ one’s
own story in terms of personal meaning. When Holbrook (2005a,
2005b) is asking his mother about certain remembrances, he is engaging
in retrospection in some sort of confirmation through his mother. But
again, he is not confirming the personal meanings of his experiences or
of his story. Thus, Woodside makes what is a major flaw [that] underlies
much of his argument in establishing the need for confirmation; he
conflates event-memory checking where confirmation may be useful
with meaning and experiential member checking where it is not.
A seemingly minor point: Because Holbrook (2005b, p. 48) does not see the need,
his report does not include asking his mother anything; without him asking she
volunteers. ‘‘I often find myself musing over what rampant lack of self-confidence
would encourage a mechanical reliance on such self-imprisoning safeguards and such
vision-restricting formulas [e.g., member checks]. (The closest I have come to a
member check has been inviting my 91-year-old mother to attend a conference where
I presented some of this material and dutifully making revisions in my comments as
she called out occasional corrections from the audience.)’’
Holbrook’s view is an example of an advanced overconfidence bias applied to
interpreting both factual accuracy and meaningfulness in his SPI. His SPI research
would benefit from replacing his musing about ‘‘rampant lack of self-confidence’’
with a more mindful, complex model of explicit–implicit thinking (cf., Weick &
Sutcliffe, 2001). His use of a triangulation of data-collection methods (i.e., his SPI,
examining 2,300 photographs taken by his grandfather, and his mother’s member
check) counters his disparaging remarks about using additional data-collection
methods beyond SPI.
Overconfidence and self-fabrication rather than a rampant lack of self-confidence
are the dominant human tendencies in explaining our own behavior to ourselves and
others (cf., Langer, 1975; Wegner, 2002; Wilson, 2002). Overconfidence bias in
cognitive science (e.g., Gigerenzer, 2000; Gilovich, 1991; Gilovich, Griffin, &
Kahneman, 2002; Lichenstein, Fischoff, & Phillips, 1982) refers to the human
tendency to overestimate the accuracy of one’s own answers. For instance, the study
by Lichenstein et al. (1982) gave participants questions such as ‘‘Absinthe is (a) a
precious stone or (b) a liqueur’’; they chose what they believed was the correct answer
and then were asked for a confidence rating in their answers, for example, 90 percent
certain. When people said they were 100 percent certain about individual answers,
they had in the long run only about 80 percent correct answers; when they were
90 percent certain, then had in the long run only 75 percent correct answers, and so
on. Lichenstein et al. (1982) identify such discrepancies as the overconfidence bias and
explain their occurrence by general heuristics in memory search, such as confirmation
biases, or general motivational tendencies, such as the illusion of validity (see
Gigerenzer, 2000).
Advanced overconfidence bias (AOB) goes beyond such cognitive science reports
of overconfidence bias. AOB includes the implicit — and often inaccurate —
assumption that the person holds unconsciously the meaning she reports in a SPI.
The main point here relates to the McClelland, Koestner, and Weinberger (1992)
evidence and conclusion favoring dual (explicit and implicit) motives and goals —
not that SPI reports are invalid but that SPI and alternative introspection data-
collection methods tap different levels of meaning and explanations. Not recognizing
this possibility and not using mixed-method designs to acquire both explicit and
implicit personal introspection data is committing AOB.
Implicit motives are needs that people acquire in childhood that have become
automatic and nonconscious. Self-attributed motives are people’s conscious theories
about their needs that may often differ from their nonconscious needs. The picture
McClelland and his colleagues paint is of two independent explanatory systems that
operate in parallel and influence different types of behaviors. ‘‘In our terms, the
adaptive unconscious and the conscious explanatory system each has its own set of
needs and motives that influence different types of behaviors’’ (Wilson, 2002).
The SPI reports in the consumer research literature (e.g., Gould, 1991; Holbrook,
2005b) are akin to explicitly interviewing yourself without recognizing the usefulness
of applying implicit interviewing methods to capture unconscious meanings and
motives.
Fundamental and Advanced Attribution Errors
The fundamental attribution error (Ross & Nisbett, 1991; Wilson, 2002) refers to
people overlooking situational influences on their actions and inferring that they
acted on the basis of their own internal states — inferring internal states via explicit
interpretation without the use of tools for learning implicit thinking. Choi, Nisbett,
and Norenzayan (1999) demonstrate that people in Western cultures are especially
prone to the fundamental attribution error, and that people in East Asian cultures
are less prone. The advanced attribution error (AAE) includes denial that the
fundamental attribution error is relevant personally and that applying additional
tools (a mixed-methods research strategy) will help overcome illusion of will and
cultural bias that occur automatically during SPI.
Bargh et al. (1996) demonstrate how situation-message treatments (i.e., marketing
manipulations) can influence behavior directly without affecting participants’
introspection. These researchers had college participants fill out a scrambled
sentence that included words such as ‘‘wrinkled,’’ ‘‘gray,’’ ‘‘retired,’’ ‘‘wise,’’ and
‘‘old.’’ These participants were thus primed with the stereotype of an old person,
whereas other participants in the study did not receive this version of the test. As
each participant left the experiment room, the person’s gait was measured
surreptitiously. The individuals who had been led to think about senior citizens
walked more slowly than did those not primed with this thought. The idea of the
action arose from the stereotype and so influenced the behavior directly, apparently
without conscious will.
Extensive post-experimental interviews suggested that the participants were not

particularly conscious of the aged stereotype after the experiment. And even if they
were, they were certainly unaware that this might suggest they should walk at a
different speed, or for that matter that their walking speed was being assessed. Yet
merely thinking of the kind of person who walks slowly seemed to be sufficient to
induce shuffling (Wegner, 2002, p. 128).
Gladwell (2005) offers several examples of how conscious and unconscious
thinking and deciding in the same person often diverge and conflict.
The winning trombone-playing performance of Abbie Conant, a professional
concert musician, in a behind-a-screen competitive audition is a case in point. Ms.
Conant played sixteenth in the 33-candidate audition. Her playing was so outstanding
that the Philharmonic music director, Sergiu Celibidache, cried out, ‘‘That’s who we
want!’’ The competition was stopped and the remaining 17 players were sent home.
There were two more rounds of auditions. Conant passed both with flying colors. But
once Celibidache and the rest of the committee saw her in the flesh, all those long-held
prejudices began to compete with the winning first impression they had of her
performance. She joined the orchestra, and Celibidache stewed. A year passed.
In May 1981, Conant was called to a meeting. She was demoted to second trombone,
she was told. No reason was given. Conant went on probation for a year, to prove
herself again. It made no difference. ‘‘You know the problem,’’ Celibidache told her.
‘‘We need a man for the solo trombone’’ (Gladwell, 2005, p. 247).
Conant took the case to court and over the next 13 years she won several rounds
of battles (in courtrooms and additional auditions), including reinstatement as first
trombone and pay equal to male colleagues (Gladwell, 2005).
Might reports of culture-trained automatic thinking differ in ways unrecognized in
existing SPI reports (e.g., Gould, 1991; Holbrook, 1986, 1995, 2003, 2005a, 2005b)?
Would the use of additional tools in a mixed-methods design (e.g., screened and
unscreened auditions) sometimes confirm and sometimes disconfirm meanings
expressed in explicit thinking reports? In two studies, Brunel et al. (2004) provide
evidence of such confirmation and disconfirmation within consumer research contexts.
Their studies compare participants’ explicit attitude reports with implicit association
test (IAT) findings. Based on computer-mediated response latency measurement
protocols, the IAT measure is computed by comparing the relative response times
associated with several categorization tasks (see Brunel et al. for details).
Study 1 reports high efficacy of the IAT as a measure of brand attitudes and brand
relationship strength toward Macintosh and Microsoft Windows-based PC
machines: Under conditions in which participants were not expected to hide their
beliefs, explicit brand attitudes were strongly correlated to implicit attitudes and
implicit brand relationship, thereby validating the IAT for brand evaluation. Study 2
demonstrates no significant differences between explicit attitudes toward ads with
White spokespersons compared to ads with Black spokespersons, suggesting
that at the explicit level, participants do not exhibit racial preferences; however,
implicit measures of attitude toward ads reveal strong preference for ads containing
White versus Black spokespersons. Brunel et al. (2004) conclude, ‘‘Consumers’
associative brand networks may include concepts and associations that a consumer
either cannot or will not report [to others or themselves], but which may surface
through the IAT.’’
A number of additional studies examine the effect of introspection, or thinking
about reasons, on attitudes, judgments, and choices (e.g., Levine, Halberstadt, &
Goldstone, 1996; Simonson & Nowlis, 2000; Wilson, Hodges, & LaFleur, 1995).
When people are asked to explain their attitudes (or choices), they tend to focus on a
subset of the reasons that would otherwise (without the need to explain) influence
their attitudes, particularly reasons that are verbalizable, accessible, plausible, and/or
self-enhancing. Having to provide reasons or introspect can affect choices, because
provided reasons are typically only a subset of the factors that would otherwise
influence preferences.
Folk-Conceptual Theory of Mind and Behavior Explanation
Heretofore, the consumer–research literature does not include the unique and
valuable advances in the literature of how people explain their own behavior (i.e.,
SPI). Malle (1999, 2004) summarizes this body of work and develops the folk-
conceptual theory of mind and behavior explanation (folk model, for short). ‘‘It is
not, however, ordinary people’s own theory of explanation (they probably don’t have
one), but rather a genuine scientific theory’’ (Malle, 2004, p. 236). Malle points out
that prior attribution theory focuses introspection on people allegedly classifying
causes of effect outcomes into two major categories: person and situation causes —
greatly simplifying the possible conceptual framework in which explanations are
embedded. Similarly, Gould’s (1991) typology of energy states focuses on a small
subset of human explanation — ‘‘bodily felt experience of everyday consumption’’
(p. 205) rather than representing a sophisticated folk model of mind and behavior.
The folk model categorizes behavior explanations into two major modes of
explanation — reason and cause — as well as two minor modes — causal history
reasons (CHR) and enabling factors (EF). Reason explanations are people’s
explanations of an intentional behavior that cite the agent’s reasons for acting that
way; cause explanations are people’s explanations of an unintentional behavior that
cite the causes that brought about the behavior.
Causal history of reason explanations provides an explanatory link between
reasons and their own causal history, citing factors that preceded and thus brought
about the reasons for an action. These explanations literally describe the causal
history of reasons, which could lie in childhood, in cultural training, in personality
traits, or in a situational cue that triggered a particular desire (Malle, 1999).
Without direct reference to the folk model literature, Allen (2002) demonstrates
the dominance of causal history of reason and enabling-factor explanations in his
‘‘fits-like-a-glove’’ (FLAG) framework choice of postsecondary education. Similarly,
Holbrook’s (2005a, 2005b) SPI application of an eight-cell value typology, extrinsic–
intrinsic, self–other oriented, and active–reactive provides a causal history of reason
explanation — rich in value interpretation but very narrow in coverage of behavior
explanation.
Enabling-factor explanations refer to the agent’s skill, efforts, opportunities, or to

removed obstacles (see McClure & Hilton, 1997; Turnbull, 1986); these explanations
take it for granted that the agent had an intention (and reasons) to perform the
behavior; what they try to clarify is how it was possible that this behavior was in fact
performed’’ (Malle, 1999, p. 31). Enabling-factor explanations only explain the
action’s occurrence — they cannot be used to explain why the agent formed the
intention in the first place (which is what reason explanations do). For example,
‘‘How come John aced the exam? — He’s a stats whiz.’’ Such enabling-factor
explanations refer to the agent’s skills, efforts, opportunities, or to removed obstacles
(Malle, 1999, p. 31; also McClure & Hilton, 1997; Turnbull, 1986). Gould’s (1991)
eight-cell typology of energy states: tense–calm, energized–tired, absorbed–not
absorbed is an enabling-factor explanations framework.
Malle (2004) contends that two broad motivations for explaining behavior exist:
finding meaning (i.e., sense making; cf., Weick, 1995) and managing social
interaction. These two broad motivations correspond to the two forms in which
explanations exist in the world: as cognitive representatives and as communicative
acts. Strong self-serving biases tend to occur in behavior explanations that favor self-
fabrication over self-revelation (Wilson, 2002). On one level of analysis (type of
behavior explained), explanations of actions are most likely to enhance or diminish
one’s self-image because they are observable, and hence accessible for evaluation by
others, and intentional (thus fully subject to either praise or blame). Conversely,
experiences are unobservable (thus difficult to evaluate by others) and unintentional
(hence easier to excuse from responsibility), so explanations of experiences should be
at least susceptible to self-serving biases. On another level of analysis (mode of
explanation), self-serving actors should explain their positive behaviors with reasons
(implying intentionality) and their negative behaviors with causes (implying lack of
intentionality) because intentionality intensifies praise and blame. On a third level of
analysis, citing situation causes for negative behaviors reduces blame and citing
person causes increases blame, both when using cause explanations and when using
enabling-factor explanations (see Malle, 1999; Malle & Bennett, 1998; Weiner, 1995).
Examining the preconditions (e.g., meta-thinking by the actor of her perceived
intentionality of a behavior — ‘‘did I really intend to do that?’’) affects which
behavior explanation category most likely applies to explain behavior. If the self-
informant perceives that the behavior was unintentional, situation/person cause
explanations most often occur. If the perception is that the behavior was intentional,
belief/desire reason explanations most frequently occur and CHR and EF
explanations also occur but with less frequency.
Alternative framing of the behavior explanation issue affects the category applied
in SPIs. ‘‘How was this possible?’’ is a frame that increases EF explanations that
otherwise are rare when the framing is a motivationally biased question — ‘‘Why?’’ or
‘‘What for?’’ Malle et al. (1998) supports this hypothesis, finding that enabling-factor
explanations occur 4–12 times more frequently in response to a ‘‘How possible?’’
question than in response to any other explanatory question. Moreover, enabling-
factor explanations should be frequent when the behavior is difficult to perform but
rare when the behavior is easy to accomplish (McClure & Hilton, 1997; Malle, 1999).
Figure 1 illustrates how different ways of framing issues influence behavior

explanations. For example, several researchers (Becker, 1998; Thompson et al., 1989;
Woodside, 2004a) caution against using ‘‘Why?’’ framing questions for several reasons;
for example, the question implies intention when the behavior occurred unintention-
ally; informants, even self-informants, become defensive because ‘‘Why?’’ requires an
answer that makes sense, one that does not reveal logical flaws and inconsistencies;
‘‘How?’’ questions give people more leeway, are less constraining, and encourage
telling a story that includes a chain of events, thinking, and evaluating. However,
Becker (1987, p. 60) emphasizes an important exception to his condemnation of ‘‘why’’
questions: sometimes the self-informant or other researcher wants to know, exactly,
what kinds of reasons the informant gives for what he or she has done or thinks he or
she might do as part of a description that guides thinking. But, in his study
investigating choice of postsecondary education, Allen (2002) provides examples of
informants finding great difficulty in answering ‘‘Why’’ questions but their difficulty
vanishes quickly when providing in situ descriptions regarding ‘‘who, what, where,
when’’ issues.
Malle (2004, p. 122) emphasizes that the influence of explanatory choices can be
found at every level of analysis, ‘‘People increase their use of causal history
explanations when accounting for negative actions (Nelson & Malle, 2003); they
increase their use of belief reasons when trying to appear rational (Malle et al., 2000),
and they explicitly add a mental marker to their belief reasons when they want to
distance themselves from the agent [or situation] (e.g., ‘‘Why is he looking at
apartments?’’ — ‘‘He thinks I am moving in with him’’; Malle et al., 2000).’’ The use
of mental state markers on reasons occurs in introspection reports, for example,
when a person distances current belief or want from reports of belief and want in the
prior situation that the researcher–informant is describing (e.g., ‘‘I searched online
because I thought I could get a better deal’’) versus not marking the reason (e.g.,
‘‘I searched on-line to get a better deal’’).
Note that Figure 1 includes a feedback relationship between belief and desire
reasons to describe the finding that valuing occurs automatically from a belief and that
humans are culturally trained to perceive and categorize information, and form beliefs,
based on evaluations — judgments of familiar and acceptable behaviors, desires, and
dislikes. Valuing is finding benefit from achieving the goal implied by a belief.
What should be done to overcome framing biases? A useful initial answer:
Because explicating descriptive details may help reduce the influence of the
fundamental attribution error and aid in uncovering the if–then contingencies in
behavior explanations (see Woodside & Wilson, 2003), self-informant behavior
explaining benefits from first asking ‘‘what, how, where, and when’’ questions before
asking ‘‘who and why’’ questions.
A useful follow-up answer: Let us recognize the need to apply multiple framing
questions for the same behavior–explanation issue because all questions include
biases (e.g., see Clark & Schober, 1992). A third answer: Because behavior
explanation includes both unconscious and conscious intentional and unintentional
thinking that the self-informant is unable to uncover adequately from any one
inquiry method, SPI benefits from employing mixed-method designs for confirming
Frame
What? When?
Who? Why? How?
Where?
Implies causal Implies belief / Implies enabling

intentionality desired intentionality intentionality Implies un-intentionality
“Stuff happens.”
V Situation
Offer situational Offer Offer
A enabling
CHR belief L
desire Offers cause: Actor
factor
reasons U reasons Mentions factors present
I that caused the behavior
N Person
Offer personal G
CHR Marker? Marker? enabling
factor Situation cause:
conscious/
Unconscious / unconscious
Trait CHR conscious knowledge Actor’s desired Trait enabling
(e.g., informant (actor) Of what the actor Outcomes; often factor Person cause:
commits act finds to be true called goal, aim, (e.g., able to do it conscious/
because she is (e.g., “I thought…”) End, or purpose Because I is a unconscious
friendly) (e.g., “I wanted professional
to…”) basketball player)
Trait cause (e.g.,
High/low self-esteem)
Key: CHR = causal history reason; valuing = positive or negative attitude toward the reason;
marker = mental state verb used to distance oneself from statement
Figure 1: Framing questions’ influence on directing subjective personal introspection in explaining behavior. Source: Based, in
Overcoming the Illusion of Conscious Will and Self-Fabrication
part, on Figure 5.1 in Malle (2004, p. 119).

137
events and personal meanings and identifying the applicability of explicit versus
implicit meanings and attitudes for specific behaviors.
Mixed Methodology in Subjective Personal Introspection

Gladwell (2005, pp. 81–84) dramatically illustrates the use of a mixed-method design
for (dis)confirming meanings expressed in his explicit SPI. He reports,
I’ve taken the Race IAT [available at www.implicit.harvard.edu] on many
occasions, and the result always leaves me feeling a bit creepy. At the beginning
of the test, you are asked what your attitudes towards blacks and whites are.
I answered, as I am sure most of you would, that I think of the races as equal.
Then comes the [implicit association] test. You’re encouraged to complete it
quickly y I took the test a second time, and then a third time, and then a fourth
time, hoping that the awful feeling of bias would go away. It made no difference.
It turns out that more than 80 percent of all those who have ever taken the test end
up having pro white associations, meaning that it takes them measurably longer to
complete answers when they are required to put good words into the Black category
than when they are required to link bad things with black people. I didn’t do quite so
badly. On the Race IAT, I was rated as having a ‘‘moderate automatic preference for
whites.’’ But then again, I’m half black. (My mother is Jamaican.)
As Bargh et al. (1996), McClelland et al. (1992), and Wilson (2002) stress and
empirically support, the most important point being made here is that explicit
attitude, motive, and meaning self-report measures do not correspond to implicit
measures (e.g., IAT, TAT, and FMET — forced-metaphor elicitation technique,
Woodside, 2004b).
Gould (2005) mistakes Woodside (2004b) as seeing ‘‘one of the problems with
introspection as being its subjectivity and the need to rein that in with various
confirmatory approaches.’’ Woodside (2004b) focuses on broadening SPI — not
trying to rein it in — by making the case for combining explicit and implicit methods
in SPI reports.
Applying the useful proposition that research to learn explicit or implicit
meanings held by an informant applies to SPI, Figure 2 is a typology of research
methods that provides insights for designing mixed-method designs and for building
SPI theory that covers both explicit and implicit meaning. The typology includes
the eight possible combinations of verbal versus nonverbal, explicit versus implicit,
and positivistic versus interpretative methods; examples of each of the eight, and
combinations of using two of each, are available in the social, psychological, and
consumer research literatures.
Cell 1, the fixed-point survey method, is the dominant research method for most
studies attempting to measure informant-held meaning. Major advantages of this
method include ease of data collection of explicit meanings held individually for a
large number of informants and the applicability of statistical hypothesis testing.
Limitations of this method include the failure to recognize that most thinking is done
unconsciously and that implicit thinking may differ substantially from explicit
Principal Method
Positivistic Interpretive
2.
1.
Existential-
Explicit Fixed-point
phenomenological
surveys
reports
Verbal
3.
4.
Implicit Automatic
TAT;
thought
FMET
Type of retrievals
Responses
5. 6.
Direct observation Direct observation
Explicit frequency Meaning reports;
recordings ZMET
Nonverbal 7. 8.
Implicit Implicit Behavioral drama
Association test enactments
Key: TAT = thematic apperception test;

FMET = forced metaphor elicitation technique;
ZMET = Zaltman method elicitation technique
Figure 2: Measuring informant-held explicit and implicit associations. Note: TAT,

Thematic apperception test; FMET, forced metaphor elicitation technique; ZMET,
Zaltman method elicitation technique.
thinking, along with social-desirability biases in answering questions explicitly (see

Fisher, 1993).
Cell 2, existential–phenomenological reports (see Thompson et al., 1989), includes
the researcher starting with a very broad question and allowing the informant to
reframe and pose additional questions as well as answers. Such interviews provide
thick descriptions of situations, thinking processes, and some amount of both explicit
and implicit meanings held by the informant. Self-editing of responses before the
informant shares them with the researcher likely limits the value of the method along
with the inability of the informant to retrieve a substantial share of implicit meaning
held in memory. However, especially when the practice includes multiple interviews
over several weeks with the same informant, the method offers the advantage of self-
revelation of implicit meanings the informant rarely becomes aware of when
completing fixed-point surveys (e.g., see Cox, 1967).
Cell 3, automatic thought-retrieval research, includes asking what benefits and
beliefs evoke which brands (e.g., see Thelen & Woodside, 1997). The method is useful
for measuring implicit meanings associated as well as not associated with a given
brand versus competing brands. The method has the advantages of limiting the
occurrence of self-desirability bias, the collection of implicit data for a large number
of informants, and applicability of statistical hypothesis testing. Limitations include

being unable to uncover the causal history reasons that support the automatic brand
retrievals.
Cell 4 includes TAT (see McClelland et al., 1992), and the FMET (see Chapter 7
in the present book). McClelland provides compelling evidence supporting the
external validity of TAT data and offers mixed-method reports comparing explicit
and implicit informant-held meanings. Cell 4 methods often require extensive
training of researchers and often substantial effort in analyzing data.
Cell 5, direct observation-based frequency recordings, includes researcher
assumptions about the implicit meanings held by consumers. For example, Wells
and Losciuto (1966) inferred by direct observation that some consumers examined
packages to learn price information and others did not. The method has all the
advantages associating with being present and not relying on an informant’s
memory. Playing back video or tape recordings of consumers in naturally occurring
situations to learn their explicit interpretations of their own behavior represent a
mixed-method design (e.g., see Taylor & Woodside, 1981). Like all interpretive
methods Cell 5 requires very substantial effort for both data collection and analysis.
Cell 6, direct observation meaning reports, includes the informant creating a
tangible expression of meanings and interpretations that he or she wishes or is able to
share with the researcher. The collages from magazine and newspaper images that
Zaltman’s informants create as part of the ZMET are examples of nonverbal data.
The Doyle and Sims (2002) advances in the cognitive sculpting technique provide
another example. Data collection usually extends to include both etic (researcher)
and emic (informant) explicit interpretations of the nonverbal expressions; thus, a
mixed-method design results from actual applications of Cell 6 methods.
Cell 7 includes IAT reports (e.g., Brunel et al., 2004; Masion, Greenwald, & Bruin,
2004). Brunel and his colleagues (2004) demonstrate the value of collecting both
explicit and implicit meanings held by informants — applying the mixed-method
strategy of combining Cells 1 and 7. Gladwell’s (2005) commentary of his own prior
explicit-survey and IAT-implicit associations represents a three-method design: Cells
1, 2, and 7.
Cell 8, behavioral drama enactments, includes videotaping and interpreting
informants’ enactments of themselves as inanimate objects, for example, ‘‘using
facial and body gestures and motions with any available props that you might care to
use, describe yourself as a typewriter (or dog, cat, sports utility vehicle, ice cream)’’;
for examples, see Dichter (1964). To expand on etic interpretations of informant
enactments, asking informants to observe videotapes and interpret their own
performances offers the opportunity of expanding from a single to a mixed-method
design, for example, combining Cell 8 with Cell 2.
Conclusions and Implications for Theory and Research
SPI is pervasive in consumer research — answering questions asked by ourselves or

others requires SPI. SPI researchers (e.g., Gould and Holbrook) need encouragement
to look beyond single-method applications to mixed-method designs, as Woodside

(2004b) illustrates in SPI reporting. This point is extremely important for building
and testing scientific theory, building in safeguards to protect ourselves from our own
biases and self-fabrications when making decisions affecting ourselves and others, as
well as increasing mindfulness (Gaither, 2002; Kozak, 1996; Weed, 1991; Weick &
Sutcliffe, 2001) while reducing arrogance.
Along with ignoring the multiple benefits of using confirmable introspection
methods, the main point Gould (2005) suggests is inaccurate: ‘‘But there is no
member checking the meaning a person assigns in the story. This is her own
introspective story.’’ Own explicit self-reports often do not express the same implicit
meaning held by the same informant — our own implicit introspection stories often
differ dramatically from own explicit storytelling.
Single-method SPI reporting and criticism of calls for using multiple method,
explicit–implicit tools in SPI studies illustrate more than the fundamental attribution
error and overconfidence bias. Such criticism is illustrative of not doing the
homework, confirming that humans are incapable of explicitly fully uncovering their
own-held meaning without resorting to mixed-method explicit–implicit research
designs. Fortunately, the work of Bargh et al. (1996), Brunel et al. (2004), McClelland
et al. (1992), Wegner (2002), Wilson (2002), Woodside (2004b), and Zaltman (2003)
advance SPI research methodology beyond single-method explications of meaning.
The work of Allen (2002), Hilton (1990), Malle (1999, 2004), McClure and Hilton
(1997), Wegner (2002), and Wilson (2002) advances behavior explanation research
well beyond person–situation attribution theory (Kelley, 1967, 1973) and person-trait
only typologies (e.g., Gould, 1991; Holbrook, 2005a, 2005b). Thus, the future is bright
and the possibilities are profound for advancing SPI in consumer research.
Chapter 7
Using the Forced Metaphor-Elicitation

Technique (FMET) in Subjective Personal
Introspections about Self
Synopsis
Chapter 7 describes research tools that permit zoomorphistic explications of

self-viewing of human self-behavior in terms of the behavior of animals.
Transference theory, archetypal, culture, and early experiences propositions
also serve to inform the etic interpretations of an informant’s zoomorphistic
self-report. The chapter describes applications of the forced metaphor-
elicitation technique (FMET) that provides case study data including story-
telling and paradox resolution by informants. The chapter closes by advocating
acceptance of Gigerenzer’s proposal that method can drive theory advance-
ment. The discussion reviews relevant literature on examining dual thinking
processes by humans — implicit and explicit beliefs, attitudes, decision
processes, and behavior. The research evidence helps to decode consumers’
implicit thinking and behavior toward products and brands; such evidence
serves to inform ourselves and brand executives of consumers’ dreams about
brands and how such dreams become reality — or what prevents consumers
from buying the brands playing roles in consumers’ stories crafted through
implicit thinking.
Metaphor is everywhere in the language we use and there is no escape

from it y . But more important than their ubiquity, the metaphors we
use structure our thinking, hiding some features of the phenomena we
apply them to, and highlighting others. (Goatly, 1997)
Introduction: Meta-Thinking or Thinking about Thinking
Using metaphors assists in explaining, interpreting, categorizing, and summarizing

a behavior, person, product, brand, or other phenomenon. Goatly (1997) reviews
the strong evidence that metaphor and the mental processes metaphors entail are
basic to language and cognition. Zaltman (2003), Coulter, Zaltman, and Coulter
(2001), and Zaltman and Coulter (1995) shows how metaphors structure consumer
thinking and serve to reveal unconscious thinking. Chapter 7 describes the use of a
forced metaphor-elicitation technique (FMET) to help surface implicit (i.e.,

unconscious) beliefs and feelings about self and to learn how such implicit beliefs
and feelings associate with product designs and brand choices (see Woodside, 2004b,
2006; Woodside, Floyd-Finch, & Wilson,.1986). Part of Woodside’s (2004)
application of FMET includes zoomorphistic explications of self. Definitions of
zoomorphism include the viewing of human behavior in terms of the behavior of
animals (http://en.wikipedia.org/wiki/Zoomorphism).
The present study supports and extends the conclusion of prior work (Woodside,
2004a, 2004b) that most informants can easily identify the animal that first-comes-to-
mind ‘‘that represents some aspects of who you are, what you are like,’’ as well as the
‘‘animal that first-comes-to-mind that you admire, might select to be if you were an
animal other than a human.’’ Self-reports of the characteristics and behaviors of these
animals enable the surfacing of product-design configurations and specific brands that
appear to match the requirements/needs that such characteristics and behaviors reflect.
The present chapter includes a ‘‘confirmatory subjective personal introspection (CSPI)’’
(see Chapter 5) that presents two sides of the author: Arch Rhinoceros and Arch Eagle.
Prior work on subjective personal introspection (SPI) (Gould, 1995; Holbrook, 1986,
1995, 2005a, 2005b) and the dual systems perspective toward explicit and implicit
thinking (Wilson, 2002) inspire the research program that this chapter reports.
Defining Metaphor
One-word definitions for metaphor include symbol, figure-of-speech, simile, image,

and allegory. An allegory is a parable, fable, symbol, story, or tale. More formally, a
metaphor occurs when a unit of discourse is used to refer unconventionally to an
object, process, concept, or colligates (co-occurrence of words) in an unconventional
way (Goatly, 1997). Goatly (1997) uses certain terms to correspond to certain phrases
in this definition:
The conventional referent of the unit is the vehicle.

The actual unconventional referent is the topic.
The similarities and/or analogies involved are the grounds.
Consider ‘‘Barq’s [root beer] — has bite.’’ The vehicle is Barq’s; the
unconventional referent (topic) is bite by a dog (see a dog with a can of Barq’s for
snout with teeth made of jagged cuts in the side of the can at http://www.barqs.com/
index.jsp); the grounds are the physical and emotional response by a human when bit
by Barq’s or a dog — Barq’s is one of the few root beers that contains caffeine.
The Adaptive Unconscious
The term ‘‘adaptive unconscious’’ conveys the proposition that nonconscious

thinking is an evolutionary adaptation: the ability to size up our environments,
Using FMET in Subjective Personal Introspections about Self 145
disambiguate them, interpret them, and initiate behavior quickly and nonconsciously
confers a survival advantage and thus was selected for. Without these nonconscious
processes, we would have a very difficult time navigating through the world (much
less standing up without constant attention y. This is not to say that nonconscious
thinking always leads to accurate judgments, but on balance it is vital to our survival
(Wilson, 2002).
Zaltman (2003) emphasizes that most thinking is done unconsciously and that
unique research tools are necessary to reveal an informant’s unconscious thinking to
self and to the researcher. The literature on unconscious thinking is extensive and
supports the conclusion that mental processes that are inaccessible to consciousness
but that influence judgment, feelings, and behavior substantially do occur. Also, most
thinking involves processes that are nonconscious, fast, unintentional, uncontrollable,
and effortless; automaticity represents unconscious thinking that satisfies all or most
of these criteria (Bargh & Chartrand, 1999; Wegner, 2002; Wilson, 2002).
Figure 1 summarizes literature relevant to providing a theoretical foundation for
the study of unconscious thinking and its influence on behavior. The three
cornerstones of Jung (1969), Hofstede (2003), and Wegner (2002) provide unique
proposals for the origins of unconscious thinking. Wilson (2002) champions the
proposition that implicit and explicit beliefs, attitudes, behavior, and mental
processing are parallel systems that sometimes result in different outcomes (e.g.,
future behavior or interpretations of recent behavior). He reviews a large body of
empirical evidence that supports this view.
The objectives here do not include providing an exposition of the literature
streams that Figure 1 summarizes beyond pointing out that some archetypes (Jung,
1969) are experienced earlier in life (Wegner, 2002) in some cultures (Hofstede, 2003)
than in others — for example, the American cowboy with a horse and gun appearing
on several TV shows weekly in the 1950s in the United States more so than in Sweden
(e.g., Gunsmoke; Paladin, Have Gun, Will Travel; Bonanza). Such a hero myth might
unconsciously enable the implicit acceptance of specific behaviors in one culture
more than another. (See Rapaille, 2006, for his SPI of experiencing the equivalent of
an American cowboy jumping down from a tank during World War II and handing
him a chocolate bar when he was three years old and living in a small French village.)
Thus, the issue is not which foundation theory of unconscious thinking is most
compelling. All contribute valuable insights that increase understanding of how and
why implicit thinking affects beliefs, attitudes, behavior, and mental processes.
Research Methods for Examining Nonconscious Thinking
Such theory and conclusions about unconscious thinking implies that identifying
nonconscious-based needs, beliefs, processes, and actions requires creating research
methods that go beyond SPI. Any context wherein an informant responds verbally to
a direct question involves some amount of SPI. The need to be politically correct,
appear sane, or feel good about ourselves, and exhibiting overconfidence that we are
capable of answering accurately are examples of restrictions of SPI and explicit
Figure 1: Origins of the adaptive unconscious.

responses. The literature is extensive and strongly supports the view that implicit
(mostly conscious) processes run in parallel with explicit (conscious) processes. (For
a review, see Woodside, 2005.) The best-seller Blink reports some of this evidence
(Gladwell, 2005). While the case is made by Wilson (2002), Wegner (2002), and
Rapaille (2006) that implicit versus explicit thinking dominates most beliefs and
actions, the more relevant points here are (1) that thinking and actions include bits
and pieces of both implicit and explicit thinking and (2) that direct questioning alone
serves a poor meal. Collecting explicit responses alone rarely provides enough
sustenance to accurately learn how and why people think and act in different
contexts. Such a conclusion is often shocking to both informants and researchers.
For example — being ‘‘half Black’’ himself — Gladwell (2005) reports being shocked
to discover his own moderately strong (implicit) bias against African Americans
versus his strong (explicit) bias favoring racial equality.
Related to implicit and explicit thinking and knowing, visual and verbal measures
of attitude and beliefs reflect different kinds of knowledge that develop at different
rates. Visual measures tend to tap a nonconscious, implicit type of knowledge,
whereas verbal measures tap a conscious understanding of the theory of mind that
takes longer to develop (Wilson, 2002). (See Malle, 1999, 2004, for a full treatment of
an explicit theory of mind and Wegner, 2002 for an exposition on an implicit theory
of mind.)
A substantial amount of research evidence confirms that implicit and explicit
motives, attitudes, beliefs, and behaviors represent two independent and mostly
unrelated systems that operate in parallel and influence different types of outcomes:
‘‘the adaptive unconscious and the conscious explanatory system each has its
own set of needs and motives that influence different types of behaviors’’ (Wilson,
2002, p. 83).
Relating to consumer behavior, Puccinelli, Braun, and Mast (2002, p. 1) observe,
‘‘While it is possible that explicit and implicit knowledge correspond, the exciting
opportunity for marketers is that often there is a discrepancy; that is, what a
consumer believes explicitly may have no bearing on their [sic] actual behavior’’:
‘‘For example, a consumer may have an association about a product of which they
[sic] are unaware, such as a positive inclination toward purchasing Tide because their
mother had used the brand (even though she might not be able to consciously
remember her mother’s usage)’’ (Puccinelli et al., 2002, p. 1).
The work by Anderson, a social/personality psychologist, and her colleagues
(Anderson & Glassman, 1996; Chen & Anderson, 1999) demonstrates empirically
that a form of transference process occurs outside of awareness and that this process
appears to be an important source of individual differences in how people react to
new acquaintances. Anderson provides a new method to study transference
systematically in controlled experiments and shows that activating chronically
accessible constructs of all sorts influences people’s judgments and behaviors —
without introducing Freud’s interpretation of transference (i.e., the way we
superimpose infantile feelings toward our parents onto new relationships). Thus,
Tide’s repeated presence in early life contexts with a daughter and her mother
transfers implicitly to present-day contexts of buying and using Tide.
Consumer research focusing on implicit awareness, attitudes, beliefs, behavior,

and processing has applied several methods. These methods include Holt’s (1991,
2002, 2003, 2004) anecdotal and case study research on how brands become icons via
storytelling (especially associating a brand with a story of rebellion) rather than
lecturing. The consumer-research literature includes applying the implicit associa-
tion test (IAT) in laboratory experiments (Brunel et al., 2004); the IAT is an implicit,
nonverbal, empirical, positivistic procedure that is available at www.implicit.
harvard.edu. The Zaltman metaphor-elicitation technique (ZMET; Zaltman, 2003;
Zaltman & Coulter, 1995) includes asking an informant to create a tangible
expression of meanings and interpretations that he or she wishes or is able to share
with the researcher.
Here is an example of one study’s procedure using the ZMET to examine
informants’ implicit attitudes toward advertising:
Instructions included in the letter asked informants to bring to the

interview pictures that illustrated their perceptions of the value of
advertising. The instructions indicated that the informant could bring
pictures from magazines, newspapers, pieces of artwork, and/or photos
taken specifically for this assignment or retrieved from photo albums.
We specifically mentioned in the letter that we were interested in their
thoughts and feelings about advertising in general and not about
particular advertisements and that their visual images should be
representative of these general thoughts and feelings.
Furthermore, the images should not be actual print advertisements, according to

the instructions. The ZMET interview employs several steps to bring key metaphors
to the surface and determine their interrelationships, including (l) storytelling,
(2) missed images, (3) Kelly Repertory Grid and laddering, (4) sensory images, (5) the
vignette, and (6) the summary image. Finally, Step 6, the summary image, was
designed to have the informants create a composite of their thoughts and feelings
associated with advertising.
Informants chose entire pictures or specific images from their pictures that they
wanted to include in their summary image. Then a graphic artist scanned the images
and demonstrated the capabilities of the imaging software to the informant.
The informant then instructed the graphic artist regarding the size, color, and
positioning of the pictures in the summary image. When the summary image was
complete, the graphic artist asked the informant to describe the image and its
meaning with regard to his or her overall opinion about advertising. (Coulter et al.,
2001, p. 11)
Woodside (2004a, 2004b) and colleagues (Woodside et al., 1986) propose an
alternative approach for surfacing and examining metaphors that consumers use
implicitly — FMET. The FMET includes four distinct steps that build first on a
zoomorphistic interpretation of self.
The first step asks the informant ‘‘to draw or select pictures of the animal that
first-comes-to-mind that you believe represents some aspects of who you are, what
you are like.’’ Second, the informant is asked to say or write the two-to-four features
that first-come-to-mind that best represent this animal. Third, the FMET asks the
informant to mention ‘‘the thoughts that first come-to-mind about what each of these
features tells about yourself.’’ Finally, the FMET asks the informant to ‘‘tell a story
or true event that illustrates the concepts and description of the features that you just
mentioned about you related to a specific choice behavior under investigation (e.g.,
selecting a type and brand of vehicle to drive).’’ Informants are asked to repeat the
four FMET steps a second time focusing on the animal that first-comes-to-mind
‘‘that you admire, might select to be if you were an animal other than a human.’’
A fifth procedural step includes asking the informant to think aloud the thoughts
that come-to-mind when viewing her/his own answers to the two sets of questions
and to relate a personal history story that reflects the likely outcome following a
comparison of the alternative choice behaviors.
If an animal that is most representative of self does not come to mind readily,
Feinson (1998) provides a nine-item quiz that informs the taker of the ‘‘animal in
you’’ and the resulting behavior that you are likely to observe of yourself. The quiz is
also available at www.animalinyou.com. For example, for the rhinoceros human,
Feinson describes, ‘‘With such a cantankerous personality, your unpopularity is
hardly surprising, and your bullying reputation is carefully cultivated to ensure the
solitude you crave.’’ For the eagle human, ‘‘Sleek and well groomed, your looks are
striking. However, you occasionally have a flawed aspect to your appearance —
perhaps a bald spot or too large a nose — but you still exude strong sexual energy
that turns heads.’’
Applying FMET and Learning Storytelling Resolutions
Figures 2–5 summarize FMET applications for four informants. The first three
informants were college students at Boston College while participating in the study.
Each of the four informants was asked to find a photograph, drawing, or other
illustration of the animal that was most representative of her or himself in real life
and the animal that she or he would want to be other than a human. The informants
completed each of the steps in the FMET procedure as the previous section describes.
Ana as a Small Dog and as a Dolphin

Figure 2 shows informant Ana’s animal-who-I-am visual to be a small dog that is
cute, fun, and friendly. Such characteristics provide clues for a storyline about brands
that seek to associate with Ana’s inner small dog life. One such brand that Ana
identifies is the VW Beetle.
Ana’s animal — who-I-would-be visual is a dolphin — an animal that she
describes as friendly, beautiful, playful, intelligent, and swims fast. Brands likely
would benefit from illustrating these characteristics in attempting to associate with
Ana’s zoomorphistic wish. She identifies the Lexus RX Hybrid as one such brand.
Figure 2: Ana: dolphin and dog.

Figure 3: Bob: Cheetah and giraffe.

Figure 4: Joseph as a tiger and panda bear [sic].

Figure 5: Arch as a rhinoceros and as an eagle.

Ana’s friends’ view that she would disappear when driving a large car confirms her
preference for a small vehicle. Ana admits to not being a very good driver,
information that might be more difficult to acquire by direct questioning.
Bob as a Giraffe and as a Cheetah
Figure 3 shows informant Bob’s animal-who-I-am visual to be a giraffe. ‘‘Tall, lanky,

laid back’’ describe both the giraffe and Bob, according to Bob. A cheetah is the
animal Bob would become; he shows a cheetah moving at full speed along with the
description of fast, strong, and agile. Bob associates the giraffe with long, big, and
not really fast cars — like a Ford Crown Victoria. He associates a fast car with a
Ferrari. The life history story Bob tells suggests problems in attracting Bob to
considering owning the Ford.
The fifth step in using the FMET quickly surfaced Bob’s high school experiences
in driving an earlier model of the Ford Crown Victoria. Ford faces a difficult task in
gaining Bob as a buyer.
Joseph as a Panda Bear and as a Tiger

Figure 4 shows a black bear that informant Joseph identifies as a Panda Bear — the
animal that he most represents. ‘‘Cute, cuddly, strong, and likes eating’’ fit both the
bear and Joseph, according to his self-report. This transference translates into a
Cadillac Escalade. Joseph’s wish to be a tiger associates with wild, beautiful, fierce,
and strong which surfaces a Range Rover. His story indicates that the bear-reality
side wins out over his tiger-wish side.
The concept ‘‘flashy’’ is something that Range Rover might best avoid in trying to
attract customers like Joseph. A question for Cadillac to consider before getting too
comfortable — how many men under 25 make ‘‘a hefty fortune’’ on Wall Street in one
summer? Joseph represents a very atypical customer even if a highly desirable one.
Arch as a Rhinoceros and as an Eagle

Figure 5 shows the rhinoceros that Arch describes as himself. Arch’s zoomorphistic
description includes ‘‘big, grumpy, adorable, and near-sighted.’’ This description
translates into a GMC Suburban — a vehicle Arch reports owning. Feinson (1998)
reports some rather disappointing behaviors that help deepen understanding of this
zoomorphistic treatment:
Getting to know a rhino can be an exhausting business, for it tends to

rely on wordless forms of communication that must then be carefully
interpreted. But, rhinos will let you know in no uncertain terms when
you have crossed its (sic) line. For they do not suffer fools gladly
and often hurt the ones they love. (http://www.animalinyou.com/
rhino.htm)
Arch recognizes his desire to be an eagle: soaring high, seeing far, catching fish
easily, and getting lots done by moving fast. This description translates into his latest
dream car — a BMW 645 CiC. A moving great work-of-art! However, Arch’s story
indicates that his mother will never give permission for such a purchase — even
though she died in 2002. So Arch reports his resolve to transfer his need to soar like
an eagle through his work while continuing to dream about the BMW.
Contributions to Theory
Transference theory regarding how past relationships can influence people’s

perceptions of any new person or object they meet serves to expand from the
zoomorphistic self-interpretations and resulting storytelling to a nonconscious, social
information-processing system — namely, what Wilson (2002) refers to as the
adaptive unconscious. Anderson and Glassman (1996), Sullivan (1953), and Klein
(1952) inform the etic interpretations of the zoomorphistic and brand-related self-
reports. Mental representations of relationships with significant others and animals
are self-relevant and are often used to interpret and evaluate unique contexts:
Because representations of relationships with significant others are self-

relevant and frequently brought-to-mind, they become chronically
accessible and are often used to interpret and evaluate new people
[animals, brands] we meet. In short, just as the construct of ‘‘honesty’’
or ‘‘kindness’’ can be activated and applied to a new person, so can the
construct of a specific person [animal] such as ‘‘my mother’’ or ‘‘Uncle
Henry’’ [or the animal I would be]. (Wilson, 2002, p. 78)
The surfacing of Bob’s and Arch’s mothers is illustrative of Klein’s (1952) and
Heimann’s (1956) treatment of transference and the possibility of unconscious
mental conflicts if not mental illness caused by such unconscious conflicts — the
clashes between instinctual impulses and what is opposed to them. To some extent,
Bob and Arch have taken their mother inside themselves (as most of us do) and the
surfacing of usually held unconscious knowledge affects self-interpretations of
animal brand associations. This application of transference theory suggests that a
fourth foundation corner may be necessary in Figure 1: psychoanalysis of dreams,
with Freud (1923) as the foundation source and Dichter (1960, 1964) as the most
well-known proponent in consumer research.
Archetypal, cultural, and early experiences propositions also serve to inform the
etic interpretations of informant zoomorphistic self-reports.
Such additional theoretical work is one suggestion for future research. The
relevance of Dichter’s (1964) contributions is important even if seemingly fallen from
view in the 21 st Century. In Strategy of Desire (1960), he reports that 70% of 200
informants participating in in-depth interviews admit to dreaming about owning
a convertible; he presents the transference argument that the convertible represents
a mistress while a sedan stands for a wife. His recommendation to the car
manufacturer sponsoring the study was to place a convertible in the showroom
window to bring in customers who would buy sedans — convertibles representing
only 2% of sales.
Dichter never bothered to fully-develop a psychoanalytic model of consumer
motivation. Advances in theory and creative empirical methods on the adaptive
unconscious support his work in transferring psychoanalytic theory to consumer
contexts. Such interdisciplinary research should pick up speed in the 21 st Century
in both theoretical and empirical contributions — a time to soar as an eagle
(cf. Holbrook, 1988, 1995).
Reducing Inbreeding
Gigerenzer’s (2002, p. 295) wisdom is worth repeating in closing this discussion.

‘‘Intellectual inbreeding can block the flow of positive metaphors from one discipline
to another. Neither disciplines nor sub-disciplines are natural categories. Inter-
disciplinary exchange has fueled the development of some of the most influential new
metaphors and theories in the sciences.’’ Thus, the explicit surfacing of metaphors
has application as an interpretive research method and for creating theory.
Gigerenzer (2002) also emphasizes the role of method in creating substantive
theory: Method can drive theory prior to data. Subsequently, data following brilliant
new data-collection methods can serve to support theory. Recent advances in implicit
thinking methods support this spirited view.
Chapter 8
Surfacing Executives Interpretations

of Self and the Roles of Co-Workers
in Enacting Front and Back Stage Strategies$
Synopsis
This chapter examines the topic of internal branding from an organizational/

behavioral science perspective, theoretically and empirically investigating how
organizational members actually enact corporate brands. A mixed method
research procedure serves to surface conscious (i.e., deliberate) and uncon-
scious (i.e., tacit) internal brand meaning enactments in an internationally
operating Austrian corporate business-to-business brand. The results are
evidence of the potential complexity of real-life internal branding processes that
limit the possibility of achieving a cohesive intended internal implementation of
corporate brands. The chapter concludes with the managerial implication that
purposeful managerial interventions necessitate an understanding of the social
system that is the target of an internal branding initiative.
Introduction
Corporate branding implies understanding brands from a corporate identity

perspective, linking brands with corporate values and beliefs. Internal branding is
the discipline dealing with establishing this link. A large body of literature approaches
internal branding from a normative, managerial perspective, proposing that internal
branding efforts should start out with a definition of intended brand meaning (i.e.,
intended values, beliefs, behaviors) or an intended corporate brand identity, that is,
‘‘what does the organization want others to think about the organization’’ (Brown,
Dacin, Pratt, & Whetten, 2006, p. 100). Managerial internal branding efforts aim to
achieve a common understanding, commitment, and behavior among organizational
members, which is in line with this centrally established corporate brand identity
$
This chapter is co-authored with Sylvia von Wallpach and Arch G. Woodside, University of Innsbruck.
(Thomson, de Chernatony, Arganbright, & Sajid, 1999). Internal branding theory

ascribes employees the role of ‘‘brand ambassadors’’ (Vallaster & de Chernatony,
2005, p. 5) which should spread intended brand meaning through every action and
interaction and thereby contribute to the development of a consistent, intended brand
image among all relevant stakeholders. The facilitation of this task is likely to be
high if externally communicated brand values are simultaneously internal guiding
principles — accepted tacitly by organizational members and directing their actions
(LePla & Parker, 1999; Tosti & Stotz, 2001; Urde, 2003).
Few prior studies approach the topic of internal branding from an organizational/
behavioral science view (Burmann & Zeplin, 2005) dealing with how organizational
members actually perceive the organization and the corporate brand, that is, ‘‘who
we are as an organization’’ (Brown et al., 2006, p. 100). The present chapter deals
with this ‘‘perceived organizational identity’’ (Dutton, Dukerich, & Harquail, 1994)
perspective, investigating how organizational members actually interpret and enact
corporate brand values, beliefs, and behaviors.
The organizational study focus for internal branding is not on what should
happen, but on what actually happens in enacting internal brand deliberate and tacit
thinking and behavior. The assumption underlying this approach to internal
branding is that purposeful managerial interventions necessitate an understanding
of the social system that is the target of the internal branding initiative. This chapter
provides insights into the potential complexity of actually implemented, real-life
internal branding processes, which limit the possibility of achieving a cohesive
internal corporate brand perception and consistent brand enactments. Hereunto, the
second section gives a theoretical overview on the social dynamics characterizing
enacted internal branding. The third section introduces a methodology that allows
to access organization members’ emic interpretations of internal brand enactments
and applies this methodology to the case of an Austrian corporate B2 B brand,
demonstrating how the brand is implemented internally. The chapter closes with
a discussion on enacted internal branding and implications for managing corporate
brands internally.
Theory of Enacted Internal Branding
Life behind the Scenes
Theory on internal branding so far mainly takes a normative, interventionist

perspective, focusing on what should be done to achieve consistency between actual
and intended internal enactments of brand values, beliefs, and behaviors (e.g.,
Kotler & Pfoertsch, 2007; Miles & Mangold, 2005; Vallaster & de Chernatony, 2006);
and assuming that this goal can be reached.
Enacted internal branding theory illuminates the life behind the scenes. Following
Goffman’s (1959) idea of backstage processes, enacted internal branding theory
forwards the proposition that there are hidden, real-life internal brand enactments
Surfacing Executives Interpretations of Self and the Roles of Co-Workers 159
that might contradict what management intends for its brand, that is, intended
corporate brand identity aspects management defines front stage. This proposition
is in-line with organizational sensemaking literature, arguing that ‘‘shared meaning
is difficult to attain’’ (Weick, 1995, p. 188) and, ‘‘There is often lack of actual
agreement, which is ignored or assumed away’’ (Brown, 1994, p. 97). One can think
of many instances where front and back stage values, beliefs, and behaviors do not
match. Consider the following example of the British B2 B brand, Gate Gourmet,
where management itself does not live up to the officially stated values, creating
a back stage reality which considerably harmed the brand’s credibility:
Gate Gourmet is the world’s second largest in-flight catering company,

serving clients such as British Airways, American Airways, or
Continental Airlines. In its official, front stage value statement as
published on the corporate website in 2005 (http://www.gategourmet.
com), the corporate brand particularly highlighted and reinforced
the element ‘‘World Class People’’ and stated that ‘‘we treat our
colleagues, customers, and suppliers with respect and dignity’’. In the
same year, company management drastically acted against this value.
Gate Gourmet unilaterally decided to cut wages and conditions
in order to avoid financial losses and refused to negotiate with the
union. The major dispute resulted from dismissing 670 employees at
Heathrow Airport in August 2005 after staff staged a two-day ‘‘illegal’’
walkout — meant to express their indignation with management
practice — and the parallel hiring of lower paid temporary workers.
Managements’ inconsistent back stage enactments led to major image
deteriorations amongst a wide variety of internal and external
stakeholders, including British Airways management and employees,
British Airways passengers, unions and the general public. (Vallaster,
de Chernatony, & von Wallpach, 2006, p. 33)
While this example pictures an extreme case of inconsistent front and back
stage brand enactments provoked by management, other more hidden internal front
and back stage conflicts are likely to exist due to various organizational members’
emic interpretations of front stage values. These interpretations influence both
how organizational members actually enact the brand, that is, how they behave
and interact with other internal stakeholders, and those organizational members’
consequent brand perception and behaviors. The result is a multiplication of
unintended value perceptions and enactments, leading to a variety of back stage
internal, ‘‘‘doppelgänger brand images’ — that is, families of disparaging images and
stories about a brand’’ (Thompson, Rindfleisch, & Arsel, 2006, p. 50) and potential
conflicts and confusion between internal stakeholders.
The consequences of disparaging internal back stage brand images are, however,
not only limited to the internal context. A variety of organizational members is likely
to interact with external stakeholders, such as customers, suppliers, investors, or
media. During those interactions, organizational members represent the brand and
contribute to the development of various external stakeholders’ brand perceptions.

Emic back stage interpretations of brand values and beliefs guide the organizational
member’s behavior in those stakeholder contact situations, resulting in respective
external Doppelgänger brand images. Problems arise if the values and beliefs the
organizational member enacts are in obvious conflict with values other brand
communication channels that are under corporate brand management’s control
officially promote. Imagine a company stating the value ‘‘environmental friendli-
ness’’ front stage in its corporate brand value statement and other communicative
media directly targeting customers, while its sales representatives make customer
visits with a vehicle obviously consuming large amounts of petrol. Inconsistent and
eventually contradictory messages of this kind confuse external stakeholders and
deteriorate the brand’s image and credibility.
In many cases, especially in strongly service-oriented businesses, the contact with
organizational members (e.g., sales representatives or key account managers) is the
most pronounced and important experience external stakeholders have with the
brand, establishing what the brand really means to the stakeholder. This actual
brand image, that is, ‘‘What stakeholders actually think about the organization’’
(Brown et al., 2006, p. 100) might be different from intended brand image but not
necessarily harmful for the brand because stakeholders might not even know or
consider what management intends its brand to be.
According to Bartunek (1993), Goffman (1959), and Stevenson, Bartunek, and
Borgatti (2003), organizational members are mostly aware of contradictions between
front and back stage processes. The theory of enacted internal branding extends this
view and accounts for the fact that organizational member’s interpretations of back
stage enactments are likely to be both conscious and unconscious (Wegner, 2002;
Wilson, 2002). While organizational members’ stories allow investigating conscious
back stage brand enactments and conflicts with front stage values and beliefs, more
sophisticated research methods are needed to uncover unconscious brand enactments
and related conflicts.
Origins of Discrepancies between Actual and Intended Internal Brand Enactments
Organizational sensemaking literature forwards several reasons for heterogeneity in

organizational members’ sensemaking that can be applied to realized internal branding.
One major reason for different brand enactments are organizational members’ own
identity needs (Coopey et al., 1997). The organization an individual is a member of can
affect the individual’s own identity. When organizational members strongly identify
themselves with the organization, the attributes they use to define the organization also
define them (Dutton et al., 1994). Individuals therefore tend to interpret organizational
values, beliefs, and behaviors in ways that best suit their individual identity generation
or maintenance. On the one hand, individuals perform this identity generation effort
for themselves in order to ‘‘make things rationally accountable to themselves’’ (Weick,
1993, p. 635). Through their emic, idiosyncratic interpretations of brand values people
can model their autobiography, attributing values and favorable outcomes to
themselves, thereby raising their own self-esteem and their individually perceived
importance for the corporate brand — a self-serving tendency referred to as
attributional egotism (Bettman & Weitz, 1983; Brown & Jones, 1998).
On the other hand, organizational members enact brand values and beliefs in ways
that influence the opinion others have of them, that is, their company internal
personal image — a self-presentation behavior known as impression management
(Schlenker, 1980). In order to maintain a desired image and to please their in-groups,
organizational members adopt group interpretations of brand values as their own,
emic interpretations. Maintaining one’s own identity might also require generating
the identities of other social actors by interpreting their internal brand enactments:
actors interpret the enactments of those individual they are interacting with most in
ways that support their own brand enactments. Both individually and socially
focused identity generation can lead to back stage interpretations of brand values
and beliefs that are different from and eventually conflicting with front stage
intended brand values.
Personal experiences outside and inside the organization can be further sources
for discrepancies between front and back stage internal brand enactments. All
individuals enter an organization with some prior interpretation of what are desirable
values, beliefs, and behaviors. This is the result of former contacts with other social
groups (e.g., family, school, or other organizations) or their national cultural
background: ‘‘The individual’s system of values is itself a product of his experiences,
and especially of his experiences in the social environment y As an adult he comes
to adopt the values and norms of the group with which he identifies himself. This
development of values and the adoption of norms constitute the process of
socialization’’ (Kelvin, 1971, p. 232).
Individuals go through this process several times throughout their lives, whenever
they get in contact with a new social group. When entering a new organization,
individuals undergo tertiary or organizational socialization, ‘‘the process of ‘learning
the ropes’, y of being indoctrinated and trained, y of being taught what is
important in an organization or some subunit thereof’’ (Schein, 1971, p. 210, quote
marks in the original). Ideally, this process leads to the accommodation of individual’s
believes with front stage defined corporate brand values. However, this ideal reaction
is only one possible outcome: the newcomer might as well adopt a back stage view of
corporate brand values after experiencing the actually enacted values, beliefs, and
behaviors within the corporation (cf., Cooper-Thomas, van Vianen, & Anderson,
2004). Depending on the position, the new organizational member adopts or the point
of time he enters the organization, he will, for instance, get in contact with different
formal and informal social groups within the organization or different leaders,
influencing the individual’s ultimate emic interpretation of internal brand enactments.
Finally, an organization member’s actual interpretation of brand enactments
depends on his or her former contacts with branding events within the organization.
Branding communication might, for instance, not reach every organizational
member with the same intensity or might be interpreted differently as a result of
incomplete or inaccurate individual information processing (Dearborn & Simon,
1958; Hedberg, 1981). Again, conflicts between front and back stage values systems
can be the result.
Conflict Neutralizing System Inherent Dynamics
The majority of meaning conflicts is likely to happen unconsciously and to remain

unresolved because of missing awareness (Wilson, 2002). Conscious conflicts,
however, represent unbalanced psychological states producing anxiety and inducing
a wish of conflict resolution in organizational members. In line with Bartunek (1993)
and Goffman (1959), organizational members contribute themselves actively
in resolving contradictions they are aware of. The degree of involvement in these
conflict-solving activities varies. Some organizational members are very active in
avoiding or diminishing front and back stage as well as various back stage brand
enactment conflicts regarding themselves but also other internal stakeholders. These
stakeholders are important brand influencers, who can have both a positive and
negative impact on intended brand enactments: they can either strive for a conflict
resolution that strengthens intended brand values and beliefs, taking on the role of
internal brand ambassadors; or they can aim to spread an alternative, back stage
brand meaning.
Most organizational members are, however, not that active in solving conflicts.
On the one hand, organizational members might not be willing to engage in conflict
resolution because of individual identity generation needs. These needs can
necessitate the maintenance of a value and belief system that contradicts front stage
values while supporting how the organizational member wants to perceive himself or
wants to be perceived by others. On the other hand, executives might simply not have
the power and miss the necessary resources to induce change — especially if the
contradictions do not concern themselves but other executives (Giddens, 1979, 1984).
Still, organizational members may take the necessary steps to avoid major
conflicts they are aware of and that would harm their own position in the company,
typically aiming to find some compromise or working agreement on how to enact the
brand internally. Even though resulting actual internal brand enactments might not
be intended by corporate brand management, they do not necessarily have to harm
the brand but might be what the brand actually represents for most internal and
external stakeholders (Figure 1).
Practice of Enacted Internal Branding
The third section gives insights into the practice of enacted internal brand, proposing
a mixed method research procedure that allows surfacing actual conscious as well as
unconscious internal brand enactments. The present study applies this research
procedure to identify actually enacted internal brand values, beliefs, and behaviors
in a business to business company. This single case approach is common practice in
organization studies and accounts for the fact that enacted internal branding
is a context specific phenomenon that cannot be understood in isolation, that is,
‘‘Apart from the perceptions of the actors and the mindsets and cultures of the
contexts in which they are embedded’’ (Langley et al., 1995, p. 261).
Facilitators Community Governments
E.g. E.g. E.g.

• Banking • Local • Local
• Consultants • Industry • National
Front Stage
Suppliers Customers
Values1 Beliefs1 Behaviors1
Line of external visibility

Line of consciousness

Backstage
Key: = agreement, harmony = contradiction, inconsistency
Figure 1: Implemented internal branding theory.
The Case
The case-organization is an internationally operating Austrian business to business

company, producing high-tech materials used as components in public transport,
industrial, and construction applications and selling all of its products and services
worldwide under the corporate brand name. The company’s original scope of
activities was the production of artificial leather products, which then shifted to
the manufacturing of high-tech materials in 1975. In the mid-1980s, the company
founded its first subsidiary in Germany and began to develop its European
distribution network. In the 1990s, the company opened offices in Amman, Jordan
and in Berlin, Germany, and intensified its distribution activities in Asia, South
America, and in Europe.
At the end of the 1990s, the company underwent a major change in leadership,
bringing a new value system and the introduction of a process organization with
four main processes (resources including the CEO, human resources, organizational
development and controlling, innovation, sales and production) with it, which
required a major restructuration. In 2005, a strategic realignment at the global level
took place, dividing the enterprise into business units in order to bundle its know–
how and expand its services to provide individualized solutions to its customers.
Today, the company is world market leader in its field, employing a workforce of 232
people, with a staff of some 172 at the Austrian headquarters. The company booked
sales of EUR 46.3 million in 2007, with exports accounting for some 83 percent of
sales. In May 2008, the company opened its own production facilities in China.
In 2007, the new CEO approached one of the two authors, asking for support in
surfacing actually enacted corporate brand values, beliefs, and behaviors within his
company. Planning for the study included agreement with the CEO that the main
focus of the project should be on organizational member’s emic interpretations of
corporate brand value enactments. For that purpose, a sample of 35 (out of 150)
Austrian employees was chosen to cover all resource processes, departments
and hierarchical levels at the headquarters (see Figure 2 for an overview on the
departments and hierarchical levels included in this study). The 35 organizational
members participated in an in-depth interview regarding the principles guiding work
in the company. All interviews took place in meeting rooms at the headquarters or —
if executives had a private office — in the executives’ respective offices. Interviews
lasted between 45 minutes and 2.5 hours. All interviews were recorded and verbally
transcribed, resulting in 530 pages (single spaced, Times New Roman, pt. 12) of
transcripts. The next section gives insights into the research procedure used to surface
internal brand enactments.
A Mixed Method Research Procedure to Surface Internal Brand Enactments
A main challenge in surfacing internal brand enactments is the fact that

organizational members are not always consciously aware of their own enactments
(cf., Wilson, 2002). Zaltman (2003) stresses that most thinking is done unconsciously
and researchers need unique tools to make unconscious knowledge accessible to both
the individual and the researcher. In order to detect both conscious and unconscious
brand enactments in organizational members’ emic interpretations, the present
study applies a mixed method approach (Woodside, 2004b, 2006), combining
storytelling about the organizational members’ immediate work environment with
cognitive mapping (Huff & Jenkins, 2002) and metaphor elicitation (Woodside,
2004a, 2004b, 2006; Zaltman & Zaltman, 2008). This combination of research
methods has several further advantages. First, the application of multiple research
methods leads to participants’ ‘‘confirmatory personal introspection’’ (Woodside,
2004a, 2004b), that is, to the confirmation of their own interpretations via multiple
research methods. Additionally, interpretations are cross-checked by other organiza-
tional members reporting their own views of actual internal brand enactments.
Second, various forms of knowledge representation and preferences with regard to
knowledge retrieval (verbal versus visual) can be accounted for. Finally, the methods
were chosen to allow individuals slowly approaching topics that might be socially
less acceptable, such as sensitive information regarding personal relationships and
interactions within their work environment.
In a first step, organizational members told stories about their time in the
company. The main assumption underlying the application of storytelling as a
research tool is that stories are the most natural way for human beings to acquire,
store, and retrieve knowledge (Bruner, 1991; Hiltunen, 2002; McKee, 2003; Mink,
1978; Weick, 1995; Woodside et al., 2008). In an organizational context, storytelling
plays a pivotal role. By exchanging stories, organizational members are socialized
and learn to distinguish between valued as well as unacceptable beliefs and behaviors
Executive
Board
CEO
Head
Head
Head Head Head Germany Head Head
Marketing,
Controlling HR Sales (not Production Innovation
Org. Dev.
interviewed)
Employee. Employee Employee Employee
Head Head Head Subsidiaries

Competence Competence Competence Competence Competence Competence Competence
Business Business Business (not
Team Leader Team Leader Team Leader Team Leader Team Leader Team Leader Team Leader
Unit 1 Unit 2 Unit 3 Interviewed)
Employee. Employee. Employee

Employee
Employee. Employee. Employee. Employee. Employee. Employee.
Employee. Employee.
Employee. Employee.
Figure 2: Sample overview: Departments and hierarchical levels included in the empirical study.
Surfacing Executives Interpretations of Self and the Roles of Co-Workers
165
(Allan, Fairclough, & Heinzen, 2002; Kelvin, 1971; Schein, 1971). Stories are also the
foundation for individual sensemaking and the way organizational members assign
meaning to the organizational environment surrounding them (Brown, Stacey, &
Nandhakumar, 2008). Organizational members might therefore store much
information regarding internal brand meaning enactments as stories in episodic
memory, containing verbal and non-verbal as well as conscious and unconscious
knowledge elements (Tulving, 1972, 2002). The most natural way to access episodic
knowledge and to gain emic interpretations of internal brand enactments is asking
organizational members to reproduce these stories. While storytelling should allow
gaining deep insights into organizational members’ social realities and internal brand
meaning enactments, the interpretations verbal, oral storytelling retrieves are
predominantly conscious/verbal, neglecting alternative ways of knowledge represen-
tation and retrieval.
In order to access eventually unconscious, non-verbal interpretations of internal
brand enactments participants had to draw a cognitive map depicting their
immediate social work environment and to comment their map with a story.
Cognitive maps are ‘‘visual representations that establish a landscape, or domain,
name the most important entities that exist within that domain, and simultaneously
place them within two or more relationships’’ (Huff & Jenkins, 2002, p. 2). The maps
are not direct personal representations of cognitive processes but ‘‘intermediate tools
that facilitate the discussion of cognitive processes that can never be directly
observed’’ (Eden, Jones, Sims, & Smithin, 1979, in Huff & Jenkins, 2002, p. 1);
mental processes that include tacit knowledge that is hard to verbalize. Cognitive
mapping is a much applied research method in organizational contexts, encouraging
organizational members to share their idiosyncratic sensemaking of organizational
reality, giving insights into actual values, beliefs, and behaviors guiding behaviors in
their immediate social context (e.g., Howard & Morgenroth, 1968; Huff & Jenkins,
2002; Woodside & Samuel, 1981).
Cognitive mapping is effective in eliciting implicit (i.e., tacit or unconscious)
knowledge ‘‘when coupled with techniques such as metaphors and storytelling,
which have been argued to help express the inexpressible’’ (Ambrosini & Bowman,
2002, p. 19). To support organizational members in accessing the most hidden
interpretations regarding internal brand enactments and crossing both the line
of unconsciousness and social desirability, storytelling and cognitive mapping
was completed by a metaphor elicitation task. Metaphors are common to human
cognition and language and allow ‘‘understanding and experiencing one kind of
thing in terms of another’’ (Lakoff & Johnson, 1980, p. 5). The main advantages
of metaphors are that they ‘‘give tacit knowledge a voice’’ (Munby, 1986, p. 198) and
allow organization members to avoid directly talking about sensitive information
regarding other social actors and their relationships.
Several metaphor elicitation techniques are known in marketing and organization
science, such as cognitive sculpting (Sims & Doyle, 1995; Doyle & Sims, 2002) or the
Zaltman metaphor elicitation technique (ZMET) (Zaltman, 2003). The present
study applies an extension of the forced-metaphor elicitation technique (FMET)
by Woodside (2004a, 2004b, 2008c) to surface organization members’ unconscious
internal branding values, beliefs, and behaviors. Building on Woodside’s (2004a,

2004b, 2008a, 2008b, 2008c) conclusion that individuals have no difficulty giving
zoomorphistic explications of self, that is, identifying themselves with animals and
that many personality traits and behaviors of humans can be easily transferred to
animals and vice versa (Feinson, 1998), participants were asked to identify
themselves and other social actors depicted in their cognitive maps with animals
and to support their metaphors with stories explaining why single animals represent
respective human beings.
Data Analysis Procedure
The research team, including the first author and a research assistant, applied
hermeneutic text analysis to analyze the organizational members’ stories. Hermeneutic
text analysis considers both language (that is the semiotic-structural characteristics of
the text, considering syntactic and semantic text elements) and content of a text as
potential loci of meaning (Arnold & Fischer, 1994; Thompson, 1997). Researchers
move back and forth between these two levels of analysis — a process referred to as
‘‘hermeneutic circle’’ (Thompson, Pollio, & Locander, 1994) — aiming to ‘‘achieve
an understanding free of contradictions’’ (Arnold & Fischer, 1994, p. 63). Still, two
researchers separately interpreting a text might come to different conclusions because
of the researchers’ different pre-understandings. This variability of researcher
interpretations is perceived as fruitful because it allows discerning multiple meanings
texts might eventually contain (Arnold & Fischer, 1994). The analysis of cognitive
maps and zoomorphistic metaphors was performed by organizational members
themselves, who had the task to explicate their maps and metaphors with stories.
The analysis of these stories followed the principles outlined above.
Internal Branding Reality
Conflicts and agreements Numerous conflicts between front and back stage
internal brand enactments characterize the company internal reality. The most
pronounced example for a conflict between front and back stage brand enactments
regards the company’s management handbook, containing front stage, intended
corporate values, beliefs, and behaviors. The following emic interpretations by
two organizational members regarding the handbook’s actual back stage enactment
illustrate this conflict:
Management has created a solid base for all working processes in our company —
the management handbook. But if we would always follow this handbook working
would become impossible because it blocks all of our natural work flows.
There are process descriptions for everything in this company but most employees
don’t even know they exist and do things their way.
Throughout the 35 participants’ autobiographic stories, a number of critical

incidents and issues arise, causing conflicts:
(a) A major leadership change toward the end of the end of the 1990s led to the
introduction of new front stage values, splitting the workforce into ‘‘old
employees’’ still following the former value and belief system and ‘‘new
employees’’ promoting the newly introduced, intended corporate brand identity.
This results in conflicts between the actual back stage internal brand enactments
of old employees and the new front stage system, as well as between old and new
employees, who enact the brand differently back stage.
The following statements by old and new organizational members illustrate the
conflicting views present in the company very well:
Old employees 1. I am strongly committed to this company but not to our new
CEO and his management team. These managers were not selected because of
their skills but because they please the new CEO and do not contradict him.
Our old CEO would never have accepted that — he was very knowledgeable
and very modest. Old employees still follow his tradition: they care about
other people, try to maintain personal contact and do not follow everything
that is formally stated, while new employees follow to the new structures
and processes without questioning them and foster impersonal, formal
communication.
Old employee 2. Today everything is different. The management team has
changed. Under the old CEO, R&D had a different standing: if we needed a
machine, we got it. Today, we are fighting with company internal bureaucracy.
Work is also much more impersonal today: for instance, we have computer
systems such as SAP replacing interpersonal contact and in a project team we
are not supposed to talk to the responsible persons directly but only via the
project manager. Due to the immense growth that took place during the last
10 years we are not like a family anymore — everything is fragmented.
New employee 1. It’s like two different worlds: with the new CEO a new
era started. The old guys are like a sect, they are old fighters which do not
integrate new employees into their groups and give them the impression to
be youngsters. They create a negative atmosphere; find negative aspects in
everything; and are against change and innovation. They will never become
supporters of the new CEO — they fight against everything he does. And they
are jealous because all important positions are filled by employees adhering to
the new CEO’s values and beliefs.
(b) A second major conflict raising issue were two waves of restructuration that took
place at the end of the 1990s and in 2005, first introducing a process organization
and then dividing the sales department into business units and organizing them
as a matrix. Both restructurations required the introduction of new processes and
related, intended brand behaviors. Front stage, the new structures still appear
to be the ideal solution. Managers who have been involved in elaborating these
structures are still in favor of them because they allow the company to grow and
to provide its customers with individualized solutions. However, especially in the
case of the new matrix organization introduced in the sales department, everyone
is aware of the fact that the back stage enactment is non-satisfying. A manager
comments the situation as follows:
I have been heavily involved in elaborating this new structure and I really
support what we came up with. But I cannot support how it has been
implemented and what is made of it today.
Other organizational members confirm this conflict.
From an organizational point of view, the new matrix organization was the
right decision. But honestly, do you know a case where a matrix organization
really works — except for large companies like Gore or IKEA? Our people,
especially our subsidiary leaders, don’t accept that they have been degraded
and that the business unit leaders supervise their employees. We are dealing
with people and these people will find their ways and change the structures to
suit their needs.
A matrix organization is always difficult and probably only works if people get
along with each other and treat each other with respect. If everybody ‘‘cooks
his own soup’’ it cannot work.
In the sales department, the restructuring led to major losses for many people:
old sales representatives lost customers to new sales representatives; many
people in leading positions were degraded; others were not promoted as
expected. These people tend to boycott the new structure and are not willing to
change their ways.
(c) Also, different national cultural backgrounds present in the company lead to
discrepant emic interpretations of brand enactments. A German and an Austrian
employee, respectively, describe the situation:
There are major differences in mentality between German and Austrian
employees, influencing their way of interpretation and behavior. The topic
‘‘just in time,’’ for instance, is interpreted very differently. For me, just in time
means delivering on exactly the right time. For Austrians, delivering early
means doing a better job. But for me early delivery is as bad as late delivery.
I am the only leader in our team, who represents the local Austrian culture.
My German colleagues are different. I am trying to make sure that our local
culture is preserved.
Organizational members try to solve various front and back stage conflicts
throughout their stories, but indicate themselves that the realistic outcome are
working agreements between employees, ensuring a smooth workflow. In the case of
the above-mentioned management handbook, organizational members engage in
conflict resolution, simultaneously indicating that the conflict is not resolved and that
working agreements are in place:
But all in all, the management handbook has been based on actual working
processes; therefore there are not too many situations where you have to act
against it to ensure a smoother work flow y Anyways, it’s not only me who does
not follow the management handbook; many of my colleagues also prefer the
informal way: this makes our collaboration very agreeable and we all very much
appreciate that.
Besides conflicts, organizational members’ stories also indicate that there are
agreements with regard to internal brand enactments. One major point of agreement
is the enactment of positive and negative incentives.
In our company you are usually neither punished nor praised. I would accept
punishment if I got praise from time to time. But in 11 years nothing like that
happened.
It is part of our culture to allow people to make mistakes and to learn from these
mistakes. Rewards do not really exist — nothing really happens if you do a good
job because that is what you are supposed to do anyways.
In their stories, organizational members confirm a strongly ego-centric stance,

reporting events that are of highest concern to themselves and picturing themselves
as the heroes of their stories. Stories allow organizational members to make sense of
their own role in the company and to generate their own identities. Old employees,
for example, tend to interpret their beliefs and values as the right way to go and
picture themselves as representatives and promoters of the ‘‘right’’ value system.
Organizational members do, however, avoid too detailed descriptions of their actual
work environment and the values and beliefs guiding their most frequent and intense
company internal behaviors and relationships. Only the combination of storytelling
with cognitive mapping and metaphor elicitation takes organizational members this
step further, encouraging them to switch from a macro to a micro-level perspective
and to cross the line of consciousness and social desirability.
Figure 3 shows two original cognitive maps two organizational members working
at different hierarchical levels in the innovation department/process forwarded. Both
maps are unique with regard to their design, the people included, and the connections
between them.
The main advantage of the cognitive mapping exercise is that the method
allows organizational members to slowly approach and to make sense of their
immediate work environment and enables them to position themselves vis-à-vis other
social actors. However, most organizational members still avoid giving details in
their interpretations of actual back stage brand enactments in their groups and do
usually not describe how the relationships and interactions in their immediate work
environment look like. The main focus is on who is involved and why these people
formally interact, but not on the principles underlying these interactions. Cognitive
mapping is still a valuable and necessary prerequisite preparing the ground for the
subsequent application of metaphor elicitation.
Most organizational members need the combination of cognitive mapping
and metaphor elicitation to verbalize unconscious or eventually socially undesirable
interpretations regarding their immediate work environment. Projecting their
interpretations of other organizational members’ brand enactments to animals
instead of talking about them directly relieves participants from the pressure to
Figure 3: Original front and back stage cognitive network maps of our executives.
171
respond in socially desirable ways. A few individuals refused to perform metaphor

elicitation and preferred giving a verbal explanation. The fact that the resulting
descriptions were similarly thick is an indicator for the existence of different
preferences in knowledge retrieval that need to be accounted for in order to gain deep
insight into interpretations of actual brand enactments.
Figures 4 and 5 again show the cognitive maps of the two representatives working
in the innovation department/process. The presentations in these two figures preserve
individuals’ anonymities by covering the original content with equivalent symbols as
the ones participants used in their maps and assigning each person mentioned in
the maps a person number (Px) and a department number (Dx) consistent across
all maps. The maps now include the animal pictures the two participants (called
Eric and Marc) chose to describe the people mentioned in their maps. The stories
Eric and Marc forwarded during their cognitive mapping and metaphor elicitation
tasks are discussed next.
Eric (P26) is one of three competence leaders in the innovation department. Eric
appears to enjoy both verbal and visual knowledge retrieval, having no difficulty
approaching topics that might be socially undesirable (such as relationships at work)
already while drawing his cognitive map. The results of the metaphor elicitation task
(discussed later in this section) show, however, that Eric still avoided details
regarding the most pronounced conflicts or other hot topics in his department.
Eric puts himself in the middle of his cognitive map, then depicting the
various teams he is interacting with. Eric starts with his own team (D53) where the
contact is most intense, then proceeding to the second team in the innovation
department (D52), mentioning the competence leader of this team (P23) he is sharing
his office with. The third team in the innovation department (D51) is more distanced
from Eric’s team, wherefore he depicts them a bit farther away, still stressing that he
needs them for his daily work. Next, he mentions P27, who is the eldest in his team,
his official representative and the direct contact to other team members (P51, P52,
and P53).
Eric is also collaborating with P24 who is in P23’s team now but was formerly
one of Eric’s employees — as many other organizational members — but ‘‘they’’
(meaning top management) took lots of employees away from him while
restructuring the company. Eric’s story accompanying his cognitive map indicates
a front and back stage conflict: Eric is not totally convinced of the new front stage
structure which reduced his power and importance in the company by diminishing
the number of employees under his supervision. Eric has the impression that he has
been degraded. In his emic interpretation of the situation Eric forwards a conflict
resolution attempt, which could also be perceived as a personal coping strategy:
Ok, while I was here I always had 10–12 employees. D52 and D53 were one team.
But one day I would have had 20 employees which is definitely too much. So they
had to split our team.
Next, Eric mentions his boss by saying no more then ‘‘well, then I should probably
not forget my boss who also belongs to my immediate environment.’’ Eric stresses
Figure 4: Front and backstage cognitive social network map of Eric (P25) including animal metaphors.
173
174
Figure 5: Front and backstage cognitive social network map of Marc (P5) including animal metaphors.
that there are many other people in the company he interacts with, for instance, in
the two business units D82 and D83. The contact with D82, especially with the
business unit leader P10 is more pronounced because of this business nits’ techno-
logical needs. D83 has more contact with his fellow colleagues in the innovation
department (D52). Eric marks the relationships between the business units and the
various teams in the innovation department with an additional connection line.
In the production department (D40) there are two competence team leaders
Eric is strongly interacting with — both were once employees in his team. In his
interpretation regarding these two employees, Eric finds a way to satisfy his own
identity generation needs, while indirectly distancing himself from the new
management team, especially his own boss. Eric is one of the old employees
supporting the point of view that the strong technology focus the old CEO promoted
was desirable while nowadays irrelevant topics such as project management are
paramount. Eric’s direct graphical connection to one of the oldest members of
the executive board (P6) underlines this strong connection to the past. The following
statement supports Eric’s identity generation attempt:
People that have worked in my department get to know the material we are
working with so well, that they want to work with it and move on to the produc-
tion department. That is the one department in our company where they really
need know–how, and they got it here. In top management positions this knowledge
is not required anymore today.
When asked to assign animals to single persons in his cognitive map, depicting
these persons’ characteristics and roles, Eric’s first comment is ‘‘wow, the idea to use
animal symbols is really not bad.’’ Eric begins with his team: P27 is a rooster. He is
proud, an alpha animal and Eric’s official representative who would like to have
more of a leading role in the team. P52 is a sheep, at times too generous with herself
and others. P51 is a fox, not an insidious fox but a cunning, smart fox. Also the busy
bee fits P51 because he is smart and hardworking. P53 is a rather new team member
and quite young, a caterpillar that still needs to become a butterfly. Eric perceives
P53 as a good developer because he is reliable and solid: he does not give up right
away when developing some new material but reliably finishes his task. Finally,
the conflict between Eric and his boss becomes obvious. Eric states ‘‘oh yes, P5, he is
guy how wanted to become my boss, but I won’t say more about that now, well,
restructuration at times provides you with a boss.’’ However, he refuses to assign an
animal to P5 before finishing the rest.
Next, Eric chooses people randomly, moving from one department to the next.
P54 (one of his former employees who now works in the production department) is
described as kind of vain, somebody who talks a lot without saying much and who
likes to present himself in the best light. Eric cannot think of an animal that would fit
this description. P6, the only member of the executive board Eric has intense contact
with is a rabbit or what Germans calls ‘‘rabbit foot,’’ meaning that he is spineless.
P10 is a lion who is stubborn and very self-conscious, that is, he perceives himself as
the best and knows everything better that anybody else, also when it comes to
technological issues.
Again, a conflict between various backstage value and belief systems is
perceivable: P10 is a new employee with much less experience than Eric and
representing the ‘‘new’’ value system. P24 is incredibly diligent and tends to loose
himself in irrelevancies. He is like a hamster in his wheel: if you give him some work
he will not stop until he is done with his task. Eric describes himself as a cat who has
its own will and does whatever it wants, stressing that also other people perceive him
like this. He is the contrary to his fellow competence team leader colleague P23
who is a dog doing whatever he is told. In the case of P22 Eric cannot think of an
animal but describes this person as a former farmer, very solid, reliable, tidy, what
differentiates him from P54 who is not reliable at all. P55 is an elephant who doesn’t
forget anything and is also comparable to P22 because he is very reliable.
Finally, Eric has to assign an animal to his boss (P5), choosing a bear and a
donkey. According to Eric, both animals are not ideal. P5 is very persistent and a
supporter of the new values. He likes to work in a very structured way which does
not fit Eric’s work approach. In the story accompanying his metaphor elicitations,
Eric overtly talks about their conflict regarding their interpretations of internal
brand enactments and states that they have found a way to get along — a work
agreement — without really accepting each other. In a last step, Eric chooses a
metaphor for characterizing his team: they are a bunch of penguins. This is meant to
describe the typically enacted brand values and behaviors as well as the climate
within the team: ‘‘on the one hand, they are very warm and friendly but on the other
hand they can be cold as ice and treat each other in a ‘‘frosty’’ way.’’ Eric states that
there are strong contradictions in his team that become most obvious when looking
at the variety of animals he chose (a sheep, a rooster, a bee that stings). People do
not always get along — an indication for the existence of back stage conflicts in the
department.
Figure 5 depicts Marc’s (P5) cognitive map. Marc is Eric’s boss. First, Marc draws
his own team, the innovation department. Marc is one of the few participants who
dares mentioning interpersonal relationships, and especially conflicts (e.g., with the
CEO, members of his team or members of other teams) already in his autobiographic
story. He appears to be very self-reflective and tries to make internal brand
enactments conscious to himself by analyzing his own beliefs and behaviors as well as
the behaviors of other organizational members in his surroundings.
Marc is also able to describe the corporate brand values that are actually guiding
company internal behaviors, mentioning respect and acceptance between employees,
willingness to move on and to change things, transparency (e.g., every employee is
personally informed about the company’s financial situation four times a year), and
everybody is empowered and receives the freedom to take decisions himself but is
also responsible for himself. Marc is an internal brand ambassador. On the one
hand, he is strongly involved (or involves himself) in defining front stage values and
behaviors and on the other hand he shows an interest in promoting the resolution
of conflicts. Marc is a very structured person and a strong supporter of project
management. This focus on structures dominates his back stage understanding of
internal brand values and behaviors. The following anecdote regarding himself and
the CEO of the company shows how Marc is actively engaging in making his
individual back stage value and belief system part of the front stage values and beliefs
and also illustrates Marc’s interest in conflict resolution:
Some days ago I had a fight with our CEO. I am a very structured person and a
strong supporter of project management and I am actively trying to spread this
way of working in our company and to convince other people of its benefits.
Therefore, I asked the CEO’s secretary to put this topic on the agenda for our next
executive board meeting because I wanted to report some new ideas and
experiences I just had while doing a seminar on innovation management, ideas
on how to do things differently in our company. For some reason our CEO was
completely against it and we had a fight on the phone. I was personally offended
and went to his office ten minutes later to get an appointment for the next day so
that we could discuss the topic after cooling down. The day after we solved the
problem and found out that it was actually a misunderstanding and that we had a
similar attitude with regard to this topic.
Marc starts his cognitive map by positioning himself in the middle and the
three competence teams of his department (including the competence team leaders
P22, P23, and P26, that is Eric) as well as P56 (who work independently from
the three teams) to his right. These are the people Marc is daily interacting with.
P56 is Marc’s assistant, who is coaching and supporting him when it comes to
strategic issues.
Marc only includes the persons he is interacting with most but stresses that he is
also in contact with the competence team leaders’ representatives and their
employees. Next, Mark includes several departments and heads of departments
(who are on the executive board with him) in his map: the production department
with P4, the sales department with P8, the controlling department with P3,
organizational development with P6, human resources with P2 and P43, and the
CEO, P1. In the sales department, Marc adds the three business units with their
heads (P10 with whom he has most contact as well as P9 and P11) and the
subsidiaries in Germany with the respective heads (P7 who is also on the executive
board, P12 and P13). Marc interacts with all of these people and there is also project-
dependent contact with their employees.
So far, Marc’s description of his social environment is rather superficial, focusing
on formal relationships but not so much on actual interactions and underlying
values. Only the metaphor elicitation task encourages Marc to share some deeper
insights with regard to actual brand enactments and conflicts in his environment.
Marc starts out with Eric (P26), whom he already mentioned in his story, indicating
that there have been some discrepancies between them. Marc chooses a running
horse, a fox and a rooster to illustrate Eric’s characteristics and his role in the team.
Eric is chaotic but visionary, like a horse that is running very quickly, with lots
of know–how that should be used. Compared to Eric, Marc is small horse, that is,
he is similar but not that quick yet. Also the fox fits Eric well: he is smart, foxy.
Finally, Eric is a rooster who likes ladies and cars. Marc’s relationship with Eric was
complicated at the beginning.
Due to the restructuring some years ago, Marc suddenly became Eric’s boss. Eric
did not accept that because he was himself head of a department for years. The two
of them also have very different approaches to work and tend to enact the brand
differently: Eric is chaotic while Marc is very structured. Additionally, Marc is a new
employee while Eric has been working for the company already under the old CEO
and strongly supports the old value system. According to Marc, the two of them have
been able to sort this back stage conflict out: ‘‘We get along really well today and are
able to collaborate. Eric is a very valuable member of my team.’’ Eric’s accounts
show that his perception of the relationship is different: he is still not accepting
Marc, the conflict is still there but they found an ‘‘agreement’’ that allows them to
work together.
P23, the second competence team leader Marc describes, is a squirrel — an animal
very diligent, hardworking, committed, quick, and dexterous. He has been working
for the company for a few years and is already in a leading position. Additionally,
P23 is a good communicator. P22 is very active and committed. He is a dog that can
be seduced but also a dog that barks and bits and runs if he does not agree with
something. Marc stresses that the combination of these three people is ideal: there is
P23 who is diligent and careful, P26 (Eric) who is very creative and P22 who overtly
says what he thinks. As individual, all of them enact the brand slightly differently
from what is expected front stage, but together they make it work. Marc can’t think
of an animal representing P56. She is the one who makes sure that things happen in
the department, who coaches all team members.
The heads of the other departments are the second major group Marc reports
interacting frequently with. The CEO (P1) is both a horse and a donkey. The horse is
visionary, galloping toward the future but also somewhat unstructured and chaotic.
The donkey is stubborn and determinative, not listening to others. To some extent,
these behaviors are contradictory. Marc stresses that the two of them get along really
well and communicate on a high level. Still, an obvious discrepancy appears in their
brand values, beliefs, and behaviors when it comes to a structure (the CEO is much
less structured than Marc) which also explains the fight Marc described before.
P3, the head of the controlling department, is a scorpion.
Marc is at times annoying, critically reflecting everything, causing troubles on the
executive board but more in the sense that he makes people aware of what goes
wrong. He could also be described as a brand ambassador who makes sure that
other managers are aware of eventual conflicts between front and back stage values,
beliefs, and behaviors. P6 leads the organizational development department. He is
the handsome guy, a peacock, elegant, noble, calm, not looking for excitement or
for challenges. He usually follows the main stream but is a very valuable team
member due to his know–how. P8, the head of the sales department, is new in the
organization and has a hard time because he does not know the ways of doing
business here. He is not yet accustomed with the company internal front stage value
and beliefs system and entered the company with prior experience in other companies
and a different national cultural background. He is a Bambi: insecure, needs lots of
support in order not to get lost or not to be killed in the high grass. P4 is head of the
production department. P4 is a hamster, very active, critical. Finally, Marc chooses
the zebra for himself. He tries to give his department and the whole company
structure (indicated by the zebra’s black and white pattern) and to maintain the
balance within his team.
Marc stresses that he is not a ‘‘typical boss’’ but more of a team member
responsible for coordination. On the executive board he is the troublemaker how
addresses hot topics. He is not looking for challenges or conflicts. However, he does
not avoid them if they arise but tries to solve them — supporting the fact that he is
a brand ambassador.
Metaphor elicitation helped participants to overcome restraints present
during storytelling and cognitive mapping and encouraged them to give insights in
actual back stage internal brand enactments within their immediate work
environment. Both organizational members include conflicts between the various
value and belief systems and conflict resolution strategies in their emic interpreta-
tions accompanying metaphor elicitation. These interpretations were obviously
not consciously retrievable or socially undesirable to mention in their stories.
Additionally, by sharing their perceptions regarding the behaviors people surround-
ing them enact and the various relationships between these people, the two
participants generate the identities of these people relative to the firm’s brand
identity (i.e., which value and belief system are these people enacting and is
this system consistent with front stage values and beliefs?) as well as their own
identities relative to other actors (i.e., do the value and belief enacted by other
social actors correspond to the participants’ own emic interpretations and do
interactions with these people support his identity?). Finally, only the combination
of all three methods is likely to give real deep insights into and thick descriptions
of conscious and unconscious, socially desirable and undesirable internal brand
enactments.
Back stage enacted corporate brand values In-depth hermeneutic analysis of the
stories, the 35 participants forwarded in every step of the research procedure led to
an identification of actually enacted values and beliefs in each of the four resource
processes and on the corporate level. For illustration purposes, the value system
enacted in the innovation process (Eric and Marc belong to) will be introduced
before discussing the company-wide value system. A total of four managers and
four employees working in the innovation process participated in this study. Figure 6
shows the values derived from these eight organizational members’ stories,
differentiating between values that only management or employees mentioned and
values that both groups enacted. A discussion of shared values follows, supported
with excerpts from their original stories.
The value fairness and support implies assisting employees and allowing them to
make mistakes. For managers, this value signifies establishing a work environment
where employees can fulfill their tasks without problems and assisting employees
when they have questions or are in troubles. Also the active support between
tolerance towards mistakes
Shared values
transparancy
fairness and support
face challanges aiming for community
teamwork
individual
regardful collaboration
responsibility
self-criticism openness and helpfulness
transparency
structured and systematic
work Personal contac
competitiveness
-- Employees - Management
Figure 6: Actually enacted value system in the innovation process.
employees is part of this value. Employees expect and value assistance from
management. The following statements support this value:
My boss has to take care that others don’t take advantage of me.
If we really want to be innovative we have to collaborate and support each other.
Aiming for community is another value both management and employees in the
innovation process share. This value contains teamwork and personal contact.
Managers, for instance, perceive themselves as team members and not as strict
supervisors and believe in the power of their teams. The innovation process also
values inter-departmental collaboration. The following statements illustrate how
participants interpret this value:
I am not a traditional supervisor. I set deadlines and make sure that work is done.
And at times I participate in meeting where I am in the role of a normal team
member. If we want to make a difference we have to do that together, as a team.
I very much appreciate that innovation and sales work closely together.
Regardful collaboration encompasses an adequate conversation style as well as

acceptance of various personalities and opinions within the team and the company:
We treat each other with respect and accept other organizational members as they
are.
To be nice with each other, we do y how should I put this, sometimes people are
not as nice with each other but that is not the desired way to collaborate.
Another value that is of major importance to both groups is openness and

transparency, implying an open discussion about problems, conflicts, and further
actions as a basis for collaboration. Statements like the following underline this
value:
I try to tell people what bothers me and what I disapprove of, or what could be
changed.
It is important to discuss different ideas and to deal with themes even if it’s
emotional and sometimes not comfortable for the persons that are involved.
Transparency is important in our company: there are quarterly held meetings
where our CEO informs all employees about the company’s financial situation.
Finally, structured and systematic work characterizes the innovation process.

The head of the innovation process strongly promotes this value and tries to install it
front stage: ‘‘I am very structured, I need this for my work and I try to convince
others of the benefits of this approach to work.’’
Not all members of the team positively evaluate this strong focus on structure.
This indicates that there are conflicts between back stage beliefs and employees only
enact this brand value as intended because they feel pressured to do so:
If I do something today I have to ask myself if I fulfill the formal criteria and
processes. It doesn’t really matter whether I do a good job or not — it’s all about
formalities and following the formally installed processes.
We have been growing and needed more formal structures but at times it’s a bit
difficult to work in such an environment, especially for someone who is not that
formal and more creative.
The above examples illustrate that employees and management agree on a core of
enacted brand values that guide the behavior within their team as well as with other
teams. Still, there are numerous values that only one of the two groups mentions.
The reasons might either be that the other group does not enact this value at all or is
simply not aware of this value. Differences in value perceptions and enactments can
but do not necessarily have to lead to conflicts between the groups. On the other
hand, there are values that both groups seem to share (e.g., structured work) but are
actually only enacted as desired due to pressure from the top while employees still
believe in another approach to work.
In order to gain insights into the company-wide enacted brand value system, the
single value systems identified for each process were aggregated. Figure 7 depicts the
values mentioned in most organizational members’ stories. Most values overlap with
those identified for the innovation process. Additional values that have been more
pronounced in other processes are individual responsibility and facing challenges.
Individual responsibility is a central enacted brand value guiding the behavior of
the majority of organizational members. Managers give their teams the freedom
to find their own way to solve problems and to make mistakes in order to learn and
Resources
Innovation
• Fairness and support

• Aiming for community
teamwork
• Regardful collaboration
• Openness / transparancy
• Face challanges
• Individual responsibility
Sales
Production
Figure 7: Company-wide actually enacted value system.
to develop themselves and the company. Employees very much appreciate this value.
The following statements illustrate the importance of individual responsibility:
Leading in this company signifies installing equilibrium between freedom and soft
pressure to achieve the goals.
I tell nobody how to do his job. In the end the results has to be ok.
Everybody is allowed to make mistakes and to learn from his mistakes.
For my supervisor it is important that the result is ok and that I reach it without
breaking any basic rules. But we have lots of freedom.
Finally, a strong will to face challenges characterizes the company. These

comments support this value:
This team has to face challenges to enhance the company.

I don’t look for challenges but if there is a challenge I will always face it.
Concluding, throughout all participants’ emic interpretations of internal brand

enactments the impression arose that the local mentality strongly dominates this
company. All values tend to support the central idea of working together to achieve
more and the personal integration of all organizational members. The following
statement by one participant summarizes the company’s effort to balance local
tradition with innovation, two values that seem to be inextricably linked and in
this synthesis crucial for the corporate brand’s success, ‘‘We honor the old while
welcoming the new.’’
Discussion and Managerial Implications
Contrary to existing literature on internal branding, the present chapter applies an

organizational/behavioral science view, discussing how brands are actually enacted by
organizational members. Enacted internal branding theory and practice illustrate that
organizational members can be considered active brand co-creators, enacting brand
values, beliefs, and behaviors during every action and interaction with company
internal and external stakeholders. Former research acknowledges this active role
of employees, picturing them as ‘‘brand ambassadors’’ (Vallaster & de Chernatony,
2005, p. 5) or ‘‘brand builders’’ (de Chernatony, 1999, p. 157), who support manage-
ment in establishing a desired corporate brand image with other stakeholders. Enacted
internal branding theory demonstrates that achieving this ideal outcome is unlikely.
This chapter shows that organizational members’ emic interpretations of actually
enacted brand values, beliefs, and behaviors are in many cases different from what
management intends for its brand. This is consistent with Goffman’s (1959)
proposition that there are hidden back stage processes that contradict how
management defines its brand front stage. Numerous conflicts between front and
back stage values, beliefs, and behaviors, as well as between various back stage brand
enactments characterize enacted internal branding, leading to ‘‘Doppelgänger brand
images’’ (Thompson et al., 2006, p. 50) inside and outside the company.
Given the enormous social complexity management encounters an obvious
question. What can management do to nurture front-stage, intended, brand values,
beliefs, and behaviors? To some extent, management can rely on system inherent
conflict resolution dynamics. The case study findings here show that unbalanced
psychological states, for instance, provoke brand enactment conflicts that organiza-
tional members are consciously aware — resulting in distress among organizational
members and lead to conflict-solving attempts that ideally promote intended brand
values. Still, management cannot rely on such a process to occur naturally because in
many cases organizational members are (a) not willing to solve conflicts, (b) unable
to solve conflicts, and (c) not even aware consciously of the conflict.
In order to face this challenge, management needs to gain an understanding of the
functioning of both deliberate and tacit social reality. Organizational reality is to be
understood as a dynamic social system, structured by inherent values, beliefs, and
behaviors/processes that are not necessarily consistent with management’s formally
installed front stage values, beliefs, and behaviors/processes (Giddens, 1979, 1984;
Goffman, 1959).
Organically grown values and beliefs influence organizational members’ indivi-
dual sensemaking and the social norms guiding their interactions and power relations
with other individuals (Giddens, 1979, 1984). Theoretically, every member of the
organization can induce change in this social system: ‘‘organizations can be moved in
some coherent and explicit direction — here lies the possibility of deliberate and
effective strategy’’ (Whittington, 1992, p. 695). Changing the underlying values
guiding the organization is, however, most difficult because they represent stable
principles that will only change under certain circumstances, for instance, due to
major events such as a leadership change (Bartunek, 1984). The actual outcomes of
attempts to influence social reality are not foreseeable (Giddens, 1984).
Before attempting to influence internal brand values, beliefs, and behaviors,
management is therefore well advised to learn more about the actual social reality in
its company. Applying the mixed research procedures in this report represent a
design to help management to surface various conscious and unconscious brand
enactments and eventual conflicts between them. This knowledge can serve as a
starting point for re-thinking and eventually redefining front stage values, beliefs,
and behaviors. Organically grown guiding principles might be much more powerful
and important for the success of a corporate brand than artificially defined values.
This becomes most obvious in the case company investigated in this study, where the
organically grown synthesis between local tradition and innovation is essential
for the corporate brand’s success. Still, management may have to consider further
corporate brand values, beliefs, and behaviors that are not yet present in the
company but are necessary for reaching the corporate brand’s vision. Integrating
actually enacted values with intended values provides a corporate brand with a solid,
credible base that is automatically lived and enforced by organizational members, as
well as with a common goal organizational members can work toward.
The study that this chapter presents confirms that social dynamic enactments to a
substantial extent may inhibit managerial efforts to spread intended corporate brand
values, beliefs, and behaviors. On the one hand, management has the advantage of
having more resources and power at its disposition than other organizational
members that allow exerting a certain influence on the system. Management can, for
instance, choose from a variety of communication devices designed to transmit
desired corporate brand values (Burmann & Zeplin, 2005); can develop incentive
systems that exemplify which behaviors are valued and which should be avoided
(Esch, Rutenberg, Strödter, & Vallaster, 2005); or can promote desired values via
leadership (Vallaster & de Chernatony, 2005) and human resource activities
(de Chernatony, Cottam, & Segal-Horn, 2006). Still, many other organizational
members actively participate in this structuration process. These social actors may
willingly or unwillingly counteract managerial internal branding efforts and limit the
possibility of implementing the brand’s meaning and behavior as intended.
Experiential Exercise in Emic/Etic Deliberate and Tacit Interpreting
Task Description
Figure 7 shows Susan’s (P56) cognitive social network map including her animal
metaphors. Susan is Marc’s (P5) assistant in the innovation process/department.
She is responsible for organizational development within the innovation process. In

order to gain further insights into the actual front and back stage brand enactments
in the innovation process, the reader is asked to interpret Susan’s emic inter-
pretations of her immediate work environment. Please analyze the following data
material taken from Susan’s original emic interpretations both from an emic
(informant) and etic (researcher) point of view and answer the following questions:
What does the emic text say and what do the emic metaphors indicate?
Are there any conflicts between front and back stage structures (etic point of view)?
Justify your answer.
Data Material
Storytelling excerpts:
At the beginning I did not realize what is really going on in this company because
I was mainly working with my boss (Marc) who is very structured. But now I am
aware of the chaos. It looks like we have structures and we actually have them but
we do not have the same structures in every part of the company. This leads to
chaos and big problems because we are growing very quickly while not having the
necessary structures.
There are some employees who have experience with project management. They
are much more receptive to our efforts to introduce project management as a
company wide tool than other employees who have been following a different
work approach for years. Officially, project management has been introduced for
quite a while but it is not lived, yet.
There is a high degree of transparency in this company and people get lots
of information but that does not mean that people understand everything as
intended.
There are some problem areas in this company: people do-at times-not get along
with each other — especially within the sales process and within the innovation
process. But I have the impression that it is slowly getting better.
Story accompanying Susan’s cognitive mapping task:

There is Marc (P5) who is my boss. Then there is P52 who is project leader in one
of the projects I am involved in. P52 is introducing a tool that allows specifying
strategic goals on the level of every single employee. Then I am interacting quite a lot
with the competence team leaders in the innovation department (P22, P23, P26 —
Eric). There is also project related contact with P3, the head of the controlling
department and P26 (Eric, who is also competence team leader) when it comes to
patents. I am also working with two of our business unit leaders, P9 and P10, but
there is more contact with P10. Susan leads a project team that is responsible for
process development. This team consists of representatives from different depart-
ments and several executive board members: P25 represents the innovation
department, P28 works in one of the German subsidiaries, P16 and P57 are
sales representatives in two different business units, P58 is responsible for marketing,
P6 is the company internal project initiator (P6 is also one of Susan’s direct contacts
when it comes to strategic issues because he is the company internal expert
for organizational development) while P5 and P8 are the steering committee.
Finally, Susan is also interacting with the human resource department, especially
with P43 and P59.
Excerpts from Susan’s story accompanying animal metaphors:
P26 (Eric) is a peacock. This animal does not represent all of Eric’s characteristics
but it exemplifies that Eric has an immense know–how and likes to talk about
things he knows well and to share this know–how. Eric is very special: he has his
own way of thinking and working. What makes sense for him often does not make
sense for other people in this company. This leads to confusion.
P23 is a squirrel. He is very diligent, ambitious, structured, and makes sure that the
collaboration with other organizational members’ works.
P22 is a dog. He does not like to contradict our boss (Marc) — he accepts his ideas
and decisions without objection. What the boss says must be right.
Susan can’t think of an animal for Marc who is very structured, not bossy but very
collegial, down to earth and knows how to deal with other people. One of Marc’s
problems (which is simultaneously one of his strengths) is that he trusts people
too much and is disappointed if somebody does not live up to his expectations
(Figure 8).
Teacher’s Note
Already an analysis of Susan’s storytelling excerpts allows identifying several brand

enactment conflicts. Similar to Marc (P5), Susan is very structured and has a clear
idea of ideal front stage brand enactments. Susan is convinced that management
makes efforts to introduce intended brand values and behaviors/processes/structures.
However, these efforts are not consistent enough. Consistency is already missing
front stage: there are many different front stage brand enactments (i.e., intended
structures) in various parts of the company. Additionally, organizational members
do not live the brand as intended, that is, organizational members enact structures
differently than intended front stage. This inhibits the achievement of a major
company goal: constant growth.
Susan also indicates that there are a variety back stage brand enactments
originating from different experiences organizational members have made outside
and inside the company. Employees who have, for instance, used project manage-
ment as a tool in other companies enact this front stage desired brand enactment as
intended while older employees that have been working in the company for a long
time do not accept this new way of working, that is, they do not enact this brand
behavior as intended front stage. Even though management is trying to communicate
Figure 8: Front and backstage cognitive social network map of Susan (P56) including animal metaphors.
187
front stage values, beliefs and behaviors/processes/structures people do not

interpret and live the brand as intended but implement their own emic back stage
interpretations.
Finally, Susan mentions some ‘‘problem areas’’ indicating that there are several
conflicts between organizational members in various departments, also in the
innovation department. From this excerpt it is impossible to tell why these conflicts
arise — one possible reason is different back stage brand enactments by various
organizational members. There is, though, a tendency toward conflict resolution:
‘‘it is slowly getting better.’’
Susan’s cognitive mapping exercise does not give deeper insights into the actual
brand enactments in her immediate work environment. Susan illustrates who she is
interacting with and why (i.e., the formal relationships between various actors)
as well as what some of these people are doing. How these interactions look like
remains uncovered. Only the combination of Susan’s cognitive map with her animal
metaphors allows an understanding of actual interactions and brand enactments.
Eric (P26), for instance, is a source of conflict and confusion within the
department and the whole company. Eric has his own way of interpreting and
enacting the brand. His emic interpretations are not consistent with front stage or
with other organizational member’s back stage brand interpretations and enact-
ments. Still, Eric is a valuable and capable team member with lots of knows how
that he likes to share with other people — thereby supporting the achievement
of departmental and company goals. P23’s way of working is very different from
Eric’s and in line with front stage structures. Similar to his boss (Marc), P23 is very
structured and likes to collaborate with other organizational members. P23 is
interested in avoiding potential conflicts hindering collaboration that may arise due
to different back stage brand interpretations. P22 is also trying to avoid conflicts and
to follow his boss’s way of thinking. It is unclear whether P22 is really convinced of
the front stage brand interpretation his boss promotes or whether he is just ‘‘a slave
to his boss.’’ Finally, Marc is pictured as the one who lives his front stage brand
interpretation and actively engages in advancing the implementation of intended
brand enactments backstage, for instance, via active collaboration with his team
members and by empowering/trusting his team in implementing the brand.
Chapter 9
Personal Exchanges, Social Behavior,

Conversation Analysis, and Face-To-Face Talk
Synopsis
A book on case study research would be remiss without a chapter that

introduces the reader to relevant literature on personal exchanges and face-to-
face talk. Chapter 9 offers a conceptual property–space analysis for theory and
research on personal exchanges including talk. The chapter describes a study
on face-to-face conversation analysis (i.e., talk) in a buyer–seller context.
The study includes examining forty transactions between actual insurance
salespersons (n ¼ 3) and prospective clients (n ¼ 57) interacting in field settings.
The chapter reports the relationships between purchase behavior and the
frequency of key orientation and bargaining statements made by the
salespersons and customers. The findings support the importance of studying
social factors, influence, and situation variables in constructing a general
conceptualization of exchange relationships.
Introduction
Personal (one-on-one) exchanges include all forms of contact by two or more entities.
This broad definition includes my late grandmother’s conversations with persons that
she thought were living in the radiators in her house. Thus, personal exchanges occur
between two or more entities — real or imagined.
Personal exchanges also include human–animal (e.g., pet) exchanges and face-to-
face exchanges among two or more persons. Human-to-human exchanges are
classifiable into two broad categories: official and unofficial. ‘‘Official exchanges’’
are interactions on topics relating to professional or job-related task such as an
instructor and student discussing a research topic. ‘‘Unofficial exchanges’’ are
interactions on topics that are social in nature such as an instructor and student
discussing a recent movie seen earlier and separately. Patient–doctor conversations
by a patient meeting with a medical doctor for an annual check-up is likely to include
mostly official exchanges and some unofficial exchanges. Most face-to-face meetings
include topics of discussion relating to both official and unofficial exchanges.
Social Behavior Personal Exchanges
Conversation Analysis
Face-to-face Talk
1 2 3 4 5 6 7
Figure 1: Personal exchanges, social behavior, conversation analysis, and face-to-

face talk. Notes: ‘‘ ’’ ¼ ‘‘and’’; ‘‘B’’ ¼ ‘‘not’’: 1. SB BPE: Social behavior (SB)
without personal exchange (PE) (e.g., sending greeting card; anonymous donation);
2. SB PEBCA Social behavior and personal exchange without conversation
analysis (CA) (e.g., sending a greeting card in response to receiving one);
3. SB PE CA without face-to-face talk (FFT) (e.g., written communication
exchanges about social topics); 4. SB PE CA FFT (e.g., face-to-face communica-
tions about social topics); 5. BSB PE CA FFT (e.g., face-to-face communications
about ‘‘official’’ topics such as a discussion by motorist with police officer about
speeding); 6. BSB PE CABFFT (e.g., written exchange a appealing speeding
ticket and receiving a denial of the appeal); 7. BSB PEBCABFFT (e.g., receiving
an invoice and sending payment).
Figure 1 is a Venn diagram showing talk as one method of exchange that spans
social behavior and personal exchange. Figure 1 shows face-to-face talk as one form
of conversation analysis. ‘‘Conversation analysis as a mode of inquiry is addressed to
all forms of talk and other conduct in interaction, and accordingly, touches on the
concerns of applied linguists at many points’’ (Schegloff, Koshik, Jacoby, & Olsher,
2002). Most researchers in the field include examining both verbal and nonverbal
exchanges as well as face and non-face interactions in the domain of conversation
analysis (CA). However, most of the CA literature includes analysis of turn-taking
involving talk (e.g., Sacks, Schegloff, & Jefferson, 1974, 1996; Schegloff et al., 2002).
Schegloff et al. (2002) underscore a single underlying premise of CA:
People use language and concomitant forms of conduct to do things,

not only to transmit information; their talk and other conduct does
things, and is taken as doing things — things such as requesting,
offering, complaining, inviting, asking, telling, correcting, and the
myriad other actions which talk in interaction can accomplish.
Personal Exchanges and Face-To-Face Talk 191
By ‘‘actions’’ here we are not referring to physical actions but to

ones accomplished through the talk; and we are referring not only
to actions with familiar vernacular names like those just mentioned,
but recognizable and describable actions without names (such as
‘‘confirming allusions,’’ cf., Schegloff, 1996). (Schegloff et al., 2002,
p. 5, italics in the original)
The central issue for case study research on personal exchanges from a CA
viewpoint is to understand language use in contexts of its deployment y ‘‘we need in
the first instance to understand how and for what it is deployed by its participants,
and how its deployments are understood by them and reflected in their own
responsive conduct’’ (Schegloff et al., 2002, p. 6).
Following this introduction, the second section describes a study designed to
understand how and why participants to a conversation use language to do things.
The study focuses on face-to-face first-time interactions between a buyer and seller
of life insurance products (i.e., ‘‘products’’ include services). The study includes a
researcher sitting-in and tape-recording real-life meetings in 40 separate case studies;
the data collections were done with the cooperation and approval of the senior
management of the life insurance firm, the three salespersons, and the 57 customers
participating in the meetings.
Figure 2 locates the study in cell 7 among 27 property–spaces that occur when
considering three dimensions of CA with 3 levels for each dimension:
unobtrusive (U) to U & O to obtrusive (O) observations of exchanges by a

researcher
frequency of parties meeting: one time, twice, three plus times
face-to-face, face-to-face plus; separate locations and by documents.
The note at the bottom of Figure 2 identifies additional studies that exemplify
other cells in the 3 3 3 (27) property–spaces. Cell 1 includes studies that record
interactions unobtrusively — the participants are unaware that are being observed in
a study, the participants for most of the interactions in studies in cell 1 are meeting
for one-time only, and they are meeting face-to-face. For example, Humphreys
served as a ‘‘watch queen’’ to collect observational data on meetings by homosexual
men in a men’s room in a park; the participants were unaware that Humphreys was
conducting a study. In the study the watch queen scanned the park area through
small windows located near the ceiling to be able to warn participants about
the arrivals of police — the presence of a watch queen being a natural and regular
occurrence in this context.
A study by Woodside and Davenport (1974) is representative of cell 4 in Figure 2;
in Woodside and Davenport study of selling and buying in a retail store context, the
seller knew that he was participating in a study but the buyers did not. The study
involved manipulating the talk content by the seller and recording the buyers’
behavior (non-purchase versus purchase of a music tape-player cleaner). The study
was conducted in a naturally occurring, non-laboratory, context of a music store.
192
Context → Unobtrusive (U) U &O Obtrusive (O)
Frequency → 1 2 3+ 1 2 3+ 1 2 3+
Faces at same
location (FSL) 1 2 3 4 5 6 7 8 9
Touch- FSL &

at separate 10 11 12 13 14 15 16 17 18
ability
locations,
documents
Separate
19 20 21 22 23 24 25 26 27
locations &
documents
Figure 2: Property–space analysis (27 cells) of research on person-to-person exchanges. Notes: Examples of exemplar studies.
Cell 1: Humphreys (1970); Cell 4: Woodside and Davenport (1974); Cell 5: List (2006); Cell 7: Buyer–seller study in this
chapter; Cell 8: Browne (1973); Cell 9: Whyte (1943/1993); Cell 12: Sirsi et al. (1996); Cell 15: Eichenwald (2000); Cell 21:
Pettigrew (1975).
Representing cell 5, a study by List (2006) includes customers buying collectable

sports cards from sales agents both in laboratory and naturally occurring contexts
with two purchases occurring in the natural context per sales agent — with the sales
agents knowing that they were participating in a study. The study indicates that sales
agents behave differently in tightly controlled laboratory experiments than in their
naturally occurring environment.
Representing cell 8 is a study by Browne (1973) of obtrusive direct observations of
meetings between salespersons and customers and salespersons and sales managers
at a used-car sales lot for 18 months. All participants knew that the researcher was
observing these real-life interactions and most of the meetings involved two meetings
with the same customer before completing a purchase.
Representing cell 9 is Whyte’s Street Corner Society. Whyte made direct
observations of the interactions among a 13-member small-informal group during
1937–1940 in New York City (identified in his book as Eastern City). Almost all the
data that Whyte collected was from direct observations of face-to-face meetings
among the group members and between members of the group and persons outside
of the group. The group members knew that Whyte was conducting a study; his
study was obtrusive yet he became nearly a member of the group.
He learned to speak Italian [some of the talk observed was in Italian];

he spent the better part of three years living in the district; he hung out
with the Nortons [the group] on their corner, won the confidence of the
leader and the rest of the gang, and became one of the gang in its
games, its political campaigns, and its other activities. (Homans, 1950,
p. 157)
The study by Sirsi, Ward, and Reingen (1996) is representative of cell 12 — many
of the participants started out knowing but came to accept the researcher as a
member who was observing and documenting their interactions; the researchers
completed observations of many meetings.
The ethnographer spent over 900 hours in the field. He approached

participant observation in each micro-culture with a commitment to
full immersion. He did not merely observe the participants but became
one of them, radically changing major aspects of his lifestyle.
Participants were always informed of his dual interest in their cultures.
Rapport was established quickly among the macrobiotics and more
slowly among the activists. To win the activists’ trust, the ethnographer
gave up eating all meat, foreswore the use of possessions implicated in
animal abuse (from leather shoes to Bic pens), and threw himself into
helping organize protests against animal abuse. Gradually, he was
accepted into the inner circle of activists. (Sirsi et al., 1996, p. 349)
Representing cell 15 is the field investigation by the FBI (United States Federal
Bureau of Investigation) and reported by Eichenwald (2000) in which nearly all of
the participants were unaware that their multiple meetings to fix agricultural

chemical prices globally were being taped-recorded and video-tapped. One senior
executive, an informant for the FBI, in one of the firms (ADM Corporation)
participating in the meetings was aware of his participation in the study; thus, the
study includes both obtrusive and unobtrusive data collection.
Representing cell 21 is a study by Pettigrew on the participation of members in
three levels in an organization: senior, middle, and junior executives. Pettigrew
analyzed the content of written documents among the participants. The main
conclusions of the study include the following points: middle executives acting as
information gatekeepers and they manipulated conclusions and decisions to favor
their own objectives by the information that shared with senior and junior executives.
The aim here is not to explicate fully representative studies for all 27 cells in
Figure 2. The aim is to introduce the reader to possible classification schemes for
locating personal exchange and social behavior studies including studies that focus
on talk.
Research on buyer–seller face-to-face interactions
Several marketing scholars propose that exchange behavior between two or more
persons is the fundamental phenomenon of marketing to be explained. Exchange
behavior includes dynamic social-process communication flows. The face-to-face
meeting of salespersons and customers represent an important type of exchange
behavior in marketing. For example, personal selling in most economies represents
larger marketing expenditures than advertising. In the United States, industry
estimates on annual expenditures range from $ 100 billion to $150 billion annually,
with personal sales calls to industrial sales customers estimated to cost an average of
$400 per call. Unfortunately, the recognition of the importance of exchange behavior
in selling and buying is coupled with a scarcity of empirical data on actual two-
person or multiple-person interactions among salespersons and customers. The
data for 210 department store selling/buying interactions taped surreptitiously via
wireless microphone are the most well-known marketing reports of exchange
behavior. Analyses of these data indicated that customer buying versus not buying of
appliances is significantly associated with the number of suggestions offered by both
salesman and customer (Willett & Pennington, 1966) and the frequency in which
both salespersons and customers engage in specific bargaining activities (Pennington,
1968). Additional research on part of this data (involving 40 complete transactions
resulting in the actual purchase of either a refrigerator or a color television) indicates
that in most cases the salesperson and not the customer who determines the extent of
search and evaluation of alternatives.
‘‘In spite of his very sketchy and initial information on the customer’s desire,
the salesman typically selects the order and the number of alternatives evaluated.
The salesman also seems to dominate the evaluation of each alternative in that his
semi prepared presentations guide the customer’s attention to various product
attributes’’ (Olshavsky, 1973).
Varela (1971) reports on additional empirical research on marketing exchanges.

Varela applies social-psychological theories to influence purchase behavior in natural
settings among retail-store buyers. In one study over 90 percent of the buyers
visited by specially trained salespersons were persuaded to agree to later travel to a
manufacturer’s showroom ostensibly to determine if the manufacturer had followed
the advice offered by the store buyer to the salesman.
Analyses of tape-recorded exchanges between life insurance salespersons and
customers in natural settings have indicated that discreetly packaged blocks of
interactions occur during the face-to-face meetings (Schenkein, 1978). The following
Puzzle–Solution–Confirmation–Comment is one example of such a discrete block
(Schenkein, 1978):
Salesman: Uhh, do you have any insurance? (Puzzle)

Customer: No life insurance (Puzzle)
Salesman: No life insurance, yeah you have car and that don’chu. (Solution)
Customer: Yeah (Confirmation)
Salesman: We all have car, an- and s- stuff like that. (Comment)
This example is one that attempts successfully to locate the customer in an

‘‘official’’ (insurance related) identity position. The definition for ‘‘official’’ identity
here refers to a characteristic of an exchange participant that relates to the main
purpose of the exchange. Schenkein (1978) describes reciprocating attempts by
customers to locate insurance salespersons into official identity positions (e.g.,
salesman’s job title, years of experience) as well as examples of reciprocal ‘‘unofficial’’
identity positioning (e.g., hobbies, choice of church affiliation).
An ‘‘unofficial identity’’ refers to a characteristic of an exchange participant that
relates only indirectly or not at all to the main purpose of the exchange. Unofficial
identity characteristics often help establish points of similarity between two members
in an exchange (e.g., ‘‘Oh, we went to the same high school’’).
For life insurance, several non-interactional studies report that the purchases
occurring during the exchanges of salespersons and customers depend upon the
economic, social, physical, and personality characteristics of each person (Evans,
1967; Gadel, 1964). For other products, the content of sales message and oral skills of
salespersons have been shown to be important influences of purchase outcomes in
several one-way studies of the effects of salespersons variables on customer behavior
(Busch & Wilson, 1976; Pace, 1962; Woodside & Davenport, 1974).
A useful choice of analysis of bargaining communications in marketing should be
two or more communication turns between two or more parties in the exchange. Such
communications are likely to include discretely packaged blocks of interactions and
breaks and connections between such blocks. The concepts of official and unofficial
identity negotiations may be useful in labeling a few of the blocks of interactions.
These concepts are defined and examples provided from tape recordings of buyers
and sellers meeting in natural settings. ‘‘Being there’’ is advocated as a necessary part
of research programs to study bargaining behavior in marketing communications.
Extant content analytic systems for analyzing bargaining communications call

for dismembering verbal or written exchanges of buyers and sellers and placing the
separate utterances into categories (e.g., Bonoma & Rosenberg, 1975; Pennington,
1968; Willett & Pennington, 1966; Angelmar & Stern, 1978; Bales, 1968; Pettigrew,
1975). For example, building directly on the work of Bonoma and Rosenberg (1975),
Angelmar and Stern (1978) develop eight categories of semantic units of each party’s
turn in a bargaining conversation.
While the half-turn analytical paradigms used by Olshavsky (1973), Angelmar
and Stern (1978), and others (e.g., Willett & Pennington, 1966; Taylor & Woodside,
1979) are useful, the violence done in uncoupling an exchange of turns between
communicators needs to be recognized. These referenced content analysis
schemes use something less than exchanges as units of analysis. Classifying a
communicator’s verbal or written turn as ‘‘a promise’’ or into another bargaining
category without relating the turn to the other party’s response is similar to listening
to one-hand clapping. Such analytic schemes fail to capture the essence of marketing
exchanges.
The choice of analysis of bargaining communication in marketing should be
two or more communication turns between two or more parties in the exchange.
Bargaining communications in marketing are likely to include discretely packaged
blocks of interactions and breaks and connections between such blocks.
Within the talking or verbal part of a negotiation, a content analytic scheme of
such exchanges should build upon the recognition of the following facts, which seem
grossly apparent to relatively unmotivated examination of conversational materials.
In any conversation:
(1) speaker change recurs, or, at least occurs;

(2) overwhelmingly, one party talks at a time;
(3) occurrences of more than one speaker at a time are common, but brief;
(4) transitions from one turn to a next with no gap between them are common,
together with transitions characterized by slight gap or slight overlap, they
make up the majority of transitions;
(5) turn order is not fixed, but varies;
(6) turn size is not fixed, but varies;
(7) length of conversation is not fixed, specified in advance;
(8) what parties say is not fixed, specified in advance;
(9) relative distribution of turns is not fixed, specified in advance;
(10) number of parties can change;
(11) talk can be continuous or discontinuous;
(12) turn-allocation techniques are obviously used, current speaker addresses a
question to another party; parties may self-select, in starting to talk;
(13) various turn-constructional units are employed, turns can be projected one
word long or, for example, they can be sequential in length;
(14) repair mechanisms for dealing with turn-taking errors and violations obviously
are available for use. For example, if two parties find themselves talking at
the same time, one of them will stop prematurely, thus repairing the trouble
(Sachs, Schegloff, & Jefferson, 1978, pp. 10–11; Morley & Stephenson, 1977;
Morley, 1978);
(15) one or both participants in a two-person meeting likely are seeking to
accomplish a specific task via the exchange.
Exchanges between insurance salesmen and prospective clients are used here to
describe what the parties do in such interactions to juggle their official with informal
and personal (unofficial) identities in the course of their conversations. Schenkein
(1971, 1978) calls attention to the ‘‘identity negotiations’’ which are likely to occur
in seller–buyer encounters. ‘‘Whatever else they might or might not share; such
encounters are made up of talk between strangers who might know one another
only as local versions of some abstract identity like ‘salesman’ or ‘client.’ For these
encounters, strangers not only conduct their business under the auspices of their
official identify relations, but they also negotiate into the unfolding of their
encounter eminently personal identities from their separate biographies’’ (Schenkein,
1978, pp. 57–58).
Both official and unofficial identity negotiations are likely to occur in buyer–seller
interactions of the type described here, that is, between a life insurance salesman and
prospective client meeting for the first time. Similar negotiations are likely to occur in
industrial marketing, distribution channels, and in retail transactions involving
expensive products and services.
Official identity negotiations refer to two or more conversational turns between
parties which provide some information about one of the parties and comment by the
other party related directly to the main purpose of the meeting (e.g., selling and
buying insurance). Official identity negotiations help to specify the rather abstract
identities of ‘‘salesmen’’ and ‘‘client’’ characteristics of the Parties in the meeting.
Unofficial identity negotiations refer to two or more conversational turns
between parties which provide some information about one of the parties and
comment by the other party not directly related to the main purpose of the meeting.
Attempts by salesmen to use ‘‘referent power’’ (French & Raven, 1959) by first
learning and commenting to the client on the similarity of a planned, unofficial,
identity negotiation.
Official and unofficial identity negotiations may be centered on characterizing
and commenting on either by the buyer or seller. Thus at least four types of identity
negotiations may occur in a customer–seller meeting: salesperson official and
unofficial and customer.
The development and testing of several propositions is possible that concern the
frequencies and sequences of occurrence of each type of identity negotiation relative
to the purchase and satisfaction outcomes of buyer–seller meetings. The purpose here
is to describe several examples of such negotiations, how they might be classified into
discrete blocks of exchanges, and report on the separate interpretations a salesman
and buyer offer later to identities earlier negotiated between them. In several
instances, conjectures and other comments are offered with the examples.
Excerpts of conversations from several buyer–seller meetings were transcribed
from a study (Taylor & Woodside, 1980) of exchanges of insurance salesmen and
customers which occurred in natural settings. The study includes a total of 40

salesman–client parties. All face-to-face meetings of the salesman and client for
each of the 40 parties were tape-recorded. James Taylor accompanied each of three
salesmen on sales calls to collect the data.
Each client was requested to permit the tape recording ‘‘for a study of
conversations among persons meeting for the first time.’’ A total of 15 customers
purchased life or health insurance and 25 did not during the meeting and taping of
the exchanges. Given the wealth of data produced, only a few of the recordings were
transcribed into written text. Taylor (1977) reports further details of this study.
Action sequences in identity negotiations
Schenkein (1978) observes the presence of a common four-turn Puzzle–Pass–

Solution–Comment action sequence or discrete conversation block in identity
negotiations. The following exchanges may be an example of such a block for an
unofficial identity negotiation (U) of a customer (C) with a salesman (S):
C: I have one son who will take over the business someday.
S: How old is he?
C: Six.
S: So it won’t be for awhile before he’s ready to take over. Ha, ha.
Such four-turn action sequences are very common in the conversations among the
40 customer–salesman parties and for both official and unofficial identity
negotiations. Though not always present, a question by the second party is a
common identifying feature of the ‘‘Pass.’’ A ‘‘Pass’’ is a response by participant two
to an opening remark by participant one that requests and/or allows participant one
to elaborate on the opening remark. The following exchange illustrates the presence
of a question in the Pass in an Us (unofficial identity negotiation of the salesman)
in the same meeting as the U just reported.
S: When I was in the military I was out at Fort Knox.

C: At Fort Knox?
S: Yeah.
C: I’ll be.
Us and Uc occurred commonly during the first 1–5 minutes of the first meeting
between the salesmen and clients (Taylor & Woodside, 1980). Us and Uc are likely to
be planned specifically to occur by the salesman as a method for developing referent
power over the customer.
Thus, Wilson (1977) hypothesizes that the initial exchanges during a dyadic
customer–salesman exchange is devoted to source legitimization attempts by the
salesman. This may include attempts by the salesman to induce Os, Oc, Us, and Uc.
‘‘Unless this basic acceptability [of source legitimization] is developed, further

communication tends to be ineffective if not impossible’’ (Wilson, 1977, p. 36).
Identity negotiations require the willingness of both parties to participate in
the exchange. While such negotiations can be stopped during the action sequence,
specific changes in the topics of conversation nearly always occurred after
participating in the full Puzzle–Pass–Solution–Comment sequence, in the meetings
of the 40 customer–salesman parties. Thus, some action sequences occur that
resist interruptions, and with even an unwilling participant, they run to completion,
as Schenkein (1978, p. 74) notes.
Examples of discrete blocks of exchanges believed to represent Os, Oc, Us, and Uc
are provided from different salesman–customer–parties in Tables 1a, 1b, 2a, and 2b.
The exchanges in the two tables are taken from the same meeting. Different customer
parties and two salesmen (A or B) are included between the tables.
Notice in comparing Oc’s in Tables 1a, 1b, 2a, and 2b that the third turns are
different. In Table 1, the Solution, or third turn, is a confirmation to the candidate
solution offered by the salesman to the customer’s first turn: ‘‘I work for State Farm
Insurance.’’ The Oc in Table 1 may be best described as an example of the following
action sequence:
C: Identity-Rich Puzzle
S: Candidate Solution
C: Confirmation
S: Comment.
More than a confirmation is provided in the third term in the Oc in Table 2a:
‘‘I make all the decisions for this shop, but he does as far as the corporation goes.’’
Thus both confirmations and/or ‘‘Identity-Rich Solutions’’ (using Schenkein’s term)
Table 1a: Official identity negotiation for salesperson A and a husband (HC) and
wife (WC) (couple 1) with salesperson’s turn first.
Salesperson related Customer related
S: I’ve been with Protection Life for 5 years, WC: I work for State Farm
and [pause] Are you familiar with Protection Insurance.
Life?
WC: More or less. S: You work for the insurance
department?
S: They’ve been in business since 1923 and they WC: Yeah.
Are the ___ ranked company in South
Carolina. They own such things as Southland
Mall and WTS TV.
S: Ah, oh! I’m in big trouble
already.
WC: I didn’t know that. WC: Complaint department.
Table 1b: Unofficial identity negotiations a customer and a salesperson early during
a first meeting with customer’s turn first.
Customer related Salesperson related
HC: I would like to think that I will be S: My wife is a big bowling fan. I use
a professional bowler in five years. to bowl when I was small and do a lot
of pin settingy but that was before
you guys time.
S: Oh, really? Do you have a program HC: Yeah, we do. They still have them
to that end? Are you getting advice? some places in Ohio.
HC: Really, I know several pros and S: Yeah, my legs use to hand down in
They are helping me but I haven’t The pits and bowlers got me y ha,
mapped out a plan. ha. They would take side bets.
S: That’s interesting. I’m going to HC: Yeah, I bet so.
follow your progress real well.
Note: S, Salesperson; WC, wife customer; HC, husband customer; S, salesperson.
Table 2a: Official identity negotiation for salesperson A and a husband (HC) and
wife (WC) (couple 2) with salesperson’s turn first.
Salesperson related Customer related
S: Are you familiar with our WC: My husband makes the decisions on
company — Protection Life? fringe benefits.
WC: Protection Life Insurance? S: He makes all those decisions, hug?
S: Yeah! WC: I make all the decisions for this shop,
But he does as far as the corporation goes.
WC: Yeah, I am. S: Ah, huh.
Table 2b: Unofficial identity negotiation for salesperson A and a husband (HC) and
wife (WC) (couple 2).
Salesperson Customer related

related
None WC: This shop serves as a showcase for our plantscaping. For
example, we did the airport.
S: No kidding?!
WC: The people saw that we can make bids at some of the banks
and they come see what we have.
S: Yeah! They can take a look.
may be offered as solutions to a prior Pass or Candidate Solution. The willingness to

provide Identity-Rich Solutions versus Confirmations only may be an indication to
the other party, especially the salesman, of the likely outcome of the meeting.
The first part of the Os in Table 1a, ‘‘I’ve been with Protection Life for 5 years now
and y’’ is a tentative presentation of a Puzzle by the salesman which receives no
verbal response from husband and wife as customers. The salesman continues the
turn then but changed the subject from himself to the customers and presented
a direct Puzzle in the form of a question: ‘‘Are you familiar with Protection Life’’
[disguised name]. The Comment by the wife (WC) in the second turn, ‘‘More or less’’
appears to be ignored by the salesman. The Pass is followed by an Identity-Rich
Solution in the third turn by the salesman. Any Pass by the customer is likely to
follow with this solution, is one conjecture which may be reasonably suggested. Only
the provision of a Candidate Solution in the second turn in place of a Pass is likely to
change the contents of the third turn.
The original start of the Puzzle in Os in Table 1a, ‘‘I’ve been with Protection Life
for 5 years now, and y,’’ includes a pause indicated by the three dots. The pause
allows time for the customer to start a conversation turn and participate in an Os.
The customer did not participate in an Os, that is, the negotiation attempt by the
salesman failed. The salesman immediately executed a restart that was a more direct
Os, ‘‘Are you familiar with Protection Life?’’ than indicated by the original start.
‘‘Are you familiar with Protection Life?’’ may be better classified as a start to a
discrete block of exchange to help establish source credibility, that is, the company is
trustworthy and capable, rather than an fs Distinguishing characteristics of company
versus salesperson specific official identify negotiation can be made. Both, one and
not the other, or neither sometimes occurred in the meetings tape-recorded.
The Comment in the fourth turn of the Os in Table 1 is followed by a comment
and a transition to a new conversation topic by the salesman leading to an Oc.
The salesman announced to the customer that he wanted to learn a great deal of
information in the next few minutes of the meeting before the Oc occurred.
‘‘You didn’t know that. They try to keep some of those things not too much
before the public notice, but they are a good, strong, substantial company. And the
people that have been policyholders of theirs for years, they like to take good care of
them. So they assigned the Buyers [name disguised] family to me, so that’s why
I called C.C. [husband] and asked him if I could come out and sit with you and go
over what you presently have and maybe talk about some things you might want to
accomplish in the future. OK? And to do that, I’ve got a little questionnaire that
takes a few minutes to answer. Now we might discuss some things that are somewhat
confidential. If you’d care not to divulge something say so, OK? But it’s basic stuff.
For instance, Linda, where do you work?’’
This presentation serves several purposes beginning with a comment to close an Os
to an attempt to legitimize the company, to provide a rationale for the meeting with
a vague implication to purchasing additional insurance, to gaining cooperation to
complete a ‘‘little questionnaire,’’ to starting an Oc. The ‘‘little questionnaire’’ did not
include a written form but the term was used to set the immediate future exchanges
to facilitate participation in several successive Oc’s.
The use of such a strategy may appear to be intuitively beneficial for a salesman to
use. However, in several other instances of the 40 exchanges no Oc’s were found. Oc’s
did not occur in all customer–salesman meetings. An Oc followed by another type
of identity negotiation (Os, Us, and Uc) often occurred but the inclusion of all four
identity negotiations occurred for less than 50 percent of the meetings.
Successful bargaining of information by the salesperson (i.e., the customer
receives, understands, and accepts information offered by the salesman) may depend
significantly on the ratio of Oc and Os. Specifically, when Oc/OsW1, then the
probability of successful information–negotiation by the salesman is likely to increase.
Both the number of Oc and Os as well as their ratio are likely to relate to the sales
outcome. The likelihood of purchase increases as (1) Oc W0, (2) Os W0, and (3) Oc/Os
W1. However, the primary effect of identity negotiations is likely to be on other
discrete blocks of exchanges during customer–seller meetings and not the purchase
outcome. Identity negotiations may be related most strongly with information
bargaining. In turn, information bargaining may relate most strongly to other
discrete blocks of exchanges occurring during the meeting.
A retrospective analysis
Following the face-to-face meetings for three customers and salesmen exchanges,
separate additional meetings were held with each customer and one of the
researchers. The tape recording of the initial meeting was played to each customer
with the request to stop the recording every few minutes and comment on what
was happening. In several instances, the researcher stopped the tape and asked the
customer why he/she made a particular comment and what did the customer think
‘‘when the salesman said that?’’ The retrospective comments by the customer to the
original meeting were cape recorded.
The same procedure was followed with the salesmen. A separate meeting was held
with the relevant salesperson and the researcher that focused on discussing how the
salesperson interpreted what was happening at various points in the original meeting.
The tape recording of the original meeting was stopped at the same locations in both
of the separate meetings with each salesman and each customer and each was asked
to describe what was happening a that point, what each was thinking at the point in
the conversation.
Tables 3 and 4 provide two brief excerpts involving identity negotiations from
the original meeting of a customer and salesman with their retrospective comments.
The Us in Table 3 is followed shortly by the Oc in Table 4.
Both the salesman and the customer report in their retrospective comments
that they were manipulating the conversation, ‘‘I was trying to build rapport y’’ and
‘‘I was just trying to find out y’’ Both report a discovery about the other party in
exchange.
Note that the original excerpt begins with a comment by the salesman. This
refers to a previous Us. Following ‘‘That’s interesting,’’ the salesman elaborates an
Identity-Rich Puzzle. ‘‘That’s where I got my MBA degree (University of South
Table 3: Unofficial identity negotiation of salesperson A and retrospective

comments of salesperson A and customer.
Conversation
S: That’s interesting. That’s where I got my MBA degree [University of South

Carolina].
C: Did you happen to know, by chance, Professor Olin Poe?
S: I worked for Olin Poe for a year and a half. I was his graduate assistant. He taught
me everything I know today. So, anything, I do wrong, blame it on him y ha, ha.
C: OK. Lay it on my head. I’ve been taking notes y ha, ha. Tell me what I need to
maybe what I don’t need.
Seller’s retrospection Customer’s retrospection
At this point, I was trying to build I was just trying to find our more about
rapport. I discovered this common the salesperson. e had both gone to
ground between us — the professor the same school, and I was surprised
that we both knew. So, I made to discover he had worked for this
several references to this fact. professor. It was good to know we
had a common friend.
Analysis
This conversation and retrospections represent one attempt by the seller to evoke
referent power as a means of influencing the customer. Similarly, the customer
sought to learn if some commonality existed with the seller prior to the
conversation turned to product-related topics. Attempt to construct seller–
customer similarities and life-history overlaps focused frequently on (1) having a
common friend, (2) attending the same or similar colleges/universities that
competed in sports programs (e.g., football), and (3) the same attitudes or hobbies
(e.g., bowling).
Carolina).’’ This is followed by a puzzle expansion plus candidate solution by the

customer and a confirmation by the salesman. Then the customer comments ‘‘OK’’
and shifts the conversation to indicate willingness to accept insurance information,
‘‘Lay it on my head y’’ A brief analysis by the researcher appears at the bottom of
Table 3 on the exchange.
The Us in Table 3 appears to serve to help permit the occurrence of the Oc in
Table 4. Notice that the Us in Table 4 represents the following scheme:
S: Puzzle-Candidate Solution,
C: Confirmation,
S: Comment-Conclusion,
C: Confirmation,
S: Transition.
Table 4: Official identity negotiation of salesperson A and retrospective comments of

salesperson A and customer.
Conversation
S: Is it true that you don’t have anyone that’s dependent on you?

C: That’s right.
S: Right? OK! So, if something were to happen to you then, who ever your estate was
left with would really have no hardship?
C: That’s right.
S: In light of that, let’s look at your present plan.
Seller’s retrospection Customer’s retrospection
What I was trying to establish here is I was wondering why he brought this
the fact that no one would suffer an point up. I remembered that my
economic loss at her death. She mortgage company recommended, but
Needs only sufficient insurance to did not require, that I purchase a
cover burial expenses. Her present mortgage protection policy when I
coverage was too much. bought the house.
Analysis
The salesperson enters the interaction with a well-defined goal to convert the
prospect’s present term policy to a whole life policy. A 3-step strategy was
undertaken: (1) establish a rapport with the buyer; (2) show the present policy was
not needed, and (3) convince the buyer that a whole policy better fit her needs.
Kathy had no prior knowledge of the salesperson; thus, the had no well-defined
goals or strategies. Only vague goals of ‘‘seeking information’’ were cited by Kathy
as a reason for meeting with the salesperson.
The coupling of a puzzle and candidate solution by either the salesman or

customer may occur most often after both Us and Uc have occurred. The coupling of
verbal actions may serve to imply to the other party that the conversation is directed
to a specific goal.
Notice in the salesman’s retrospective comment in Table 4 that he has a hidden
agenda that he expects the customer to discover. The customer fails evidently to
discover this agenda. The salesman evidently does not realize this, or he believes the
discovery to be unimportant to future agendas since he shifts the conversation:
‘‘In light of that, let’s look at your present plan.’’ The salesman may be about to
elaborate on the original Oc in the conversation about the present plan to accomplish
his goal mentioned in his retrospective analysis. Later in the meeting he specifically
tells the customer the conclusion he believes she should reach based upon this and
other Oc’s.
Conclusions
Some discrete parts or blocks of buyer–seller interactions may represent identity

negotiations of both the seller and buyer. Such identity negotiations are classifiable
as official or unofficial. Possibly more categories than two should be used to classify
such negotiations, e.g., semi-official, semi-unofficial.
Identity negotiations represent only a part of the total exchanges in meetings of
buyers and sellers. As theorized by Wilson (1977) and Taylor and Woodside (1980),
attribute delineations and attribute value negotiations are likely to occur in such
meetings. The important point is that interactions appear to occur in discrete blocks
of exchanges linked by transitions. The separating of the seller’s and buyer’s
interactions into different classification categories may be useful but also violates
the nature of the exchange. Both the analysis of chunks of exchanges and complete
exchanges, as well as classifying buyer’s and seller’s turns into separate categories are
advocated for use within the same research program.
Detailed analysis of salesman–customer conversational turn-taking in natural
settings is a necessary step to understanding bargaining behavior in marketing
exchanges. Case-by-case research programs of salesman–customer meetings are
needed for marketing exchanges in several settings. Several empirical studies
incorporating such ethno-methodologies are available (e.g., Lombard, 1955; Browne,
1973; Varela, 1971) which reduce the arguments that (1) the approach is too time
consuming and too much work, (2) the customers and salesman will not agree to be
observed or tape-recorded, and (3) it is too obtrusive to be valid.
Lombard (1955) and his associates report substantial numbers of meetings of
20 salespersons and customers can be observed meaningfully in six months. Browne’s
(1973) participant observation study of salesmen and customers meeting in a used-
car lot indicated that nearly all the customers and salesmen were willing to be
observed. The rates of sales of the salesperson during the studies by Browne (1973),
Lombard (1955), and Taylor (1977) were not significantly more or less than the sales
recorded before the study nor the sales expected by the salespersons. Given these
findings, the recognition of the need to learn ‘‘theory in use’’ (Zaltman, LeMasters, &
Heffring, 1982), and the serious problems with self-reports (cf., Wilson & Nisbett,
1978), research that includes ‘‘being there’’ is necessary in most research designs on
marketing exchange behavior.
Chapter 10
Constructing Thick Descriptions of

Marketers’ and Buyers’ Decision Processes in
Business-to-Business Exchange Relationships
Synopsis
A central finding in the relationship marketing/buying literature is that the

thought and decision processes by both marketers and buyers include a series of
branching, if-then, questions and answers. For example, will customer X accept
a 7 percent price increase? The correct answer: acceptance depends on the
changes in the other attributes on the table (i.e., in the bid proposal or product–
service design). Consequently, from designing and evaluating bid-purchase
proposals to evaluating the current state of the overall seller–buyer relation-
ship, the perceived value of the level of any given attribute depends in part of
the value perceived in the levels of several other attributes. Possibly, business-
to-business decisions and outcomes may be understood best by constructing
thick descriptions of the multiple contingency paths that marketers and buyers
think about and sometimes enact when deciding. Chapter 10 reports the use of
two ‘‘think aloud’’ methods to learn the contingency thoughts and decisions of
marketers and buyers of industrial solvents. The main conclusions of the study:
designing generalized ‘‘gatekeeping,’’ contingency, models of if-then decision
paths can be achieved; these models are useful for constructing accurate
behavioral theories of marketer–buyer relationships.
Introduction: The Contingency Thinking Proposition

Buyer behavior, marketing behavior, and organizational research includes tracing and
generalizing contingency models to include the thoughts and decisions of the multiple
interacting actors (e.g., see Howard et al., 1975; Bettman, 1979; Nutt, 1993, 1998).
Applying multiple-person, contingency models to marketer–buyer relationships may
increase understanding of the triggering thoughts, decision frames, and events
resulting in alternative paths and outcomes occurring in such relationships.
Such modeling may be useful for predicting outcomes to multiple links in chains
of if-then thoughts and decisions: if we do X and they respond with Y, so our
response is T and their further response to W, then what happens next? Such
modeling is complex and complicated further by the involvement of at least three sets
of additional sets of person-relationships:
(1) The buyer’s relationships with the marketer’s competitors;

(2) Persons in multiple departments in the buying organization intimately involved
with some of the details in the process; and
(3) third-party participants (e.g., consultants) in the seemingly two-way relationship
(e.g., see Choffray & Lilien, 1978; Biemans, 1989).
Weitz, Sujan, and Sujan (1986) emphasize the contingent relationship between
adaptive selling and effectiveness. They define adaptive selling as the altering of the
sales behaviors during a customer interaction or across customer interactions based
on the perceived information about the nature of the selling situation. The concept of
adaptive selling can be generalized to describe adaptive buying: altering buying
behaviors during an interaction with a marketer or across marketer interactions
based on the perceived information about the nature of the buying situation.
Similarly, Montgomery (1975) provides an insightful generalized ‘‘gatekeeping’’
model of such adaptive buying behavior for supermarket buyer decisions.
Chapter 10 describes the use of collecting thick descriptions (Geertz, 1973a;
Sanday, 1979) of marketer–buyer relationships for building contingency models of
chains of if-then decision paths. The main objective is to achieve deep understanding
of the thought paths and decisions: actually enacted by the marketer and buyer in
these relationships in response to the communications and actions proposed and/or
taken by other party; and the possible alternative paths they are prepared to take,
and sometimes do take depending on other responses foreseen by the other parties in
the relationship.
Here is the central research proposition that is relevant to achieve these objectives:
learning the thoughts, decisions, and actions of the participants in buyer–seller
relationships is necessary for achieving deep understanding of marketer–buyer
relationships.
The study this chapter reports applies ‘‘the think aloud method’’ (van Someren
et al., 1994) and ‘‘direct research’’ (Mintzberg, 1979) to demonstrate a field study
application to support this central research proposition. The think aloud method
consists of asking people to think aloud while solving a problem and analyzing the
resulting verbal protocols (van Someren et al., 1994, p. xi). Direct research is tracing
the patterns in streams of decisions over time by ‘‘measuring in real organizational
terms’’ — getting out into the field into real organizations: ‘‘Research based on
description and induction instead of implicit or explicit prescription and deduction’’
(Mintzberg, 1979, p. 588).
The next section includes a limited review of extant literature on contingency
modeling relevant to buyer–seller relationships. Then, the chapter offers five core
propositions based on the findings from this literature. The third section then describes
a data collection approach that includes the two research methods. The following
section presents results from a field study to understand and describe contingency
Constructing Thick Descriptions of Marketers’ and Buyers’ Decision Processes 209
thinking. The final section includes implications for theory building on exchange
relationships.
Literature on Contingency Modeling of Exchange Relationships
Substantial literature has emerged on modeling the thought and behavior processes
(i.e., through time) of buyers, sellers, and buyer–seller relationships. For example, Cox
(1967) presents verbal models of the thoughts and actions of two consumers on their
household buying decisions 7 data were collected using weekly interviews over 15 and
16 sessions. Cox reports that the concept of perceived risk emerged from the data
collected in this study (see Bauer, 1967). Based on think aloud, supermarket-shopping
interviews of two consumers, Bettman (1970, 1971b) develops a graph theory approach
to comparing consumer information processing models. The consumers’ development
and use of simple decision rules to aid in making choices quickly to reach satisfactory
outcomes is a key finding from Bettman’s work (also see Bettman, 1979; Payne,
Bettman, & Johnston, 1993). Gladwin (1989) uses ‘‘ethnographic decision tree
modeling’’ to describe the contingency paths in buyer–seller relationships for Malawi
farmers’ credit buying thoughts and decisions. Gladwin emphasis the need for multiple
revisions in the field (while the data are being collected) for model-building to improve
accurate description and prediction of the resulting contingency model — a similar
observation is made by Morgenroth (1964) who urges making several rounds of
revisions of contingency thought-decision models via the think aloud method and
direct observation.
From perspectives of marketing thoughts, decisions, and actions of relationships
with customers, Browne (1973, 1976) provides verbal and diagram models of used car
‘‘salesmen’s’’ thoughts, decisions, and actions with customers, and sales managers.
Browne provides the following insights about this direct research study:
Model building is an ongoing process. Because a participant-observer

does not go into the field with hypotheses, the end point of such a study
is not always obvious y One of the advantages of this method of
model building is that you can use the group’s opinion for verification;
that is, various participants can be queried for their opinions on the
accuracy of the model. (Browne, 1976, pp. 81–82)
Within the gasoline distribution industry Morgenroth (1964) and Howard and
Morgenroth (1968) provide detailed maps of the contingencies in thoughts
and decisions of two marketers (working in the same firm) based on the known
and possible behaviors of customers and competitors. The resulting ‘‘decision
systems models’’ (DSA) both described and predicted accurately price-shift decisions
made by the marketers for different contingency situations. The Morgenroth (1964)
and Howard and Morgenroth (1968) reports are based on a research method
combining thinking aloud with document analyses of past decisions.
Capon and Hulbert (1975) review several early studies in the literature using
DSA to learn the contingencies in marketers’ thoughts and decisions. The earlier
work of Montgomery (1975) in marketing is similar in method to the studies reported
by Gladwin (1989); Montgomery develops a composite ‘‘gatekeeper analysis’’ of
the contingency decisions of supermarket buying committees to proposals made by
manufacturers. The proposals were offers of new grocery products. The result
of Montgomery’s gatekeeping analysis is a contingency model of the questions,
answers, and decision paths representing the buying committee’s deliberations.
Montgomery (1975, p. 263) describes the advantages of this research method:
The gatekeeper analysis makes no assumptions as to the scale of the

data or its underlying distribution. It also allows for nonlinear
conditional interaction among the variables. Further, the gatekeeper
approach would seem to be a closer approximation to a buyer’s
decision process [compared with multiple discriminant analysis]. It is
difficult to imagine that a buyer cerebrally forms a weighted linear
combination of variables and compares this score to a cut-off level as is
done in linear discriminant analysis. It seems more plausible that they
exhibit a decision process something like the gatekeeper tree where
serious failure [contingency path leading to a reject decision] at some
point spells difficulty for a product offering.
The literature reviewed serves two intended purposes. First, this literature
illustrates related literature streams in marketing, buyer behavior, sociology,
psychology, and anthropology focused on providing ‘‘thick descriptions’’ (Geertz,
1973a) of the thoughts, decisions, and actions of persons involved in buying and
selling behaviors. Second, the review provides grounding for developing several core
propositions for the study of thinking and decisions occurring in seller–buyer
relationships.
Finally, Nutt (1998) provides a valuable normative implication from his research
on how 352 strategic decisions were framed: success improved when claims (made by
decision participants on how problems should be framed) were performance based.
Nutt’s work is important for at least two reasons: relationship research efforts often
overlook how problems are framed; even though the decision frame used
substantially influences both the direction and outcome quality of decisions.
Five core propositions Based on reviewed theoretical and empirical literature, we
developed the following five core propositions to guide a field study on industrial
seller–buyer behavior:
P1. Individually, both sellers and buyers identify and work with contingency
sub-routines in thinking and deciding relationship issues with each other.
P2. Individually, both sellers and buyers build-in specific objectives to attain
within their contingency decision processes; these objectives relate to attaining or
surpassing specific levels for specific attributes. For example, achieving ‘‘5 per
cent + cost reductions’’ not ‘‘cost savings’’ may be a specific level of a specific
attribute guiding the contingency thinking for some customers. (The expression,
‘‘cost savings,’’ is used in industrial purchasing to refer to avoiding some share of
price increases due to inflation; ‘‘cost reduction’’ is a purchasing term referring to
achieving price decreases for purchases in the current contract year versus the
previous year.)
P3. Individually, both sellers and buyers use a limited number of thought paths to
classify the other party into a limited number of types. Buyers often classify sellers
first according to the current state of the relationship: ‘‘preferred supplier,’’
‘‘secondary supplier,’’ and ‘‘vendor used to check market prices.’’ Sellers often type
buyers by the thinking and actions the seller expects to encounter in meetings with
buyers: a cost reduction buyer; a cost savings buyer; an insure delivery buyer.
P4. The majority of sellers and buyers can be segmented into a limited number of
thinking-deciding groups based on data collected from think aloud and direct
research methods. Both sellers and buyers use a few ways-of-thinking variables to
assign the other party in the relationship; for example, a seller may reflect that
buyer X always works hard and she is successful in achieving cost reductions with
my firm; I have five other customer firms like buyer X.
P5. Applications of both the think aloud method and direct research both
complement as well as uncover unique information about the thinking and
deciding processes of sellers and buyers.
Applying both research methods permits confirmation of processes uncovered by

each approach; such applications may result in learning important nuances left
uncovered by the use of only one research method.
Method
The steps in the research methods used are summarized in Figure 1. The cooperation
of the marketing and sales departments of an international chemical manufacturer
was gained. (For competitive purposes, the names of the firms in this report are
fictitious.) The data collection started with several direct research steps: individual
interviews with the vice president of marketing, two sales managers, a pricing
manager, and four sales representatives. The direct research steps included document
analyses and direct observation of seller–buyer meetings at customer manufacturing
plants and distribution centers.
These initial steps were taken to achieve several objectives: to learn how sales reps,
sales managers, and the VP of marketing classify customers; to develop several
problem scenarios in the form of conjoint exercises for the buyers to solve while
thinking aloud; to learn the industry language used in selling and buying industrial
solvents (the product category selected for the study); to learn the contingency paths
developed and then selected in real-life conversations between sellers and buyers.
(Models of individual preference functions based on preference rankings are not
included in this report.
212
1
2 3
Researcher meets
Write-up narratives of Arrange meetings with sales
individually with
Initial meetings and reps selected for high
VP Marketing
give copies to expertise in adapting selling;
and sales managers
VP Marketing and sales interview selected sales
Start to gain descriptions
managers; request each to reps (7 hours each) and
of conversations,
make corrections and observe their face-to-face
decisions, and events
elaborations of the conversations with
with sales reps, and
narratives customers
with customers
7
Schedule individual 6 4
5
meetings with customers Receive corrections and Receive corrections/elaborations form
Individually give copies
selected as representative elaborations from VP VP Marketing and sales managers;
of scenarios and negotiation
of different seller-buyer Marketing, sales managers, write-up draft scenarios of: seller-
process routines to VP
scenarios and negotiation and sales reps of written buyer-encounters based on
Marketing, sales managers,
routines; use think aloud scenarios and negotiation interviews with sales reps and direct
and sales reps; ask for
methods to learn how process routines; revise
corrections, and observations of seller-buyer

customers solve different both the scenarios and conversations, negotiation process
elaborations.
seller buyer scenarios routines routines for alternative selling-buying
-
and buying processes situations.
Figure 1: Applying the think aloud method for learning categorized knowledge of selling–buying situations.
Buyers in three of four buying firms were willing to describe their thinking
processes only partially for the conjoint exercises; complete rankings for both buying
exercises were provided by the buyer in only one firm — the buyer in the smallest
customer firm studied.) Using direct observation, full-day interviews and customer
visits were made with each of the four sales reps. Three six-hour interviews were held
with the VP of marketing; these interviews included detailed reviews of the
relationships with 20 customer firms, including document analysis of the histories
with these customer firms.
Following the development and revisions of alternative, written selling-buying
scenarios, a second group of four buying firms (not including firms observed in
selling-buying meetings in the direct observation data collection stage) were selected
and interviewed. Data from applying both the think aloud and direct research
methods were collected in this step (shown as step 7 in Figure 1).
Two-hour interviews were scheduled with each of four buyers at their business
sites (two in Ohio, one in Pennsylvania, and one in South Carolina). The interviews
included asking open-ended questions and reviewing documents to learn the detailed
steps used in buying solvents (the direct research method). Second, each buyer was
asked to think aloud when solving two seller–buyer scenario problems. As an
inducement for participating in the study, each buyer was given a published
monograph focused on mapping industrial buying processes. All four buyers first
contacted agreed to participate in the study; the buyers agreed that the need existed
for detailed knowledge of the thinking and deciding process buyers actually use in
working with suppliers.
The attempt was made to select buyers for the separate, individual, on-site interviews
to represent different customer firms segmented by contingency thought processes.
Limitations
The study reported here does not include an attempt to generalize findings to a
population, nor to generalize findings to theory. The data presented are limited to
describing problem framing and relationships for a limited number of buyer–
marketer interactions focused on selling–buying one industrial commodity.
Some of the buying heuristics described in the findings reflects a cost-driven
industrial environment. While perhaps useful for understanding the nuances in
relationship marketing-and-purchasing, our emphasis is on the research process in
providing thick descriptions rather than on generalizing the results. The reported
study and findings are exploratory.
Findings
The results of the study are summarized in Figures 2–6. Figures 2–5 are maps of the
thoughts and decision processes of the four buyers. Figure 6 is a composite summary
of the thinking and contingency paths found to be used by the sales reps and sales
managers in adapting their pricing decisions to the thoughts and decisions of

customers.
Decision Frame of Buyer Focused on Achieving ‘‘Cost Reduction’’
How a professional buyer focused on achieving annual ‘‘cost reductions’’ frames for
buying and relationship decisions is summarized in Figure 2. A cost reduction is
paying less for the same quantity and quality purchasing requirements in the current
year compared to the previous year; achieving a cost reduction differs from a ‘‘cost
avoidance’’ objective. Cost avoidance is paying a price increase lower than a noted
and an agreed-upon view of the industry price increase for a product. Framing a
decision to focus on achieving a cost reduction objective likely is viewed by nearly all
professional buyers as a more aggressive stance compared to focusing on achieving a
cost avoidance objective.
The greater value of framing decisions using a cost reduction strategy versus cost
avoidance or other strategies (e.g., vendor performance rating strategies) was
supported by several claims made by the buyer, Robert Elwood. See Nutt (1998)
for research on claims made by decision participants to influence how decisions are
framed.
Mr. Elwood claimed: ‘‘[I’ve] achieved $2 to $4 million annually in net savings since
becoming department head of purchasing — my predecessor and most of the buyers
reporting to him were downsized [transferred or fired] before I came to the
department; I do not use vendor management rating forms because they are not
focused on purchasing performance;’’ and ‘‘my department has grown in size — four
buyers now report to me, only one buyer reported to me when I became department
head three years ago.’’
In a separate interview, both a senior marketer (the solvents product manager) who
meets face-to-face with Mr. Elwood and a local sales representative focused their
annual and monthly discussions on the following topics: reaching specific price-points;
creating unique value-added projects; and gaining/offering ‘‘preferred supplier
participation’’ by receiving/offering 75–100 percent of the annual purchase require-
ments.
Note in Figure 2 that Mr. Elwood uses a ‘‘second look’’ and additional looks
strategy (box 6) as a decision tool to achieve his cost reduction goals. Such look
strategies include the buyer providing limited competitive bid information to
competing vendors followed by the vendors responding usually with additional price
reductions. Mr. Elwood pointed out that he ensures that senior management is aware
of cost reductions being achieved with his unique ‘‘Cost Reduction Reports’’ he
sends to senior management on a quarterly, and sometimes monthly, basis.
Framing decisions using vendor performance rating systems Figure 3 is a
summary description of a contrasting decision frame enacted by buyers in the same
industry as Mr. Elwood. Related to Figure 3, the two buyers interviewed based their
evaluations of competing suppliers and vendor proposals using a ‘‘New Generation
Division Vendor Performance Rating System.’’ This system includes a written survey
3
1 2
Decide on cost reduction goals; seek 2 to 3
“Master Prepare and
bids for each product line; include estimated
Requisition send request
requirements by plant locations; inform
Reports” for proposals
vendors of potential requirements for
received from (RFPs)
“preferred supplier” participation
plants to vendors
4 5
Yes
No Receive bids: Will vendor support (fund) unique
are cost reduction value-added projects or provide other
and value-added unique and valuable terms and conditions?
goals achieved?
8 Yes No
Award contracts;
6 Notify suppliers of
Offer vendors “second look” 7 annual quantities
and seek price decreases due Vendors awarded by plant 9
to efficiencies of long-term revise locations Continue
contract if “preferred supplier” and submit negotiation
status award made to vendor; new
continue with “third look” proposals
when necessary
10
Prepare “Cost Reduction Reports”(CPRs);
monitor suppliers for contract conformance;
send CPRsto senior management
Constructing Thick Descriptions of Marketers’ and Buyers’ Decision Processes
Figure 2: Robert A. Elwood: Decision systems analysis for buying solvents at Smith Metal Works Finishing Plant.
215
216
1 2 3
Receive production Develop RFQs and send Review performance of supplier X using “New Gen Division
budget and requirements to vendors; receive Vender Performance Rating System”:
forecasts from logistics proposals from vendors • Delivery on-time? Specs in line?
• Quantity objectives met? Strong tech support?
No • Provided exclusives in NPD?
Is supplier willing and
No Yes
Can we drop vendor able to improve
X as a supplier and not performance to
have to single-source? achieve certification?
No Is a price increase included in vendor X’s proposal?
Yes
No
Did supplier X have Yes
substandard reports in
recent previous years?
Request X to create Is price increase less than 14% and
self-improvement Justified by increases in market prices? No
No
plan to achieve
re-certification Notify X that price
quote is above market Yes
price; receive lower
price quote from X
Yes
Award vendor X same share of requirements
as previous year or increase share award
Drop vender X Reduce share of business

as a supplier award to X
Figure 3: Solvents buying process by Willard F. Boys and Dale Apple at Smith New Generation Division.
form that requires plant buyers at multiple locations to rate each supplier on 20
attributes on a monthly basis.
The five attributes reported by Mr. Boys and Mr. Apple, the buyers identified in
Figure 3, as core objectives are described in box 3. Note that cost reduction is not an
objective in this decision frame. A trigger price-increase of 14 percent is part of a cost
avoidance strategy used by Mr. Boys and Mr. Apple (see box 10 in Figure 3). These
buyers are not focused on achieving cost reductions in the form of price reductions;
rather, they are focused on evaluating suppliers and other vendors using an elaborate
rating system and ensuring that their suppliers achieve ‘‘certification’’ — high-
performance scores across all attributes in the rating system. Note that this
relationship includes the requirement made by the buyer for suppliers not achieving
certification to provide a written plan on the self-improvement strategy the supplier
will create and implement to achieve certification (box 7).
Framing relationships focused on achieving materials-delivery requirements.
Related to Figure 4, fear of not having the necessary raw materials for keeping the
plant operating for 3 shifts per day, 365 days per year, is the overriding issue for the
two buyers interviewed. Consequently, multiple sourcing and never single-sourcing
with one solvents manufacturer was described first in how these two buyers framed
decisions and relationships with suppliers.
Price increases for solvents were acceptable for the buyers at this firm if such price
increases met three criteria:
(1) the supplier’s delivery and quality record was void of major problems (box 7 in
Figure 4);
(2) the supplier could ‘‘justify’’ the price increase (box 8); and
(3) the price increase did not result in an increase in the ratio of the material’s cost
to the buyer’s product price (not shown in Figure 4).
These buyers reported not using a second look strategy but do sometimes call-in
suppliers to negotiate revised price points (box 10). A small manufacturer’s buying
relationship with a much bigger supplier Figure 5 is a summary of the ongoing
buyer–marketer relationship and self-questions asked by the buyer in a small (one-
plant location) firm. Solvents are a critical raw material for this firm; the buyer
prefers to single-source with one of the two known manufacturers of the solvent
being purchased.
From time-to-time (twice each year), the buyer inquires or receives a quote from a
regional distributor of solvents (box 3); such a quote may be an unrequested bid, but
the buyer does sometimes ask the distributor to bid. ‘‘I always give BIGCHEM a
second look because I know BIGCHEM will beat the distributor’s price (box 4). It is
a way for me to be sure that I am getting market price.’’
When this small manufacturing customer came out of bankruptcy protection
(Chapter 11 filing) 3 years prior to the interview, BIGCHEM was the first supplier
permitting the small customer to pay on net 30 days credit terms for tank care
deliveries; this action was cited by the buyer as the reason causing her to switch from
awarding 50 percent of the required solvents purchases to awarding 100 percent to
2 3
218
1
Develop purchasing requirements and Use informal RFQ to request price &
Receive production forecasts
identify 3 possible mfg suppliers and up annual volume commitments from
from mfg at plant
to 2 distributors on AVL selected vendors
4
Panned purchasing strategy: 5
• (1) Do not single source with one solvents mfg Receive responses from approved
• (2) Select a “primary”supplier who has capability to supply total annual volume vendors and apply planned purchasing
• (3) Select a large secondary, mfg, supplier (award 30%+ to this supplier) strategy
• (4) Select a distributor for additional back-up insurance for product delivery
• (5) Call in supplier for negotiation if price increase is “out-of-line”
6
Evaluate first: the solvent mfg having
the production capacity to meet
100% of our requirements (but this
9 mfg will not receive 100% award)
Reject No
supplier Supplier taking acceptable
corrective actions?
7
Yes Any major problems with product
No quality, delivery, or unjustified price
increases with this supplier’s current
12 year performance?
Evaluate second solvents 10
mfg’s capability of providing Call-in supplier and negotiate; No
nearly 100% of our firm’s is price reduced and No
annual requirements; now acceptable?
8
search for, find, select a

Is this supplier’s proposed price for
back-up distributor to
new contract period justifiable?
receive a 10% share
Yes
Yes
11
Offer letter of intent to award
share of business t supplier
Figure 4: Solvent buying process at three production shift manufacturer. Notes: Lee and Baur’s (Purchasing at Customer firm)
first priority is to fulfill requirements for plant to operate 3 shifts per day, 365 days per year; their nightmare: non-delivery of
required solvents; and ‘‘cost reductions’’ are not an objective. AVL, approved vendor list.
1 Yes 2
Start of annual solvent Does BIGCHEM appear Agree to year contract with
TXR contract cycle Competitive in prices to TXR? BIGHCEM
No
4
Give BIGCHEM a
3 Yes “second look”
Is a local distributor offering me
a better price than BIGHCEM?
No
7
Will BIGCHEM beat Yes
local distributor’s
5 price offer?
Are BIGCHEM’s delivery and
No
terms and conditions acceptable?
8
Agree to a one
No year contract with
local distributor
6
Seek a bid from COMCHEM Yes
Figure 5: Buying process for TXR by buyer at Petro Division of Widget Corporation.
219
BIGCHEM. The second supplier, COMCHEM, was dropped; no explanation was

given to COMCHEM. Until the time of the study, BIGCHEM was unaware of why
the share of business they received from this customer had increased and BIGCHEM
was uncertain what share of business the firm was receiving from this customer.
(BIGCHEM is the marketing firm participating in the study.)
Even though the specific solvent being purchased was crucial for manufacturing at
the customer plant, the buyer was unaware of the availability of a third manufacturer
of the solvent. She had done no search of alternative solvent suppliers. Her framing
of the buying problem was focused explicitly on the question, ‘‘Am I paying market
prices?’’ Implicitly, the buying problem was focused on availability of supply; she
reported never having had a problem with delivery from BIGCHEM in a span of
more than 10 years. Thus, she did not raise the issue of availability when framing her
thoughts and decisions when buying solvents.
The Framing Process for the Marketer
The marketer’s framing process in describing relationships with customers and in

pricing is summarized in Figure 6. While seemingly complex, the thoughts and
actions of the product managers and sales representatives in this firm are centered on
asking a short series of questions: how much business does the customer represent
(box 2)? How does the customer frame key aspects of his/her firm’s relationship with
us and our competitors (boxes 3–7)? Which objectives should dominate our response
to the customer’s response to our proposal (boxes 15 and 16)?
If the customer firm is a key account (i.e., large business for the marketer) and the
customer insists on achieving a price reduction, the marketer is likely to respond with
a ‘‘creative proposal’’ that includes: first, a low price; second, funding for storage
equipment or related facilities at the customer’s sites; and third, ‘‘price protection’’
against price increases during some of the contract period. Whether or not such an
outcome occurs depends on the marketer’s belief that ‘‘preferred supplier
participation’’ was received from the customer — a euphemism for being awarded
the largest share or 100 percent of the customer requirements for solvents.
Results Relating to the Five Core Propositions
P1 supported: contingency sub-routines. The results support the first proposition:

individually, both sellers and buyers identify and work with contingency sub-routines
in thinking and deciding relationship issues with each other. Only a few self-questions
are asked and answered when buyers and marketers decide on actions related to such
relationships.
Partial support for P2: specific objectives planned in contingency decision
processes. Individually, only some buyers may build-in specific objectives to attain
within their contingency decision processes; these objectives relate to attaining or
2. Is customer large, 3. Is customer focused on 4. Will customer singe- 13.
1. Current customer Key “consumer” Yes achieving cost reduction Yes source requirements Yes Use lowest
requests proposal from Customer for for new contract if price reduction price strategy
BIGCHEM BIGCHEM? period? offered? (set price
at Z = -1.0
No No No
14.
7. Will customer Yes Attempt to set
5. Is BIGCHEM No 6. Is customer
BIGCHEM’s use 50-50 split of price equal
sole source focused on
pricing objectives: business? to known
supplier? cost avoidance?
competitor’s
•Retain customer at price
No
a profitable p rice point No No Yes
•Capture 100% of 8. 9.
customer’s business Is customer receiving a Will customer Yes 12.
competitive bid one cent sole-source if price Set price increase 15. Observe customer
•Design proposal that below our current price? increase is less at less than response to price point
customer prefers with than inflation? inflation rate set in bid proposal;
a higher, more Yes No adjust price in second
profitable price No
round if necessary
10. 11.
Set price one cent below Increase price 10%
competitor’s bid to signal to above current
competitor that BIGCHEM price point
refuses to lose
customer’s business
Figure 6: Summary pricing, and sales negotiations, decision model for BIGCHEM based on customer decision profiles.
221
surpassing specific levels for specific attributes. Such framing behavior is illustrated
in Figure 2 for price reductions; Figure 3 for meeting certification objectives; Figure 4
for maintaining a current cost/price ratio; but no such numerical objective was found
related to the data in Figure 5.
P3 partially supported: limited number of thought-paths found. Possibly because
both buyers and sellers must focus some attention to several ongoing relationships
each day, P3 received some support: individually, both sellers and buyers use a
limited number of thought paths to classify the other party into a limited number of
types. However, the thought-paths uncovered were not focused on classifying the
other party in the relationship but rather on framing and solving specific problems
related to the relationship.
P4 untested: segmenting sellers and buyers by thoughts and decisions. In the
exploratory study reported, the number of relationships and thought-paths examined
is too limited to test P4. However, the findings described do lead to the conclusion
that such research would support this proposition — even though collecting such data
is intensive. Nutt (1993) suggests that we should not be deterred because of the great
amount of work required by such direct research. Certainly, he has not been put off.
P5 untested: combining think aloud and direct research uncovers unique
information. The results presented do not compare and contrast data from the two
research methods used. Thus, the following proposition is not tested: applications of
both the think aloud method and direct research both complement as well as uncover
unique information about the thinking and deciding processes of seller and buyers.
For the one buyer (in the small manufacturing firm, see Figure 5) who did complete
all steps related to the think aloud tasks, unique information was uncovered from
each research method.
Only when using think aloud in solving the conjoint scenario tasks did this buyer
report not being aware of the third manufacturer-supplier of solvents; the conjoint
attributes used included a factor involving three alternative names of manufacturers.
Her thoughts related to this supplier, ‘‘I would never use this supplier because no one
calls on me from this firm and I have never heard of them,’’ provided valuable
information on how she goes about framing the buying problem and relationships
with solvent suppliers.
Implications for Theory Building on Exchange Relationships
Deep understanding of buyer–seller relationships may be gained only from thick

descriptions of the thoughts, decisions, and actions that occur naturally in such
relationships. Both sellers and buyers exhibit adaptive thinking and behavior in the
relationships. While all relationships include seemingly idiosyncratic nuances, using
only a few relationship description variables, sellers and buyers may be segmented
usefully into a few groups according to the contingency thoughts they use.
Thick descriptions of marketing–buying relationships should include the thinking/
contingency processes used by both parties in such relationships. In examining the
thoughts and behaviors of industrial marketers and buyers in real-life relationships,
the adaptive selling strategies found include varying responses to the customer firm
based on buyer responses to proposals made by the marketer, and considering the
value of the ongoing relationship with this individual customer. Based on the limited
results of exploratory study described, adaptive buying strategies includes asking a
limited number of problem framing questions; these framing questions vary among
buyers in different firms but most customer firms may be segmented into a limited
number of categories (five to ten) by problem frames.
Within marketer–buyer relationships, finding customers focused on achieving
price reductions and marketers focused on achieving profit maximum price points
may be rare occurrences. Such relationships often include problem-framing steps
only obvious after collecting data from direct research.
Managerial Implications
Both marketers and buyers are likely to follow the human tendency to frame
problems and people using a limited number of issues and categories. Most often
such framing processes are done implicitly rather than explicitly (Huff, 1990; Senge,
1990). Several researchers point out that the quality of ‘‘sensemaking’’ (Weick, 1995)
is improved by making explicit our usually implicit (i.e., automatic) thought
processes (see Huff, 1990; Weick, 1995).
Improvements include increasing the abilities of sales representatives to perceive
and respond appropriately to nuances and complexities in customers’ buying
processes that would otherwise be overlooked (see Weitz et al., 1986). Thus, the
results of constructing decision maps such as those illustrated in this chapter may
serve as sales training and buying training tools. The resulting thick descriptions in
such maps identify mental markers for categorizing customers (suppliers) and
appropriate response routes to take when such mental markers appear. For example,
issues in boxes 3, 4, and 6 in Figure 6 are identified as important mental markers in
customers’ thinking — the answers to these issues dramatically affect the courses of
actions taken by the marketing firm, in this case, BIGCHEM. Complicate yourself !
Learning about such multiple decision routes, and when to take them, complicates
thinking and performing when engaging in selling and buying behaviors. Achieving
such complication is the final suggestion in the ten pieces of advice offered by Weick
(1979, p. 261): ‘‘Complicate yourself!’’ Otherwise, two natural tendencies occur:
reliance on oversimplified, novice, mental models of customer (supplier) behaviors;
and failure to build in the requisite variety to achieve ‘‘adaptiveness’’ (Campbell,
1965) — a trait necessary for healthy marketing/purchasing relationships.
Micro Segmentation and Target Marketing
Thick descriptions and mental mapping are tools useful for micro segmentation
(see Wind & Cardozo, 1974). Macro segmenting variables, such as SIC grouping,
company size, and geographical location, are fine tuned by sales representatives’ use
of micro segmenting variables. Micro segmentation includes grouping customers by
their relationships with the competing suppliers and other decision-related variables.
Knowledge of how customers frame and solve buying problems serves as the
foundation in most attempts to micro segment customers. While each customer firm
performs buying processes uniquely, most customer firms may be categorized into a
limited number of segments by how they frame and solve buying problems (e.g.,
Choffray & Lilien, 1978; Nutt, 1998).
The decision-process mapping procedures described here can be used to cluster
customers into segments according to similarities and differences apparent in the
resulting decision maps. Nutt (1998) provides a detailed empirical example of such a
micro segmentation process of buying decisions. Product managers and sales
representatives can use such maps to adapt product offerings and sales calls to match
the processes and features (i.e., specific combinations of attributes and attribute
levels) that are shown in the decision maps of different customer micro segments.
Chapter 11
Case Study Research on Means-End

Laddering Chains
Synopsis
Means-end chain (MEC) theory proposes that knowledge held in individuals’

memory is organized in a hierarchy with concrete thoughts linked to more
abstract thoughts in a sequence progressing from means (i.e., brands and
product features) to psychological and social consequences and finally to ends
(i.e., fulfillment of personal values). This chapter proposes several advances in
the theory. First, specific buying and consumption situations serve as frames of
reference when consumers are thinking about products and alternative features
of products and brands. Second, states of psychological imbalance may occur
in consumers’ minds among linkages retrieved automatically for features/
consequences and consequences/values; thus, Heider’s balance theory incorpo-
rates MEC theory and research. The theoretical and practical usefulness of
means-end research increases by asking consumers to name an acceptable
alternative to the product and brand used in a recent consumption situation, as
well as an unacceptable option and to describe the features/consequences/
values of these options. Consequently, alternative relationships of consumer/
brands (e.g., casual friendships, marriages, enmities) become relevant for MEC
theory. To examine the propositions empirically, this chapter describes
psychological schemata for four MECs that combine two consumers’ recent
consumption situations with personal values.
Introduction: Associating Brand/Features/Consequences/

Benefits/Values
In consumer research on MECs, the meanings that a consumer associates with a

product, service, or specific brand are represented hierarchically. The lowest level
depicts an object’s attributes — those that are physical or concrete, and those that are
abstract (e.g., a brand’s image). The remaining levels are consumers’ outcomes
including functional and psychosocial consequences, followed by consumers’
instrumental and terminal values (see Gutman, 1997). Laddering is a research

method frequently used for electing to do MEC studies.
Traditionally, laddering is a metaphor representing a respondent’s answers to
a series of ‘‘Why?’’ probes, typically starting with the attributes that distinguish more
desired from less desired alternatives. Asking for successive elicitations creates
a chain of features and concepts leading from product or brand attributes to one or
a few terminal values: each successive concept learned becomes a sub-goal for the
final goal. Gutman and Reynolds (1978) illustrate a resulting ladder elicited from
a respondent in a study of breakfast cereal: crunchy -has body -stays with me -
avoid snack -aids weight loss -improves appearance -romance.
In the empirical research that this chapter describes, the questions, ‘‘What makes
you say that?’’ and ‘‘What does that lead to?’’ are used in place of ‘‘Why’’ probes.
Becker (1998) emphasizes that ‘‘Why?’’ requires a ‘‘good’’ answer, one that makes
sense and can be defended. Such answers are unlikely to reveal logical flaws and
inconsistencies. Becker recommends using, ‘‘How?’’ questions.
When I asked them, [such questions] people gave more leeway, were
less constraining, invited them to answer in any way that suited them,
to tell a story that included whatever they thought the story ought to
include in order to make sense. They didn’t demand a ‘‘right’’ answer,
didn’t seem to be trying to place responsibility for bad actions or
outcomes anywhere. (Becker, 1998, p. 59)
The scientific literature for avoiding ‘‘why’’ questions in most contexts is

compelling because most thinking is unconscious and consumers typically are
unable to bring up and report unconscious processes (unconscious processes often
drive behavior that consumers are unaware that they performed; for example, see
Bargh, 2002; Zaltman, 2003). Advancing theory and research that inform knowledge
of consumer unconscious processes is worthwhile (see Malle, 1999 for a complete
theory of folk explanation of behavior, that is, a thorough theoretical development of
how humans explain everyday behavior and how the question asked affects the cause
people assign in answering).
The value for consumer theory and research of MECs and the laddering method
relates only implicitly to unconscious mental processes — and how laddering may
help to surface unconscious thoughts. For example, direct questioning of how a
cereal being crunchy aids a consumer’s love life may cause a scoffing response from
the consumer. Asking the consumer a sequence of probing questions is a form of
‘‘auto-driving’’ (Heisley & Levy, 1991); research may enable the consumer to surface
unconscious thoughts that rarely enter conscious processing.
Surfacing unconscious processes may result in tension and stress because
unbalanced associations occur among concepts now in a consumer’s working memory
(Heider, 1958). The thought, ‘‘A Jewish couple buys a German car’’ (Figure 1; see
Woodside & Chebat, 2001) implies an example of such stress resulting from the
surfacing of negatively valued concepts and images — some held unconsciously —
with the positive experiences associated with buying a car highly prized for its design
Case Study Research on Means-End Laddering Chains 227
Franz Stangl +
(Treblinka death-camp Wartime (WWII)
Commandant, VW worker slave labor
+ +
in Brazil after war)
+
+
+ VW
+ + +
+
Hitler German
-
-
- - -
- +
+ Jew - Holocaust
Posner
-
Figure 1: Extended balanced state for Posner case study. Source: Original drawing
created from the written story by Posner (1998).
features. Advancing Heider’s balance theory to person, brand, and attribute (and
attribute to benefit to instrumental and terminal values) enables more robust MEC
laddering research because the resulting ladders may uncover and display stress and
stress resolution, as well as harmonious associations.
Pieters, Baumgartner, and Allen (1995) advocate following an alternative
empirical method for MECs rather than the better-known laddering approach that
Gutman and Reynolds (1978) demonstrate. Pieters et al. (1995) elicit people’s reasons
for wanting to do something or to pursue a goal rather than to name concrete
features that they associate with a product. Bagozzi and Dabholkar (2000) apply
Pieters et al.’s approach for mapping people’s cognitive schemata regarding a specific
target, President Clinton. While Pieters et al. (1995) and Bagozzi and Dabholkar
(2000) refinements are useful for describing how specific schemata accurately predict
attitudes and intentions regarding future events, the following advances focus on
framing means-end theory and research by consumers’ recent experiences.
Heider’s Balance Theory
Heider (1958, Chapter 7) distinguishes two types of relations between separate

entities: unit and sentiment relations. ‘‘Separate entities comprise a unit when they
are perceived as belonging together. For example, members of a family are seen as a
unit; a person and his deed belong together’’ (p. 176). ‘U’ denotes the cognitive unit
between two entities, and not U the fact that the two entities are segregated’’ (p. 201).
A sentiment relation refers to the positive (L or + ) or negative (DL or –) feelings or

valuation that one gives to an entity, such as a person, activity, or object. These
relations may be for dyads, triads, or more complex cases but all relations are from
the perceiver’s subjective point of view. Thus, while a brand may possess a given
attribute or provide a specific benefit, if a consumer perceives the opposite, a not U
relation results between a brand and such an attribute or benefit. Heider (1958)
describes the two relation concepts to result in four possibilities between two entities:
‘‘U, not U, L, and DL.’’ He emphasizes:
By a balanced state is meant a situation in which the relations among

the entities fit together harmoniously; there is no stress toward change.
A basic assumption is that sentiment relations and unit relations tend
toward a balanced state. This means that sentiments are not entirely
independent of the perceptions of unit connections between entities
and that the latter, in turn, are not entirely independent of sentiments.
Sentiments and unit relations are mutually interdependent. It also
means that if a balanced state does not exist, then forces toward this
state will arise. If a change is not possible, the state of imbalance will
produce tension. (Heider, 1958, p. 201)
Unit and sentiment relations represent independent theoretical and empirical

propositions grounded in cognitions (i.e., units) and affections (i.e., sentiments).
When tension caused by imbalance arises in the mind of the individual, then the
individual is likely to exercise some mental and physical effort to eliminate the
tension.
Unbalanced situations stimulate us to further thinking; they have the character of
interesting puzzles, problems which make us suspect a depth of interesting
background y . Stories in which the stress is laid on unbalanced situations are felt
to have a deep psychological meaning. Dostoevski, for instance, describes again
and again feelings full of conflict resulting from just such situations. (Heider, 1958,
pp. 180–181)
Advancing Means-End Chain Laddering Research

with Heider’s Balance Theory
Figure 2 summarizes the application of Heider’s balance theory and related
extensions to MEC laddering research. Note that the grounding in Figure 2 includes
‘‘prequel to action’’ to emphasize that if consumer research focuses on existing
consumer/brand relationships (Fournier, 1998); such relationships depend on specific
purchase or use situations. A situation is defined as a conjunction of time, place,
objects, and person(s). Consequently, the experiences, attributes, and benefits that a
consumer identifies (to herself and the interviewer) are contingent on the structure of
the situation leading to brand purchase or use.
Terminal value Y Terminal value Z Unique terminal value
Ask: for alternative X
“What first comes-to-mind
about [reject] product/ - + + + + +
brand [the informant Instrumental Value
mentions]? What else?” Instrumental value of each benefit:
Value of L Value of M Value of Q Value of B + +
-
+ + + + Attribute C Attribute D
Attribute 1 Attribute 2 Benefit: L M Q B
+ -
+ + Ask: “What is the first
+ + + -
The reject alternative alternative (X) that
that comes-to-mind Attributes: A F H I comes-to-mind
during the interview? if you did not have P/B?
+ + - - + in this situation?”
3. Description of product P and brand B actually consumed in the act.
2. What immediate feelings/thoughts/actions were drivers to act?
1. Prequel to action: What events influences the buying/using act/context?
Figure 2: Means-end laddering research: Updating the theory and data dollection method.
Case Study Research on Means-End Laddering Chains
229
Consequently, the situation informs the attributes/benefits/value associations.

The implication for research design is that MEC data for a given respondent
should be collected for multiple situations (some of which may be relevant to the
focal brand). A focal brand is defined as the one that the consumer names as used
most often or identifies as the one used most recently; alternatively, a focal brand
may be defined as the specific target chosen by the researcher — the primary focus of
the study for which the researcher asks a respondent to describe — its product
features and consequences in experiencing such product features and how these
experiences relate to personal values. Asking informants individually to identify
a focal brand for a given product category is a less reactive procedure; the researcher
naming a target brand indicates possible self-validation problems — informants
report perceptions about the brand mentioned by the researcher that may not relate
much to the informants’ past or future lives (Feldman & Lynch, 1988). Thus, asking
the informant to report on her most recent consumption behavior and collecting
means-end data with respect to situation X as well as for situation Y may be more
relevant for the informant and reduces the occurrence of self-validity problems.
Figure 2 includes the assignment of sentiments (positive and negative signs) to
associations in the resulting laddering steps. In Figure 2, note that product P and
brand B associate negatively with attribute I and lack of I relates negatively with the
benefit B (e.g., assume benefit B is an informant’s comment, ‘‘good tasting
mouthwash,’’ for a mouthwash attribute I). Figure 2 advances MECs by suggesting
collecting data on the first alternative that comes to mind for the consumer for the
recent experience in the focal situation. For example, ‘‘If you had not consumed
brand B in situation X, what might you have consumed instead?’’ Figure 2 also asks
for reject alternative information and implies the proposition that the respondent is
able to easily remember a product/brand that would have been unacceptable for
consumption in her recent experience.
Such brand rejection data are then followed by means-end probes to bring up
brand/attributes/values associations for unacceptable alternatives for use in the given
situation. The brand strategist might ask here: how often, and for what situations,
does my product and brand surface as the reject alternative? And when rejection
occurs, what attributes and personal values surface in the respondent’s mind?
Fournier’s Consumer/Brand Relationship Typology

Fournier (1998) extends the two-party social relationship metaphor to encompass the
consumer and brands. She develops the consumer/brand relationship proposition,
including anthropomorphizing the brand as an active relationship partner — at the
level of consumers’ experiences with their brands. Her proposal of 15 consumer/
brand relationships includes arranged marriages, casual friends, marriages of
convenience, committed partnerships, best friendships, compartmentalized friend-
ships, kinships, rebounds, childhood friendships, courtships, dependencies, flings,
enslavements, enmities, and secret affairs.
Case Study Research on Means-End Laddering Chains 231
Advancing Means-End Chain Laddering Research by Applying

Fournier’s Consumer/Brand Relationships
Use of the advanced MEC laddering research that Figure 2 summarizes serves to
complement and inform Fournier’s consumer/brand relationship typology: Multiple
consumer–brand relationship types are uncovered by collecting data within given
situations of consumer/brand experiences, acceptable alternatives to these lived
experiences, as well as rejected alternatives. Equally valuable are the data uncovering
association streams of the three laddering paths (done, alternative, and rejected) to
terminal values.
Applying the Advances in Means-End Chain Laddering Research

The following four case studies illustrate applications of advances in MEC laddering
research. The four cases report two experiences for each of two informants (subjects, Ss).
Method
In order to collect the data to examine the propositions, two subjects, Eric and Peter,
were chosen to participate in two separate interviews each. Each subject is a senior in
college, in his early twenties, living in an apartment on campus at a large university in
the eastern United States. The interviews were conducted at the residences of the
respondents at two different times during the day. The two situations under
consideration for this study were the consumption of a beverage in the morning after
waking up and the beverage consumption between eight o’clock and midnight on a
Friday night. For the MEC laddering process described previously, each respondent
was asked a series of 34 questions for each situation. The questions were designed to
elicit responses that would provide the perceived attributes, benefits, and
consequences of the products consumed.
The respondents were also asked to provide alternative products that they might
have chosen in each situation as well as rejected products that would not be
considered. The subjects were also asked questions pertaining to prior actions to the
consumption of the beverage and about the situation itself, such as where it occurred
and if anyone else was around. With the responses gathered from each subject, an
extended MEC was created for each subject in each situation using the theoretical
model shown in Figure 2.
The extended MEC begins at the bottom with the prequel to the action discussed.
This prequel is the event(s) that is believed to influence the action to occur. From this
prior event, there must be some sort of recognition, or immediate feelings or thoughts
that act as drivers to act. Stemming from the driver is the action itself, or the product
usage. The consumer’s description of the product (what it was, how much was
consumed, and what brand) as well as a brief description of the situation (when and
where the action occurred, others present at the time). Completion of these steps
results in completion of MECs, including the identification of terminal values.
Identifying a specific consumption experience and situation provides theoretical
ground for developing MECs based on the proposition that the consumption of a
specific product and brand is acceptable in only certain situations. In fact, the same
brand and product form may associate positively to one situation and negatively to
alternative situations for the same consumer. Also, different attributes and benefits
may come to mind for a consumer for the same brand depending on the usage
situation that the consumer is thinking about (Bearden & Woodside, 1978; Belk,
1974). Consequently, resulting MECs may be expected to vary contingent on the
focal situations being described in consumers’ experiences. From the description of
the product and the situation comes the three distinct series of the extended
MEC. On the left is the reject alternative and on the right is the alternative that
would be considered. The center of the model focuses on the product that was
actually used or consumed.
The separate attributes of the product or brand are laid out in order to get an idea
of the benefits or consequences of each. This same questioning process is used to
collect information for the alternative and reject products. The benefit of the
attributes is where the MEC begins to probe for the instrumental beliefs and values
of the respondent. These beliefs are often unconsciously stored thoughts that the
consumer may not connect but that actually play a role in the decision-making
process. Some benefits may lead to the same values and benefits from the alternative
and the reject products can also tie into these same values. The model shows that the
alternative and reject products may tie in with the entire MEC, which is why they
should be considered when looking at unconscious thinking of consumers. Finally,
the chain concludes with terminal values or their core values and beliefs in the S’s life.
Findings
Beverage Consumed on Monday Morning
For this situation, each respondent answered questions about the first beverage they
consumed on the day that the interviews took place. Eric went to bed around one in
the morning and did not have anything to drink before going to sleep. After waking
up at ten o’clock on Monday morning with a dry mouth and morning breath, he
went down to the kitchen in his apartment and poured a 16-oz. glass of Tropicana
orange juice. Eric proceeded to have another glass of orange juice while he ate a bagel
alone at the table. The attributes that he associates with Tropicana orange juice are
that it tastes sweet, kills morning breath, and contains vitamin C (Figure 3). No
benefit was linked with the sweet taste, but sweet taste is an attribute that Eric
remarked that he prefers. The benefit of killing morning breath was having better
breath throughout the day, which was also a benefit that he did not believe any
alternative drinks offered. Eric explained that his mother had always told him that he
Terminal Values
Unique terminal value for
Saves alternative Gatorade
Feel comfortable Money
in my surroundings More attractive to females
+
+ No doctor +
visits ? Keeps me in shape
+
Makes me jittery
Instrumental Socialize w/o Stay healthy +
Values: embarrassment
+ + Stay hydrated
+ +
Caffeine Have to go Calories +
to bathroom Benefits: Better breath Prevents colds
Quenches Thirst
+ + +
+ + +
Reject that comes-
Attributes: Sweet Kills morning Vitamin C Alternative: Gatorade
to mind: Coffee
breath
- + +
+ +
3. Action: Eric goes downstairs to the kitchen for a drink of orange juice; uses plastic Solo cup that
holds 16 oz.; Eric drinks two servings; each serving14-15 ounces; Eric alone in kitchen, other roommates
are sleeping; Eric eats bagel with cream cheese while drinking juice.
2. Action: Eric awakens at 10 AM with feeling of dry mouth and morning breath.
1. Prequel to action: Went to bed at 1 AM after eating some pretzels while doing homework.
Figure 3: Means-end laddering chain for Eric’s first beverage on Monday morning.
233
should drink a lot of orange juice to prevent getting a cold, which is a benefit of a
product containing vitamin C. By preventing illness he believes that he will not have
to visit the doctor (which was a positive relationship, because he does not like paying
the extra money or dealing with doctors in general). As an alternative Eric suggested
that he also likes to drink Gatorade because it quenches his thirst and keeps him
hydrated. Again, by staying hydrated he believes that he will be able to stay in better
shape and remain healthier, which ties in with the benefits of the vitamin C in orange
juice.
Having good breath is important for Eric because one of his main goals is to feel
comfortable in his surroundings. By not having bad breath he felt that he could
socialize with others and not worry about others smelling his breath. Coffee is the
product that Eric reports that he rejects because of caffeine, calories, and causing him
to have to go to the bathroom. The consequences of all of these attributes negatively
connected with the value of feeling comfortable because each one posed a problem
with comfort. Interestingly, there were two other benefits that surfaced in this
interview stemming from staying healthy and staying in shape. As a result of staying
in shape Eric felt that this would make him more attractive to females, this seemed to
be a common theme with the four different situations. Saving money came up as a
core value because in the interview he explained that he bought Tropicana orange
juice even though it tended to be more expensive. He was willing to pay the price
premium for the brand but overall he is looking to save money.
Peter spent Sunday night watching movies and eating popcorn with his girlfriend
until about 2 a.m., when they went to bed. When he woke up around 11 on Monday
morning, his mouth was very dry from the popcorn and he had bad breath. He got
out of bed and went to the refrigerator in the bedroom and pulled out the gallon jug
of orange juice. He drank three 10-oz. servings directly from the jug while his
girlfriend and roommate were asleep in the room. Peter does not recall the brand
because he bought whatever was cheapest at the store, but he does know that it was
not a major brand like Tropicana. His alternative product was apple juice, and the
rejected product was cranberry juice (Figure 4).
The attributes that Peter described for the orange juice were that it was refreshing,
it killed germs, and it contained vitamin C. In the interview he remarked that he was
not sure if the orange juice actually killed germs but this was what he guessed gave
him better breath after drinking it. As a result of having better breath he said that he
would actually be able to kiss his girlfriend because she did not like to taste his bad
breath. Besides the fact that Peter does not like the taste of cranberry juice in general,
he also felt that it stained teeth with its red coloring, and caused saliva buildup, which
made him spit. These consequences can be negatively linked with Peter’s being able
to kiss his girlfriend. All of this leads up to the core value of romance and the feeling
of closeness that Peter believes is important.
Peter mentioned that he believed that vitamin C prevents colds and would help
him to stay healthy. He also felt that the alternative, apple juice, could help him to
stay healthy because it is ‘‘good for you.’’ Staying healthy is important to Peter
because another core value is that he likes to keep up with his responsibilities in life.
He does not like to fall behind in schoolwork or to miss days at his job.
Terminal Keep up responsibilities
values
More romantic, +
feeling of closeness
Don’t miss classes
- or work Instrumental value
+
+
“Makes - Can kiss girlfriend Stay healthy
me Spit
cringe” +
+ +
Benefits: Better breath Prevents colds
Bitter
Saliva “Good for you”
Build -up Stains + +
teeth
+ + + Attributes: +
Refreshing Kills germs Vitamin C Alternative: Apple juice
Cranberry Juice:
the reject that
- +
comes-to-mind + + +
3. Action: Peter goes to refrigerator in the bedroom for a drink of orange juice out of a gallon container;
Peter drinks 3 servings; each serving 8-10 ounces; girlfriend and roommate asleep in the room
2. Action: Peter awakens at 11 AM with feeling of dry mouth and bad breath.
1. Prequel to action: Went to bed at 2 AM after eating popcorn and watching a movie with girlfriend.
Figure 4: Means-end chain laddering findings for Peter’s first beverage on Monday morning.
235
Both informants chose the same beverage for quite similar reasons, but it did not
turn out that their core values were also the same. There was the similarity of staying
healthy, but that is one of the only similarities. Also, each informant had a different
relationship with his brand of choice. Using the relationships developed by Fournier
(1998), Eric is in a ‘‘committed partnership’’ with the Tropicana brand, while Peter is
a ‘‘casual friend/buddy’’ with his brand of orange juice. Eric is in a long-term,
voluntary union with Tropicana even though it is higher priced because it is a brand
that he knows and trusts. Peter buys more based on price. As long as a certain brand
is the cheapest, he will continue to purchase it, but if a competitor suddenly becomes
cheaper, he will buy that brand. There is no long-term commitment in the brand
relationship.
Beverage Consumed Last Friday Night between 8 P.M. and Midnight
Both Eric and Peter are self-described heavy drinkers, drinking ten beers on average
four or five nights a week. This consumption has increased somewhat over the last
couple of years, but all throughout they have been ‘‘beer drinkers.’’ Eric’s situation
on Friday night began when he invited friends over for pizza and to watch the game
on TV (Figure 5). He had gone out earlier and bought two 30-packs of Busch Light
for the evening and his friends were expecting to drink when they arrived. As the
night progressed the group began to play drinking games and consume even greater
amounts of beer. By midnight Eric recalls having about 12 cans of beer before going
to sleep.
The attributes suggested for Busch Light beer are a smooth taste, it contains
alcohol, and that it contains calories. The fact that the beverage contained alcohol
seemed to be the most prominent reason for consuming the beer. The benefit of the
alcohol is that it gets the person drunk, which makes it easier to socialize, but much
like the calories in the beer, the alcohol is also not healthy. The calories lead to
getting fat and being less attractive to females.
The alternative product to drink on a Friday night for Eric was vodka. Vodka was
an alternative because it was stronger than beer, which would get him drunker, but
this leads to having a hangover. The negative consequence of being so very drunk
leads to being unattractive to females and goes against his core values of feeling good
both physically and emotionally.
The rejected product for Eric was wine, no specific brand, just wine in general.
He reports not liking the taste of wine and he also perceives wine as a ‘‘girly’’ drink.
The bitter taste of the wine causes him to feel sick to his stomach, which has a
negative relation to the core value of feeling good physically. The alcohol in beer,
which makes it easier to socialize, has a positive relationship with the value of having
fun. Eric stressed that at his age he was more concerned with having a good time than
with worrying about too many responsibilities but he did see this mentality changing
in the near future because of graduation.
Terminal
value
Have fun Feel good physically
- - and emotionally
+ -
Less attractive
to females Bad hangovers
Makes me Instrumental +
feel sick +
values
Easier to Get Get drunker
Socialize fat
+
+
+ +
Bitter Girly Benefits: Get drunk Hot healthy Stronger than beer
taste drink
+ + + +
+
+
Attributes: Alternative: Vodka
Wine: reject that Smooth Alcohol Calories
comes-to-mind
+ + + + +
3. Action: Eric opens the refrigerator for a drink of beer; grabs a silver-and-blue 12-oz aluminum can of
Busch-Light; Eric drinks about 12 servings between 8 PM AND 12 PM; friends and roommates are also drinking
while playing card games and Beirut.
2. Action: Friends arrive around 8 PM and immediately ask, “Where is the beer?”
1. Prequel to action: Invited friends to cover over for pizza; play drinking games; and watch TV
Figure 5: Means-end laddering findings for Eric’s beverage consumption on Friday night.
237
The situation for Peter was different because of the setting, but for the most part
the MEC looked similar to that of Eric. Peter had taken a test earlier in the afternoon
that he did not believe had gone very well. After the test, he went down to the local
bar, Mary Ann’s, for ‘‘happy hour.’’ He met up with friends at the bar and they
began to drink beer. By 8 p.m., Peter was still at the bar and he was still drinking
bottles of Busch Light beer. He had about eight beers between 8 p.m. and the time he
left the bar at 10 p.m.
The attributes that Peter describes in Busch Light beer are that it is carbonated,
and that it contains alcohol and calories. He could not explain what he thought the
benefits of carbonation are, but this could be investigated in a follow-up interview if
one were to be performed (Figure 6).
Alcohol was the major attribute that Peter described in beer. The alcohol led to
getting drunk and acting immature, which was negatively related to his core value of
thinking rationally and responsibly. Peter mentioned rum as the rejected beverage
because of a bad experience that he had in the past. Due to this, bad experience he
cannot drink any rum because it makes him vomit.
Calories were not a major concern for Peter but he did acknowledge that they are
not healthy. Because of this he would need to work out to be more attractive to
females. Wine was given as an alternative to beer as a beverage to consume on a
Friday night. Peter felt that wine is more of a sophisticated drink and makes him
look more mature. He felt that this would make him more attractive to females — in
this case, one of his core values.
The brand relationships between both Peter and Eric and Busch Light beer are
about the same. Both respondents remarked that Busch Light beer was popular with
them because it was cheaper than most beers, but that it was not the cheapest. This is
a ‘‘casual friendship/buddy’’ type of relationship because there is not a long-term
commitment to the brand. The consumers are loyal but they do buy other brands and
will likely not stay with this brand after graduation because of their increased
financial situation. Another brand relationship can be seen between Peter and rum.
This is an ‘‘enmities’’ relationship because he has a desire to avoid the product due to
its negative effects.
Applying Advanced Means-End Chain Laddering

to Learn Unconscious Processes
Research on consumer thinking processes (see Zaltman, 2003) indicates multiple
levels of mental processing. Figure 1 in Chapter 1 summarizes five levels of thinking
processes that demonstrate MEC laddering research. Level 1 represents conscious
thinking that is verbalized between two or more parties. Level 2 thinking includes
conscious handling of thoughts before and after verbalizing thoughts and surfacing
thoughts. Level 3 thinking includes surfacing thoughts that are mostly not under
conscious control. Level 4 thinking represents unconscious thinking between two or
more persons. Level 5 thinking represents unconscious processing including
Terminal
values
Think rationally More attractive

and responsibly to females
Instrumental +
- +
Makes values
me vomit Look more mature
Act Have to work out
immature to stay in shape
+
+
+ +
Benefits:
Strong Alcohol Get drunk Not healthy More sophisticated
+
+ + +
+
Attributes: Alternative: Wine
Rum: reject Carbonated Alcohol Calories
that comes-to- - +
mind + + + +
3. Action: By 8 PM Peter is still drinking beer at Mary Ann’s; orders a 12-oz bottle of Busch Light from the bartender;
Peter drinks about 8 servings between 8 PM and 10 PM when he leaves the bar; the bar was full of friends and classmates
and music was playing loudly.
2. Action: Peter goes to Mary Ann’s for happy hour around 5 PM.
1. Prequel to action: Had a test earlier that afternoon that didn’t go so well.
Figure 6: Means-end laddering findings for Peter’s beverage consumption on a Friday night.
239
spreading activation of relationships among concepts that the individual is unaware

are occurring.
Other empirical findings (see Bargh, 2002; Wegner, 2002) support a core
proposition about how the mind thinks: the most processing occurs unconsciously.
Such unconscious thinking influences consumers’ actions in ways mostly unknown to
the consumers’ themselves.
Such observations are likely to cause stress among consumer researchers who
mainly apply the current (early 21st century) dominating logic of asking closed–
ended (e.g., 7-point scale items) questions because such questioning mostly reaches
only level 1 processing, that is, verbalized thoughts (Zaltman, 2003). Some deeper
form of questioning is necessary to reach into level 2 to 5 mental processing.
The advances in MEC laddering research that Chapter 11 discusses may help achieve
this objective.
Limitations and Suggestions for Future Research
The four MEC laddering cases serve to inform theory development and are not
intended as representative of college students in general or for specific segments of
beverage consumers. This report is limited by not taking the step of describing the
results in follow-up interviews with the two case study informants and having the
informants confirm and elaborate on the researcher’s interpretations (Hirschman,
1986, argues for such a validation step). Such an additional ‘‘auto-driving’’ step will
likely prove useful for achieving further advances in MEC laddering theory and
research.
Chapter 12
Building in Degrees of Freedom Analysis

in Case Study Research: Empirical Positivistic
Testing of Data to Alternative Theories
Synopsis
A valuable, although little-used, case data analysis technique, degrees-of-

freedom analysis (DFA), is the subject of Chapter 12. Given the richness of case
data and its prevalence in business marketing research, DFA has the potential
to become an important addition to one’s ‘‘research workbench.’’ Donald
Campbell (1975) first proposed this theory testing.
This chapter presents three business-to-business marketing applications;
the first two involve use of the technique to compare the extent to which four
theories of group decision making are manifested in organizations. The third
application illustrates how the technique is useful for theory development in
the context of manufacturer–distributor relationships. The contribution is in
demonstrating how researchers can link ‘‘traditional’’ (i.e., logical positivistic)
hypothesis testing procedures to examine theoretical propositions in case study
research. This approach is one way of achieving a critical test (Carlsmith,
Ellsworth, & Aronson, 1976), that is, testing the relative empirical strengths
of competing theories. The chapter highlights the value of generalizing case
data to theory versus the inappropriate attempt to generalize such data to a
population (Yin, 1994). The explication and demonstration of this technique is
not available elsewhere to the degree that Chapter 12 provides.
Introduction
Degrees-of-freedom analysis (DFA) was introduced by the noted psychologist,

experimental methodologist, and philosopher of science, Donald Campbell, in the
late 1960s. While other case methodologists (Miles & Huberman, 1994) mention
DFA in passing, there are few published examples of applications of this technique.
One reason why this technique is so interesting and unique is that DFA employs a
quantitative framework to gain insight and understanding about qualitative case
data. The essence of the technique is the idea of ‘‘pattern-matching’’ between

theoretical propositions and observations in a set of data.
A simple example to readily illustrate DFA is a doctor–patient interaction. Upon
examining a sick child, the doctor, after a series of questions, determines symptoms
of fever, irritability, loss of appetite, nausea, and a dull pain in the lower, right
quadrant of the abdomen. The pattern of observed symptoms (quantitative data)
leads the doctor to diagnose her patient as suffering from appendicitis (the
theoretical condition). In the same fashion, case data collected in social science
contexts can be examined to note the degree of match to a pattern that is set forth by
theory.
Related to the medical diagnosis example, a key problem found in the literature is
the tendency of medical doctors to use only one or two points of observations and
their most easily retrieved knowledge (Gilovich, 1991) this two-point process may
not represent sufficient coverage of issues to indicate a pattern of responses.
Consequently, too often the result is jumping to inaccurate conclusions and
misdiagnoses. In short, the more patterns that can be ‘‘matched,’’ the more confident
one is that the diagnosis is accurate and not subject to systematic bias. Campbell
(1975) maintains that this pattern-matching activity is analogous to having degrees-
of-freedom in a statistical test:
In a case study done by an alert social scientist who has thorough local
acquaintance, the theory he uses to explain the focal difference also
generates predictions or expectations on dozens of other aspects of the
culture, and he does not retain the theory unless most of these are also
confirmed. In some sense, he has tested the theory with degrees of
freedom [emphasis added] coming from the multiple implications of
one theory. (Campbell, 1975, pp. 181–182)
Such analysis considers case data quantitatively because the researcher notes
the degree of match to the theory in terms of ‘‘hits and misses.’’ How many hits
are necessary to ‘‘confirm’’ the theory? Simple statistical tests are useful for
noting whether or not the number of hits or misses is greater than that expected by
chance. Or, the researcher may conduct DFA purely to note the absolute number
of confirmed predictions for the sake of basic knowledge development (without
worrying about whether or not results are ‘‘statistically significant’’). This aspect of
DFA is consistent with Denzin and Lincoln’s (1994) observation, as follows.
Nor does qualitative research have a distinct set of methods that

are entirely its own. Qualitative researchers use semiotics, narrative,
content, discourse, archival, and phonemic analysis, even statistics
[emphasis added]. (Denzin & Lincoln, 1994)
The contribution of Chapter 12 is in demonstrating how researchers can link

‘‘traditional’’ (i.e., logical positivistic) hypothesis testing procedures to examine
theoretical propositions in case study research. This approach is one way of
Building in Degrees of Freedom Analysis in Case Study Research 243
achieving a critical test (Carlsmith et al., 1976), that is, testing the relative empirical
strengths of competing theories.
The contribution is valuable for two reasons. First, Chapter 12 provides the nitty–
gritty details in applications, where case data are generalized to theory versus
the inappropriate attempt to generalize such data to a population (Yin, 1994). The
explication and demonstration of this technique is unavailable elsewhere to the
degree here. Second, DFA utilizes strong features from both the logical positivism
and critical relativism traditions in terms of its scientific method. That is, rich
qualitative case data are used to examine whether or not theoretical propositions are
supported; thus, a substantial degree of objectivity and validity are achievable via the
DFA procedure.
These examples show DFA to be a very flexible technique that can accommodate
case research studies of varying goals (e.g., theory development versus theory
comparison) and contexts (e.g., individual versus group decisions, discrete transac-
tions versus ongoing relationships). In addition, DFA offers researchers some
advances over existing techniques. For example, content analysis (Kassarjian, 1977)
is a technique often associated with case data. Results of content analysis are
often expressable as counts, means, or frequencies of the phenomenon of interest.
DFA takes the researcher a step further by subjecting the counts, or patterns, in a
qualitative dataset to an a priori set of predictions (i.e., hypotheses, propositions, and
conjectures) so that theories can be compared, tested, or constructed, according to
the researcher’s purpose.
The next section, a brief overview of the steps involved in conducting DFA.
Doing Degrees-of-Freedom Analysis (DFA)
The heart of DFA is the development and testing of a ‘‘prediction matrix.’’ The
prediction matrix sets up the ‘‘pattern,’’ based on theory, to be either confirmed or
disconfirmed by the case data. The statements in the prediction matrix are analogous
to hypotheses in the sense of traditional statistical hypothesis testing. Campbell
(1975) states, ‘‘One should keep a record of all the theories considered in the creative
puzzle-solving process. To represent the degrees of freedom from multiple
implications, one should also keep a record of the implications against which each
was tested, and the box score of hits and misses.’’
Any research study, even the most exploratory, has some grounding in the extant
literature. An established theory or a ‘‘theory-in-use’’ may motivate a new research
study. A theory-in-use is the set of propositions guiding the behavior of a decision
maker, and theories-in-use are usually stated implicitly rather than explicitly (Senge,
1990). Other examples of developing theories-in-use include script-theory research
(e.g., Leigh & McGraw, 1989) and research on perceptual processing and acquiring
cognitive skills (Anderson, 1982; Chase & Simon, 1973). Following the development
of such theories, the predictions made explicit in the theories may be tested
empirically after collecting additional case data.
So, the first step in DFA is for the researcher to be familiar with the existing
knowledge base about the phenomenon of interest. If one or several theories on the
topic exist, the prediction matrix can be constructed with relative ease. If no theory
has been proposed in an area, individual studies may generate rival explanations
(via theories-in-use) that may be incorporated into a prediction matrix for testing.
This use of DFA would then lead toward the goal of theory development.
Upon careful development of one or more prediction matrices, the researcher is
ready for fieldwork. Data may be in the form of personal interviews, document
analysis, participant or nonparticipant observation, or other case data collection
methods (Yin, 1994). Care must be taken by the researcher to conduct the data
collection in such a way to avoid introducing bias into the data. Chapter 12 discusses
this point in some detail later and offers several strategies for maintaining data
integrity.
After the data are collected, trained judges then review the information (interview
transcripts, for example) to note hits or misses to items in the prediction matrix.
The box-score of hits and misses can then be subjected to statistical tests to note the
significance of the ratio of confirmed versus unconfirmed predictions found in
the data. Such tests might be a sign test (Siegel, 1956), a chi-square test, or a z-test for
differences in proportions.
Testing rival theories, that is, doing a comparative theory test (Sternthal,
Tybout, & Calder, 1987) or critical test (Carlsmith et al., 1976), via DFA deepens the
value of case data. That is, when several theories exist, the number of confirmed
predictions can be noted to see which theory tends to be supported relative to others.
Or, if theory development is the goal, the researcher can look at the confirmed
predictions to examine the variance in findings across the original empirical studies.
Constructing a table of ‘‘benchmark,’’ findings of confirmed predictions may be the
first step at formulating an initial theory.
Based on the discussion above, Figure 1 is a summary diagram that provides an
overview of DFA research process. Upon completion of a DFA study, the researcher
may continue with additional programmatic investigations into the phenomenon of
interest. The feedback loop in Figure 1 represents this refining process. Depending
on the outcome of one study, the researcher may go back and repeat to some or all
of the activities in the process. The remainder of the chapter examines three DFA
applications (Dean, 1986; Wilson & Vlosky, 1997; Wilson & Wilson, 1988).
Using DFA for Theory Comparison: Group Decision

Making in Organizational Behavior
Dean (1986) applies DFA to examine the degree of support for four theories of
organizational decision making in the context of adoption decisions of advanced
manufacturing technology. Because Dean’s research focuses on adopting and
acquiring new manufacturing technologies, his empirical application of DFA may be
of particular interest for industrial purchasing and marketing researchers. From the
Figure 1: The research process for degrees-of-dreedom analysis (DFA).
literature, the four theories include: (1) the rational model of decision making
(Allison, 1971); (2) the bounded rational model (Cyert & March, 1963); (3) the
political model (Pettigrew, 1973); and (4) the garbage can model (Cohen, March, &
Olsen, 1972) Dean’s central finding was that while no single theory was supported
in all cases, one theory, the bounded rationality model, tended to have more of its
predictions confirmed while the garbage can model tended to have the fewest
confirmed predictions.
In a partial replication of Dean’s work, Wilson and Wilson (1988) examine these
four theories in the context of organizational buying decisions for copier equipment.
Since they replicate Dean’s study in a marketing context, the details of the Wilson
and Wilson study are presented here as an example of DFA as a comparative theory
test (Carlsmith, Merrill, & Aronson, 1976; Sternthal et al., 1987).
Four Models of Organizational Decision Making
Briefly, each of the four models of decision making offers different explanations for
behavior in terms of outcomes and processes. The rational model, derived from
microeconomics, posits that members of organizations make decisions that provide
maximum benefit (i.e., utility) to the firm. The bounded rationality model proposes
that while decision makers try to be rational, they are constrained by cognitive
limitations, habits, and biases (i.e., human nature). According to the political model,
decision makers are competing to satisfy their own goals, and choice is a function of
an individual’s power. Finally, in the garbage can model, decisions are the result
of an unsystematic process — that is, problem definitions can change, preferences are
unclear, and people may come and go from the decision group (Dean, 1986).
Development of a Prediction Matrix
The four theories are a mixture of similar, complementary, competing, and orthogonal
predictions about organizational decision-making behavior. The prediction matrix is
developed based on seven basic decision activities (Dean, 1986). They are:
1. Problem definition — the conceptualization of the decision problem or process by

buying center members
2. Solution search — the existence, degree, and type of search for alternative
solutions to the problem(s)
3. Data collection, analysis, and use — the extensiveness, type, and function of
attempts to collect and use information
4. Information exchange — the ways in which buying center members share
information during the decision process
5. Individual preference formation — the existence, nature, and resistance to change
of buying center members’ preferences
6. Evaluation criteria — how decision criteria are developed and used
7. Final choice — how when and why choices among alternative products are made.
Thus, each theory or model of organizational decision making has predictions for
buying center member behavior in each of the seven facets. The following discussion
of these behaviors is adapted from Dean (1986) for the present business marketing
context. According to the rational model (Allison, 1971; Kepner & Tregoe, 1965),
buying center members would be expected to develop comprehensive problem
definitions, to conduct an exhaustive information search, develop a priori evaluation
criteria, and to exchange information in an unbiased manner. Individual preferences
and final buying center choice should reflect the alternative that offers the maximum
benefit to the organization.
Under the bounded rationality model (Cyert & March, 1963; Kepner & Tregoe,
1965), buying center members simplify the problem definition, search is sequential
and limited to familiar areas, and information exchange is biased by individual
preferences. Preferences originate from either personal or departmental sub-goals for
each buying center member. Evaluation of alternatives follows a conjunctive decision
rule, where criteria are expressed in terms of cutoff levels. Choice depends on which
alternative first exceeds the minimum cutoff levels of the evaluative criteria.
The political model (Pettigrew, 1973; Pfeffer, 1981) proposes that buying center
members will compete for decision outcomes to satisfy personal and/or departmental
interests. Preferences are based on these interests and formed early in the decision
process. Problem definition, search, data collection, and evaluation criteria are
weapons used to tilt the decision outcome in one’s favor. Choice is a function of the
relative power of buying center members.
Finally, the garbage can model (Cohen et al., 1972) suggests that decisions are
analogous to garbage cans into which problems, solutions, choice opportunities,
and buying center members are dumped. Problem definitions are variable, changing
as new problems or people are attached to choice opportunities. Data are often
collected and not used. Preferences are unclear and may have little impact on choice.
Evaluation criteria are discovered during and after the process, and choices are
mostly made when problems are either not noticed or are attached to other choices.
Given the propositions of each model across the seven decision phases, a
prediction matrix can be constructed. Rather than have a general statement for each
model and decision phase, operational items may be developed to make the data
judging task clear. Such is the case for buying copiers; two operational items for each
decision phase were developed. The resulting 56-cell table (2 statements 7 phases
4 models) contains the predictions that a theory is either confirmed (Y), partially
confirmed (P), or not confirmed (N). Table 1 shows this prediction matrix.
Collection of Case Data
The researcher must carefully design data collection forms in order to avoid
including items that favor one of competing theories described in alternative
predictive matrices. Alternatively, to insure that the data have a high degree of
nomological validity (i.e., empirical pattern of findings fits theoretical pattern), the
research might incorporate alternative questions that favor each theory (e.g., several
different scenarios illustrating alternative theories could be evaluated by the
respondent to see which best matches his/her ‘‘reality’’). The first approach — to
achieve bias reduction in questioning — benefits by having independent experts
check several revisions of the open and closed-end questionnaire. The same checking
procedure may be used for the second approach to insure that theory biases are built
into the scenarios accurately.
Possibly, because DFA and the idea of formulating competing prediction matrices
are relatively new methods in case study research, the literature provides scan
direction to issue of bias favoring one theory over competing theories (Wilson &
Woodside, 1999). To allow for objectivity and verifiability in data collection and
analysis, the actual survey forms used to collect data for DFA should be available
for independent examination. In Wilson and Wilson’s study, the set of questions is
available in the original article and is not included here for the sake of brevity
(Wilson & Wilson, 1988, p. 590).
Wilson and Wilson collected data from members of four buying centers regarding
office copier decisions in different departments across their university. Buying centers
typically consisted of two, and sometimes three, persons. In-depth interviews were
Table 1: Predictions of four models on decision process activities in organizational

behavior.
Decision phase and operating Rational Bounded Political Garbage

mechanism model rationality model can model
model
1. Problem definition
Do the participants view the problem Y P N P
in the same way?
Does the problem definition represent Y Y N Y
the goals of the organization?
2. Search for alternative solutions

Is search limited to a few familiar N Y P P
alternatives?
Are potential solutions considered Y P N N
simultaneously and compared with
one another?
3. Data collection, analysis, and use

Is information collected so that an Y N N N
optimal decision can be made?
Is control over data collection and N N Y N
analysis used as a source of power?
4. Information exchange
Is information biased so as to conform N Y Y N
to the preference (position) of the
person transforming it?
Is information exchange negatively N P N Y
affected by people entering and
leaving the decision process and
changing their focus of attention?
5. Individual preferences
Do preferences change as problems N P N Y
become attached to or detached from
the decision?
Are individual preferences a function N Y P P
of personal goals and limited
information about the alternative?
6. Evaluation criteria tradeoffs

Are criteria for a solution agreed Y P P N
on a priori?
Do tradeoffs across solution criteria Y N P N
occur?
Decision phase and operating Rational Bounded Political Garbage
mechanism model rationality model can model
model
7. Final choice
Is the first alternative that exceeds N Y P N
the cutoff level(s) selected?
Is the alternative chosen one that is Y P N P
expected to maximally benefit the
organization, compared with other
alternatives?
conducted by the first author after the supplier choice decision had been made.
In addition, documents pertaining to each decision (i.e., purchase requests for
quotes, supplier quotes, and purchase orders) were analyzed.
The interviews were semi-structured; similar questions were asked of each
respondent, but questions were open-ended. The questions were across broad areas
of decision activities and as such, the interviewer could ask for details on relevant
points. In other words, the question order and probes did not follow exactly the same
route for all interviews because of elaborations by respondents when answering. The
interview format and questions were not designed to operationalize any one theory.
The transcripts and archival material were then reviewed by three trained judges
(the two authors plus an ABD graduate student in business marketing) to note the
extent to which tenets of the four models were supported by the data.
The use of one interviewer only in the collection of case data may increase the
possibility of interviewer bias. Several strategies are possible to reduce such bias.
First, in some cases two interviewers working alone, and sometimes as a team, can
conduct multiple interviews in the same case study and then, the two or more
interviewers can compare their mental and written notes. Second, additional data
from written documents and direct observations can be collected to verify or
disconfirm the reported reality presented by respondents to any one interviewer.
Third, a trusted informant within the organization may act as a ‘‘consultant.’’
Interview notes may be reviewed by the consulting informant to verify facts and
eliminate apparent paradoxes.
A criticism often leveled at studies using case data is that generalization is difficult,
if not impossible. Yin (1994) countered this argument by stating that each case
should be considered a study within itself — just like an experiment. Thus, the buying
decision data are viewed as an initial case study with three replications. Yin makes
the point that multiple cases should not be considered as a ‘‘sample’’ and external
validity issues are not so problematic as some logical positivists might argue.
Critics typically state that single cases offer a poor basis for generalizing.
However, such critics are implicitly contrasting the situation to survey research, in
which a ‘‘sample’’ (if selected correctly) readily generalizes to a larger universe.

This analogy to samples and universes is incorrect when dealing with case studies.
This is because survey research relies on statistical generalization from a sample to
a population, whereas case studies (as with experiments) rely on analytic general-
ization. In analytic generalization, the investigator is striving to generalize a
particular set of results to some broader theory (Yin, 1994).
Inter-Judge Reliability
An examination of the level of agreement among the three judges offers information
about the reliability of the findings. Based on their review of the interview transcripts
and archival material, a judge could say that a theory is confirmed (Y), partially
confirmed (P), or not confirmed (N). Four levels of agreement exist for the three
judges — perfect (YYY, PPP, NNN), near perfect (YYP, YPP, NNP, NPP), some
(YYN, YNN), or none (YPN). This scheme is adapted exactly from Dean (1986).
The three judges did display a pattern of agreement greater than one would expect
by chance. Each judge made 56 evaluations (7 phases 2 statements each 4
cases); see the prediction matrix in Table 1. Judges were in perfect agreement for
55 percent of the evaluations and nearly perfect agreement for 23 percent. Thus, near
perfect to perfect agreement occurred for 78 percent of the evaluations. Judges
were in some agreement for 20 percent of the evaluations and in total disagreement
in only 2 percent of the evaluations. By chance, the distribution would be 11 percent
total agreement, 44 percent nearly perfect, 22 percent some agreement, and 22 percent
no agreement. By a chi-square test, these distributions of agreement levels are
significantly different (w2 ¼ 103.77, 3 DF, po0.001).
Theory Comparison by ‘‘Box-Scores’’
For brevity, results from one buying center case are presented along with a summary
of the results across the four cases. These abbreviated findings serve the present
purpose of illustrating applications of DFA; the complete set of tables is available
(Wilson & Wilson, 1988). Table 2 provides details about one copier buying decision
case. For Judge A, 8 evaluations of a possible 14 confirmed the predictions of the
rational model of organizational decision making. In other words, for the rational
model there were eight hits and six misses; a 57 percent hit rate. Raw scores for the
rows in Table 2 may not sum to the same total because multiple hits were possible
given overlapping aspects of the theories, as indicated in the prediction matrix.
Looking down the column for the rational model, 17 evaluations of 42 possible
(3 14) confirmed the predictions of the rational model. That is, the rational model
had 17 hits and 25 misses, a 40 percent hit rate. Similarly, the bounded rationality
model had a 62 percent hit rate while the political and garbage can models had
Table 2: Box-score results for buying case x1: absolute and percentage matches (hits)
to predictions.
Organizational decision-making Rational Bounded Political Garbage

model rationality can
Judge A 8 (0.57) 8 (0.57) 2 (0.14) 4 (0.29)

Judge B 2 (0.14) 9 (0.64) 8 (0.57) 3 (0.21)
Judge C 7 (0.50) 9 (0.64) 3 (0.21) 6 (0.43)
Total observed matches 17 (0.40) 26 (0.62) 13 (0.31) 13 (0.31)
Total expected matches 17.25 17.25 17.25 17.25
Note: w2 ¼ 6.53, 3 DF, p ¼ 0.10. Numbers in parentheses are percentages of matches (hits).
31 percent hits, respectively. On an absolute basis, the bounded rationality model had
more predictions supported by the case data than the other decision-making models.
Even though care was taken by Wilson and Wilson to insure that the survey
questions did not favor one theory over others, a possibility exists that respondents
wanted to appear rational to the interviewer. This view supports, in part, an
explanation of findings in support of the bounded rationality model. Consequently,
consider using alternative survey methods, such as scenario problem-solving by
respondents and the forced-metaphor elicitation technique (FMET; see Chapter 8),
to examine for the contextual possibilities for applications of competing theories.
To evaluate this result statistically, a chi-square test is used to test whether there
is a significant difference between the observed distribution of ‘‘hits’’ (i.e., confirmed
predictions) and the distribution one would expect by chance. Use of the chi-square
test in this manner is appropriate since we are examining the extent to which two
distributions (observed and expected) are different from each other (Siegel, 1956).
Since the a priori assumption is that any model may be as good as another, all
four models have an equal chance (25 percent) of having their predictions confirmed.
The absolute number of confirmed predictions across the models in this case is 69.
The expected distribution would be 69/4 ¼ 17.25 hits per cell. The chi-square statistic
is significant at po0.10 which indicates that the two distributions are significantly
different and a systematic pattern occurs in the data. The pattern is that the bounded
rationality model has more of its predictions confirmed compared to the other
models (see Table 2).
When the matches to the predictions are considered as proportions, a z-test can be
used to evaluate the results. The highest proportion of matches is for the bounded
rationality model (0.61) which is significantly larger than that for the rational model
(0.40); z ¼ 2.01, po0.05. It follows that the number of matches for the bounded
rationality model are also significantly greater than the political model and garbage
can model.
Table 3 presents a meta-analytic summary of the results across all cases. Again, a
chi-square test indicates that the distribution of matches to the prediction matrix
is significantly different from that expected by chance (w2 ¼ 15.9, 3 DF, po0.01).
Specifically, the bounded rationality model tends to have more predictions confirmed
Table 3: Meta-analysis across all cases: observed matches to predictions.
Organizational Rational Bounded Political Garbage

decision-making model rationality can
Case x1 17 26 13 13
Case x2 13 30 15 22
Case x3 27 15 6 20
Case x4 14 28 17 17
Total observed matches 71 (0.42) 99 (0.59) 51 (0.30) 72 (0.43)
Total expected matches 73.25 73.25 73.25 73.25
Note: w2 ¼ 15.9, 3 DF, p ¼ 0.01. Numbers in parentheses are the percentage matches (hits).
from the case data compared to the other models of decision making. The proportion
of matches for the bounded rationality model is also significantly greater than the
proportion of matches for the political or garbage can models (z ¼ 2.93, po0.05).
Theory Comparison Conclusions
As Figure 1 shows, the final step in DFA is to assess the findings in light of the
existing knowledge base. Questions to ask at this point might be ‘‘how do our results
compare to those of other studies?’’ and ‘‘how do our results confirm or disconfirm
rival theories?’’
In the example, the bounded rationality model has the greatest number of
predictions confirmed by the case data compared with rival models. This conclusion
supports Dean’s contention that tenets of one theory may dominate while rival
theories may receive less support (Dean, 1986). In the context of a modified rebuy
decision (Robinson, Farris, & Wind, 1967), a post-hoc look at the findings indicates
that the bounded rationality model seems to provide the best framework for
understanding organizational decision making. This makes intuitive sense given that
modified rebuy decisions are characterized by limited problem-solving behavior with
buying center members drawing on past buying experiences. In other contexts, the
results might be different. The results of this comparative theory test helps to clarify
the context where one theory might apply to a greater degree than others, but more
research on other contexts is necessary before more definitive conclusions can be made.
For example, if the decision had been for new technology with little information
available (a typical new task situation), the political model may have had more
predictions confirmed since buying center members might be motivated to ‘‘protect
their turf’’ (Pettigrew, 1975).
Similarly, the rational model may have had more predictions confirmed in a new
task situation in order to reduce financial risk of making a high-dollar purchase.
Thus, decision context is proposed as strong moderator on the applicability of a
particular decision theory within a specific case study.
The following view may be best: decision-making realities may reflect bits and
pieces of competing theories. No one theory may dominate. The issue becomes
one of learning the contingencies that generate the occurrence of support for portions
of each theory. Next, we illustrate of the use of DFA for the purposes of theory
development in the area of relationship marketing.
Using DFA for Theory Development: Manufacturer–Distributor

Relationships
While there is a wealth of empirical studies on the topic of relationship marketing,

there are no established ‘‘middle-range’’ theories (Merton, 1957); thus, this field is
ripe for the application of DFA. Wilson and Vlosky (1997) and Fontenot and Wilson
(1997) conducted a large-scale case research project to gain deep understanding
about partnering relationships between firms as opposed to typical (transactional
and often adversarial) relationships. The context of their work is in the area of
manufacturer and distributor relationships in the wood products industry. The
motivation for this work was to identify the activities that tend to be associated with
the more beneficial (‘‘win–win’’) partnering relationships as opposed to short-term
adversarial (‘‘win–lose’’) relationships. This section presents a brief review of their
work to illustrate a theory-building application of DFA.
Grounding in the Extant Literature

Webster (1992) proposes that business marketing relationships span a continuum
from discrete transactions to legal partnerships (vertical integration). In between,
these two endpoints are relationships with varying degrees of cooperation, trust,
and dependence. Discrete transaction relationships, historically the most common
form in the U.S. marketplace, are usually characterized as adversarial with
both the buyer and seller attempting to achieve the best economic position at the
expense of the other. As one moves toward the opposite end of Webster’s continuum,
relationships are characterized as being more ‘‘partner-oriented’’ with firms
exhibiting higher degrees of cooperation, trust, commitment, and communication
(Webster, 1992).
Four models in the literature (Anderson & Narus, 1990; Dwyer, Schurr, & Sejo,
1987; Mohr & Spekman, 1994; Morgan & Hunt, 1994) offer a good base for
developing a prediction matrix. These studies include some overlapping constructs
(e.g., trust, functional conflict, communication) but also have their own unique
constructs and hypotheses. From these four studies, we get an idea of the most often
studied features of business-to-business relationships, and based on the findings,
we can make predictions accordingly.
The Prediction Matrix
Table 4 is the prediction matrix for the relationship marketing study. Table 4
includes predictions (most in the form of ‘‘yes’’ or ‘‘no’’ answers to questions in the
matrix) for 11 relationship activities. Predictions are for both partnering and typical
relationships so that a contrast between these two forms can be observed. The 11
relationship activities (i.e., joint programs, pricing, logistics) were distilled from the
literature, and the indicators shown in Table 4 are the operating mechanisms used
during the personal interview. These statements are similar to those used in our
previous example (see Table 1). Operating mechanisms in a prediction matrix are
helpful because they help the interviewer and respondent to focus on issues rather
than talk in generalities.
Data Collection
Members of research teams interviewed key informant respondents at distributor

organizations about activities with partnering manufacturers and typical manufactur-
ing principals. A standard list of questions was provided to each team for conducting
semi-structured interviews and again, Dean’s (1986) methodology was followed
(Wilson & Vlosky, 1997). Specifically, questions were asked about the activities firms
engage in (or don’t engage in) with what the respondent considers a ‘‘partnering’’
supplier and a ‘‘typical’’ supplier. Wilson and Vlosky’s (1997) complete questionnaire
is included in their article, and while questions may seem to require simple ‘‘yes’’ or
‘‘no’’ responses from participants, the research teams were instructed to use the
specific questions to get respondents talking about their business relationships.
Interviewers were expected to, and did, use their own probes with respondents in
order to get fine detail and nuance in order to write up their final cases.
The research teams were used since informants were spread across the United
States and Canada. Team members were identified as being active academic
researchers in either marketing or wood science and in the general geographic area
of the potential distributor respondents.
After conducting the interviews, team members worked to transcribe their field
notes and write cases on the companies they studied. Based on the complete set of
field notes, the team members made judgments to complete the prediction matrix for
their particular case. While most teams interviewed only one distributor, some (e.g.,
Paun, 1997), compile data from multiple cases. The cases written by the research
teams have additional in-depth nuance on manufacturer–distributor relationships
that is beyond the scope of reporting for this chapter. Interested readers may consult
the May 1997 issue of Journal of Business Research to view the cases in their entirety.
Degrees-of-Freedom Analysis Results
Table 5 contains the DFA results for the partnering suppliers to the wood products
distributor informants. For most cells of the prediction matrix, 10 cases could be
Table 4: A prediction matrix of relationship activities.
Relationship Indicator Partnering supplier Typical/average supplier

activity
Programs Development of new product or service programs? Yes No

Joint programs to market manufacturers’ products? Yes No
Involvement in product deletion decisions? Yes No
Pricing Offer trade discounts? More generous than normal Industry norm only
Special pricing problems? No Yes
Claim policy? Better than industry norm Industry norm
Payment terms? Better than industry norm Industry norm
Logistics Typical shipment size? LTL accommodated Full truck load required
JIT Inventory management? Yes No
Method of transportation Truck or other options Truck only
FOB mill or FOB delivered? FOB mill FOB delivered
Dealer Supplier featured in promotional literature? Yes No
promotion Sharing of customer lists with supplier? Yes No
Sales volume incentives offered by supplier? Better than industry norm Industry norm
Advertising Co-op advertising? Yes No
Sales force Joint sales training? Yes Little to none
activities Joint sales calls to distributors customers? Yes No
Joint performance reviews of suppliers’ sales force? Yes No
Joint performance reviews of distributors sales force? Yes No
Joint customer lead development for distributor? Yes No
Building in Degrees of Freedom Analysis in Case Study Research
Marketing Conduct joint marketing planning with supplier? Yes No

planning Does supplier request a written marketing plan? Yes No
255
256
Relationship Indicator Partnering supplier Typical/average supplier
activity
Performance Conduct annual performance reviews with the Yes No

reviews supplier?
Manufacturing Does supplier configure shipments Yes No
to your specs?
Does supplier use/offer UPC bar coding? Yes No
Does supplier manufacture products to your specs? Yes No
Does supplier offer special packaging services? Yes No
Communication Does distributor visit supplier? Yes No
Does supplier visit distributor? Yes No
Seek out supplier at trade shows or association Yes No
meetings?
Information Does supplier have access to distributor’s computer Yes No
exchange files
Does distributor have access to supplier’s computer Yes No
files
Face-to-face communication frequency Multiple times per week Less than once per week
Telephone communication frequency Multiple times per day Once per day or less
Electronic communication frequency Multiple times per day Once per day or less
Which department mostly communicates with Multiple departments Purchasing
supplier?
Other departments that communicate with supplier? Multiple departments Senior management
Use of EDI between supplier and distributor? Yes No
Table 5: Box-score results for partnering supplier relationships.
Relationship Indicator Prediction Number Significance*
activity hits (%) of cases
Programs Development of new product or service programs? Yes 70 10 +

Joint programs to market principal’s products? Yes 80 10 ++
Involvement in product deletion decisions? Yes 30 10 –
Pricing Offer trade discounts? Yes 80 10 ++
Special pricing problems? No 80 10 ++
Claim policy? Better than industry norm 60 10 +
Payment terms? Better than industry norm 20 10 –
Logistics JIT inventory management? Yes 10 10 –
FOB mill or FOB delivered? FOB Mill 20 10 –
Dealer Supplier featured in promotional literature? Yes 20 10 –
promotion Sharing of customer lists with supplier? Yes 20 10 –
Sales volume incentives offered by supplier? Better than industry norm 50 10 0
Advertising Co-op advertising? Yes 50 10 0
Sales force Joint sales training? Yes 70 10 +
activities Joint sales calls to distributors customers? Yes 70 10 +
Joint performance reviews of suppliers sales force? Yes 20 10 –
Joint performance reviews of distributors sales Yes 10 10 –
force?
Joint customer lead development for distributor? Yes 100 10 ++
Marketing Conduct joint marketing planning with supplier? Yes 60 10 +
planning Does supplier request a written marketing plan? Yes 0 10 ++
Building in Degrees of Freedom Analysis in Case Study Research
Performance Conduct annual performance reviews with the Yes 70 10 +

reviews supplier?
257
258
Relationship Indicator Prediction Number Significance*
activity hits (%) of cases
Manufacturing Does supplier configure shipments to your specs? Yes 100 5 **

Does supplier use/offer UPC bar coding? Yes 20 5 **
Does supplier manufacture products to your specs? Yes 100 5 **
Does supplier offer special packaging services? Yes 40 5 **
Communication Does distributor visit supplier? Yes 100 10 ++
Does supplier visit distributor? Yes 100 10 ++
Seek out supplier at tradeshows or association Yes 100 10 ++
meetings?
Information Does supplier have access to distributor’s computer Yes 20 10 –
exchange files
Does distributor have access to supplier’s computer Yes 20 10 –
files
Face-to-face communication frequency Multiple times per week 20 5 **
Telephone communication frequency Multiple times per day 90 10 ++
Electronic communication frequency Multiple times per day 40 5 **
Which department mostly communicates with Multiple departments 40 5 **

supplier?
Other departments that communicate with supplier? Multiple departments 40 5 **
Use of EDI between supplier and distributor? Yes 30 10 –
*The test of statistical significance is a sign test (Siegel, 1956) by using the following indicators: 11, the partnering relationship prediction is supported both
directionally and statistically (P, 0.05); 1, the partnering relationship prediction is supported directionally; 2, the partnering relationship prediction is not
supported directionally; 22, the partnering relationship opposite to that predicted is supported statistically (P, 0.05). **No statistical testing inferences are made
in these cells due to the small number of cases. The weighted average hit rate of predictions to observations is 53 percent; that is, our predictions for partnering
relationship activities are confirmed approximately half the time. By a z-test, this proportion of hits to misses is not significant.
evaluated. In some cases, though, particular questions did not apply to the
distributor respondent; thus, the number of observations ranges between 5 and 10
cases. From Table 5, the specific activities present for partnering firms (e.g., joint
marketing programs, trade discounts, joint sales force performance reviews, specially
configured shipments, etc.) can be observed.
Wilson and Vlosky use a combination of statistical tests used to evaluate the
compiled results of prediction matrices completed by the research teams. First, a sign
test (Siegel, 1956) is done for each row/prediction. ‘‘The sign test gets its name from
the fact that it uses plus and minus signs [or ‘yes’ ‘no’ responses] y as its data’’
(Siegel, 1956). We can evaluate each row of the prediction matrix by assessing the
number of cases where the partnering prediction is confirmed (yes) or not confirmed
(no). The resulting ratio can be evaluated in terms of an associated p-level for a
binomial test (Siegel, 1956, pp. 68–75 and Table D, p. 250).
In Table 5, 7 of 10 cases, partnering suppliers participate in development of
product/service programs. While this result is in the expected direction as Wilson
and Vlosky propose, p ¼ 0.17 (from Table D in Siegel); thus, the finding is only
marginally significant, at best. Table 5 provides a note to this point. For the next row
of the prediction matrix, 8 of 10 firms reported they did participate in joint programs
to market the partnering manufacturer’s products. This result is statistically
significant ( po0.05) and is so noted in the explanation at the bottom of Table 5.
Results for the remaining rows are interpreted similarly.
To evaluate the results over the entire table, a weighted average of the proportion
of matches to the prediction matrix is calculated (53 percent). A z-test is used
to estimate whether this proportion is not significantly different from chance
(50 percent). The computational formula for this test is available in Bruning and
Kintz (1968). The difference between 53 and 50 percent is not statistically significant
for the table as a whole.
Even so, gaining interesting insights about partnering behavior follow from
examining the rows of Table 5. In so doing, a post-hoc profile of partnering
relationship activities emerges. For example, in the context of one specific industry
(wood products), partnering activities with manufacturing principals are manifested
in terms of joint marketing programs, superior pricing arrangements, cooperation
between the sales forces, and joint planning activities, to name a few. Relatively, little
partnering activity occur between manufacturers and distributors in more ‘‘sensitive’’
areas such as information exchange (e.g., use of EDI and having computerized access
to information in the partner’s organization).
Such detailed insights allow for industry specific refinements in thinking about
business-to-business relationships. Wilson and Vlosky’s findings corroborate work of
earlier studies. For example, in the wood products industry, manufacturers and
distributors still have some progress to make in terms of investments in technology
before partnering activity in communication will occur. Vlosky (1995) notes that
the industry is fraught with distrust between firms; this would contribute to an
unwillingness to share information and communicate freely.
In this narrow industry example, several studies have been conducted as initial
explorations (Vlosky, 1995; Vlosky & Wilson, 1994); subsequent corroboration of
such findings in Wilson and Vlosky (1997) provides an initial step in the inductive
theory building process for a research area (Eisenhardt, 1989).
The next step in the research program of Wilson and Vlosky would be to test their
posterior (i.e., post-hoc) profile of partnering relationship activities for additional
verification. The partnering activity statements/predictions in the posterior profile
(Dean, 1986) would become new hypotheses for an additional study that may use
DFA or other analysis techniques, as necessary.
DFA as a Research Tool
As evident from the two examples of DFA in this chapter, the technique has a lot
to offer to researchers in business marketing. Many phenomena in this field are
very complex and a case methodology is needed to uncover and confirm nuances of
organizational and/or individual behavior. Case studies are also more feasible
logistically in terms of data collection, compared with experiments or even surveys,
given the time demands on respondents in the business-to-business context. With
multiple case observations, the researcher can use DFA to meta-analyze his/her data
for purposes of theory building, theory comparison, or theory testing.
A major advantage of DFA is its flexibility. Data from one case or many cases
may be used. While case research methodologists maintain that studies with n ¼ 1
can be perfectly valid in terms of analytical generalization, researchers need multiple
data points for any sort of ‘‘statistical’’ generalization. However, multiple cases
in a DFA should not be considered as data points/observations in a sample, but
separate replications in the same way that multiple experiments about a common
phenomenon are considered.
DFA is flexible in how results are evaluated. One or several judges may evaluate
the data to tally the theory box-score results. When multiple judges are used, inter-
judge reliability computations offer additional evidence of the reliability of the
evaluations and the validity of this approach.
The purpose of the researcher’s study (theory building or theory comparison) is a
third dimension on which DFA demonstrates flexibility. In theory building, disparate
findings from studies in the literature can be explored, propositions can be
formulated and assessed, and the developing theory can be refined through post-
hoc examination of DFA findings. On the theory comparison side, Campbell (1975)
and Sternthal et al. (1987) both noted the need for considering rival explanations
against each other to note which theory offers a better explanation for phenomena.
DFA is a way to conduct such comparative theory tests.
Research Limitation Considerations
In these examples of studies employing DFA, note several limitations to avoid

where possible. For example, in Wilson and Wilson (1988), respondents were not
contacted for a follow-up review of the decision that the interviewer had recorded.
Thus, replication of this research is needed to determine if the results obtained are
stable. This correction was made in case data collection as reported in Wilson and
Vlosky (1997); in writing the case reports, researchers often had multiple visits with
respondents to verify and clarify information obtained at earlier visits. In addition,
the case data analyzed by Wilson and Vlosky were collected in a semi-structured
depth interview, as described earlier. Case writers were asked post-hoc to complete
the prediction matrix based on the information gathered from wood products
distributors. In other words, the case writers did not have the prediction matrix until
all data had been gathered in order to minimize data contamination and theory
confirmation bias.
As this tutorial on DFA notes, strategies to reduce the potential for bias in data
collection should be given consideration by researchers planning to use this technique.
Wilson and Wilson (1988) employ both personal interviews and document analysis
in an effort to achieve triangulation. However, assurances about data reliability and
validity would be increased if post-hoc interviews had been done or by employing some
of the other strategies mentioned earlier (use interview teams, present respondents with
alternative decision scenarios, use a ‘‘consultant’’ respondent for verification).
Similarly, more data integrity checks would benefit Wilson and Vlosky’s (1997)
work since multiple interviews were the only source of data. However, in their
defense, most of the cases were written by teams of academic professionals who have
a relatively high degree of knowledge about and experience with validity issues.
We would have less confidence in their conclusions if a ‘‘student worker’’ or other
novice individual had been employed to collect the data. The bottom line, though,
is that more formal controls on data collection (as suggested earlier) would only
improve DFA studies and should be given serious consideration.
Conclusion
The contribution of this chapter is to illustrate and advocate the use of DFA in
business marketing research. Detailed examples (Wilson & Vlosky, 1997; Wilson &
Wilson, 1988) are presented as a tutorial to illustrate the technique in theory
comparison and theory building applications. This explication of the technique is
important because although the approach was originally described by Campbell
(1975), he never, in any of his writings, provided a field study application of the
method. Similarly, other case methodologists (Miles & Huberman, 1994; Yin, 1994)
mention DFA briefly but provide no examples.
Much potential exists for DFA in business market research. For example, a DFA
examination of the major paradigms used to frame relationship marketing studies
would be interesting. Research in relationship marketing has been grounded in
overarching, global paradigms such as exchange theory, transaction cost analysis,
game theory, and organizational governance theory. When are these models
applicable and in which contexts in relationship marketing? By developing a
prediction matrix from these paradigms, theory comparison studies (similar to those
this chapter describes) are possible.
Chapter 13
Applying the Long Interview

in Case Study Research$
Synopsis
The long interview is an intensive questioning of informants selected for their

special knowledge, experiences and insights (or ignorance) of the topic under
study. The objectives of the long interview include learning the thinking,
feeling, and doing processes of the informants, including an understanding of
the informants’ worldviews of the topic under study in their own language. The
chapter compares the strengths and weaknesses of the long interview to other
primary data collection methods. The chapter describes a research application
of the long interview in integrated marketing. The study was designed to
(a) learn about the rich complexities in the lives of household gardeners buying
and using seeds plants after responding to direct marketing appeals and
(b) resolve two conflicting ‘‘theories-in-use’’ of how and why different customer
types purchase products. These competing theories were proposed by different
executives in the firm sponsoring the study. The development and critical
testing of competing theories-in-use are described. This chapter reports a
study to learn the behavior of five customer types. The results include thick
descriptions of the processes of buying and using seeds and plants purchased
through direct marketing offers and store visits.
Introduction
The long interview is a qualitative research method — not to survey the lives of
respondents — but to mine them (Cox, 1967; McCracken, 1988). The long interview
has been reviewed and used extensively in research on buyer behavior. The purpose
here is to describe its features and usefulness for collecting data about key
informants; the chapter includes an application in the field of integrated marketing.
$
This chapter is co-authored with Elizabeth Wilson, Suffolk University, Boston.
‘‘Key informants’’ are persons with special knowledge or are members of a specific
sub-population of immense interest to the researcher or strategist (for example,
extremely frequent casino visitors (Perfetto & Woodside, 2009) or drivers who
commit road rage nearly every week of the year (Woodside, 2008c). Thus, the long
interview is a method that selects informants who are representative of unique sub-
populations using pre-qualification selection routines and not by random selection
from a general population.
Chapter 13 compares the feature characteristics of the long interview versus other
primary data collection methods. Second, the chapter presents a brief literature
review of long interview applications in research in marketing. Third, the chapter
describes details of McCracken’s (1988) long interview paradigm within the context
of a study of an integrated marketing program. From applying the paradigm,
theoretical propositions are developed that have important strategic implications
for an integrated marketing firm. Next, Chapter 13 describes details of the research
method, and fifth, the findings are discussed. Finally, the chapter offers suggestions
for applying the long interview in studying the lives of key informants for testing
theories-in use and improving management and marketing strategies.
Features of the Long Interview Method

The features typical of long interviews include: (a) a two- to six-hour, face-to-face
meeting with the interviewer and respondent; (b) interviewing the respondent in his
or her life space, that is, the environment related to the topic under study; (c) asking
open-ended, semi-structured questions with deeper exploration of unexpected topics
related to the study as opportunities occur; (d) tape recording of responses (when not
disruptive) during the interview; (e) verification of responses by triangulation of
research methods (e.g., comparing answers with data from direct observation and
documents); and (f ) developing thick descriptions of individual cases (individual
customers’ buying and using behaviors). A detailed description and brief research
examples of the long interview method are provided by McCracken (1987, 1988).
The long interview is held in the respondent’s product/service buying or using
environment. Table 1 summarizes comparisons of strengths and weaknesses of the
long interview with four other primary research methods. The feature profiles of the
five research methods in Table 1 were developed from several sources on research
designs; see Churchill (1991), Dillman (1978), and Miller (1991).
In Table 1, the profile of strengths and weaknesses for the long interview is most
similar to the profile for participant observation. Participant observation in
consumer behavior studies includes direct viewing of a customer’s buying and/or
using of products and services by an observer. Usually these studies include an
extended period of observation, ranging from several days to several months
(Arnould & Price, 1993; Celsi, Rose, & Leigh, 1993). In this section, we provide a
brief commentary on the strengths and weaknesses listed in Table 1 that are
especially relevant to the long interview as applied to direct marketing.
Applying the Long Interview in Case Study Research 265
Table 1: Feature comparisons of primary data collection methods.

Feature Mail survey Telephone Mail Long Participant
interview intercept interview observation
1. Ability to verify Low- Low Low Lowest High

responses moderate
2. Response Rate No Slow Lowest High Moderate
3. Ability to probe, Low- Moderate Moderate Low Moderate
learn response moderate
why responses
4. Cost per Yes Fast Low High Low
completed
interview
5. Ability to describe Low High Moderate Moderate Moderate
purchase and
use(what, when,
who, how,
where)
6. Possibility of Yes Fast Moderate Low Low
interviewer bias
7. Speed in High Very High High High High
completing
study
8. Capability of Yes Very Slow High Very Low High
auto-driving
9. Ability to Very High Very High Very High Very High Very High
generalize
results to a
population
10. Ability to Yes Very Slow Very High Very Low High
generalize
results to a
theory-in-use
Verifying Responses
The ability to verify respondents’ answers (the first feature in Table 1) is a strength of
the long interview. Answers may be verified by one or more of several potential
sources: documents offered by the respondent in describing his or her responses,
direct observation of products purchased and usage environments, confirmation by
other persons in the household on key events, and quantities and varieties of
purchases and prices paid. Some ability to verify responses to mail and telephone
surveys is supplied by partial knowledge of the respondents’ buying histories
(based on records in a direct marketer’s customer database), and/or by repeating key
questions on two occasions during the interview schedule.
Response Rates
Very high cooperation–completion rates may be expected for long interview studies
compared to mail or telephone surveys. The long interview usually includes a three-
step contact method: (a) initial letter describing the purpose of the study (and often
an offer of payment for participating); (b) a telephone call to arrange an
appointment; and (c) the interview. Four to five telephone attempts are usually
made to reach each subject sampled before replacing a non-response with an
alternative subject. The perceived importance of the study by the subject (because the
interviewer requests an on site interview) and the offer of a substantial payment (e.g.,
$30) for completing the survey, are two explanations for the high response rates.
Ability to Learn Reason Why Responses
The use of probing questions to learn deeper and deeper reasons why specific
products and brands are purchased, and why certain usage behaviors are performed,
is a key strength of the long interview compared to the broad but relatively shallow
mail and telephone approaches. The reasons for direct buying versus other forms of
buying, as well as the reasons for buying from one direct marketer instead of another,
are more likely to be uncovered using the long interview.
Cost per Completed Interview
The long interview approach is relatively more expensive than mail or telephone
surveys. The costs per completed interview may range from 10 to 20 times higher
than for a returned mail survey. Consequently, the number of long interviews
completed may be expected to be 10 to 20 times smaller for the long interview than
for a mail survey.
McCracken (1988, p. 37) emphasizes that informants selected for long interviews
‘‘are not a sample,’’ and that their selection should not be governed by sampling
rules. The numbers selected for long interviews are usually four-to-eight informants
for each sub-population type of research interest; for example, first-time customers
buying two or more product varieties from firm X may represent one customer type.
Ability to Describe Purchase and Use
In Table 1, all five research data collection methods are described as moderate-high
in ability to collect purchase and use descriptions. Woodside and Soni (1991) provide
a detailed example of collecting such information in a direct marketing research
study using a mail survey and empirical positivistic hypotheses testing.
Possibility of Interviewer Bias
An important strength of mail surveys is the absence of interviewer bias; such bias
includes tone of voice, speed of asking questions, interruptions made, and many
others (e.g., for long interviews and mall intercepts, facial expressions, body
mannerisms, dress, and demographic characteristics).
Speed in Completing the Study
The long interview is a rather time-consuming method compared to the telephone

and mall intercept methods. Contact time, travel time, interviewing time, and
answering open-ended responses usually result in four to six interviews per day, the
upper limit of a long interviewer’s typical day in the field. For each completed
interview, data interpretation, classification, process analyses, and computer analyses
(see Fielding & Lee, 1991) of long interview data can require 5 to 10 times the
amount of time needed to analyze of mail, telephone, and mall intercept survey data.
Capability of Auto-Driving
‘‘Auto-driving’’ means asking the respondent to comment on photographs, a video,
or some other stimulus directly relevant to the respondent; he or she is then asked to
provide an account of what is seen in these visuals, and whether or not the depiction
relates to him or her (and, if so, in what ways). The long interview is suited
particularly to showing respondents competing catalogs (some of which they are
known to have used) and asking them to comment on, and compare, the catalogs.
Similarly, respondents may be shown competing products available through direct
marketers and be asked to comment on their awareness and use of each product, and
to compare the strengths and weaknesses of each.
Ability to Generalize Results to a Population
Because of low numbers of completed interviews and the purposive selection of

households for study, the long interview research method is poorly suited for
generalizing results to a population. Mail and telephone surveys often include
drawing large sample sizes selected randomly that enable the researcher to generalize
results to a known population.
Ability to Generalize Results to a Theory
Long interview reports of individual household buying and using behavior represent
detailed case studies. Yin (1989, p. 47) pointed out that a rationale for single case
studies is substantiated when the case ‘‘represents a critical case in testing a well
formulated theory.’’ A ‘‘critical test’’ (Carlsmith et al., 1976) can be made in case
study research when the propositions of two or more competing theories are
examined empirically, using the data from one or more cases. Such a critical testing
involves ‘‘pattern matching’’ [also referred to as building-in degrees-of-freedom
(Campbell, 1975)] of the propositional predictions made from each theory with the
findings observed in one or more case studies. Rarely is one theory supported entirely
and a second or third rejected entirely; however, the case data are likely to support
one theory to a substantially greater extent than other theories.
Given that most direct marketers are able to specify how and why different
customers buy their products and services, such specifics, described in detail, can
serve as theories in-use in long interview data analysis. Thus, the predicted answers
and theoretical buying profiles of different customer types (suggested by direct
marketers) can be compared to the actual answers provided by customers fitting
different customer typologies.
In many instances, different senior and middle managers in the same direct
marketing firm propose very different theories-in-use for the behavior and
explanations of a given customer type. Data from the long interview are used to
resolve such differences. In the empirical example described later, different
expectations regarding each proposition were offered by different managers. Two
competing theories-in-use could be identified from the patterns of propositions
advocated by different senior managers in the firm. We labeled these two theories
in-use ‘‘Stay-the Course’’ and ‘‘New Wave’’ theories. Table 2 summarizes the
propositions advocated by proponents of the respective theories. A later section
describes alternative propositions taken on the ten issues.
In Table 2, the two patterns of responses to the ten issues illustrate building in
degrees-of-freedom in case study research for generalizing case data to theory, that is,
the tenth feature in Table 1. Examining the data from the long interviews to learn if
the answers advocated by one theory are supported more often than the competing
theory is an example of a critical test.
After reviewing related literature and describing the long interview method,
we describe details of the long interview research method and the results of its
application for 26 customers of direct merchants of gardening products (seeds,
plants, equipment).
One of the five largest mail order gardening suppliers, Boone, Inc.,1 agreed to
work closely with us in sponsoring the study. The applied objective of the study was
to learn rich descriptions of the thinking and feeling processes of each of five distinct
customer groups: loyal customers, new customers, non-buying catalog requesters,
buying once-only customers, and ‘‘divorced’’ customers. The principal objective of
the study was to learn how the products and services of direct merchants in one
industry (gardening products) fit into the lives of each customer segment.
1. Name disguised.
Table 2: Theory-in-use application.

Issue Theory 1: Stay- Theory 2:
the-Course New Wave
1. Customers with young children involve children No Yes

in buying gardening products
2. Our heavy-repeat customers prefer our products No Yes
versus competitors’ products
3. Former customers stop buying mainly because No Yes
of competitors’ actions
4. Substantial regional differences in purchasing exist No Yes
5. Most customers follow planned buying Mostly Yes Mostly No
strategies when using catalogs
6. Our customers prefer four-color photographs Yes No
for all products shown in catalog
7. Customers like the idea of buying a planned garden No Yes
8. Some important requests of customers are No Yes
concerned about buying gardening product
from environmentally friendly firms
9. Customers perceive differences among direct No Yes
merchants by product/service guarantees
10. Customers requesting our catalog and who No Yes
do not buy from us are buying from
competitor catalogs
Marketing and Consumer Literature Relating to the Long

Interview Method
Tucker (1967) provides an early application of a long interview method in consumer
behavior, including detailed case studies on the buying behavior of specific
individuals and families. Tucker’s work represents an early reporting of thick
descriptions of the complex relationships among cultural, sociological, psychological,
and marketing variables involved in need creation, buying, and consuming of
products and services. Tucker (1967, p. 134) offers two simple, yet profound,
propositions for modeling consumer behavior:
Proposition 1. Someone goes through some process and acquires something with
some effect.
Proposition 2. Someone uses something in some way with some effect.

In Tucker s opinion, not enough reliable information is available for the construc-
tion of an operational theory about the four elements in these propositions: someone,
process, something, and effect. The long interview is a research method suitable
particularly to provide reliable information for examining Tucker’s propositions.
Woodside and Fleck (1979) demonstrate substantial differences in need

recognition, buying, and consuming complexities in their study of two beer drinkers.
Each beer drinker was interviewed for six-hours over three meetings. More micro-
studies focusing on the thinking and decision processes of consumers who are
actively touching and buying products have been reported by Alexis, Haines, and
Simon (1968) and Bettman (1970, 1971b). This second body of work and a related
study (Bettman & Zins, 1978) leads to a fully developed theory of consumer
information processing and choice (Bettman, 1979a).
Visual research and auto-driving data collection methods are usually integral
parts of long interviews (Heisley & Levy, 1991, p. 36). In one of the most insightful
auto-driving studies, Taylor (1977) audio-taped the conversations of several real-
life insurance salesperson–customer encounters. Later, in separate meetings, Taylor
played the audiotapes to the two participants and asked them to explain their
thinking processes while speaking and listening to the original encounters. Thus,
detailed interpretations of the participants’ verbal exchanges at many points of the
original encounters were collected from the participants themselves. Chapter 9 in
the present book provides an in-depth discuss on the theory and research on
conversations.
McCracken (1987) uses photographs showing different styles of interior design
to stimulate provocative and useful data from respondents. Heisley and Levy (1991)
also photographed and audio-taped three families as they prepared and ate their
evening meal. Later, Heisley and Levy asked the respondents to look at the
photographs and ‘‘tell me whatever you think about when you look at [these
photographs].’’ Heisley and Lev emphasized that such auto-driving makes it possible
for people to communicate about themselves more fully and more subtly and,
perhaps, to represent themselves more fairly.
The Long Interview Research Method
McCracken (1988) offers a useful four-step method of inquiry for long interviews.
Figure 1 summarizes the four steps. The present chapter describes an application of
this method in a study of home gardeners who came in contact with direct merchants
of gardening products and services.
Stage 1: Review of Analytic Categories
Stage 1 calls for a review of both the scholarly and applied literature related to the
long interview and gardening behavior. Research questions central to this stage of
research include: (a) What analytic categories of customer types might be expected in
the field? (b) Have counterintuitive data and findings been reported related to the
planned study?
Analytical
Data
Stage 1 Stage 4
Review of analytical Discovering of analytic
categories and interview categories and analysis/write
design: predicting counter - up; glimpse and systematically
expectational data reconstruct S's view of world
Review Discovery
Processes Processes
Stage 3
Stage 2 Discovery of cultural categories
Review of cultural categories and interview: "grand tour":
and interview design: "what do category questions";
own, how did 1 come to own it" "auto-driving"
Cultural
Data
Figure 1: Long qualitative interview: four-part method of inquiry applied to direct

marketing research. Source: Developed from discussion in McCracken (1988,
pp. 29–45).
From the scholarly marketing literature (McCarthy & Perreault, 1993), several
analytic customer categories may be proposed as relevant for examining customers of
direct merchants. For example, a direct merchant may group customers
broadly: heavy-loyals (customers who buy almost every year with annual purchases
in the top quartile of all buying customers); light-loyals; buying once only customers;
customers requesting a catalog but not buying; divorced customers, those buying in
several previous years but who have stopped buying in recent years; non-customers
buying similar products from retail stores; noncustomers with demographics
and general life styles similar to loyal customers but not buying products from
any sources.
Such analytic categories have been applied in large-sample, national survey
studies in direct marketing research and, specifically, to direct response gardening
customers (Woodside & Moore, 1983; Woodside & Soni, 1991). Such studies support
the general proposition that the demographic and general life-style characteristics
between groups of customers, who have been categorized by differences in
purchasing behavior, are unique from one another. Such ‘‘backward’’ segmentation
provides useful information about the different lives of heavy loyals versus divorced
customers, for example.
Stage 2: Review of Cultural Categories and Interview Design
Cultural categorization of customers follows and blends with analytic categories.

For designing the long interview survey form, cultural categorizing is asking how
each analytic customer category might think and feel about each step in buying and
consuming a product/service. In addition, descriptive questions of current and past
buying and consuming behavior are included in developing the survey form.
McCracken (1987, 1988) began his study on personal possessions by asking himself,
‘‘What do I own? How did I come to own it?’’
A core issue in cultural categories is identifying responses of members of different
customer analytic categories to specific marketing tactics. For example, how do
customers of each category interpret the organi7.ation (format) of competing direct
merchant catalogs? Which customer group, if any, prefers ‘‘planned gardens,’’ that is,
a designed garden where the customer orders the design and an architectural
plan and seeds are purchased? Which customer group, if any, is concerned with
environmentally friendly products, catalogs, and packaging materials? Do such
concerns affect their buying decisions, and if so, how are decisions affected?
One to three intuitive and counterintuitive propositions were developed related to
each of these, as well as other, macro- and micro-issues. Examples of specific
propositions will be discussed in the next section.
Generation of Research Propositions
Following Stages 1 and 2, several intuitive and counterintuitive propositions may be

generated and considered. Different senior executives within the same direct marketing
firm often advocate opposing ‘‘intuitive’’ propositions about the demographic and life-
style characteristics of customers in different analytic segments, and about the buying
behavior of different customer groups. Similarly, executives may have conflicting
propositions (views) about the thinking and feeling processes of customers within
specific customer segments. Resolving such conflicts is important because key elements
of marketing strategy build on core propositions held to be valid.
Next, we offer a sample of 10 sets of such opposing intuitive propositions that
were examined empirically in this study. Each specific proposition was advocated by
one or more senior executives of Boone, Inc. Distinctive patterns in positions held
by different senior managers occurred. For example, a manager who believed that
families with young children involved the children in buying gardening products also
proposed that former customers stopped buying because of competitors’ actions
(Propositions 1 and 4 of the New Wave theory, see Table 2). These propositions are
offered to illustrate the usefulness of the long interview format. Many other strategy
issues of interest to Boone, too numerous to include here, were also examined.
1. Children and gardening. P1a: Customers include one or more of their young
children in buying gardening products and in related gardening activities.
P1b: Customers do not include any of their young children in buying gardening
products and in related gardening activities.
An understanding of opposing propositions will help resolve the following
marketing strategy issue: Should Boone offer products designed for children to
promote (position) gardening as a family activity?
2. Preferences of heavy-loyal customers. P2a: Our heavy loyal customers prefer our
firm versus our leading competitor. P2b: Heavy-loyal customers do not have
a preference for our firm versus our leading competitor.
Among gardeners buying from direct merchants, few differences in
customer attitudes toward direct marketing firms have been found among
heavy-loyal customers in national survey research studies (Woodside & Moore,
1983). However, senior management at Boone wanted more evidence on
this issue. Should Boone be positioned as the direct merchant most preferred
by its repeat customers for long periods of time (from parent to child to
grandchild)?
3. Why do former customers quit buying? P3a: Former customers stopped buying
mainly because competitors have made attractive product and promotional
offers that they prefer over our offers. P3b: Former customers have stopped
buying mainly because they have stopped gardening.
An examination of these propositions may affect a direct merchant’s response
to competitive moves; if many customers are being lost to competitors, then a
very aggressive response to new competitor actions is warranted. However, if loss
of customers has little to do with new competitor actions, it probably is not
worthwhile to spend much time and money on trying to win back these lost
customers.
4. Region of-country influence on purchase. P4a: Substantial differences in purchase
and consumption behavior occur across all regions of the country: North, South,
East, and West. P4b: Regional effects on purchase and consumption behavior are
more substantial in some regions than in others.
Empirical evidence on these opposing propositions helps to resolve strategy
issues including the following question: Should Boone include unique, regionally
oriented inserts in its national ‘‘Big Book’’ mailing (the largest sized catalog
mailed to all customers)?
5. Catalog shopping process by customers. P5a: Many, if not most, customers follow
planned buying strategies from the catalogs they buy from. P5b: Many, if not
most, customers buy products they did not plan to buy from the catalogs they
buy from.
These propositions relate to the issues of share of customers doing unplanned
buying, and their relative amount of planned versus impulse buying. The use of
‘‘reason why’’ copy and four-color photographs on all pages of the catalog are
affected by the shares of unplanned purchases reported by customers.
6. Customer preference for four-color photographs of products. P6a: Most customers
prefer seeing four-color photographs for all flower and vegetable products on all
pages of the catalog. P6b: Customers do not expect, or prefer, to see four-color
photographs of all products on all pages of the catalog.
Substantial cost reductions would occur if black and white photographs could
be used for one-fourth to one-half of the pages in the Big Book.
7. Customer acceptance/preference of garden plans in catalogs. P7a: Some customers
like the idea of buying a planned garden, which includes a detailed architectural
plan of the gardens and the seed/plant products. P7b: Few, if any, customers like
the idea of buying planned gardens.
In the 1990s, several competitors were starting to offer planned gardens in
their catalogs. Should Boone devote a little, a lot, or no space in the catalog to
planned gardens?
8. The issue of being environmentally friendly. P8a: Some important segments of
customers are concerned about buying from direct merchants who are environ-
mentally friendly in using recycled paper in their catalogs and biodegradable
packaging materials. P8b: When buying seeds and plants from direct merchants,
few, if any, customers are concerned about whether or not the direct merchants
are environmentally friendly.
Empirical evidence on customer thought processes and preference for direct
merchants positioned as being environmentally friendly is likely to affect the
speed in which direct merchants adopt environmentally oriented marketing
strategies. Do some customers recognize certain seed packaging to be more
environmentally friendly than other packaging materials? Would some customers
like to see special sections in catalogs of seeds and plants that are grown in
completely organic (no chemicals) conditions?
9. Customer differentiation of product and service guarantees by direct merchants.
P9a: A substantial number of customers perceive differences among direct
merchants in the product/service guarantees offered and how well direct
merchants live up to their guarantees. P9b: Few, if any, customers perceive
differences among direct merchants in the guarantees offered, or how they live up
to their guarantees.
Several senior executives at Boone were quite proud of their firm’s guarantees
of product quality and company service offered to customers; these executives
wanted to display prominently the guarantee in the main catalog and in
advertising. Empirical support for one of the opposing propositions would help
answer the question of whether or not promoting the guarantee would affect
customer perceptions and behavior.
10. Customers requesting, but not buying from the catalog. P10a: Customers requesting
a direct merchant’s catalog, but who do not purchase from the catalog,
are buying from competitors’ catalogs. P10b: Customers requesting a direct
merchant’s catalog, but who do not purchase from the catalog, are not buying
from any competitors’ catalogs.
Some senior executives at Boone believed that non-response catalog requesters

were not converting into customers because of powerful actions of competitors.
Other executives believed that most non-response requesters simply enjoy receiving
and looking at catalogs: These customers were not being lost to competitors because
receiving and looking at the free catalog was the end consumption activity for these
customers. This issue has substantial impact on marketing expenditures: Should

Boone allocate substantial funds to convert non-response requesters to buyers?
Whereas some intuitive findings may be supported empirically, many (if not most)
predicted relationships made by direct marketing executives often are ‘‘best case
scenarios,’’ atypical of reality; the opposite propositions match reality. Thus,
competing, counter intuitive propositions should be generated and become an
integral part in Stages 1 and 2.
For the empirical study we report here, a total of 82 sets of macro- and
micro-intuitive versus counterintuitive propositions were developed for empirical
examination. These propositions were developed from both analytic and cultural
categories of customers. The 10 example proposition sets were included among the
82 sets; many of the others relate to the micro-issues described in Stage 2.
Following Stages 1 and 2, the resulting initial draft of the survey form for the long
interview of gardeners was 54 pages. After a pretest with two customers from two
different analytic categories, and several discussions with a team of eight managers
from different departments within Boone, the survey form was reduced to 38 pages.
Stage 3: Discovery of Cultural Categories
I1ie questionnaire is finalized in Stage 3 — An important decision in this stage is

deciding if the study will include auto-driving and, if yes, the specific forms of the
auto-driving to use.
A second important issue settled in Stage 3 is the selection of respondents. For
the long interview procedure, McCracken emphasized that respondents are not a
‘‘sample,’’ and that their selection should not be governed by sampling rules. There
are, however, a few rules of thumb.
For our direct marketing research study of gardeners, we modified these heuristics
to contrast known customer groups. A central objective was being able to contrast
the thinking and feeling processes of customers likely to have special knowledge
about buying from direct marketing firms.
Stage 4: Discovery of Analytic Categories
Analyzing the results and writing up the report are principal activities in Stage 4.
Some analytic categories of organizing the data need to be created. The major
analytic categories for the data that we used for interpreting and writing up the
gardening study are: (a) synopsis; (b) plans and actions in 1993; (c) equipment and
space used in 1993; (d) buying sources used for 1993 and prior years; (e) favorite
catalogs and reasons; (f ) looking at, handling, and thinking about eight competing
catalogs; (g) favorite carriers for receiving product; (h) keeping catalogs; (i) new
variety buying; ( j) examining the contents of catalogs; (k) environmental issues;
(l) examining seed packages; and (m) a constant-sum evaluation exercise.
Care should be taken in Stage 4 to include all the steps found in thinking processes
important to customers or to direct marketing executives. For example, whereas
customers in several categories reported being concerned with environmental issues
and recognized that some seed packets were environmentally friendlier than others;
these concerns did not affect their selection of seed packets or use of catalogs.
Method
Mail-order buying of gardening products is an interesting area for an application of
the long interview technique. Direct marketing of plants and seeds is one of the oldest
mail-order industries in the United States. Before World War I, most plants and
seeds for home gardening were purchased using mail order; mail-order buying of
plants and seeds now represents less than 10 percent of total purchases in the United
States. Total Industry sales are less than $1 billion. The mail-order gardening
industry is not dominated by one or two firms, however, the top 10 firms in sales
represent about 40 percent of total mail-order industry sales. In the past 20 years, the
industry has stabilized its share of total gardening sales. However, the market shares
and profits of individual firms fluctuate, sometimes substantially, from year-to-year.
Competitors in the industry need to be concerned with two marketing objectives:
(a) maintaining close relationships with their core customers and (b) attracting/
retaining new customers.
Meeting with Executives of a Mall-Order Gardening Firm
The senior executives of Boone worked closely with us in designing the study. During
three meetings early in 1993, a market research project team was formed and
23 customer groups were identified for possible study. In consultation with the
project team and Boone’s CEO, five customer categories were selected for the study.
1. Core customers. Persons buying a higher than average dollar amount in three or
more years during the most recent five years from Boone’s main annual catalog.
2. New customers. First-time buyers in 1993, spending a higher than average dollar
amount, using the major annual catalog, which they got from Boone’s direct
response advertising offer.
3. Non-response requesters. Persons requesting the annual catalog from direct
response advertisements but not buying (i.e., non-convertors).
4. Once-only customers. Persons buying more than the average dollar amount in one
of the two years prior to 1993 but not in 1993.
5. Divorced customers. Persons buying more than the average dollar amount in two
or more years, but not since 1990.
As well as the product strategy issues highlighted earlier, several intuitive and
counterintuitive propositions for comparisons among the groups were created. Given
that large-scale, national surveys had been conducted on these customer groups
(see Woodside & Soni, 1991), these planned comparisons were exploratory at tempts
to offer rich insights into the details of known differences among the groups.
Participants
Informants for the study were selected from three urban areas of the United States
that are important geographic markets for Boone: Rochester, New York, Raleigh,
North Carolina, and Dallas, Texas. For each of the five customer groups, 10 names
and addresses of customers were selected from the firm’s database.
Initial Contact
Three weeks before visiting each of the selected geographic markets, two to three
subjects for each customer group were initially sent a letter on university stationery;
Boone was not identified as a sponsor of the study. The subjects were informed that
they had been selected for the study from the names of households that had either
requested and/or purchased products using mail order. The letter included an offer of
$30 for participating in an in home, personal interview.
Telephone Contact
Four to five days after the letter was sent, each subject was telephoned and asked
to participate in the study. With the exception of the divorced customers, over
70 percent of the subjects agreed to be interviewed. Those refusing to participate were
replaced; letters and phone calls were made to new subjects in an attempt to interview
one or more persons in each of the five groups in each of the three cities. Close to
60 percent of the divorced customers reported by telephone that they did not wish
to participate; most of those subjects reported not gardening in 1993.
Confirmation Letter and Interview Schedule
Following the telephone contact, a confirmation letter of the personal in

home appointment was mailed to each subject who had agreed to be interviewed.
The interviews were scheduled during Friday, Saturday, and Sunday in March
(for Dallas), May (Raleigh), and June (Rochester). Two to five interviews were
completed each day over the three-day period per city. The planned and implemented
interviews scheduled per city are summarized in Table 3.
Table 3: Research design for 1993 plant and seed customer study:
planned/implemented long interviews.
Customer group City

Dallas Raleigh Rochester Total
1. Current core: Combo vegetable/flower 2/1a 2/3 1/2 5/6

2. New customer: 20 + multi-products 1/2 2/4 2/2 5/7
3. Non-response requester: 2/2 1/0 2/2 5/4
4. Once-only customer 2/2 2/3 1/1 5/6
5. Divorced Customer 1/0 2/2 2/1 5/3
Total 8/7 9/11 8/8 25/26
a
The first number represents the interviews planned within each customer group: the second number
represents the interviews completed. For example, two interviews were planned in Dallas with current core
customers: one interview was completed. ‘‘Divorced’’ refers to customers not buying from the firm for the
most recent two years.
Personal, In-Home Interviews
Both authors participated in each of the 26 completed personal interviews.

Audiotape recordings were collected for 24 of the 26 interviews (electrical outlets
were not available for 2 interviews). Two to four photographs were made for all
26 respondents; 25 were photographed in their gardens.
Auto-driving data were collected during two parts of the personal interviews. In a
latter part of the interview, subjects were shown eight competing gardening catalogs
and asked several questions, such as:
‘‘Do you recall receiving any of these catalogs in 1993?’’
‘‘Did you order any seeds or plants for growing in 1993 from any
of these catalogs? If yes, which ones?’’
‘‘Which of these catalogs in front of you looks most interesting

to you?’’
‘‘What makes you say that?’’
‘‘Which of these catalogs looks least interesting to you?’’
Numerous other questions pertaining to issues about the catalogs were

asked. All of the questionnaire items are not included here for the sake of brevity.
Second, later in the interview, subjects were shown seed packages from eight different
mail-order firms. A series of open ended questions were asked, such as:
‘‘Please look at these seed packets before you, what do you notice
about them?’’
‘‘Please pick up three seed packets that are most appealing to you.’’
During an early part of the interview, several ‘‘grand tour’’ questions were asked
(see Stage 3 in Figure 1). For example, subjects were asked to describe their
gardening activities for 1993. This was to learn general behavior related to
gardening. Many specific probing questions followed (e.g., where were seeds/plants
purchased, who in the household did the buying and gardening, time spent per week
gardening, etc.).
Top-of-mind awareness data were collected. Questions were asked about what
mail-order firm first came to mind for each of several core benefits and mail-order
buying problems, if any. Each subject was also asked to complete a constant sum
exercise of selecting up to five core benefits she or he sought from a set of 30 benefits
printed on cards. After choosing the core benefits, subjects divided 10 points among
the benefits to represent the importance of each.
Completion Time
The average time to complete each interview was 90 minutes. Three interviews were
completed in less than 45 minutes; all of these brief interviews were completed with
once only and divorced customers.
Findings
Because our principal objective is to illustrate the application of the long interview in
direct marketing research, only a few highlights of the findings are presented here.
Following a discussion of findings for the 10 sets of opposing propositions, we offer
a detailed summary of one core customer and brief synopses of one customer in each
of the other four customer segments.
Findings for the Sets of Opposing Propositions
Children and Gardening
Nine of the twenty-six house holds interviewed included young children or teenagers.
Joint gardening activities by parents and children were reported by none of these nine
households; P1b not P1a is supported. Interestingly enough, some of the parents
reported being influenced by their parents in their buying from direct merchants
versus retail stores and in their choice of specific direct merchant firms. Thus, family
influence was found to occur, but not as expected by the intuitive proposition of
parent–child influence. The advantage of being a very well established mail-order
gardening firm in affecting word-of-mouth recommendations between parent and
adult child was an unexpected finding of the study. The decision was made, based in
part on the findings of this study, not to place substantial funds in promoting
gardening as a parent–young–child activity.
Preferences of Heavy-Loyal Customers
All of the heavy-loyal customers interviewed reported preferring one or two direct
gardening merchants over others; thus, P2a, not P2b, is supported. However, these loyal
customers did not report negative feelings toward any particular direct merchants
(with one exception). Rating scales would not likely capture their preferences
toward a particular direct merchant because preferences were based on experience.
For example, ‘‘I always buy from Boone, that s the reason I prefer them,’’ is most
representative of heavy-loyal preference. The strategy implication for direct marketers
is that an extensive trial period of different suppliers is unlikely to occur among heavy-
loyals. In other words, a direct merchant’s catalog must be readily available during the
strategic window of opportunity when the heavy-loyals start to garden.
Why Do Former Customers Quit Buying?
Former customers reported a decrease in gardening activities during the most

recent years, or preferences to buy from local retail stores, P3b, not P3a is supported:
Trying to win back such lost customers is doomed to failure.
Region-of-Country Influence on Purchase
Regional influences were found in one region versus the other two: Customers in the
Dallas area reported the need to buy from direct merchants located in the Southwest
because of special growing problems. Thus, P4b not P4a is supported. Four of seven
respondents in Dallas reported buying from the same regional direct merchant
because they believed this supplier offered products uniquely suited to the growing
requirements of the Southwest. A similar finding did not occur among respondents in
the Raleigh and Rochester areas. Given the substantial size of the Texas market
and the possibility of substantial regional bias of customers, the development and
advertising of products uniquely suited for dry, hot climates may be a viable strategy.
Catalog Shopping Process by Customers
All respondents who did buy gardening products from direct merchants in 1993
reported buying some unplanned products. To a greater extent, P5b not P5a, is
supported. All respondents buying from the catalogs reported examining each
page of the catalogs they requested and received. Heavy loyal customers reported
buying some favorites every year, plus buying some unplanned items each year.
Hence, providing reason why copy and four color photographs throughout the
catalog are necessary.
Customer Preference for Four-Color Photographs of Products
A Strong consistent bias in favor of four-color photographs for all plants was found
across all five customer types. P6b not P6a, is supported. This finding was supported
especially in the auto-driving portion of the interview for the sponsoring firm’s
catalog. The respondents were asked to examine two catalogs in-depth during the
interview, one of the two catalogs always being the sponsoring firms catalog.
According to customers, a competitive advantage of Boone’s catalog was the
completeness in providing four-color photographs of all products. For Boone, the
savings from switching to using black-and-white photographs on some pages may
weaken brand equity in the minds of customers.
Customer Acceptance/Preference of Garden Plans in Catalogs
None of the respondents bought planned gardens and none reported willingness to
buy planned gardens. The findings support P7b not P7a. Devoting substantial catalog
space to planned gardens is unlikely to be profitable. They are unlikely to gain
widespread customer acceptance. ‘‘The whole idea for me is to come up with my own
garden arrangements,’’ was one response to planned gardens. Several respondents
reported that they could see why someone would like the idea of planned gardens,
but such plans were not for them.
The Issue of Being Environmentally Friendly
With the exception of one respondent in the Rochester area, all reported that helping
the environment was not something they think about when looking at seed and plant
catalogs. No one company came to mind as being environmentally friendly. P8b not
P8a is supported. In the auto-driving portion of the interview, respondents did
recognize that some seed packet materials were more environmentally friendly than
others; however, none reported that they would be influenced to buy for this reason.
Uniquely capturing the environmentally friendly position in the minds of
customers is unlikely to increase sales.
How Do Customers Differentiate Product and Service

Guarantees by Direct Merchants?
Of the respondents who purchased gardening products by mail in 1993, all reported
that the merchants they were familiar with offered money-back guarantees. ‘‘All the
companies do a good job in guaranteeing their products,’’ was the usual response
to this issue. Thus, P9b not P9a is supported. Attempting to focus customer attention
on a guarantee is unlikely to be effective in increasing customer loyalty.
The majority of respondents indicated some past communication with direct
merchants to complain about mistakes related to order fulfillment. In all instances,
prompt responses were made to the complete satisfaction of these customers, with
the exception of one direct merchant (this particular company was criticized by
four respondents in two regions of the country as offering inferior products
consistently). Except for that one firm, all respondents reported that they would
purchase again from the same direct merchants because of prompt handling of their
complaints.
Customers Requesting but Not Buying from the Catalog
Non-response catalog requesters prefer to buy locally. P10b, not P10a is supported:
Such customers are not being lost to other direct merchants. All four of the non-
response requesters interviewed reported being most familiar with Boone’s catalog.
None reported actually buying from any gardening catalogs. ‘‘I like to just look
at the catalogs, especially the ones in color,’’ is one statement that best summarizes
the customer relationship with any direct merchant of gardening products.
Identification and purging of non-response requesters from receiving future catalog
issues may be best; trying to convert non-response requesters into buyers may be too
expensive to justify.
Critical Testing of the Competing Theories-in-Use
Referring to Table 2, the Stay-the-Course theory in use is supported by the case

study results for 7 of the 10 propositions (Propositions 1, 3, 6, 7, 8, 9, and 10).
The New Wave theory-in-use is supported by the results for Propositions 2, 4,
and somewhat 5. Critical testing of 10 additional propositions not discussed in
this article included support for 8 additional propositions advocated by Stay-the-
Course, with 2 additional propositions advocated by New Wave. The 15 of 20
‘‘hit ratio’’ for the Stay-the-Course theory is statistically significant ( po.04 by a
two-tailed sign test, Siegel, 1956). A more useful view for strategy develop-
ment would reflect that, although the Stay-the-Course theory receives greater
support, a few important propositions advocated by the New Wave theory are also
supported.
Descriptions of Two Customer Types
Two exhibits provide summaries of detailed descriptions of gardeners from two of

the five customer types. Additional detailed summaries of gardeners from the other
three customer types are available from the authors.
A Core Customer
We begin with a summary description of a core customer couple, Bruce and

Dorothea. Appendix A summarizes the discussion. Their major gardening activities
occur on a small family farm located several miles from their home. Appendix A is a
detailed transcription of the audio-taped responses of this couple.
The demographic and life-style profile of Bruce and Dorothea closely matches
the profile of about one-third of Boone’s core customers: retired homeowners and
high school and college graduates, who garden every day of the week. Note that this
couple does have favorite direct merchants they buy from each year; however, they
are not 100 percent loyal to one or two suppliers. They buy from three to four direct
merchants each year.
Of the three geographic areas included in the study, only in Texas was a
regional direct merchant mentioned as being especially able to meet local growing
conditions. Given the importance of the Southwest market, this study served as an
early warning system of a possible regional threat to Boone. The regional direct
merchant in Texas was mentioned by customers in other segments as a supplier
during the study. Before data collection, this particular merchant was unknown
to Boone.
The data on this couple and other core customers revised the concept of customer
loyalty for several senior executives. Although a few customers are close to 100
percent committed to one direct merchant supplier, core buyers usually rely on three
to four direct merchant sources, and one to two local stores.
A New Customer
Exhibit 2 summarizes the demographic, life-style, buying, and consuming profile of

Bert, a new customer. Bert does all of his gardening at his home. Bert buys most
of his plants and seeds from Boone, but he also buys from a local nursery. Note in
Appendix B that Bert was influenced by his mother to prefer and use one particular
plant and seed catalog. This Tran generational communication represents a major
form of brand equity for the senior mail-order plant and seed direct merchants; this
finding is confirmed by examining long interview research studies commissioned by
Boone in the 1980s.
A Non-Response Requester
Elaine prefers to buy from local retail sources. Her buying history indicates that her
reasons for not becoming a customer after sending an inquiry are unrelated to
preferring another direct merchant’s products. The typical profile of non-response
requesters includes most of the characteristics reported by Elaine: They buy locally
and their garden space is small compared to core and new customers.
A Once-Only Customer
Janice, a once-only customer of Boone, reported that she is an active catalog user but
tends to use some plant and seed catalogs as reference guides, not to buy from.
Similar to non-response requesters, once-only customers prefer to buy locally.
Different from non-response requesters, once-only customers sometimes make trial
purchases from direct merchants.
Marketing to once-only customers is not profitable for most plant and seed direct
merchants, including Boone. After careful study of the long interview reports in this
study and prior studies, Boone’s senior executives concluded that substantial
investments should not be made in attempting to convert once-only customers into
long-term customers. Greater paybacks with less investment were believed possible
from increasing marketing efforts to attract new customers and increase sales to core
customers.
A Divorced Customer
Sara Beth, a divorced customer, became a lost customer not due to any particular
failure of Boone, or to the marketing actions of competing direct merchants.
She perceives particular mail order catalogs to offer less variety in recent years than
in the past.
After examining the detailed findings for Sara Lee and other divorced customers
(see Woodside & Soni, 1991), executives for Boone decided not to try to win them
back. The sensitivity of these customers to special marketing offers was concluded to
be too low and not cost effective.
Implications for Direct Marketing Theory-in-Use and Strategy
Certainly, the long interview as a research tool is not a panacea for direct marketing
strategy. The method is most useful in complementing and extending large-scale
survey research studies. By applying the long interview in direct marketing research,
the direct marketing strategist likely will gain deep understanding of the thinking and
feeling processes of members of important customer segments.
Managers often make the mistake of assuming they know their customers and that
their particular view (i.e., theory in use) is valid, without being able to refer to any
supporting evidence. Long interview research is useful for a direct marketing
manager to test competing theories-in-use about customers’ behaviors and find
out what they are really thinking, feeling, preferring, and doing. In addition, one
manager may be sure he knows what customers want, whereas another manager is
equally sure customers want something else. Studies employing long interviews can
address such opposing propositions (as illustrated here) and provide directions
for marketing strategy.
The objective for this chapter includes demonstrating the use of the long interview
(a case study research methodology) for developing descriptive and predictive
theories of several direct marketing customer types. The discussion of the results for
the theoretical propositions for different customer types in the mail-order gardening
industry leads to the conclusion that different sets of propositions are likely to be
necessary to describe accurately the behavior of different customer types, and for
designing effective marketing strategies to influence the buying behaviors of these
customers.
Appendix A. Current Core Customer
Bruce and Dorothea, Dallas, TX
Synopsis
Bruce and Dorothea are both retired and in their sixties. They own and maintain a
modest home that they have occupied for 35 years on an income in the $60,000–
$80,000 category. They grow vegetables on a five acre farm 59 miles from their home.
They experiment with plants at home and have some flowers and vegetables that are
grown around the house. Most — 90 percent — of their gardening purchases are for
vegetables. They use the Porter and Sons catalog extensively, especially for tomato
seeds. Reason: Porter is local and knows the Texas growing conditions, which tend to
be hot and dry in the Dallas area. Dorothea reported buying squash seeds from both
Boone and Walnut. Boone commands a small amount of their annual purchases.
Plans and Activities in 1993
Actions Mostly vegetables will be grown this year in the yard and at the farm; a few
flowers will be used in beds around the house. Both Bruce and Dorothea work in the
yard and at the farm in gardening. Of the crop that is harvested, they tend to freeze
black eyed peas, corn, and squash. They tend to spend at least an hour per day in
their home yard garden and go to the farm twice per week during the summer.
Equipment and Space The backyard area used for planting and experimentation is
about 20 20 feet. A small utility shed is also in the backyard for storing garden
tools and supplies.
Sources for 1993 and Prior
Top-of-Mind Sources Porter and Sons was mentioned as the main source of supply
for seeds and plants. This company has products that have been developed for the
Texas climate and conditions.
Catalog Sources Other catalog purchases include Russian Giant sunflower seeds,
squash, cucumber, and green peas from Walnut, squash from Boone, and German
Giant radish seeds from Gurney. They reported buying from these catalogs for the
past 15 years because they have had relatively good luck with the seeds from these
sources.
What Catches Your Eye (Unaided)? Walnut and Boone were identified as the
prettiest catalogs. Porter and Sons is distinctive since it is all black and white and
very plain compared to other mail-order catalogs.
Favorite Catalogs and Reasons
Favorite Porter and Sons and Walnut were identified as favorite catalogs. Porter
and Sons was chosen because of its long history of supplying products specially
suited for growing in Texas and because of prior success with Porter products. The
Walnut catalog was described as pretty and well illustrated. The pictures of tomatoes
were described by Bruce as ‘‘beautiful but they won’t grow here.’’ Dorothea
mentioned that she likes Walnut because ‘‘the squash and beans do stand the heat
pretty well.’’
Reject No catalog was identified here. They mentioned that they do not buy much
from Gurney; of all the catalogs they buy from, they get the least amount of
merchandise from Gurney. The reason for not buying much from Gurney is that
Bruce feels the seeds are better suited for the climate in the northeast rather than
Texas.
Looking at and Handling Eight Catalogs
Looking and Handling The Boone Catalog was picked up first, followed by Walnut
and Gurney. The Walnut catalog was mentioned as one whose arrival is most
anticipated.
From Ordering to Receipt Merchandise ordered from Boone and Gurney arrived at
the right time for planting whereas products from Walnut were reported to be slow in
arriving. The seeds from Boone and Gurney arrived in about 7–10 days but the
Walnut order took three weeks to arrive.
Catalog Appearance
Most Interesting and Why Walnut was identified as most interesting and Boone was
chosen second. These were chosen because of the level of familiarity with these
catalogs. When prompted. Stokes was mentioned because of the display of vegetables
on the cover of the catalog.
Least Interesting and Why Shepherds and Harris were identified as least interesting.
Shepherd was described as not colorful and no specific reason was given for why
Harris was judged least interesting.
Most Appealing Cover Dorothea chose the Walnut catalog as most appealing due
to its colorful display and the new varieties shown. Bruce chose Walnut’s cover as
most appealing because it features a big tomato.
Favorite Seeds
Porter and Sons’ tomato seeds and black eyed pea seeds were identified as favorites
purchased from year to year. These were purchased due to good performance in
Texas conditions.
Buying Exclusives
Porter and Sons’ Celebrity tomatoes were mentioned as exclusives bought in 1993.
Other exclusives bought include German Giant radishes from Gurney and ‘‘Heat
Wave’’ tomatoes from Walnut. These companies were mentioned as having exclusive
varieties for these types of vegetables. Boone and Walnut were mentioned as having
exclusive varieties of squash and green beans.
Catalog Contents for Most Familiar Catalog
The catalog identified as most familiar was Walnut. The following comments are
offered regarding the Walnut catalog.
People in Catalog Bruce and Dorothea do not particularly enjoy seeing people in
the catalogs or reading captions about people shown. They said that they don’t pay
attention to that aspect of the catalog content.
Information The amount of information contained was thought to be enough.
Writing Style The Writing style was judged easy and enjoyable to read and the
catalog was thought to be easy to use. Information is precise and to the point. The
amount of technical detail is generally fine except Bruce would like more information
to account for the hot dry conditions in central Texas, if possible.
Photographs The size of photographs was thought to be just right. Bruce mentioned
that he did not believe the photographs to be particularly realistic in showing plants
as they might grow in Texas. He mentioned that some plants don’t do well in the dry
heat compared to how they might look in more temperate climates. Showing pictures
of each item offered in the catalog was important.
Planned Garden Planned gardens were thought to be included in the catalog for
flowers, in particular. These are not of interest to either Bruce or Dorothea.
Regional Gardening Concerns Bruce said he would like to have more detailed
regional information if possible. Again, growing conditions in Texas are unique in
some ways. Bruce mentioned that he reads this information when available, but that
it is usually not furnished in catalogs.
Children’s Garden This is of no interest to the couple.
Organization They both recognized new varieties of vegetables and flowers were
offered on the first few pages of the catalog. The organization scheme of new
offerings, flowers alphabetical, and vegetables alphabetical was thought to be fine.
Guarantee The guarantee for seeds was not thought to be of major importance,
A lot of variation in performance may be due to weather and or soil conditions
rather than the seed product itself. A guarantee would be important for live plants,
but no live plants are bought from catalogs. No mail-order catalogs came to mind as
having the best guarantee.
Seed Packets
Notice Walnut was noticed as having a lot of information on the seed packet.
Least Appealing The seed packet from Boone was selected and described as best
looking but difficult to open. Thus, this difficulty makes it relatively unappealing.
Harris seeds were also judged unappealing. The pack was described as not colorful
and a picture is desirable (as on the Gurney’s dill pack).
Best Quality Seeds Walnut and Boone were judged to have the best quality seeds.
The couple mentioned that these were familiar and known products.
Appendix B. New, Multiproduct Customer

Synopsis
Bert, a trial lawyer, is an active gardener who gets a little help from his wife, Bert is in
the 35–44 age group and earns above $100,000. He works full time and his wife works
part time. They have three children living at home. They have been in their current
home for 10 years. Bert has ordered from Boone in the past after learning of the
catalog from his mother. He has bought both plants and seeds from Boone this year.
Plans and Activities in 1993

Actions
Bert has both vegetables and flowers in his yard. He will be growing strawberries,
Lima beans, peppers, carrots, onions, peas, okra, squash, asparagus, artichokes,
radishes, garlic, lettuce, and beets. Flowers include pansies, snapdragons, petunias,
cosmos, impatiens, and periwinkle. He also plans to have rosebushes and gladiolus
bulbs. Mostly flowers and vegetables will be grown and have been bought; Bert
mentioned that he did buy some azaleas from a local nursery to plant in the yard.
Bert estimated that he does some gardening six days per week for an average of 15–20
hours per week. He freezes some of his crop. Bert has an area for herbs. He will grow
parsley, cilantro, garlic, chives, and dill. He grows these herbs from seed.
Equipment and Space
Bert’s back yard is the site of most of his gardening area. It measures approximately
40 60 feet; Bert also has a greenhouse in his backyard.
Chapter 14
Tipping-Point Modeling in Case Study

Research$
Synopsis
Micro-tipping point (MTP) theory includes the proposal that a specific stream
of unconscious and conscious thoughts result in a go/no-go discretionary
action in a given context (e.g., whether or not to visit a given destination in a
given season or year involving particular persons being included or excluded
from the trip; whether to buy a Ford, BMW, or Toyota). The specific stream
represents a stream or conjunctive combination of thoughts-in-context that
results in a tourist party actually taking the steps that include booking the trip
and experiencing a destination first-hand. Building such contingency models
that are applicable to real-life combinations of unconscious and conscious
thinking requires collecting data from informants on both implicit and explicit
beliefs, attitudes, and thinking rules relevant for a specific yes/no context.
Chapter 14 illustrates applying the long interview for collecting such data and
using quantitative comparative analysis for constructing MTP models.
Introduction: Micro-Tipping Theory in Travel Research
Chapter 14 proposes building theory and doing case study research for examining
conjunctive combinations of facilitating and constraining conditions that affect
discretionary decisions (cf., Bargeman & van der Poel, 2006; Becken & Gnoth, 2004;
Woodside, Caldwell, & Spurr, 2006). The chapter builds on prior fit-like-a-glove
theory (see Allen, 2002, for an application of FLAG research in post-secondary
education institution choice behavior) and research focusing on implicit thinking
(e.g., Bargh, 2002; Wilson, 2002). The chapter describes how long interview
(McCracken, 1988) applications are useful for examining micro-tipping point theory
$
This chapter is co-authored with Arch G. Woodside, Boston College, and Drew Martin, University of
Hawaii.
in travel research focusing on potential travelers’ unconscious and conscious

thoughts that relate to visiting specific destinations.
Even though the dominant logic in research into individual behavior is to ask
informant questions, Rapaille (2006) argues consumers do not know the reasons for
their own behavior and they only will offer answers that they think the researcher wants
to hear. Rapaille (2006) collects information using his ‘‘discovery session’’ method. The
final third of the three-hour discovery sessions includes five to ten informants relating
their first experiences with a focal product or service while they lie in horizontal
positions. Rapaille’s method attempts to breakthrough consciousness and conscious-
editing of automatic retrievals from memory to learn unconscious associations and
processes. Creating methods to achieve such breakthroughs is worthwhile given the
substantial evidence that most thinking occurs unconsciously (for reviews, see Bargh,
2002; Gladwell, 2005; Wegner, 2002; Wilson, 2002; Woodside, 2006; Zaltman, 2003).
The literature on the dominance of unconscious thinking is persuasive.
Informants’ limited abilities and self-editing tendencies when answering direct
questions (e.g., to appear sane, good, and helpful), along with the substantial value of
alternative research methods), imply that replacing or complementing the dominant
logic with additional research methods is necessary. Early work in consumer research
recognizes that understanding consumer unconscious thinking requires methods
other than asking direct questions (Cox, 1967; Dichter, 1964; McCracken, 1988).
For example, Dichter (1964) video-taped informants’ behavioral responses to his
requests to present themselves as a product or perform a service-ritual involving a
service (e.g., prepare and serve a meal). Cox (1967) guided two consumers separately
in weekly conversations with themselves over more than 14 weeks in order to
understand and explain their own grocery shopping behavior. McCracken (1988)
suggests using the long interview to enable informants to become aware and to report
on how cultural and socio-historical forces affect their thoughts and actions.
Applications of the long interview include face-to-face meetings with informants in
physical contexts that occur naturally for the informants; attempting to permit the
informant to drive the topics and explanations of how events unfold for her or his
experiences; avoiding the use of ‘‘why’’ questions to lessen interviewer biases and
likelihood that the informant will present answers the she or he perceives the
interviewer would like to hear; and attempting to increase accuracy and thick
description of conversations, events, and timelines (cf., Thompson, 1997).
Bargeman and van der Poel (2006) apply the long interview method in a qualitative
study on a qualitative study among 32 Dutch households, which is part of a larger
research project on the routinization of vacation choice behavior of Dutch vacationers;
Bargeman and van der Poel conclude from their study that ‘‘the vacation decision-
making processes of the interviewed households are much less extensive and far more
routinized than described in the rational choice models.’’ In routinized thinking
processes external search, developing a consideration set of possible destinations and
travel activities, and evaluation and selection of specific destinations and activities all
tend to be minimal in comparison with extensive problem solving. The conclusions by
Crotts (1999), Decrop (1999a, 199b), and Fodness and Murray (1997, 1998, 1999) in
separate research studies all emphasize a bias by tourists toward minimizing effort,
Tipping-Point Modeling in Case Study Research 293
using internal search and automatic thought retrievals versus external search and
extensive evaluations in taking steps leading to making trips.
Using a large scale survey of visitors to New Zealand, Becken and Gnoth (2004,
p. 376) support a ‘‘tourist consumption system’’ hypothesis that ‘‘although each
holiday component could become the starting point for a tourist’s holiday plan,
choices soon become contingent on choices made previously.’’ Becken and Gnoth’s
research provides evidence that most tourists to a given destination may be
segmented by the decision processes that they employ in extensively or routinely plan
different options in their resulting trips.
Ecological systems theory (e.g., Bronfenbrenner, 1986, 1992; King & Woodside,
2000; Raymore, 2002; Woodside et al., 2006) states that understanding an individual’s
environment is essential to understanding individual’s behavior. An ecological
perspective of human development recognizes the importance of understanding the
contexts in which an individual finds herself. This approach ‘‘incorporates the
interactions between the individual, other individuals, and the social structures of
society to explain human development’’ (Raymore, 2002, pp. 41–42). Central points in
ecological systems theory include the following two observations: (1) the presence/
absence of specific factor levels may be neither sufficient nor necessary for an
occurrence of a specific outcome and (2) however, the combination of specific chains
of events may nearly always lead to a given outcome (see King & Woodside, 2000;
Ragin, 2000). From an ecological perspective, individuals interact with the contexts in
which they live their lives.
The interaction or occurrence of specific contexts triggering automatic thoughts
often represent a tipping point resulting in a seemingly discretionary decision
that reflects little to no conscious deliberation before the individual commits to a
specific action. Government advocacy of ‘‘cooling-off periods,’’ where customers can
cancel a purchase agreement within three business days after the purchase (Office
of Fair Trade, 2003) is one example of attempting to introduce substantial conscious
thinking into prior mostly unconscious-driven tipping-point responses. Therefore,
tourism behavior research needs to consider the specific contexts in which an
individual lives in order to fully understand the individual’s behavior. The present
chapter proposes that a conjunction of context and unconscious/conscious thinking
results in a micro-tipping point (MTP) that influences the occurrence of specific
outcome (e.g., visit to a given destination). A core proposition to MTP theory is that
neither context nor person factor alone is sufficient to result in a given outcome. The
use of ethnographic decision tree analysis (Gladwell, 1989) informs the conjunctive
nature of MTP processes. The sections following this introduction develop these
proposals and case study research illustrates their application.
An alternative ecological systems theory applies constructive choice theory and
the FLAG model (Allen, 2002) to describe the travel and leisure behavior of
informants. Using these theories, Allen (2002) describes the ‘‘causal historical wave’’
in informants’ lived experiences in selecting a college or university by recent secondary
school graduates. Allen reported the finding that informants did not compare the
relative strengths and limitations of alternative colleges and universities — each
option was evaluated automatically as a go or no-go outcome.
Micro-Tipping Point Theory
Micro-tipping point (MTP) theory proposes that individuals automatically engage

mentally in combining unconscious and conscious thoughts. This proposal builds on
the work of Bargh (2002) and his colleagues (e.g., Bargh et al., 1996) that purport
both forms of thinking are relevant when an individual focuses on a particular task.
McClelland et al. (1992) advance the related proposition that implicit and explicit
motives differ and operate simultaneously in regards to a given topic; empirical
evidence supporting this proposition; Brunel et al. (2004) provide substantial
additional evidence supporting the independence of implicit and explicit motives.
MTP theory proposes conjunctional causation of specific levels of multiple
features (i.e., factors) for a given binary outcome (e.g., leisure-only overnight trip of
three or more nights away from home versus no such trip). While MTP theory
recognizes that some specific feature-levels are necessary, the instances are rare when
a level of one feature is sufficient to cause a given outcome. This proposition is a core
tenet of MTP theory. Relating to conjunctional causation is the point that data
outliers often are difficult to interpret. For example, a travel-related data set includes
a few cases of very low income households engaging in one or more annual
leisure trips away from home of three or more and a few high income households
engaging in no leisure trips away from home of three or more nights (e.g., Woodside
et al., 2006). While these data points appear to be counter-intuitive, conjunctional
causation can explain these go/no-go outcomes in thinking paths that may not fully
planned or consciously reflected on by travelers. Usually, combinations of four or
more binary, demographic variables are necessary and sufficient to explain such
results.
MTP theory adopts a case-based research orientation that includes the attempt to
go beyond qualitative and quantitative arguments. MTP theory advocates the use the
methods of qualitative comparative analysis (QCA) (see Ragin, 1989) for building,
comparing, and generalizing conjunctional causation models.
While it is standard practice for case-oriented researchers to search for

constants (e.g., high levels of foreign capital penetration) across
positive [e.g., positive ¼ recent trip to a destination; negative ¼ no
recent trip] cases in their attempts to identify the causal forces behind
an outcome (e.g., anti-neocolonial revolutions), the typical case-
oriented inquiry does not assume or even anticipate causal uniformity
across positive cases. On the contrary, the usual expectation is that
different combinations of causes may produce the same outcome.
That is, case-oriented researchers often pay special attention to the
diverse ways a common outcome may be reached y When examining
similarities and differences across cases, case-oriented researchers
usually expect evidence to be causally ‘‘lumpy.’’ That is, they anticipate
finding several major causal pathways in a given body of cross-case
evidence. (Ragin, 1989, p. 38)
MTP theory proposes and focuses on identifying the few (o10) conjunctural
causal paths (also referred to as ‘‘streams of behavior,’’ see Gardner, 1990) that result
in a positive outcome for the majority of cases as well as often reporting conjunctural
causal paths resulting in negative outcomes. MTP adopts the QCA view that
multiple conjunctural causation challenges the very idea of ‘‘relative strengths’’ of
independent variables in affecting a dependent variable. ‘‘It is not possible to assess a
variable’s ‘unique’ or separate contribution to the explanation of variation in some
outcome unless the model in question is a simple additive model’’ (Ragin, 1989,
p. 41). MTP theory proposes multiplicative relationships among several (e.g., four or
more) antecedent variables as both necessary and sufficient to lead to acceptance
or rejection of a specific outcome.
MTP theory is applicable for explaining both the multiple conjunctural causations
leading to a positive outcome, negative outcome cases, and both positive and
negative outcomes. MTP theory adopts the QCA view that alternative paths of
constants leading to a constant outcome is worthy of study as well as constructing
models that explain and predict binary outcomes. Montgomery (1975) provides
an early example of an MTP model (using the label, ‘‘gatekeeper analysis’’) for a
supermarket new-product buying committee in the context of business-to-business
marketing to explain and predict go and no-go committee decisions.
MTP theory does not expect to explain all the cases in a study (e.g., households
traveling to Hawaii) with a single model (even when the model incorporates
multiple conjunctural causation). MTP and comparative case analysis are applicable
for analyzing both positive and negative outcomes (e.g., a destination visit versus
non-visit), these research methods are suitable for explicating the alternative multiple
combinations of conditions necessary to occur (i.e., path A versus path B y versus
path Z) in reaching either positive or negative outcomes alone. The case studies in the
present chapter focuses only on examining positive outcomes; other applications of
MTP are available in the literature that focus on explicating paths leading to both
positive and negative outcomes (e.g., see Woodside et al., 2006).
The following generalized example illustrates causal explanations in case-oriented
research that MTP theory proposes for a positive-outcome case study. When
conditions A, B, and C are present, X causes Y, however, if any one of these
conditions (A, B, or C) is absent, and X is also absent, then Z causes Y. This
argument is multiple and conjunctural in form because the proposition cites alternate
combinations of causal conditions. The hypothetical causal argument states four
combinations of conditions occur that result in outcome Y. Using Boolean algebra,
the causal argument can be formulated (see Ragin, 2000) as follows:
Y ¼ ðA B C XÞ þ ðA B c x ZÞ þ ðA b C x ZÞ þ ða B C x ZÞ (1)
In Eq. (1), the upper-case letters indicate the presence of a condition and lower-
case letters indicate its absence. Also, multiplication (mid-level dots) indicates
causal conjunctures and addition indicates alternative causal pathways (for further
discussion, see Ragin, 2000).
Fits-Like-a-Glove (Flag) Model
Practice theory forms the basis of the FLAG, or fits-like-a-glove model (Allen, 2002).
Practice theory advocates that, ‘‘People and psychological processes are embedded in
and inseparable from their physical and social contexts y . Time, continuity, and
change are intrinsic aspects of psychological phenomena’’ (Altman, 1992, pp. 268–269).
This perspective understands that human beings acquire knowledge

through their transactions with their social and physical environments.
Furthermore, human beings cannot be understood as apart from their
social and physical settings, but rather as embedded within them.
Development is an inherent quality of environments, which afford
individuals the opportunity to advance their understandings about
people and places. (Lippman, 2007)
FLAG theory states that social and historical forces, or habitus (cf., Bourdieu &
Johnson, 1993), shape the human experience. FLAG choice emphasizes the role of
the body in perception and comprehension which underlies the feelings, under-
standings, and actions of the consumer. The FLAG model describes embodied
sensing, where a person’s body functions as an integral unconscious and conscious
sensing organ. A person’s experience is a mixture of the meanings resulting from the
combination of the person’s touching with all things encountered in the world. For
example, embodied experience is illustrated by a traveling violation in the game of
basketball. A player’s recognition of this violation often stems from an instantaneous
sensing of the body, rather than an understanding of the rules defining a traveling
violation. Additionally, practical experience comprises all of the understandings,
feelings, and actions that are induced while a person is in motion and engaging in a
specific context. For example, athletic performance is the result of the athlete’s feel
for the game that comprises both bodily and cognitive states in relation to context
and environment, rather than the mind giving the body commands.
The social shaping of practice takes two forms. The first form of social shaping
includes low-involvement socialization of the understandings, feelings, and actions
which make up the habitus of members of a certain group. Because these members
have been exposed to similar social conditions and relations, they share a similar
habitus. For example, differences between street basketball and formalized basket-
ball illustrate socialization effects. Because different social conditions cultivate
these games, the style of play differs greatly even though both games appear to be
essentially the same. The second form of social shaping entails the way in which
external factors shape practice. These external factors include family, peer groups,
institutions, and mass media. Low-involvement socialization and external factors
combine to shape practice.
The FLAG model elaborates on practice theory and integrates elements of social
context into a model of choice (Allen, 2002). People make decisions based on what
feels right or seems natural given the conditions and circumstances surrounding
them. Therefore, consumers make decisions in which the object of choice seems to
be a predestined, perfect fit. FLAG choices are present in daily experience. Choices
made for friends, occupation, particular styles of clothing, and even travel can be
explained by the FLAG framework. Reflecting FLAG model contexts are problem/
opportunity discussions where informant responses express an inability to provide
reasons or explanations beyond simply reporting, ‘‘It just felt right’’ for one option
and ‘‘It didn’t feel right’’ for many rejected options.
The FLAG model lends support to Rapaille’s (2006) view that consumers are often
unable to answer direct questions that accurately explain the reasons or causes of their
own behavior. Possibly without being aware of the literature, Allen’s FLAG model
and Rapaille’s discovery sessions build on tenants of ecological systems theory. Such
tenants also enable researchers go beyond proposing that ‘‘it just feels right’’ as Allen
describes and a national ‘‘culture code’’ explanation of behavior that Rapaille offers.
The analysis in Chapter 14 includes detailed descriptions of the lives of the infor-
mants and how surroundings and upbringing shape their current implicit and explicit
thoughts relating to a go versus no-go outcome toward visiting one destination.
Ecological systems theory and the FLAG model inform these analyses. Following this
introduction, the second section describes unconscious thinking theories and the
development of these theories that is useful for case study research of discretionary
travel behavior. The third section offers details of the method used for examining how
the models may apply to tourists interpreting their own thoughts and behavior.
Method
The following two case studies serve as examples of inductive theory-building that
applies MTP theory. Data collection includes applying [411]McCracken’s (1988) long
interview method in 90-minute face-to-face discussions separately with travel parties
visiting Hawaii’s Big Island (BI) in July 2006. The survey instrument was structured
to provide latitude for interviewers to ask probing or follow-up questions in the event
that unexpected issues or experiences surfaced during the interview process.
The informants include members of two households visiting BI in July 2006. Both
first-time and repeat visitors participated. In the case of first-time visitors, care was
taken to assure informants were interviewed at the end of their visits. Informant
selection was by convenience sampling.
Prospective informants were approached and pre-screened with general questions
about their visit and whether they would be willing to participate in an interview.
Most interviews were conducted at Kailua-Kona (a resort city located on the
west side of BI) in hotels, or at a tourist shopping mall located at the Waikoloa
resort area — 20 miles north of Kailua-Kona. Each informant received 50 USD and
a Hawaii-themed t-shirt for their cooperation.
The questionnaire includes questions asking for: (a) demographic information
about members of the traveling party, (b) pre-trip planning and sources of
information, (c) activities and destinations — both planned and unplanned,
(d) issues surrounding flights (e.g., accommodations), (e) eating and dining
experiences, and (f) overall impressions of the travel experience. The survey includes
open-ended questions with calls for probing and pausing to enable the informants to
elaborate upon their answers and to tell stories completely that relate to their inter-
pretations of events. Informants were told that their compensation was not dependent
on answering all the questions, and they could end the interview at any time.
Nearly all questions were answered by all the informants. Written, thick
descriptions were completed for each informant. Each case study report was read
and revised by the research team. A copy of the complete long interview form is
available by contacting either author.
A German Couple Visits Hawaii (Big Island) for the First Time
This case study of a husband and wife from Bonn, Germany, visiting the State of
Hawaii for three weeks (July 2006) supports the central proposition of the study — a
tipping point (Gladwell, 2002) combination of specific conjunctive relationships
results in triggering or rejecting leisure trips to specific destinations. While the theory
may appear to be obvious, the literature rarely explicitly discusses the central
proposition and rarely describes real-life streams of connecting on–off switches
leading to or preventing leisure trips.
The husband is in his early 60s and works full-time as a professional scientist
(Ph.D. in physics); the wife is in her late 50s and manages their home full-time which
includes three children (17, 18, and 21 years old). While away on the July 2006 to
Hawaii, the couple kept in daily contact by telephone with their children back home
in Bonn. This trip signifies the first time that the couple was confident that all three
children were old-enough for the couple to travel away from home without the
children.
Figure 1a–b summarize details of a 90-minute face-to-face interview with the
couple on their last full-day during their Big Island stay. The following discussion
offers highlights and insights relating to this long interview.
The husband’s big island dream trip The husband reported that he always wanted to
visit the Big Island (BI) because of the volcanoes. He reported learning about Hawaii
and its volcanoes in the general media over many years. His training and work
history in physics likely affected his vigilance toward media reports on the BI over
earlier decades in his life. He reported that his wife was agreeable to visit the
destination, but BI was his dream rather than hers.
Delta frequent flyer points The husband’s accumulation of frequent flyer points in
Delta Airlines loyalty program became an enabling step toward transforming the
dream into reality. The points were accumulated due to business trips over 20+ years.
The couple waited to accumulate enough points for two first-class round-trip tickets.
Leaving the children alone at home The husband and wife were unwilling to travel
away from home as a couple until the children were old enough to manage for
themselves alone and unsupervised at home. The children’s advancing ages became
3. Pre-Framing and Pre-Planning Trip Issues
• Children old enough to care for themselves
• Trip needs to be 3-weeks in length to get over jet lat; 4-weeks is too long away from work
• Regarding the volcanoes in Hawaii, “If not now, when? Older, need to do this trip now.”
• Husband mentions, “I did not have the money to make this trip when I was younger.”
• Delta frequent flyer mileage program enables trip
2. Framing Leisure Choices 5. Choice of

• Husband thinks consciously Destinations 6. Key Activity 8. Situation On-
about dream visit to “South • Trip included 3-night Drivers for BI Site Influences
1. Demographics Seas Paradise” stopover in San Diego Visit • Rough coasts
• Husband and • No specific media with two hours in • See volcanoes • Not many
wife influence; generally aware Tijuana (“My wife breakers
• Teenage of Hawaii has never seen • Did not spend
children living • Report no search for before.”) 7. Key Activities much time on
at home information prior to visit • Three nights on Kauai on BI beaches
• Used Internet for • Four nights on Maui • Talked to children
accommodations and rental • Four nights on BI at home daily
car • Five nights on Oahu; • Saw lava tubes; 9. Consequences
• Delta mileage award from “End in Honolulu so no helicopter ride • “America has a
over 20 years for first class we can get home fast • Snorkeling solution for
tickets if necessary.” • Visit church everything while
• “Volcano alone is the • Hilo day trip traveling (e.g.,
4. External Influences reason to come to the taxis and buses)
• General press news stories about BI.” • “Wife happy
volcanoes in Hawaii • Alternatives: Asia… now (having
• Bought travel guide book Thailand, but cannot visited BI)”
• Studied websites travel alone there
Figure 1a: German husband and wife (Early 60s) visiting Hawaii in July 2006.
Tipping-Point Modeling in Case Study Research
299
300
Attractions
Decision Area Destinations Route/mode to Accommodations Activities in BI Regions visited
and in the BI while in BI BI Visited including
restaurants
Only State of Delta mileage Kona Seaside Hotel in Spent a day at 1-day at Hawaii Volcanoes was
Hawaii but did points saved over central area of Kailua- Hawaii Volcanoes Volcanoes National principal attraction
Consideration mention 20 years; stopover Kona. National Park; a Park; 1-day in sought and visited;
Set and Thailand; all in San Diego; car Did price comparison day road trip to Kailua-Kona; 1- beaches; Hilo;
major islands in rentals—Avis and location shopping Hilo; visited day driving and strolling around
Choices
Hawaii compact (“a good on internet. Mauna Loa Visitor visiting Hilo; 1-day Kailua-Kona; Big
deal”) Center; snorkeling beaches/snorkeling Island Grill
Hawaii volcano; Delta mileage Save money by staying Fulfill main motive Experience a Gain a feel for the
trip idea in mind points enabled trip; at middle to low priced for visiting the BI: variety of activities entire BI. Eating at
Motives unconsciously & car rentals in hotel that is centrally volcanoes as well even though Big Island Grill
consciously for keeping with being located. Location and as visit beaches. volcanoes was was “very good.”
20+ years freely independent price were key prime motive
travelers motives.
Information Low- No substantial Internet used to search No extensive Very limited search Low amount of
involvement search for in and book search about for information search
Search and dominates search formation. accommodations. locations visited indicated.
Use over 20 years prior to visits.
Joint decision by Couple successfully “What we expected for Satisfied that visit Satisfied with Satisfied by did not
husband and mapped out 80 Euros.” was worthwhile experiences and show/display joy
Outcomes wife to use Delta route/modes to go (even though no overall visit to the with visiting
“Make second elevator

FF points for trip to 5 overnight stops work” [broken]. [Need] spectacular lava BI. attractions; a low-
to Hawaii in this one trip. Car coffee maker in room.” flow views). key interpretation
blinkers broken. of their visit
Substantial A combination of Couple avoided paying Couple did not No shopping for Close to being
indirect impact several stops with $200+ per night hotel attempt to fill their clothing and other ironic that couple
Commentary of news stories, low number of room that several days with lots of products by this spent so much time
novels, about activities per stop. nearby hotels charge in activities. Slow couple—fits and effort to get to
Hawaii on Kailua-Kona. pace in doing a European profile. BI but spent little
attracting few activities each effort in studying
visitors. day. BI before visit.
Figure 1b: Summary of first-time visit to the Big Island (Hawaii) by German couple (age: early 60s).
an enabling trigger for their leisure trip. The couple reported planning their trip with
the final four days away being in Honolulu so that they could cut the trip short a few
days if they sensed the children needed their early return home. Each day, the couple
telephoned the children daily during the three-week trip.
Husband agreeing to wife’s request for San Diego stopover clinches trip decision The
wife was reluctant to fly directly to BI. She wanted to visit a place on the U.S.
mainland during the trip. The husband suggested a visit to San Diego — which
included a half-day visit to Tijuana, Mexico. They both reported enjoyed visiting
the old town area of San Diego. The San Diego stopover also served the couple by
partially overcoming jet lag before visiting Hawaii.
Visiting multiple islands in hawaii Even though coming to the Big Island and seeing
volcano related sights was a key driver for the trip, the couple included overnight
stays to each of the four major islands in the State of Hawaii. This finding supports
the proposition that while the BI may be a stand-alone destination, multiple-island
locations/sights are likely to influence first-time visitors in planning their visits to
the State of Hawaii. Consequently, marketing BI as a stand-alone destination may
be less effective than promoting the complementarily fulfilling benefits that visits to
several Hawaii Islands provide.
Figure 1c represents an ethnographic decision tree (EDT) model (Gladwell, 1989)
of the questions and alternative paths the couple thought about taking that resulted
in their decision and trip to the BI. EDT includes mapping the go/no-go conditions
that informants relate the occur in a conjunctive stream that lead to the occurrence
of a specific outcomes — the assumption of EDT is that no one condition with a
conjunctive path is sufficient to cause a specific outcome. EDT modeling across
multiple cases is a method analogous to Montgomery’s (1975) gatekeeper analysis.
An EDT model is built using the emic (informant) description of self-questioning and
answering in reflecting on how their lived experiences came about to include a specific
outcome (e.g., current trip to the BI).
Four levels of four factors combine to allow for a positive outcome for the 2006
trip to the BI. A conjunctural causal equation of the combination of conditions
affecting the positive outcome (Fig. 1d):
S ¼ Success ¼ Trip to the BI

A ¼ Dream
B ¼ Children home alone is acceptable
C ¼ Airline frequent flyer mileage account permits two free tickets to the BI
D ¼ Couple’s negotiation over trip details successful.
S ¼ ðA B C DÞ (2)
Eq. (2) implies that the trip would not occur in the absence of any one of the four
antecedents. For example,
F ¼ Failure ¼ ða B C DÞ (3)
302
3b. Awareness of nearby yet uniqueness

of Oahu, Maui, Kauai, and BI
2. Low-involvement
awareness of news Yes 4. Are children old enough
reports, novels, TV 3a. Dream of visiting to be left alone?
programs, and magazine No BI?
articles describing
Hawaii No
Yes
5. Does Delta mileage account

No have enough points for two
first-class tickets to Hawaii?
1. Husband is a trained Ph.D.
physicist with interests in
natural forces such as volcanoes Yes
No 6. Will the wife buy into

transforming the dream

into reality?
Yes
No 7. Can the Delta mileage
8. Trip to Hawaii by German tickets include a stopover
Yes
couple with a stopover in in San Diego to make the
San Diego wife happy?
Figure 1c: (Un)conscious decision process of German husband and wife (early 60s) visiting Hawaii in July 2006.
Notes.
S = Tipping-point success
(i.e., book trip to Big Island)
B. Children old enough S = (A· B ·C ·D)
to be home alone
C. Delta mileage awards

account has
A. Husband’s dream of S
accumulated enough
Big Island visit points for two first
class tickets
D. Wife buys into idea of

trip after negotiating
stopover in San Diego
Figure 1d: (Un)conscious feature-level intersection for German husband and wife
(early 60s) visiting Hawaii in July 2006.
F ¼ ðA B C dÞ (4)
Using Boolean algebra is a step toward generalizing across cases as well as

achieving parsimony in reducing the presence of specific antecedents when predicting
whether or not a given positive outcome can occur without a given antecedent. For
example, if visits by older couples to the BI can occur with or without successful
husband–wife trip-detail negotiations, then factor D can be eliminated from the
equation — if the aim only is to predict whether or not a visit can occur for such a trip
party. The same proposition applies for the other antecedents (i.e., A, B, and C) as
well. Thus, the argument is not that only four antecedents are necessary for all travel
parties for leisure travel to the BI; instead, the occurrence of all four antecedents in
combination results do result in a trip to the BI (go versus no-go outcome).
Using this grounded theory (Glaser & Strauss, 1967) approach of inductive model
building likely will result in a few stories (e.g., o10) that are highly similar for the
majority of visitors to a specific destination. ‘‘Story’’ is the narrative presentation of
a conjunctural causation equation of three or more antecedents. Storytelling by
an informant often results in thick descriptions that include relevant surfacing of
unconscious thinking and conscious elaborations (see Rapaille, 2006; Shank, 1990;
Schank & Abelson, 1977; Sood, Woodside, & Miller, 2007).
The conjunctural method serves to focus attention on explicating the fits-like-a-
glove proposition that several pieces (i.e., factor levels) fit together to permit a ‘‘yes’’
or ‘‘no’’ choice favoring BI visit. This explication includes the husband consciously
recognizing a long-time unconsciously held dream to visit a volcanic island in the
South Pacific — never mind that Hawaii is not in the South Pacific. The realization
that his children had become older teenagers and his wife’s willingness to negotiation
conditions for such a trip triggered plans to transform the dream into reality. MTP
research and theory describes and focuses attention on the combination of factor
levels resulting in the positive outcome that builds from conjunctive path analysis,
ecological systems, and FLAG modeling.
An Older American Couple Visits the Big Island for the First Time
Figure 2a–c present the analysis of a second case study — a long interview with an
older American couple on their first visit to the BI. This couple lives in Long Island,
New York. Their long interview data include information on one prior group-tour
visit to the State of Hawaii that included three islands — but not the BI. Their son’s
decade-long conversations about his two visits to the BI created a high level of
tension in this couple’s minds — they needed to get back to Hawaii to experience the
son’s reported BI experiences that they missed on their prior trip.
Conversations with their son and a previous visit to neighboring islands motivated
the couple to conclude, initially unconsciously and then consciously, to return to
Hawaii — if the conditions were right for such a visit. Children at home preventing
(factor level B) or not preventing a trip (factor level b) was not relevant in this case.
The ability to trade their time-share unit for a time-share unit on the BI was relevant
as a conjunctural causal factor — a necessary but not sufficient condition in enabling
the visit (see Figure 2c).
Like the German couple, this American couple reported a frequent-flyer airline
mileage award as a requirement — an enabling contingency — for the trip to the BI.
Similar to the German couple, the American couple negotiated between themselves
to fulfill additional enabling requirement — with the wife offering to buy the
husband’s round-trip flight tickets (Fig. 2d).
At a micro level these two couples are unique in their plans and behaviors
regarding their visits to Hawaii; however, the couples’ unconscious and conscious
thinking display substantial similarities at a mid-range level of theory. One or both
members of each couple became cognizant of their desire to visit the BI after several
years of hearing about the BI as a destination. Both couples set special trip financing
conditions as enabling conditions for their BI trips (e.g., free airline ticket(s) and a
time-share trade). Both couples referred to their children in their thoughts regarding
a trip to the BI. In both case, reported negotiations between the husband and wife
enabled the trip.
Note further in reading the summary reports in Figures 1a–b and 2a–b that both
couples downplay the use of external information sources both to influence their
decision to visit the BI and to affect their activity plans during their BI visits. Both
couples did report searching online for accommodations; however, the Internet was
3. Pre-Framing and Pre-Planning Trip Issues
• Husband and wife use time-share clearinghouse to trade their one-week unit with someone else
• Wife achieved required frequent flier mileage with American Airlines; she has money to buy air
tick for husband
• Use AA credit card to “build miles”
• “We did very little research before the trip—we watched the travel channel on TV.”
• No use of travel guide books prior to trip; received a folder of activity information in hotel lobby
• Booked accommodations (Hotel.com) and car rental online; adamant about not using a travel agent;
separately mentioned that the daughter-in-law is a travel agent
6. Key Activity Drivers

for BI Visit
1. Demographics 2. Framing Leisure • Volcano 8. Situation On-
• Husband and Choices • Black and green sand site Influences
wife • Alternative destination 5. Choice of • Visit all
Destinations beaches
• Both in late 50s considered: Great Lakes extreme points
• Own one-week area in Wisconsin • Time-share on BI for
time-share unit • Vague idea about trade for BI 7. Key Activities on BI sightseeing
• Live on Long visiting the Caribbean visit became • Stayed in Hilo; close to
Island, NY someday available Hawaii Volcanoes
9. Consequences
• Married son in • Highest preference was a National Park (HVNP)
• “We now
his 20s visit to BI • Walked in HVNP
know that Hilo
• No organized activities
is the place to
• Snorkeling tour
4. External Influences stay”
• Hapuna Beach
• Son and his wife traveled to BI for • “We now
• Driving around BI
honeymoon and 10th wedding anniversary know it’s best
• Full-size Pontiac rented
• Son talked to parents about BI’sblack and not to rent a
from Alamo
green sand beaches every year for the last car; rent a jeep
• Hotel “nice-and-close
ten years to be able to
to airport”
• Three years ago visited Hawaii, but not BI travel on lava
roads.”
Figure 2a: Husband and wife in late 50s from Long Island visiting BI for one week
305
306
Attractions
Decision Area Destinations Route/mode to Accommodations Activities in BI Regions visited
and in the BI while in BI BI Visited including
restaurants
Consideration Big Island; American Airlines Hotel near airport; low- Volcano; Green • Hawaii Parker Ranch;
Great Lakes; only due to frequent price important sand beach; Black Volcanoes Kalahikiola
Set and sand beach National Park;
Caribbean flier mileage award Church; Hapuna
Choices Beaches; Tour of Beach; Hawaii
all BI regions by Volcanoes National
car Park
“Pleasure” Use AA mileage Little driving from Follow son’s Achieve son’s Sightseeing all
Make time- points—avoid airport to hotel recommendations; recommendations regions important
Motives spending money for influenced choice activity for couple
share trade Put to rest belief Fulfill
successfully 1 of 2 air tickets; that they missed beliefs/feelings of
Cheapest full-size much by not missing something
car rental visiting the BI
Information Low- Searched for AA Searched on-line for Pride themselves in Word-of-mouth Minimizing search
involvement; trips to BI online; accommodations using doing minimal reports by son over effort and use of
Search and Hotel.com search and use of ten years prior to information
Listen to son On-line search for
Use talk about BI car rental information couple’s visit
Successful Had to pay for food “Not bad” Perceive Satisfied with A visit of marginal
time-share trade on plane; $5 for a “An older hotel” themselves to be visit; No report of value—couple was
Outcomes to BI; One- sandwich; “we staying in wrong wanting to return; pleased but not
“Had old brass key for
week BI visit brought our own room” area of BI (not Many activities passionate in
food and drank free Hilo); Wrong not done—due to reporting on their
sodas”; Unhappy Had swimming pool; vehicle rented costs visit
with car Hotel has time-share
units
Conversations Couple reported “We made a mistake; Couple counts on Minimizing Couple displays
with son and having good seating Hilo is the place to stay— on-site experiences, expenses during low involvement in
Commentary H&W prior visit on the trip but flight closer to Hawaii not seeking the visit and not search and use of
to State sans BI to LA was 1 hour Volcanoes National information, to do using professional information;
visit were major late leaving NY and Park” things better if help were Trip outcomes
influences same for LA Couple stayed on Kona there ever is a next important factors were good—couple
side of the BI time. reports
Figure 2b: Summary of first-time one-week BI visit by husband and wife in late 50s from Long Island.
4. Can we make a one-week Yes 5. Does wife have
3. Couple consciously senses
time-share trade to enable a sufficient mileage
that they are missing
BI visit? points for a trip to BI?
something by not visiting BI
No No Yes
6. Is wife willing to
7. Consider trip to
2. Couple takes 11-day tour to No buy an air ticket for
Great Lakes
Hawaii three years ago that does the husband to
(Wisconsin)
not include visit BI enable BI visit?
Yes
9. Identify factors
1. Son talks about his honeymoon and that ar sufficient and 8. Book flight for
10th wedding anniversary visits to BI necessary conjuctively one-week BI
with his parents for a trip to the visit
Great Lakes
Figure 2c: (Un)conscious decision process of couple in late 50s from Long Island visiting the BI.
307
Notes.
S = Tipping-point success
(i.e., book trip to Big Island)
E. Time-share unit can be
S = (A· E ·C ·D)
traded for unit on BI
A. Husband and wife C. Wife’s American

dream of BI visit (son’s Airlines mileage
S
influence over ten account permits one
round trip to Hawaii
years)
award
D. Wife willing and

does buy air ticket
for husband
Figure 2d: (Un)conscious feature-level intersection for couple in late 50s from Long
Island visiting the BI.
not used to learn about local BI destinations, activities, or special events. The low
level of search and external information use is signified by a lower case ‘‘i’’ in the
summary equation that covers both couples visit to the BI:
S ¼ ðA B C D iÞ þ ðA E C D iÞ (5)
Future research would be valuable for learning whether or not high versus low
levels of information search/use affect the presence/absence of factors A–E as
antecedents for visits to the BI. The likelihood is that information does not affect
the equations (the go versus no-go conjunctural causal paths) but does affect the
expenditures, length-of-stay, and number of activities visitors experienced; see
Woodside, MacDonald, and Trappey (1997) for evidence supporting this proposi-
tion. If examinations of several cases both high and low in search/use of external
information further confirm the proposition, then the go and no-go models can
exclude information as a contingency factor — at least among older couples from
visiting the BI from developed nations.
Destination-Marketing Strategy Implications
Several implications for effectively marketing visits to the Big Island follow from
analyzing this case-study long interview report. For example, rather than using only
specific campaign advertising and promotions programs, general media reports have
a long-term influence on trips such as the one taken by the couple from Germany.
Having a continual presence in news media and popular press, movies, short
stories, and novels has a long-time future impact on trips to a leisure destination.
Consequently, such case studies support the proposition that public relations
programs encouraging such story-reporting are effective.
Second, the findings from this case study support the effectiveness of destination-
marketing special offers made in cooperation with airline frequent-flyer programs.
The key is that some specific tipping point must be reached involving one or more
factors — such as an accumulation of sufficient frequent flyer points in combination
with a dream to visit the destination — to transform the travel dreams into realities.
Third, while visitors to Hawaii from some countries are known to shop for
clothing and other products for themselves and to give to others (e.g., Rosenbaum &
Spears, 2006), not all visitors (and likely many Europeans) are motivated to visit
Hawaii for nature-related experiences and not to shop. Promoting shopping may be a
turn-off for such visitors.
Fourth, the Big Island is unlikely to be a stand-alone destination magnet for
first-time visitors to the State of Hawaii. Cooperative marketing programs across
several islands and state-wide information brochures likely will be more effective in
influencing first-time visits.
Fifth, the tipping point proposition receives strong support in case study research,
such as the long interviews of the German couple and the American couple. Not only
does a specific level of one factor need to be achieved (e.g., accumulation of many
frequent flyer points), but specific outcomes of several factors must occur at the
same instance (e.g., children old enough to be home alone and unsupervised, the
accumulation of the necessary frequent flyer miles, and a successful husband–wife
negotiation of a stopover visit to a secondary destination). The storytelling episodes
with their son appear to be a necessary but not sufficient condition for the American
couple in finally triggering a visit to the BI. Such case study data provides clues for
destination advertising strategy on the importance of presenting storytelling dramas
in commercial communications versus only on destination sights and experiences.
Are the stars in alignment, all the ducks in a row, and the old saws expressing such
tipping point conjunctions? Advertising using storytelling scenarios may effectively
express such tipping point scenarios that case study research uncovers — ‘‘The kids
can manage without us, we have the mileage points, we can stop-off in San Diego
and Mexico, and we’re still young enough to enjoy it! If not now, when?’’
Conclusions
MTP theory is valuable in response to recognizing that consumers have limited ability
and willingness to explain their own behavior. MTP theoretical tenants include
a combination of the following points. (1) Most thinking occurs unconsciously.
(2) Direct questioning of informants to learn the reasons and perceived causes
(see Malle, 1999, 2004) of their own behavior is unlikely to explicate unconscious
thinking (see Zaltman, 2003). (3) Face-to-face methods that go beyond direct
questioning are necessary for uncovering unconscious thinking — such methods
include applying McCracken’s (1988) long interview method and Rapaille’s (2006)
discovery sessions. (4) Combinations of antecedent conditions are necessary and
sufficient for explaining and predicting behavior — the issue is not the relative
importance of independent variables but rather on identifying alternative conjunc-
tural causal paths leading to go and no-go outcomes (see Ragin, 1987). (5) The
seemingly opposing goals of capturing/reporting complexity and achieving generality
are achievable by collecting thick descriptions of several cases in each of several
theoretically interesting combinations of antecedent factors. (6) All cells in full
factorial design of four-or-five binary factors likely represent a limited number to
many cases assuming that a study includes collecting a large number of cases
(nW100); thus, a researcher should seek to empirically examine seemingly outlier
factor-level combinations — such combinations provide data for theoretically rich
information. (7) QCA tools are useful for building and generalizing MTP models
relevant for specific topics including describing and explaining discretionary
destination behavior.
During the long interviews, the informants were initially unable to respond to the
question as to what triggered this particular visit to the Big Island. The informants
initiated the telling of stories leading to their present visits rather than reporting a
conscious choice process or plus and minus judgments of attributes or experiences.
The stories in their reports describe life-changing events (children moving away;
reaching a travel award level; a move signaling the occurrence of job retirement)
leading to an epiphany from unconscious to conscious knowing — a sudden
realization that this was the year and Hawaii was the place.
Limitations and Suggestions for Future Research
This chapter does not describe the several alternative conjunctural causal paths
relevant for trips to the BI. The study does not include descriptions of household
members who are aware of the Big Island as a destination but never unconsciously
or consciously thought about discretionary travel to the BI. Certainly, different
additional paths exist among older American and German couples visiting the BI
for the first time. Future research should consider suggesting alternative scenarios
representing consumption systems not occurring for a travel party and asking
travelers how representative or realistic such scenarios might be for the informants’
travel party. The objective is to stimulate thinking, use, and justification of building
theory and applying alternative data collection methods via case study research.
Future research would be useful focusing on conscious and unconscious constraints
that combine to restrain individuals aware of destination that they consider as
a possible place to visit but rarely or never actually visiting. Research on nontravelers
having the discretionary funds and time to travel is sometimes a topic of major
importance for government destination organizations and firms seeking to stimulate

domestic and international travel.
Future applications of MTP theory and QCA tools in travel and consumer
research likely will achieve substantial success in reporting the complexity found in
case study research and the generality necessary for applied research. Such research
studies should uncover the alternative hidden (seemingly unimportant) factor levels
in combinations of causal paths leading to outcomes that the destination-brand
executive seeks or seeks to avoid.
Appendix. 2006 Hawaii Visitor Self-Report Study
Part 1. Type of Trip that You are Now Taking
1. How would you classify the trip that brings you to Hawaii? Is this a pure pleasure
trip, partly, or entirely a trip related to work, does the trip include attending
a conference? [Record verbatim response.] If a conference, what is the name of the
conference that you attending?
2. Did you start this current trip that brings you to Hawaii from you home in another
State or in some other country other than the U.S.? Please name the city, state,
and country where this current trip that you are on began.
3. Please describe the members of your immediate travel party for this trip. For
example, are you traveling with family members? If yes, what are the relationships
within the travel party?
4. Is your immediate travel party part of a larger group? For example, are you
visiting Hawaii on a group tour? If yes, please name/describe the group?
5. What are some of things that happened a few years, months, or weeks ago that
brought about this visit that includes you coming to Hawaii? [Use prompts: Please
provide details.]
Seek more details: Prompt: How did it come about that you are on this current trip
in Hawaii?
1. Is this your first visit to the State of Hawaii or have you been to Hawaii before
this current trip?
2. Has any other member of your immediate travel party been to Hawaii before this
current visit? If yes, please name the persons and describe the prior visits that
have been completed.
3. Are you visiting one or more of the islands here in the State of Hawaii on this
current visit or just the Big Island of Hawaii? If more than the Big Island, what
other islands in Hawaii are you visiting on this current trip and how many nights
are you staying on each island?
4. What made the difference in making this current trip to the State of Hawaii come
about versus not making the trip or doing some other activity or just staying at
home? Please discuss in some detail.
5. Are you visiting other destinations outside of Hawaii on this current trip?
If yes, please name these other destinations and tell how many nights you have
spent or will spend at each destination outside of Hawaii.
6. Please describe the things that happened and thoughts you may have had during the
years, months, or weeks before the trip about how many nights that was scheduled
for this current visit in Hawaii and the total number of nights away from home.
7. What made the difference in the number of nights, weeks, or months for this visit
to Hawaii for you?
8. Is this visit to the Big Island a minor or major part of your visit to the State of
Hawaii? Please describe how the Big Island came to be included in this current
trip to the Big Island.
9. Is there anything in particular here on the Big Island that brings you here on this
current trip? If yes, please describe. [Prompt. Anything else?]
10. Please describe the things that you are doing here today on the Big Island. What
have you done this morning and what will you be doing today for the rest of day?
[Prompt. Anything else?]
11. Please describe the things you and members of your immediate travel party have
done before today here in the State of Hawaii on this current visit. [Prompt.
Anything else?]
12. Please describe the things that you will be doing tomorrow and for the rest of
your visit to the State of Hawaii on this current visit. [Prompt. Anything else?]
Part 2. Flights, Accommodations, and Ground Transportation
1. What steps/events/thoughts occurred that relate to you getting flight tickets for
this visit to the State of Hawaii? For example, did you use frequent flyer miles,
did you reserve tickets on-line, or visit a travel agent? Please describe the steps.
2. Was paying for the flights paid for as a business expense or was did you pay as a
personal expense?
3. Was the airline company that you flew on something you thought about before
getting tickets to visit the State of Hawaii or the Big Island? If yes or if no, please
describe your thoughts about different airlines relating to the current visit to
Hawaii.
4. Please describe your actual flights to the State of Hawaii and to the Big Island.
How would you describe the flights? Prompt: Did anything surprising/somewhat
unusual happen during the flights or between flights?
5. If you could change something about the flights to Hawaii or the Big Island, what
would you change?
6. What steps/events/thoughts occurred that relate to you getting accommodations
for this visit to the State of Hawaii? For example, did you reserve
accommodations on-line, telephone for a place to stay, or visit a travel agent?
Please describe the steps.
7. Please name and describe the accommodations that you are using here on the Big
Island.
8. Was paying for accommodations on the Big Island paid for here as a business
expense or was did you pay as a personal expense?
9. Were the accommodations that you are using here on the Big Island something
you thought about before the visit the State of Hawaii or the Big Island? If yes or
if no, please describe your thoughts you had before this trip about different
accommodations relating to the current visit to Hawaii. Prompt: Did anything
surprising/somewhat unusual happen during your use of accommodations here
on the Big Island?
10. If you could change something about the accommodations that you are using in
Hawaii or the Big Island, what would you change?
11. What steps/events/thoughts occurred that relate to you getting ground transporta-
tion (such as taking bus tours, renting a car, taking taxicabs, for this visit to the
State of Hawaii? For example, did you reserve ground transportation on-line,
telephone, or visit a travel agent? Please describe the steps.
12. Please name and describe the ground transportations that you are using here on the
Big Island.
13. Was paying for ground transportation paid here on the Big Island as a business
expense or did you pay as a personal expense?
14. Were the ground transportations that you are using here on the Big Island
something you thought about before the visit the State of Hawaii or the Big
Island? If yes or if no, please describe your thoughts you had before this trip about
different ground transportations relating to the current visit to Hawaii.
Prompt: Did anything surprising/somewhat unusual happen during your use of

ground transportation here on the Big Island?
Prompt: If you could change something about the accommodations that you are
using in Hawaii or the Big Island, what would you change?
Part 3. Sources of Information about Hawaii and the Big Island
1. Before this visit to Hawaii, did you talk with friends, travel agents, family
members, and/or local Hawaii persons about this current trip to Hawaii?
2. If yes, who did you talk with and what were the topics of the conversations?
3. What information did you learn and/or was useful before visiting Hawaii? Please
describe.
4. Before coming to Hawaii, what comments were made to you by friends or family
members about this visit, if any?
5. Did you and/or any one else in your immediate travel party get a travel book/
guide in the months or weeks before this visit to Hawaii and the Big Island?
Prompt: If yes, what travel/guide books did you get?

Prompt: Was any particular travel/guide book useful for this current trip to
Hawaii? If yes, how was the information helpful?
1. Did you talk to a travel agent or visit travel agents’ in person on-line websites
before starting this visit to Hawaii?
2. If yes, what information did you look at? Please name the travel agents that you
visited on-line or in person. Were the visits useful?
Part 4. Eating out and Dinning Experiences
1. Please describe some of your eating out and dinning experiences during your visit
to the Big Island.
2. Please name some of the restaurants and fast food places you have eaten at here
on the Big Island.
3. Please describe any unusual or surprising events relating to eating out or dinning
here on the Big Island.
Part 5. Places Visited and Doing Activities on the Big Island
Please circle the following places here on the Big Island that during this current visit
you have visited or definitely will visit:
1 Cloud Forrest-Kaloko Drive

2 Kailua-Kona
3 Makalei Golf Course
4 Parker Ranch
5 Waikoloa Village
6 Holoholoku
7 Kohala Mountain Road
8 North Shore of Kohala
9 Kapa’au
10 Kalahikiola Church
11 Kauhola Point Lighthouse
12 Pololu Valley
13 Pololu Beach
14 Mo’okini Heiau
15 Kapa’a Beach Park
16 Mahukona
17 Lapakahi State Historical Park
18 Kawaihae
19 Kohala resort area
20 Hapuna Beach
21 Mauna Lani Bay Hotel
22 Kohala Lava Desert Area
23 Ke-awa-iki
24 Kiholo Bay
25 Makalawena Beach
26 National Energy Laboratory
27 Wawaloli Beach
28 Kaloko-Honokohau National Historical Park
29 Honokohau Harbor
30 The Gallery of Bamboo
31 Tropical Dreams
32 Shady’s
33 Kohala Book Shop
34 Nanbu Courtyard
35 Ackerman Galleries
36 Hilton Waikoloa
37 Mauna Lani
38 Shops at Mauna Lani
39 Kings’ Shops
40 Louis Vutton
41 Ukulele House
42 Ahu’ena Heiau
43 Kamakahonu Beach
44 Hulihe’e Palace
45 Palace Gift Shop
46 Pohaku Likanaka
47 Mokuaikaua Church
48 Kona Inn
49 Huggo’s Bubba Gump
50 Hard Rock Café
51 Cloud Forrest up Kaloko Drive
52 Kailua-Kona
53 Makalei Gof Course
54 Parker Ranch
55 Waikoloa Village
56 Holoholoku
57 Kohala Mountain Road
58 North Shore of Kohala
59 Kapa’au
60 Kalahikiola Church
61 Kauhola Point Lighthouse
62 Pololu Valley
63 Pololu Beach
64 Mo’okini Heiau
65 Kapa’a Beach Park
66 Mahukona
67 Lapakahi Historical Park
68 Kawaihae
69 Kohala resort area

70 Hapuna Beach
71 Mauna Lani Bay Hotel
72 Kohala Lava Desert Area
73 Ke-awa-iki
74 Kiholo Bay
75 Makalawena Beach
76 National Energy Laboratory
77 Wawaloli Beach
78 Kaloko-Honokohau National Historical Park
79 Honokohau Harbor
80 The Gallery of Bamboo
81 Tropical Dreams
82 Shady’s
83 Kohala Book Shop
84 Nanbu Courtyard
85 Ackerman Galleries
86 Hilton Waikoloa
87 Mauna Lani
88 Shops at Mauna Lani
89 Kings’ Shops
90 Louis Vutton
91 Ukulele House
92 Ahu’ena Heiau
93 Kamakahonu Beach
94 Hulihe’e Palace
95 Palace Gift Shop
96 Pohaku Likanaka
97 Mokuaikaua Church
98 Kona Inn
99 Huggo’s Bubba Gump
100 Hard Rock Café
101 Hang Loos
102 Pahoehoe Beach Park
103 White Sands Beach
104 St. Peter’s Catholic Church
105 Little Blue Church
106 Ku’emanu Heiau
107 Kahalu’u Beach Park
108 Ke’eku Heiau
109 Keauhou Shopping Center
110 Holualoa
111 Greenwell
112 Bayview
113 Kona Historical Museum
114 Kealakekua Bay

115 Place of Refuge
116 Napo’opo’o Beach
117 Captain Cook Monument
118 Kings’ Trail Rides
119 Ke’ei Beach
120 Moku-a-Kae Bay
121 South Point
122 Green Sand Beach
123 Whittington Park
124 Punalu’u Black Sand Beach
125 Volcanoes National Park
126 Kilauea Volcano
127 Chain of Craters Road
128 Hiking: _______
129 Volcano Village
130 Hilo
131 Glenwood
132 Pahoa
133 Kapoho
134 Kalapana
135 Lava Tree State Park
136 Kea’au
137 Cape Kumukahi
138 Pe’epe’e Falls
139 Kaumana Cave
140 Prince Kuhio Plaza
141 Nani Mau Gardens
142 Pana’ewa Rain Forest
143 Hawai’i Tropical Botanical Gardens
144 Mauna Loa Visitor Center
145 Stainback Highway
146 Puna
147 Ahalanui
148 Isaac Hale Beach Park
149 Honoli’i Beach Park
150 Akaka Falls
151 Umauma Falls
152 Nanue Stream
153 Laupahoehoe Train Museum
154 Laupahoehoe Point
155 Honoka’a
156 Waipi’o Valley
157 Old Mamalahoa Highway
158 Waiamea
159 Hamakua & Waiamea Shopping

160 Saddle Road
161 Mauma Kea State Park
162 Kaumana Cave
163 University of Hawai’i Telescope
164 Lake Wai-au
165 Keanakako’i
166 Mauna Loa
167 Kekaha Kai State Park
168 Ka’upulehu Beach
169 Kuki’o Beach
170 Manini’owali/Kua Bay
171 Makalawena
172 Makole’a Beach
173 Wawaloli Beach
174 Old Kona Airport Beach Park
175 White Sands Beach Park
176 Kailua Bay
177 Kahalu’u Beach Park
178 Pahoehoe Beach Park
179 Napo’opo’o Beach
180 Ke’ei Beach
181 Pu’uhonua o Honaunau
182 Pebble Beach
183 Ho’okena Beach Park
184 Miloli’i Beach Park
185 Honomalina Bay
186 Biking
187 Camping: ______
188 Fishing: ______
189 Golfing: ______
190 Helicopters: _______
191 Mauna Loa Summit
192 Punahaha
193 Kayaking
194 Scuba
195 Parasailing
196 Snorkeling
197 Snuba
198 Whale watching
199 Stargazing
200 Tennis
201 Submarine
202 Surfing
203 Kit surfing
204 Piloting your own boat:_________

205 Other: ___________
Part 6. What Do You Think Will be Your Memories of Your Visit to the Big Island
after This Trip Is Over? Please Describe in Some Detail:
Part 7. If you had the chance to not make this trip to the Big Island and spend your
time and money on something else, assuming that you would do so, what would you
spend your time and money on? Please describe in some detail.
Part 8. Which one would you do if you had a choice between this visit and the other
activity? Please comment with some details.
Part 9. Gifts
A. Please describe gifts that you bought for yourself during this visit to the Big
Island.
What are the specific purchases that you made for yourself ?
How much did you pay for each gift?
Did you plan to buy each of these gifts before your arrival or after your arrival?
B. Please describe gifts that you bought for other persons traveling with you on
this trip.
Did you buy any such gifts?

If yes, what did you buy and for whom is each gift for?
C. Please describe gifts that you bought to take home for family members or
friends.
Did you buy any such gifts?

If yes, what did you buy and for whom is each gift for?
Thank you for your cooperation. [Make out check payable to informant for $50.00;
give check to informant.]
Chapter 15
Participant Observation Research

in Organizational Behavior
Synopsis
At its best, participation observation (PO) includes the researcher living inside a
formal or informal organization long enough to actually observe first-hand
how the organization makes sense of its environment, frames problems and
opportunities, plans and performs actions, evaluates outcomes, rewards and
punishes its members, and celebrates and commiserates sacred, climatic, and/or
exceptional events. The core feature of PO is being there — the researcher’s
presence in the same context as participants as events happen and not relying
mostly on participants retrospections about what happened and the causes and
consequences of what happened. In some studies PO data collection occurs
unobtrusively — the researcher does not inform the organizations’ participants
that she is conducting a study of their thinking and behavior — for example, in
The Tearoom Trade (Humphreys, 1970) the researcher becomes a ‘‘watch
queen’’ (lookout watching for police) in a men’s room in park while others
engage in homosexual acts; in The Informant (Eichenwald, 2000) an executive
in an international manufacturing firm becomes an undercover researcher (with
hidden cameras and listening devices) to collect data showing his colleagues
planning and doing illegal price-fixing deals with executives in other firms.
In most studies PO data collection is obtrusive with the organizations’ members
knowing that a researcher is present for the purposes of observing, describing,
and explaining what is occurring in the organization — for example, in Coming
of Age in Samoa (Mead, 1943) the American researcher lived among the natives
in the south Pacific island to describe rituals relating to the transformation of
child to adult; in The Used Car Game (Browne, 1976) the researcher directly
observed interactions of salesmen, customers, and sales managers for seven-to-
ten hours a day for 15 months with all participants knowing that the researcher
was ‘‘being there’’ to collect data to describe and understand their thinking
and behavior. The intent for this chapter is not to present a full review of the
PO literature; the focus here is to illustrate an obtrusive PO study in a formal
organizational context in-depth. The main goals include (1) illustrating
doing PO and (2) describing the value of PO research. This chapter serves
to introduce the reader to relevant organizational PO literature and provides
details of applying participant observation to the study of organizational
behavior. The study applies an ethnographic approach to develop flow
diagrams of the information processes and decision-making stages of corporate

and plant executives in developing corporate purchasing agreements with
suppliers. Participant observations of the processes to develop corporate
purchasing agreements were conducted along with extensive personal inter-
views of plant buyers at seven plant locations of Epsilon Corporation — a
multinational electronics firm with headquarter offices in New York City.
The results indicate that valid and useful descriptions are possible of the
information processes and decisions actually used to produce corporate
purchasing agreements. Several diagnostic comments are provided to each of
the four phases in the processes used to develop corporate purchasing
agreements. A template for applying participant observation methods in case
study research concludes the chapter.
Introduction
Procedures for descriptive analysis of information processes and decision making

have been developed and applied at the organizational level. The resulting descriptive
models are useful for gaining insights into the functions of business firms and as
descriptive tools to find and solve problems. Few studies of management, marketing,
finance, and purchasing decision processes have been published despite the advocacy
for such research (Howard & Morgenroth, 1968). However, descriptive analysis
of organizational decisions has not been ignored. Wind (1966) was a pioneer in
describing the purchasing processes of component parts in firms; Mintzberg,
Raisinghani, and Theoret’s (1976) classic description of decision processes focuses on
firms’ purchasing processes. These study findings support the following four general
propositions.
1. Executives use local rationality (rather than general rubrics) to resolve organiza-
tional conflicts involving goals (i.e., delegation and specialization in decisions
and goals serve to simplify decisions and goals that really involve complex sets of
inter-related problems and conflicting goals).
2. Persons involved in the buying process (i.e., an informal or semi-formal ‘‘buying
center’’) attempt to avoid uncertainty in decisions. This avoidance of uncertainty
is achieved by using decision rules that emphasize short-run reaction to short-run
feedback and by arranging a negotiated environment.
3. Any search carried out by the members of a buying center is stimulated by a
problem and is directed toward finding a solution to the problem. Search is
motivated, biased, and follows the simplest route (search to find ‘‘black swans’’ or
negative cases that refute views occurs rarely).
4. The buying center changes its goals; shifts its attention, and revises its procedures
for search as a function of its experience.
Participant Observation Research in Organizational Behavior 323
Recent research supports and refines these propositions. For example, Na,
Marshall, and Woodside (2009) extend the four propositions with five additional
propositions through their participant observation studies of decisions–out-
comes–evaluations among advertising agencies and their clients.
5. The organizational structure (team or functional system) influences the structure
of decision-making groups enacting decision processes. This proposition is very
straightforward. For example, in the team system an account supervisor would
usually be in charge of the group. However, in a departmental structure system,
it is more usually a copy director or an art director who would be in charge of the
group. Although we would expect the decision roles to be constant no matter what
the structure (i.e., influencers, buyers, gatekeepers, and deciders), the functional
roles involved will almost certainly differ.
6. The types of advertising decision (creative, media, promotion, and campaign)
affect the structure of decision-making groups. This, again, makes intuitive good
sense. It is to be expected that different types of advertising service require
different types of expert for that particular problem-solving situation. Thus, the
structure of decision-making groups will vary across each of these decision-type
situations.
7. The decision situation (complexity, novelty, or importance) influences the decision
process and the size of a decision-making group. The buying situation determines
the size, role structure, and decision process within group decisions. Clearly,
factors such as intangibility, unfamiliarity, complexity and simply the costs of
getting the decisions wrong will all have a structural affect upon the decision-
group composition.
8. Conflicts are rare and are resolved in ways perceived to be rational. Advertising is
simply one kind of professional service, and a professional association provides
support and a code of conduct in almost every market, as for other professional
services. Professional services require a great degree of confidentiality and trust on
the part of the purchaser, who lack the detailed information and skills of the
professional being hired.
9. The function, types of decisions, and the decision situation (complexity, novelty,
and importance) affect the structuring criteria and decision rules adopted
(for evidence, see Bettman & Park, 1980; Vyas & Woodside, 1984). The
proposition has face validity as well, in that very simple decisions only involve
a small number of people, for instance, whereas a larger group of managers are
more likely to share more risky decision situations. As each role-player brings
her/his own decision criteria to the table, then the nature of the risks involved in
the particular decision process are reflected in the decisional criteria.
Na et al.’s (2009) full report shows that real-life decision processes often depart
dramatically from textbook explanations. For example, similar to all other
descriptive studies, the findings consistently reject the proposition that decision
makers weight antecedents by their importance and make choices by summing values
representing strengths and weaknesses for each option before them. ‘‘It appears as
if the participants used non-compensatory decision rules in every situation; at least,
we could not identify a situation where they applied compensatory rules. Decisional
evaluative criteria varied dramatically among individuals, who tend to carry their
criteria with them as they participate in the decisions’’ (Na et al., 2009, p. 165).
Witte (1972), Mintzberg et al. (1976), and others (Na et al., 2009) do not find a
sequence of five phases in decision processing, as most of the normative literature
suggests does and should occur (i.e., problem recognition to gathering information to
development of alternatives for evaluation to deciding on an alternative). Witte
(1972) was able to identify distinct phases in decision processing but found that all
occurred in each time interval in the processes. The frequency of occurrence of the
decision phases was different across time intervals. The total level of activity peaked
at the beginning and end of the process and the number of choices peaked at the end.
Mintzberg et al. (1976) delineate distinct phases of decision processes, but similar to
Witte does not postulate a simple sequential relationship among the phases.
Mintzberg et al. (1976) identifies three phases and seven routines in strategic
decisions:
1. The identification phase: decision recognition and diagnosis routines.

2. The development phase: search and design routines.
3. Selection phase: screen, evaluation-choice, and authorization routines.
They further found that decision processes fell into seven groupings according to
the path configurations through the seven routines. The purchasing decision processes
involved extensive cycling back from evaluation-choice to design and search routines.
Using PO research, Chapter 15 reports on how the planning, doing, and outcomes
occur in corporations for corporate purchasing agreements for commodities. The
study describes the subdivisions (phases), routines, and problems in coordinating
activities between routines. This study complements the descriptive research on
marketing decisions by suppliers in participating in annual contracts with corporate
customers that Capon and Hulbert report (1975).
Corporate purchase agreements (CPAs) or annual contracts are defined as
statements of intention to purchase and supply a quantity of product at a set price
usually for a 12-month period. Such agreements are not formal legal commitments
but, their provisions are usually adhered to by suppliers and customers. Corporate
purchase agreements are often used by large multinational corporations to combine
the purchasing requirements of several plants for negotiating to receive large
quantity discounts from suppliers. For suppliers, CPAs are an aid in learning the
total product requirements of a customer. Suppliers use customer’s request for CPAs
as a basis for planning strategy to achieve a greater share of major customers’
requirements. CPAs also aid in production scheduling for suppliers and help reduce
the risk to customers of shortages and ‘‘not getting product.’’
Capon and Hulbert (1975) report that a sales forecasting system was the core of
marketing planning and budgeting of a large multinational British firm specializing
in the highly capital intensive conversion of basic raw materials into products which
are sold to secondary processors. Capon and Hulbert studied how the marketing
department of this firm interacted with customers and processed information to
make decisions. They found that salespeople visited important customers accom-
panied by a sales manager where appropriate. Face-to-face supplier customer
negotiation on price and quantity was central to Capon and Hulbert’s descriptive
model of how the supplier reached annual contracts with customers.
Method
Similar to most research on descriptive models of executive decisions, the study was
focused on a single firm: Epsilon Corporation (name disguised). Epsilon Corporation
has plaints through the world — 80 in the United States and 40 in Canada, Europe,
Latin America, the Caribbean Area, and the Far East — with sales of $8 billion.
Epsilon has five product groups. Each has a president reporting to the Executive Vice
President for the Products Group. Six staff vice presidents report to the President of
Epsilon Corporation. Each of these staff vice presidents has some supporting line
responsibilities.
The corporate purchasing staff responsible for developing CPAs report to the
Vice President of Materials and Facilities who reports to the Vice President of
Manufacturing Services.
The study was directed only to directly describing the decision processes of CPAs
for production materials — raw materials and semi-finished products used in
manufacturing products by the five-product groups. A total of $2.9 billion
commodity purchases in CPAS were planned for 2010. Total purchases of the
Epsilon Corporation for commodities are over $2.7 billion annually. The Director of
Purchasing has responsibility for developing commodity CPA’s. The Director of
Purchasing reports to the Vice President of Materials and Facilities. The Vice
President of Materials and Facilities reports to the Vice President of Manufacturing
Services.
A multiple method approach was used to permit a ‘‘triangulation’’ (1979) of data
collection to provide valid observations of coordinated purchasing activities and
decisions. Data from direct observation, a review of company records, personal
interviews, and a literature review were compared and contrasted in an attempt to
produce clarity in understanding information processing and decision making in
corporate purchasing agreements.
With the cooperation of Epsilon Corporation, an ethnographic approach was
incorporated in the study. The ethnographic approach includes the use of anthropo-
logical tools and is commonly labeled as participant observation in sociology.
As practiced, this approach allows a fieldworker to use the culture of

the setting (the socially acquired and shared knowledge available to the
participants or members of the setting) to account for the observed
patterns of human activity. (van Maanen, 1979)
Conklin (1968) describes the ethnographic method as involving a long period of

intimate study and residence in a well-defined community employing a wide range
of observational techniques including prolonged face-to-face contact with members

of local groups, direct participation in some of the group’s activities, and a greater
emphasis on intensive work with informants than on the use of documentary of
survey data. The fieldworker must constantly check the reality of the descriptions
of events and decisions provided by informants in the defined community. This
checking includes comparing the various features of interviews with different
participants concerning the same events, re-interviews of the participants to verify
and elaborate previous statements, comparing correspondence of observations
with the verbal reports of participants, and observing first-hand the activities of
individuals and groups in the defined community.
This type of fieldwork, despite the best of intentions of the researcher,

almost always boils down to a series of endless conversations
intersected by a few major events and a host of less formidable ones.
The information as recorded by the fieldworker is then primarily talk-
based not only because this is what occupies the vast majority of the
ethnographer’s time but also because understanding the concrete
activities taking place in the field is grounded largely upon what
members have to say about what such activities mean to them.
(Conklin, 1968, p. 203)
For the present study, for seven weeks, the researcher became a participant
observer of Epsilon Corporation’s purchasing staff at the company’s international
headquarters in New York City. He attended staff meetings, committee meetings,
meetings with vendors, and interviewed eight members of the staff briefly three to
seven times each. A total of 18 members of Epsilon Corporation’s coordinated
purchasing program were interviewed at seven plant locations during the data
collection period. Each of the 18 participants was interviewed face-to-face in two-
hour, individual, meetings. Personal interviews were held with salespersons of six
suppliers of Epsilon Corporation. Five of views were held at the sales offices of
the international headquarters of the suppliers. One interview was held near the
headquarters building of Epsilon. These interviews were held to collect data to
elaborate and verify the descriptions of the purchasing processes descried by the
buyers and purchasing staff at Epsilon’s headquarters.
During the data collection period of the study, flow diagrams of the information
processing and decision making were developed and shown to several members of
Epsilon’s corporate staff and plant purchasing managers. The diagrams were revised,
shown again to the same corporate staff and plant purchasing managers, and then
revised and presented for a third time to the same corporate staff and plant
purchasing managers. The cooperating purchasing personnel were asked to describe
the actual processes and decisions used in developing corporate agreements. They
were asked to verify each step in these processes and decisions whenever possible with
correspondence, memoranda, published reports, and their personal records.
The resulting diagrams shown in the results section are composite summaries of
the descriptions provided and verified by the purchasing personnel cooperating in the
study. The attempt has been made to depict the processes and decisions used for most
of the production materials coordinated across plants.
Findings
Epsilon Corporation uses a committee approach in their coordinated purchasing

program. The Director of Purchases (Peter Reuter) and two purchasing specialists
(Dale Thomas and Bill Matthews) each co-chair production materials committees,
with a senior buyer at a plant location appointed as a co-chair on each committee.
Both Thomas and Matthews are in their mid-20s. Thomas was a buyer at an
Epsilon Corporation plant for four years before transferring to the headquarters.
Matthews is an engineer by training; he has worked with purchasing as an engineer
for four years in another corporation before moving to New York City. Both
Thomas and Matthews were purchasing specialists for approximately two years
at the time of the study. Reuter has been in purchasing at Epsilon Corporation for
approximately 25 years. He is in his early 50s.
Reuter co-chairs two committees directly: glass and Far East procurement.
Matthews is co-chair for four committees: semiconductors, capacitors, resistors,
and printed circuit boards. Thomas co-chairs eight committees, including ferrites,
plastics, magnet wire, and hybrid inks.
Several committees have two co-chair persons at the plant locations — one in the
United States and one in Europe. For example, the capacitor committee has Allan
Bates in Chicago and H. Shaw at Minneapolis as co-chairpersons from plant
locations. Committees with two plant locations co-chairpersons (United States and
Europe) rarely meet as an entire committee. The U.S. segment of such a committee
negotiates with U.S. suppliers and this segment of the committee will meet separately
from the committee members in plants in other parts of the world. However, a
co-chair (Reuter or Matthews, for example) from headquarters will attend non-
United States meetings of the committees. For example, Matthews went to Germany
for two weeks in July to attend negotiation meetings one of the semi-conductor
committee with European suppliers. A total of 14 committees for production
materials existed at the time of the study. Several new coordinated purchasing
agreements were planned for the next calendar year that might require the formation
of new committees, for example, connectors and cable agreements.
The Director of Purchases (Reuter) has no line responsibility to direct purchasing
activity or decisions at the plants. Persuasion, appeals for cooperation, and written
directives from the Executive Vice President of the Products Group are used to
produce cooperation from the purchasing management at plant locations to work on
CPAs. A total of 76 plant purchasing managers, including 42 in the United States,
report to the plant manufacturing organizations within their Business Groups.
Each of the plant managers have 2 to 12 senior buyers and buyers reporting to them.
A total of 2 to 15 assistant buyers and order processors also work in purchasing
management organizations at the plant level.
Coordinated Purchasing Savings at Epsilon Corporation
The purchased material of $2.4 billion of Epsilon Corporation is divisible into four
material groups:
Type Share (%) Value ($)
Production material 53 1.0 B

Inside material 21 0.4 B
Finished goods 16 0.3 B
Nonproduction material 11 0.2 B
Total 1.9 B
For production material, 52 percent was spent in the United States, 24 percent was
spent in Europe, 11 percent was spent in the Far East/Latin America, and 12 percent
was spent in Canada. Of that $1.0 billion in production materials, 81 percent was
spent on all types of purchases which were common in all locations. The $0.8 billion
of potential coordination has been estimated for the purchases of the five Business
Groups as follows:
Group Share (%)
Communications products business 41

Consumer groups 25
Group 3 18
Group 4 9
Group 5 7
Total 100
Of the $0.8 billion worldwide: potential, purchasing has identified for world-wide
coordination 25 percent ($200 M) in 2010. Of the $210 M potential in nonproduction
material, 15 percent is planned for coordinated in United States/Canada and Europe.
At the start of 2010, the Director of Purchasing (Reuter) provided the following
evaluation of Epsilon Corporation’s program to coordinate commodity purchases.
The value of coordinated commodities increased for a gain of 28%.

The net reduction from previous price paid was $11.1 million
compared to $8.7 million in the previous year — also a gain. Four
new commodities were coordinated on a world-wide basis for the first
time in 2009.
The coordinated commodity agreements with suppliers re-viewed as firm business

arrangements by purchasing at Epsilon manufacturing locations, by Epsilon,
New York, and by suppliers. Thus, the opportunity to report the actual dollar
amounts being coordinated and net reductions is substantial for the coordinated
commodity agreements.
Decision Systems Analyses
Figures 1–4 show the flow of events and decisions for the coordinated purchasing
program for commodities at Epsilon. The events and purchase decisions are divided
into four parts:
1. Developing and analyzing requirements

2. Preparing RFQs and analyzing quotations
3. Committee–supplier negotiations
4. Post-negotiation evaluation and reporting.
Figure 1 is a description of activities and decisions that take place prior to the
corporate purchasing staff issuing a requirements survey to the plant locations.
The development of a commodity plan is the central event of this process. The
development of such plans was first proposed in 2009. The details of he benefits and
format of commodity plans were explained in a statement prepared by Matthews
(one of the purchasing specialists). Exhibit 1 shows part of this statement.
The preparation of commodity plans was a new activity in the summer of 2009.
Such plans were not available yet for any commodity. However, the purchasing
specialists and trainees (‘‘Associates’’) were assigned to develop such plans for
specific commodities.
The need for some type of planning similar to commodity plans was identified by
three plant purchasing managers during the field interviews. The preparation of
commodity plans had not yet been discussed with plant purchasing by corporate
purchasing in the summer of 2009.
The survey of requirements near the bottom of Figure 1 was of direct concern to
plant purchasing managers. Several buyers interviewed in the study reported that the
‘‘high number of man days to respond to corporate requests’’ and the ‘‘rush request
by New York’’ were problems. To overcome these problems, a computer data system
was created in 2008 for requesting requirements from plant purchasing. However, the
continual need to plan and keep to a specific time schedule to send requests for
requirements, prepare the requirements at the plants, and receive and analyze the
requirements at New York should be recognized as the major immediate concern of
the initial stage in producing coordinated agreements.
Preparing RFQs, issuing RFQ, and analyzing the resulting quotes are the major
activities shown in Figure 2. A serious problem was reported by several of the buyers
interviewed in the issuing of RFQs to vendors. Some buyers disagreed with the
practice of sending an RFQ to a vendor who has not been qualified at several plants.
330
Develop commodity
Time to initiate Is this a new No plan to identify
Start of Stage 1 contract cycle purchase market conditions
commodity? and supply strategy
Perform initial requirements

survey to determine size and Is plan
nature of requirements acceptable
No to staff
management and
commodity
Perform market research Revise plan
committee?
and commodity analysis
Yes
Has
Go to Stage 2
Issue requirements survey to plant product

locations to determine requirement changes mix caused
Yes
commodity
requirements to
change from
Analyze requirements for value previous
and commodity No year?
Figure 1: Stage 1 of corporate purchase agreements: Developing and analyzing requirements.

Does commodity Prepare Supplier selection
Start of Stage 2 Yes competitive Process fro competitive
Plan call for competitive
bid solicitation? RFQ solicitation
Negotiate contact No
Eliminate Criteria:
extension with Does each
from bid No 1. Previous performance
current suppliers supplier meet
process 2. Committee member
Selection
request
criteria
3. Other
Go to negotiation Yes
Process in Stage 4
Suppliers submit Prepare and submit

Issue RFQs t suppliers Quote Analysis Report
quotations
Does supplier
meet criteria for Committee members review Quote Analysis Report
selection to Quote Analysis Report and prepare Issued t Location Committee
Yes
negotiation Individual Location Analysis Members
process?
Eliminate Retain Develop supplier

No proposal in negotiation schedule Go to Stage 3
Vendor
file and issue schedule to
committee members
Criteria:
• Competitive quotes;
• Meets specifications
• Multiple-locations capability
Participant Observation Research in Organizational Behavior
Figure 2: Stage 2 of corporate purchase agreements: Preparing RFQs and analyzing quotations.
331
332
Committee member
Will sends proxy strategy
Start of
Committee No recommendations
Stage 3
member personally to the committee
attend negotiations
Pre-negotiation strategy
Is there committee meeting Go to
an overall Stage 4
negotiation strategy
that satisfies Yes
individual
Negotiation process
location
with suppliers
objectives?
Agree on
terms
Committee and
No
develops award
counter- Is supplier Yes contract
Re-evaluate strategy No
offer and proposal accepted
until consensus is presents by committee?
reached to the
Yes
supplier
Yes
Does committee
Does supplier accept accept counter

counter-offer? Supplier presents counter- bid?
offer to committee
Does probability
No of Agreement justify No
Yes another round of
Supplier caucuses negotiation?
Committee caucuses for
to create counter bid Eliminate supplier; go to
Strategy revision
next supplier
Present next counter

offer to supplier
Figure 3: Stage 3 of corporate purchase agreements: Committee–supplier negotiations.

Post negotiation committee meeting

Start of Stage 4 to evaluate results of process
Inputs to Yes
Were negotiation
Negotiation
objectives met?
Report
No
Identify reasons, failure factors,

and recommend actions
Reporting
System
Committee members report

to their line (plant) management
Bulletins prepared
and issued
Contracts prepared and signed
Bulletins
received by all
company buyers Contracts sent to suppliers and
Epsilon system-wide committee members
Figure 4: Stage 4 of corporate purchasing agreements: Post-negotiation evaluation

and reporting.
Exhibit 1. Commodity purchase planning (CPP).

I. What is commodity purchasing planning? CPP is a specific short/long-term
written action plan for purchasing profit sensitive material, emphasizing value
improvement so as to obtain the lowest overall delivered cost consistent with
service and quality requirements for the commodity under plan. Plans are
typically either written or updated on a yearly basis.
II. What benefits are gained from such a plan? (1) CPP provides a means to
organize the thinking relating to buying strategies and methods which result
in enhanced supply continuity, supplier service, and lower total costs for the
purchased material y (9) Conveys management orientation and concentra-
tion of profits improvement.
III. Sample CPP Format. (1) Commodity — description and computer code
number of material y (7) Supplier analysis: A. Evaluate performance of
current and past suppliers y B. Existing contracts: tabulate essential data
including: names of suppliers; using locations; y reasons for order splits y
10. Action plan — proposed administration of contract, committee, member
involvement, and schedule of events for the forthcoming year.
In some instances, the buyers reported that a potential supplier was sent an RFQ and
quoted who was not qualified at any Epsilon plants. Two buyers preferred to see
RFQs send only to suppliers who were qualified at two or plants. However, two
buyers preferred the practice of sending RFQs to nonqualified vendors — ‘‘a good
way to test the waters’’ was a reason offered for this procedure.
Whatever the approach taken, the committees need to discuss this issue
before RFQs are sent to suppliers. The question of which suppliers should be
sent RFQs should be answered in pre-negotiation meetings of the commodity
committees. This step would be a recommendation for changing the decision system
currently in use.
A problem occurs in the last step shown in Figure 2: develop supplier negotiation
schedule and issue to committee members. Final agreement on which suppliers to
invite to face-to-face negotiations is often not decided by the committee but by
headquarters acting alone. One committee member reported that as supplier had
arrived at the negotiation who was unknown by all the committee members.
An initial pre-negotiation meeting of the committee may be necessary before the
week of the negotiations with suppliers to jointly analyze suppliers’ quotes. The use
of the committee approach appears to call for several meetings of the committee:
(1) an initial planning meeting to evaluate supplier performance across plants and to
develop a time schedule, (2) a second meeting to evaluate quotes received and plan
negotiations strategies with the different suppliers, and (3) a pre-negotiation meeting
at the start of the week of negotiations with suppliers to review strategies and make
last minute changes.
For some commodities, the first and the third meeting or only, the third meeting
was held by a commodity committee. Having three meetings of a committee for a
commodity was reported to be a rarity. A second meeting may be necessary to reach
a consensus on which suppliers to invite to the negotiations and to plan negotiation
strategies. Substantial conflicts sometimes occur and are left unresolved within the
committee on what strategy to use in negotiating with vendors.
The degree of aggressiveness with suppliers was an area of conflict among
committee members. The majority of members reported that their negotiation skills
had improved because of their participation in committee negotiations with
suppliers. They credited this improvement to Peter Reuter’s aggressive style in
negotiating price concessions from suppliers. ‘‘Before Reuter we were too close to
suppliers. Now we are more professional’’ is a statement from one committee
member that was expressed in similar ways by other committee members who were
interviewed. However, not everyone agreed that a very aggressive approach in
negotiating with suppliers should be taken. The following comments were made by
plant purchasing personnel on this issue.
1. ‘‘Making counteroffers to suppliers’ quotes below the price the committee would
ultimately expect to agree won’t work.’’
2. ‘‘Offers should be made on what the committee expects to receive.’’
3. ‘‘Suppliers XY and LIPB were pissed off at a purchasing specialist saying,
‘Apparently you don’t appreciate our business.’ It was a bad [situation].’’
4. ‘‘Making counteroffer 5 to suppliers’ quotes below the price expected to &:

received is done now. It works!’’
5. ‘‘Don’t try to drive a guy down so that he is not going to make any money. Look
for the low quote and accept it. It’s got to be good for both parties.’’
6. ‘‘Maybe we start too low. We got an initial balk from suppliers.’’
7. ‘‘Yes, we are justified in making low counteroffers. At least we asked-should
try.’’
8. ‘‘No-see who quoted low. Assume the low quote is the market price. It’s not
ethical to dream up a price 10 percent below the low quote. When you do this
you destroy long-term relationships. Credibility is destroyed. They believe me
when I tell them that this is the market price.’’
9. ‘‘Well yes — start low and compromise if needed — if in actual negotiation
sessions.’’
10. ‘‘Yes. I will never make a counter-offer that does not go below the price I am
currently paying.’’ This buyer would go lower in the counter-offer in a buyer’s
market but not in a seller’s market.
11. ‘‘Yes, but I don’t want to violate the Robinson–Pactman Act. This is a good
reason to bring in a small vendor who will low ball you — bring him in first.
Never play brinkmanship unless you have the aces. You don’t have to be
obnoxious; a common tactic now is yelling, table pounding, and them making up
with suppliers.’’
12. ‘‘Yes, because you really don’t know where the vendors’ prices are.’’
13. ‘‘Yes, but should be made from knowledge. Don’t say you have it when you
don’t. You hurt yourself. Corporate purchasing staff is concerned only with the
bucks.’’ This buyer went to explain that he has to be concerned with availability
and quality in addition to price.
14. ‘‘Yes — must have some tell you no. But — New York — Reuter — undercuts
chairman sometimes, Ralph Jones [name disguised] specifically.’’ This buyer
went on to say that his negotiation skills have improved from his committee
participation, ‘‘I’m tougher now — less a feeling that it’s important to be nice —
we don’t want to be called ‘nice’ by suppliers.’’
The important point here is that serious differences in opinions on which type
of negotiation strategies should be used with vendors sometimes occur within
committees. These differences need resolving by each committee before the face-to-
face meetings with suppliers. Details of the events and decisions in the committee and
supplier negotiations are shown in Figure 3. The need for consensus of the basic
negotiating strategy before the start of the meeting is the most serious concern with
the negotiation process.
If the supplier’s proposal is not accepted by the committee, the committee will
present a counteroffer to the supplier at the meeting. This is the most likely action.
Supplier representatives will usually ‘‘caucus’’ on this counteroffer, that is, the
two-to-five representatives will meet in a private room by themselves to discuss the
counteroffer. The suppliers usually present a counter-offer to the committee’s
counter-offer. The supplier’s counter-offer may or map not be accepted. If it is not
accepted, the committee caucuses for a strategy revision and then presents a
counteroffer to the supplier. Rejection by the supplier of this offer results in an
unsuccessful negotiation.
Figure 4 describes the post-negotiation evaluation and reporting activities and
decisions for coordinated agreements for commodities. A problem at this stage in the
process is that plant purchasing participating as members in the committees do not
learn how well the program has performed or learn months after the negotiations
are completed. A reporting system to plants management, purchasing, and other
functional areas from corporate is not included in the post-negotiation evaluation,
and reporting stage.
One committee member in Atlanta remarked, ‘‘We don’t see any report on savings.
[New York says] ‘Tell us your requirements, vendors, and other information.’
But we don’t get anything back from New York.’’ This member reported that the
negotiations with suppliers were completed in less than two weeks on semiconductors
but the contracts were not received for months. ‘‘A lot of it was New York being slow.
Same problem with vendors.’’ The member wanted to know, ‘‘Why New York did not
send me corporate agreements for semiconductors. I got them from another plant.’’
We provide New York with tons of information but get nothing back.
Not even a timetable nor a corporate-wide RFQ, or quotes back from
suppliers. After a negotiation let us see the results: ‘‘As a corporation
we committed so much.’’
Another committee member made the following comments related to the post-
negotiation stage. ‘‘We don’t get feedback. New York has problems.’’ He then
went on to discuss scheduling problems and strategy problems with the coordinated
program.
‘‘Not enough time is allowed by New York to work up requirements for

RFQs. This is a minor problem. Then it takes too long for New York to
get RFQs out — 3 to 4 months to get RFQs out. New York is not
concerned with service, quality, or lead time. They are concerned with
price alone. We are concerned with product and service. The Purchasing
Specialists try to go from coordinator to head negotiators. Suppliers get
mad. Reuter needs a guy with field experience. The RFQs on capacitors
sent to all suppliers. Most suppliers did not supply all plants. We lose
some of our integrity of negotiation on capacitors by requesting them
[suppliers] to quote on one price for a certain share of business and
then tell them they can get only a much smaller share of the business.
Suppliers do want to know where they stand. What percent of the
business they will receive. Tell them a quantity, not a percent.
This member offered several recommendations. ‘‘Give us feedback on inquiries

and savings. We do not receive corporate-wide RFQs in all cases. We should receive
copies of suppliers’ quotes a month ahead of the negotiations. Don’t ask suppliers to
quote on what they are not approved. On capacitors, RFQs went to 20 suppliers.
Don’t jerk suppliers around!’’ Another member reported that his plant publishes a
‘‘Purchasing Savings Program’’ report quarterly. A similar corporate-wide publica-
tion would serve a useful purpose.
Related to post-negotiation activities several committee members interviewed at
the plants indicated that the coordinated program ‘‘has not been sold to their plant
managers or other functional areas.’’ One member said, ‘‘The plant manager
grumbles about the time waste of my participation on the committees.’’ Another
member reported that he does not get the time to work on committee activities.
A third said that his plant manager wants to know what the payoff is for having the
buyer spend so much time on committee activities. A fourth member was leery about
becoming a committee chairperson because ‘‘of all the work I have to do here.’’
Several members of the committees voiced the impression that Epsilon had a
limited but adequate number of well-trained and insightful buyers to chair the
committees but not enough depth of such personnel. One buyer remarked that the
company has no program to train and develop buyers. He said that training and
development programs were limited to entry level personnel and purchasing
managers at the company.
Table 1 is a summary of ten activities, purchasing problems associated with the
activities, and solutions to these problems. Most of the activities shown in Table 1
have been discussed. Supplier commitment to delivery is a problem with contract
execution that not been discussed.
A Committee Meeting
At a meeting of the semiconductor meeting attended by the PO researcher in June, a

problem with supplier commitment to an existing agreement was the topic of
discussion for one of the six hours of the meeting. One committee member said,
‘‘Each supplier is picking one Epsilon Product Group which it is supporting with
product in the agreement but the suppliers are not fulfilling the agreement across
Epsilon plants. The supplier is cherry picking during the contract on high profit
items.’’ Another member told the committee that suppliers to the member’s plant
location are saying, ‘‘When you reach the commitment level in the agreement, we will
raise prices.’’ Both problems with the suppliers were narrowed to two suppliers. The
Director of Purchasing (Peter Reuter) told the two complaining committee members
that the problems would be solved that day or the next day. Reuter mentioned to the
PO researcher on the day following this meeting that the problems were solved by
telephone calls to the national account managers of the two suppliers.
One week later, a member of the semiconductor meeting reporting that he was
upset about all the time taken in the semiconductor meeting to solve the supplier
problems of two relatively small plants. ‘‘It was ridiculous,’’ he said. ‘‘It wasn’t
the place to get into problems which they should have handled before the meeting.’’
Yet he believed that the meeting was useful and progress was made in the
Table 1: Purchases, activities, problems, and solutions.
Activity Major purchasing Problem solution
1. Survey usage High number of man-days Computer systems and

patterns of and rush requests by scheduling control
locations corporate
2. Developing Bring-in consultant after Participant in project at
requirements requirements are specified inception
3. Prepare RFQs Quality and quantity of Extensive search
supplier list are weak
4. Quote analysis Lack of participation of all Telephone and face-to-face
report persons involved meetings
5. Pre-negotiation Time constraints Schedule two pre-
meeting negotiation meetings
6. Negotiation Conflict over strategy Agree on strategy in pre-
with suppliers negotiation meeting
7. Agreements Use of agreements by Coordination and
locations; suppliers communication by
honoring commitments corporate and locations
8. Evaluation of Lack of agreement by Post-negotiation meeting
cost avoidance committee members
and reductions
9. Contact Supplier commitment to Location and corporate
execution delivery monitoring programs
10. Reporting Delays in release Responsibility assigned to
systems chairpersons
current year. While the semiconductor meeting was held on June 27, the minutes
were distributed on July 23. A total of ten members of the committee were present —
all from the United States. The minutes were distributed to all 28 members of the
committee including the members at the plant locations in Europe. The following
actions were decided upon at the June 27 semiconductor meeting and reported in the
minutes of the meeting.
‘‘Because the semiconductor pricing continues to trend downward

particularly in the linear circuit devices which constitute a large portion
of Epsilon’s requirements, and because of the need of assurance of the
suppliers in 2010, it was decided to follow this resultant course of
actions:
1. Forecasts must be submitted to Bill Matthews by July 6, 2009.

2. Generate a RFQ by the beginning of August.
3. Have a committee meeting to review proposals by the beginning of
September.
Action items to be considered:
1. Enforce 2009 contracts with suppliers.’’

2. Monitor any allocation programs and present orders with suppliers.
The minutes of the meeting were prepared by Bill Matthews.

This meeting illustrates several points. Progress in the decision process for
coordinated agreements on semiconductors was accomplished. Specific problems
with suppliers were discussed and plans developed to solve them. Committee
meetings can frustrate, in part, due to differences in opinions of members concerning
the topics covered and length of time taken for each topic.
While a telex was sent shortly after the meeting to the members of the committee
on the actions agreed to by the members attending the meeting, the minutes were not
distributed until nearly a month after the meeting. The minutes contained details
of the meeting and ratings of each semiconductor supplier for each Epsilon
plant location represented at the meeting for five attributes (price, delivery, quality,
service, and distribution support). A 10-point scale was used for each attribute and a
total score was computed for each supplier by summing the five ratings. Thus, the
five attributes were equally weighted. A discussion of whether or not to weight
the attributes unequally did not occur during the meeting. Bill Matthews went
to Germany for the first half of July for duties as a co-chairman of a committee
negotiating with European suppliers. Given that he had the primary responsibility
for coordinating the actions agreed to at the June 27meeting, most of the
coordination did not occur. An RFQ was not generated by the beginning of August.
This process–outcome example illustrates one of the central problems of the present
coordinated purchasing program reported by several committee members inter-
viewed at plant locations. The existing committee organization was inadequate to
perform the coordination necessary for each commodity.
Summary of Committee Members’ Evaluations
Table 2 is an attempt to summarize whether or not each committee member

perceived that substantial savings were realized in purchases for his/her plant. The
majority (10 of 18) of the members interviewed in-depth reported that substantial
savings were realized from the coordinated program which would not have been
realized with the coordination.
Given the reports and plans of savings by the Director of Purchases, interviews
with the purchasing specialists, the committee members of the plants, and the
interviews with six Epsilon suppliers, an important conclusion of the study is that
the coordinated committee program is producing significant savings and contributing
to net profits of Epsilon Corporation. Table 3 provides an overall evaluation
of headquarters performance by the members of the committees interviewed at the
plants. The majority of the committee members do not evaluate New York’s
performance positively.
Table 2: Evaluation of savings realized from committee approach to corporate

purchase agreements.
Location Evaluation
We help out Substantial No savings It’s hard to tell
the little guys savings
Chicago 1 3
Atlanta 1 2 1
Pittsburgh 3 2
Columbus 1
San Francisco 2
New Orleans 1
Los Angeles 1
Total 2 10 5 1
Table 3: Overall evaluation of corporate purchasing by 18 purchasing personnel.
Location Evaluation of corporate purchasing performance

Positive Neutral or careful Negative
Chicago 2 2
Atlanta 2 2
Pittsburgh 1 1 2
Columbus 2
San Francisco 1 1
New Orleans 1
Los Angeles 1
Total 8 2 8
The number of members classified as giving a negative overall evaluation to

New York’s performance implies that some changes to the present coordination
program should occur. Such changes have been discussed in explaining the decision
system used in the coordinated program.
Summary of Recommendations from the Committee Members
Several recommendations made by committee members to increase the performance

of the coordinated program need emphasizing.
1. More timely coordination of committee actions is recommended. Someone is

needed to continually move the committee t rough the decision process.
2. The program needs to be sold to plant management and related functional areas.
Face-to-face meetings between corporate staff members and committee chairmen
with plant managers are necessary. Effective internal marketing programs on the
benefits of CPPs are lacking at Epsilon.
3. Time must be taken for committee chairmen and the New York purchasing staff
to visit and listen to plant purchasing.
4. A reduced role for the purchasing specialists is recommended. The purchasing
specialists should not be co-chairs on the committees.
5. New York needs some senior purchasing staff people with several years of plant
experience to coordinate committee: activities and decisions.
6. The committee members should have more say as to which suppliers are sent
RFQs and invited to negotiate.
7. Negotiation strategies with suppliers should be agreed upon by the committee
before meeting with the suppliers.
8. Post-negotiation reports prepared by the chairman on New York are
recommended. These reports should be distributed to all members of the
committee.
9. A company-wide report on savings produced by the coordinated program by
commodities should be distributed to plant purchasing, including buyers. The
plant managers and other functional areas besides purchasing should receive
these reports.
10. Two separate pre-negotiation meetings are recommended — one at the start
of the decision process and one after quotes are received from the suppliers.
A half-day committee meeting should occur immediately before the start of the
negotiations with suppliers.
Evaluation of Key Informant Interviews
The cooperation and insight provided by 13 of the 18 interviews with members of

the coordinated committee members were excellent. The answers and comments
provided by these members provided most of the data collected in the study. A total
of four additional interviews were helpful but not completely satisfactory — either
the PO researcher interpretation of the answers, vagueness of the questions, or some
other factors limited the usefulness of responses. One interview was unsuccessful.
Thus, 17 of the 18 interviews with committee members were judged as worthwhile.
The majority of the committee members interviewed were active members of their
respective committees. Several chaired the committees. Most attended one or more
committee meetings annually. The committee members interviewed were involved with
the majority of savings realized from the coordinated committee program. All members
contacted and requested to be interviewed did cooperate and were interviewed. Thus,
no non-response bias occurred in this study. Thus, even though the sample size was
small, the research method used is believed to provide valid observations on how
members of Epsilon’s coordinated committees evaluate the program.
Also, the observations of corporate staff, activities and records, and interviews
with vendors provided a triangulation of information to support the major
conclusions of the study.
Conclusions
Four central conclusions were emphasized in report of the findings of this study
to the corporate purchasing staff at Epsilon. (1) The committee approach for
corporate purchasing has produced substantial dollar savings that would not have
been realized in 1976–1979 without the program. (2) The committee approach has
produced a number of intangible benefits — the most important being the increase
in professionalism of the buyers serving as members of the committees. (3) The
coordinated program is now inadequately staffed. (4) The program has not been sold
to the management and related functional areas at the plant level — this lack of
effective internal marketing limits the benefits realized from the program.
An ethnographic approach for the study of coordinated purchasing programs is
a useful method for learning the flow of decisions and activities involved in the
administration of such programs. The study reported here should be considered a
pilot research effort. Analogous to Whyte’s (1943) Street Corner Society, six-month
to two-year participant observations are necessary for acquiring a deep under-
standing of corporate and plant purchasing decision processes. Hopefully, the study
reported here will reduce the time required by other researchers to develop a clear,
initial, focus on what is actually happening in purchasing.
Chapter 16
Systems Thinking and System

Dynamics Modeling
Synopsis
Chapter 16 is an introduction to systems thinking and analyzing the system

dynamics of relationships within an organization or between organizations.
Systems thinking builds on the propositions that (1) all variables or conditions
have both dependent and independent relationships, (2) lag effects occur in
relationships, (3) feedback relationships occur (e.g., A-B-C-A), and
(4) seemingly minor relationships (i.e., ‘‘hidden demons’’) have huge influence
in causing a set of relationships (i.e., a system) to implode or explode. The
propositions of building and testing a set of relationships apply in many
contexts; this chapter examines systems thinking and system dynamics in one
context as an introduction to this stream of case study research. Hall (1976)
provides details of an advanced application of systems dynamics research – do
not be fooled by the date of the study; Hall (1976) is an exceptional up-to-date
case research study using system dynamics modeling. This chapter describes the
issues and criticisms concerning golf, tourism, and the environment and
considers how golf–tourism–environment relationships might achieve economic
well-being for a region while avoiding vicious cycles of destruction to local
environments and the quality of life of local residents. The examination
proposes the use of systems thinking, cause mapping, and system dynamics
modeling and simulations of golf, tourism, and environmental relationships to
help achieve workable solutions agreeable to all stakeholders. Sustainable
relationships that include golf, tourism, and environmental objectives require
crafting government policies via stakeholder participation of all parties that such
relationships affect – recognizing and enabling this requirement needs to be done
explicitly – to reduce conflicts among stakeholders and avoid system failures.
Introduction
A cause map is an illustration of sense-making propositions that include mutual
dependencies of variables in a system of relationships. A cognitive map that displays
various concepts connected by causal relationships is called a cause map; it is a

summary of the cause–effect inferences of our sense-making processes (Gilmore &
Murphy, 1991; Weber & Manning, 2001; Weick & Bougon, 1986). Positive (+) and
negative ( ) signs on directional arrows inking nodes (i.e., variables) in a cause map
indicate the nature of the relationship between the variables (see Weick, 1979).
Cause Maps of Unsustainable and Sustainable Golf Tourism
The use of principles of cause mapping permits visualizing systemic propositions

of relationships that indicate unsustainability and sustainability of golf tourism for
a given geographical region – the same principles are applicable for other tourism
categories besides golf such as casino tourism, hunting tourism, and medical tourism.
Such visualization helps to capture complexity of feedback relationships and
‘‘hidden demons’’ (i.e., seemingly minor and subordinate paths) in relationships
(cf. Tufte, 2001). Cause mapping is a useful step toward building system dynamic
models that permits what-if simulations of the eventual outcomes when one variable
changes slowly or dramatically within a system of relationship.
Simple Unsustainable Models of Golf Tourism

Figure 1 includes two models illustrating a positive relationship between the number
of golf courses in a region and tourism and tourism’s positive influence on the
1. Golf Courses 2. Tourism
+
(a)
+ +
3. Property 1. Golf Courses 2. Tourism

Development
+ +
(b)
Figure 1: (a) An unsustainable simple model of golf tourism. (b) An unsustainable

moderately complex model of golf tourism.
Systems Thinking and System Dynamics Modeling 345
number of golf courses – these two variables are mutually dependent. Advocates of
increasing the number of golf courses in a region point to the first influence – golf
courses bring tourists. Increasing tourists’ numbers increases the need for golf
courses – the second arrow in Figure 1a.
Any variable that has arrows both coming and going out from it is
interdependent; the variable is affected by and affects other variables. Any variable
that only has arrows coming to it but has no arrows going away from it is a
dependent variable. And any variable that has arrows going away from it but has
no arrows coming into it is an independent variable (Weick, 1979, p. 72).
A causal or feedback loop is a relationship or chain of relationships among
variables whereby starting with one variable an arrow from that variable eventually
leads back to the same variable. The direct mutual dependency of golf courses and
tourism in Figure 1a indicates a causal loop.
Figure 1b indicates that tourism has an indirect influence on property
development. The increase in tourism increases the number of golf courses, which
in turn increases property development, which in turn increases the number of golf
courses.
Note in both Figure 1a and 1b that a starting variable (i.e., an element or node
with a path [arrow] leading out) increases even further on completing a path circuit,
and the same variable continues increasing with each completion of a cycle. In a
causal loop with an even number of negative signs (e.g., zero negative signs in both
parts of Figure 1), the variable will continue to move in that same direction until the
system is destroyed or until some dramatic change occurs (Goldsmith, 1971).
The loops in the two parts of Figure 1 are amplifying loops that represent vicious
or virtual circles (Weick, 1979). Building in necessary complexity to reflect reality
occurs by identifying additional variables that build in an odd number of negative
signs in a cause map. ‘‘These loops are particularly interesting because they impose
stabilities on organizing processes’’ (Weick, 1979, p. 74).
Complex Sustainable Models of Golf Tourism
Figure 2 shows an attempt to increase understanding by building in necessary

complexity of how golf tourism potentially provides both benefit and harm to a
geographic region. Such modeling is illustrative of integrative thinking and the value
of nurturing an ‘‘opposable mind’’ (Martin, 2007). ‘‘Integrative thinking shows us a
way past the binary limits of either-or. It shows us that there’s a way to integrate the
advantages of one solution without canceling out the advantages of an alternative
solution. Integrative thinking affords us, in other words of the poet Wallace Stevens,
‘the choice not between, but of’’’ (Martin, 2007, p. 9).
Such integrative thinking and building cause maps that include necessary
complexity reflect Chamberlin’s (1890, p. 93) wisdom: ‘‘In following a single
hypothesis, the mind is presumably led to a single explanatory conception. But an
adequate explanation often involves the coordination of several agencies, which enter
346
+ +
3. Property +
development
+ +
_ +
_
_
+
7. Arable 8. Water resources
6. Herbicides/
land +
insecticides
+ +
9. Agriculture
+
_ +
+ +
_
+
5. Residents’ Health / 4. Employment
Quality of Life
_ 10. Government regulations

(e.g., regional land use regulations)
Figure 2: A causal map of a sustainable complex golf tourism system.

into the combined result in varying proportions. The true explanation is therefore
necessarily complex. Such complex explanations of phenomena are specially
encouraged by the method of multiple hypotheses, and constitute one of its chief
merits.’’ Cause mapping is a parsimonious approach that captures ‘‘the method of
multiple working hypotheses.’’
Note that Figure 2 includes and expands on the relationships in Figure 1 and both
deviation-amplifying and deviation-dampening loops. Any causal loop that has an
odd number of negative signs serves to dampen (decrease) levels of one or more
variables in the model. The loop 1-6-5-10-1 includes three negative signs –
a dampening loop. The loop 1-3-5-10-1 includes three negative signs – another
dampening loop. The negative relationship from property development to residents’
health and quality of life (3-5) might be a point of conflict.
Initially, property development may serve to increase the quality of life of
residents by increasing the value and selling prices of land; further increases in
property development may serve to exclude local residents from living in the region.
The cause mapping and system dynamics literature demonstrates the occurrence
of such reversals in relationships (from positive to negative) in different time periods
(e.g., Hall, 1976).
Weick (1979) predicts that any system will survive as a system only if it contains
an odd number of negative loops. If the system contains an even number of negative
loops, then their effects will cancel one another and the remaining positive cycles will
amplify whatever deviations may occur – this view assumes that all loops are of equal
importance, an assumption that is revisable. The next section discusses this issue.
Working through all the loops in Figure 2 indicates that the cause map includes
three positive and eight negative loops (Table 1). This finding indicates that such a
geographic region will become overloaded and not fluctuate around some middle
value because the system of relationships includes an even versus an odd number of
negative causal loops.
The cause map serves to illustrate the ‘‘squeaky wheel gets oiled’’ principle to be
necessary to achieve such a dampening outcome – decline in residents’ health and
quality of life propels an increase in government regulations to prevent further
Table 1: Net positive and negative loop relationships in golf tourism causal map..
Net positive Loops Net negative Loops
1-2-1 (+) 1-7-9-5-10-1 ()

1-3-1 (+) 1-7-9-4-5-10-1 ()
1-2-4-5-10-1 (+) 1-3-5-10-1 ()
1 1-8-9-5-10-1 ()
1 1-8-9-4-5-10-1 ()
1 1-6-5-10-1 ()
1 1-7-5-10-1 ()
1 1-7-1 ()
declines in health and quality of life. Not shown directly in the map are the necessary
steps in organizing and prodding and promoting government regulations by residents
to achieve the dampening effect. Such cause mapping recognizes and justifies
the participation of multiple stakeholders and explicit planning with government
involvement and regulations in regional recreational development – including
building and maintaining golf courses.
Strategies for Analyzing Systems
The cause mapping and system dynamics literatures suggest three general strategies in
analyzing outcomes of the complexity that always occurs in building realistic systems
that include elements of the environment (e.g., land, water, and health-harming
pollutants). First, strategist could assume that the loops are of unequal importance.
Making this assumption, the strategist could scale the different loops on their degree of
importance and predict the fate of the system by the nature of its most important loop.
Figure 3 is a visual of a cause map showing three loops of greater importance
(in bold) than the remaining seven loops. The three high-importance loops are all
deviation amplifying. All three indicate positive outcomes from increasing the
number of golf courses: increasing tourism and employment, increasing health and
quality of life, and increasing property development. Figure 3 suggests that tolerant,
laissez-faire government approaches to golf course development are best.
However, given that all three of these seemingly most important loops are
deviation amplifying, the system is unsustainable and will eventually become
uncontrollable and breakdown or reverse its direction because of some dramatic
changes. No system continues to grow unabated forever (Ford, 1999, p. 7). Figure 3 is
one map showing the good news of building golf courses in a region. This good news
view often includes minimizing the importance of the other causal loops in the system.
The difficulty of identifying some loops as more important than others is that the
judgment of a loop’s importance may often be purely arbitrary (Weick, 1979, p. 75).
Weick (1979) describes ways to solve the arbitrariness of identifying more and less
important loops. A strategist might define importance in terms of the number of
inputs to and outputs from an element – the greater the number of inputs to and/or
outputs from an element, the more important the element is. The strategist then
identifies the loop that contains the greatest number of the most important elements.
Figure 4 shows this loop in bold. This loops goes from 1 to 8 to 9 to 4 to 5 to
10 to 1. Since the loop has an odd number of negative signs (three), it is deviation
dampening, and therefore the primary prediction is that the entire system is deviation
dampening and will stabilize around some midlevel quantity of golf courses versus
some maximum number.
Both Figures 3 and 4 serve to make the point that assumptions make a difference.
It is conceivable that both conclusions about this golf course and tourism system are
correct for different time periods. The system will continue to grow golf courses for
some period of time, but such things as the differential speed with which the loops
+ +
3. Property +
development
+ +
_ +
_
_
+
7. Arable 8. Water resources
6. Herbicides/
land +
insecticides
+ +
9. Agriculture
+
_ +
+ +
_
+
Quality of Life
1-3-1 (+) 1-7-9-5-10-1 (–)
1-2-1 (+) 1-7-9-4-5-10-1 (–)
_ 10. Government regulations 1-3-5-10-1 (–) 1-8-9-5-10-1 (–)
(e.g., regional land use regulations) 1-2-4-5-10-1 (+) 1-8-9-4-5-10-1 (–)
1-6-5-10-1 (–) 1-2-4-5-10-1 (+)
1-7-5-10-1 (–)
Systems Thinking and System Dynamics Modeling
Figure 3: A view of unequal importance among causal loops resulting in an unsustainable complex golf tourism system.
349
350
+ +
3. Property +
development
+ +
_ +
_
_
+
6. Herbicides/ 7. Arable 8. Water resources
land +
insecticides
+ +
9. Agriculture
+
_ + + +
_
+
Quality of Life
_ 10. Government regulations

(e.g., regional land use regulations)
Figure 4: The possibility of a dominating loop resulting in a sustainable complex golf tourism system.
are completed, the magnitude of the changes at each variable, the tightness of the
couplings among variables, the number of times each loop is activated, and the
effects of exogenous variables introduce increasing amounts of instability and
bubbling of the system (cf. Weick, 1979, pp. 76–77).
The third general strategy to analyze outcomes of the complexity is to use cause
maps as a step toward system dynamics modeling. Such modeling recognizes and
responds to real-life greater complexities in systems that cause maps alone do not
capture. System dynamics modeling with available software programs (e.g., see
STELLA at High Performance Systems, 1995) permits the running of multiple
simulations of alternative system assumptions to test the efficacy of these assump-
tions under which the system may operate. System dynamic modeling permits the
construction and testing of multiple loops with nonlinear feedback relationships and
delays in activating noticeable increases or deceases (i.e., tipping points; see Gladwell,
2002) among the elements or nodes in a system.
System dynamic modeling permits the recognition of time as an important factor
in the operation of all complex systems – the different loops in complex systems are
operating at different rates of speed. Consequently, bad news outcomes may follow
good news outcomes and achieving sustainability in a system may never occur – an
oscillation from relatively short best periods to longer worst periods or the
combination of best and worst of times may continue.
The primary objective of system dynamics modeling is to deepen understanding,
not point prediction, of how a given system operates under alternative sets of
assumptions. ‘‘To avoid misinterpretation, you should resist drawing any conclusions
from a single simulation. A single simulation teaches us much about a system. Its
purpose is usually to provide a starting point for comparison with additional
simulations’’ (Ford, 1999, p. 10). Running multiple simulations deepens knowledge of
likely outcomes of a given system under different sets of assumptions; such knowledge
enables executives to create useful rules of thumb (i.e., heuristics; see Gigerenzer,
Todd, & The ABC Research Group, 1999) to help manage complex systems.
Running simulations of system dynamic models increases knowledge about the
momentum and direction of environmental outcomes. Outcomes of some simulations
often suggest that the strategist may need to ‘‘hit the brakes’’ (Ford, 1999, p. 11)
to prevent disaster and to allow slow and sluggish loops the time to operate to
dampen seemingly uncontrollable expansion. System dynamic modeling often leads
to recognition among all stakeholders of the value of nurturing the opposable mind
(Martin, 2007), and most important, such modeling is a how-to method for applying
such mindfulness for achieving sustainable systems.
Simulations of Golf Course Development, Tourism,

and Residents’ Quality of Life
The analysis includes constructing mathematical models (see appendix) reflecting the
propositions in Figures 3 and 4. The model for Figure 3 assumes that no controlling
governance of golf course development – golf course development occurs unabated

during a 33-year time span (going from 0 to 1 to 2 additional golf courses annually
beginning in year 4). The resulting systems output indicates that golf course
development has both positive influences on quality of life (via employment) and
negative influences on quality of life (via degrading the availability and quality of
potable water and decreasing land use for other purposes). The results of running this
simulation include both substantial growth in residents’ quality of life (using lifetime
income estimates as a metric as a surrogate for quality of life) and a dramatic decline
in quality of life in the final four years of running the simulation.
The model for Figure 4 includes a control governing mechanism on golf course
development that affects the number of golf courses operating annually. The metric
for control mechanism includes estimated monetary costs of buying, installing, and
maintaining government-mandated equipment for monitoring and supporting
objectives relating to vegetation, water potability, and residents’ physical health.
The resulting systems output illustrates that one or more variables do not continue
to increase or decline only – the principal that growth and dampening mechanisms
operate on variables in real-life complex systems applied to golf course development.
Explicating or planning for the influence of government regulations on golf course
development serves to identify a sustainable number of golf courses that does not
result in a massive breakdown in the residents’ quality of life.
Figure 5 shows the results for both simulations for the two variables in this
discussion. Such visualization in Figure 5 is to help increase understanding of
likely patterns in the relationships. As Ford (1999, p. 10) recommends, ‘‘Our purpose
is improved understanding, not point prediction.’’ Describing the equations for
several simulations is best to avoid misinterpretation and to decrease the tendency to
advocate one view only in how a system might operate.
Limitations and Future Research
Chapter 16 serves to introduce cause mapping and system dynamic modeling

as useful tools for deepening understanding of how to achieve sustainable golf
tourism in particular, and to show how such tools are useful for understanding
complex, lagged, relationships having feedback loops that occur among interacting
individuals and organizations. Such tools recognize the value of nurturing an
opposable mind view that golf tourism is likely to substantially both benefit and
harm a geographical region in different time periods – if stakeholders do not deepen
their understanding about the important elements that occur in growing and
controlling golf tourism.
Possibly using Figure 2 as a workbench draft for system dynamics modeling of
golf tourism is a step for consideration for future research. Such modeling should
increase the likelihood of amplifying the benefits and dampening the harm of golf
tourism by moving the industry toward a sustainable system.
Residents’ quality of life with
Government regulations of land use 100
Number Residents’
of golf 30 Residents’ quality of life 90 Quality of life
Courses Without government regulations (Lifetime income
of land use 80 net of health costs)
25
70
Golf courses without 60

20
Government regulations
of land use 50
15 Golf courses with
government regulations 40
of land use
30
10
20
5
10
00
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
22 23 24 25 26 27 28 29 30 31 32 33
Time (Years)
Figure 5: Simulation of system responses to number of golf courses in annual operation without and with government control
mechanism.
Systems Thinking and System Dynamics Modeling
353
Advancing Multiple Hypotheses and Multiple Methods Approach
The findings in this chapter using theoretical non-optimizing cause mapping and
system dynamics approaches receive support in a review of prior literature from a
variety of perspectives. Applying a theoretically dynamic optimizing modeling
approach, Johnston and Tyrrell (2005) describe a dynamic model illustrating the
interrelated behavior of tourism-related economic and environmental conditions
throughout time. They characterize fundamental notions of sustainable tourism from
the perspectives of both a profit-maximizing tourist industry and permanent residents
of a tourist community. Their model illustrates findings relevant to the search for
sustainable outcomes and characterizes potential conflicts implicit in different
sustainable and unsustainable paths. ‘‘The model demonstrates that (1) in all but the
rarest of circumstances, no single, universal sustainable optimum is possible; and
(2) a policy that maintains overly pristine environmental quality may be just as
unsustainable – from the perspective of either the tourism industry or residents – as a
policy that causes excessive environmental decay’’ (Johnston & Tyrrell, 2005, p. 124).
Johnston and Tyrrell’s (2005) view about the value of using an optimality
approach suggests greater limitations occur than applying a system dynamic modeling
approach. Because the literature includes a number of industry environment applica-
tions (for a review, see Ford, 1999), because cause mapping and system dynamics
modeling include a nonoptimality approach, and because of problems with acquiring
empirical data to test the assumptions of optimality models, cause mapping and
system dynamics modeling offer greater promise in empirical field research on
sustainable golf tourism. However, the value of theoretically testing optimality
modeling may be substantial, and such a theoretical approach receives support in the
relevant literature on organizational survival (see March, Sproull, & Tamuz, 1991).
Insightful reviews of the sustainability tourism literature (e.g., Briassoulis, 2002;
McKercher, 2003) suggest both the possibility and the necessity of multiple
stakeholders working together and recognizing the usefulness of applying systemic
methods to prevent the ‘‘tragedy of the commons’’ (Hardin, 1968), that is, system
failure because of stakeholders acting individually to maximize only their own
returns rather than acting collectively to maintain the functionality of resources
common to multiple parties.
Applying here the terminology of systems thinking (Gharajedaghi, 2006) relevant
to the present chapter, Hardin (1968) emphasizes the necessity of creating some
functional form of regulation to apply policy design principles (Briassoulis, 2002) to
achieve a sustainable commons. This view supports the proposition that government
has a legitimate role as a direct participant in sustainable tourism – including golf
tourism. In his seminal essay, Hardin stresses that no technical (i.e., technology)
solution to the tragedy of the commons is possible in the long run because of the
limitations of natural resources.
Ultimately, multiple stakeholders need to come together – work together – to
recognize the necessity of adopting systems thinking and to apply systems research
methods to deepen their knowledge about their individual and collective influence
within tourism industries, social, and natural environments. Chamberlin’s (1890)
wisdom into the value of applying the multiple hypotheses method is extendable
by recognizing the value of using multiple methods rather than relying on a single
dominant data collection and one method of analysis. Incorporating multiple
hypotheses and multiple methods paradigm (cf. Campbell & Fiske, 1959) in
sustainable versus unsustainable golf tourism research is likely to provide both
confirmatory and disconfirmatory findings for one or more of the hypotheses – such
multiple evidential outcomes are likely to enhance the building of opposable minds
(Martin, 2007) and to deepen understanding of short- and long-term influences of
golf tourism and all other interdependent elements in the golf–tourism–economic–
social–environmental system. The analysis in Table 2 indicates that economic value
analysis, environmental impact analysis, and possibly other research methods are
unlikely to examine all relevant hypotheses relating to golf tourism.
Economic value analyses often do consider time or multiple decade environmental
impacts as important factors in estimating the influence of golf tourism.
Environmental impact analysis may include only modest consideration of the
short-term economic benefits to local residents of golf course and tourism
development.
A limitation of Table 2 is that an extensive literature review does not
support its presentation. Table 2 makes the point that the method applied frequently
affects the efficacy (e.g., complexity of the findings) and the likelihood that a given
set of findings supports the hypotheses. Table 2 advocates explicit recognition that
multiple methods provide fresh eyes for examining findings when testing multiple
hypotheses.
Conclusions
This chapter illustrates the value of causal mapping and suggests applying system
dynamic modeling for examining inputs and outcomes of golf tourism development.
The use of the multiple hypotheses and multiple methods paradigm is likely to
nurture the development of the opposable mind. Such integrative thinking
processes are more likely to recognize both the economic value of golf tourism
and the need to regulate the industry’s social and environmental impacts – and
subsequently create and maintain sustainable, systematic, and long-term relation-
ships in golf tourism.
Rather than assume that a systems breakdown will not occur in the foreseeable
future and assume that golf tourism critics will continue to have little or no influence,
golf tourism executives and green advocates need to embrace a highly mindful,
systems view of the multiple positive and negative influences of golf tourism
development. The adoption of systems thinking and the adoption of system
dynamics modeling are the operational steps necessary for achieving integrative
versus advocacy thinking. Nurturing integrative thinking is more likely to achieve
sustainable golf tourism and high quality-of-life objectives than is taking an
intractable stance.
356
Table 2: Issues examined by specific methods within a multiple hypotheses and multiple methods paradigm for sustainable
tourism research.
Example hypotheses Economic Environmental Optimization System

value impact analysis modeling dynamics
analysis modeling
1. Golf courses m-tourism m Yes Yes Yes Yes

2. Golf courses m-property development m Yes Yes Yes Yes
3. Tourism m - Employment m Yes No Yes Yes
4. Golf courses m - Herbicides m No Yes Yes Yes
5. Golf courses m - Arable land k No Yes No Yes
6. Golf courses m - Herbicides m - residents’ health/quality-of-life k No Yes No Yes
7. Residents’ health/quality-of-life k -government regulations m No No No Yes
8. Government regulations m - golf courses k No No No Yes

Note: Each ‘‘yes’’ and ‘‘no’’ in Table 2 indicates whether or not the hypothesis is likely to be tested by the method and not that the findings support the
hypothesis. For example, the yes in the first cell is to indicate that economic value analysis includes estimating the impact of increasing golf courses on increasing
the level of tourism.
Appendix
Variable Definition
At Number of golf courses in year t
Bt Number of tourists in year t
Ct Total property (development) value in $millions in year t
Dt Employment in number of persons in year t
Et Residents’ quality of life in estimated annual lifetime income for year t
Ft Herbicides and insecticide usage in $thousands per golf course in year t
Gt Arable land usage in millions of hectares in year t
Ht Potable water usage in billions of liters in year t
It Agriculture sales in $millions in year t
Jt Government regulations/control in $thousands per month for total
number of golf courses for purchases, installation, and maintenance
for environment and water treatment and monitoring in year t
Variable Model without Jt
At Zero for first three years; continuous annual growth of 1 to 2 units to
year 33
Bt ¼ 10000 + (A2*2000) + (A2*10)^2
Ct ¼ 300 + (A2*10)^1.2 + (A2*2)^1.6
Dt ¼ A2 + (0.001*C2*A2)
Et ¼ (E2*0.02) ((G2*0.005)^3.5) + ((H2*0.004)^1.5) + ((J2*0.02)^1.3)
Ft ¼ 50 + (A2*2)^1.6
Gt ¼ 50 (A2*0.2)^1.9
Ht ¼ 5 (A2*0.1) (A2*0.05)^2
It ¼ (H2*1.1)^1.6 + (I2*18)
Variable Model with Jt
At ¼ (1 + At 1) (At*0.0001*(1.1^Jt))
Bt ¼ 10000 + (A2*2000) + (A2*10)^2
Ct ¼ 300 + (A2*10)^1.2 + (A2*2)^1.6
Dt ¼ A2 + (0.001*C2*A2)
Et ¼ (E2*0.02) ((G2*0.005)^3.5) + ((H2*0.004)^1.5) + ((J2*0.02)^1.3)
Ft ¼ 50 + (A2*2)^1.6
Gt ¼ 50 (A2*0.2)^1.9
Ht ¼ 5 (A2*0.1) (A2*0.05)^2
It ¼ (H2*1.1)^1.6 + (I2*18)
Jt ¼ ((((E2*0.04)^2)*At 1) + (3*At 1))
Chapter 17
Fuzzy Set Social Science and

Qualitative Comparative Analysis
Synopsis
Prior reports on theory and research focusing on describing and explaining

national cultural influences on purchase and consumption behavior use a net
effects approach (i.e., theory and analysis relying on main and interaction
effects via statistical analysis). Theory and research in this chapter advances
qualitative comparative analysis (QCA) of a configuration perspective of
culture’s consequences on consumption behavior. This research informs the
view that national cultures represent causal recipes (conjunctions) of cultural
values; the study of main and interaction effects offer meager representations
of national culture’s consequences in comparison to adopting a cultural
configuration stance. The configuration research here includes transforming
Hofstede’s country cultural scores into fuzzy set values and applying Boolean
algebra to estimate the relevancy of alternative cultural configurations for each
of 14 nations to consuming experiences during visits to Australia. The findings
support primary and additional hypotheses that specific cultural configurations
are sufficient (but not necessary) for describing substantial culture’s con-
sequences on consuming tourism experiences. For example, the animus
(i.e., Carl Jung’s unconscious masculine personality-force) configuration —
the combination of high power (P), high individualism (I), high masculine (M),
and low uncertainty avoidance (BU) (i.e., P I M BU) — is sufficient in
indicating not-shopping-for-gifts while visiting Australia. Western national
cultures (e.g., United States) have higher fuzzy set scores than Eastern national
cultures (e.g., Japan) for the animus configuration.
Introduction
National cultures always represent complex conjunctions of values rather than a

collection of distinct individual value dimensions. This chapter show how to use
qualitative comparative analysis (QCA, Ragin, 2008) to consider how cultural
recipes — conjunctive configurations of national cultures affect international
experiential behavior. QCA is applicable for analyzing both small and large sets of
data — that is, data that include ten cases to a 1000 + cases and data that includes a
reasonable number of antecedents (e.g., three to eight) in predicting an outcome
condition (a term analogous to a dependent variable in statistical analysis).
The QCA advocates modeling asymmetric relationships and reporting conditions
that are sufficient (but not necessary) to cause an outcome condition (e.g., high dollar
expenditures and short versus long destination visits). Using measures of consistency
(a measure that is analogous to a correlation) and coverage (a measure of effect size),
QCA provides estimates of how well alternative conjunctive models explain behavior
rather than relying on symmetric-based methods of data analysis (i.e., correlations,
multiple regression, and structural equation modeling).
This chapter describes the influences of gestalt national cultures on outcome
conditions for international experiential consumption. This study supports the
proposition that conjunctively complex causal statements using Hofstede’s country
value indexes do provide high nomological validity in explaining international
tourism behavior. The analysis applies QCA software (fsQCA.com) and the
presentation includes XY plots of country-level value configurations and group-
level consumption for total visit expenditures, length of visit, shopping behavior, and
group touring behavior.
Tylor (1871, in McCort & Malhotra, 1993, p. 97) provides an early definition of
culture, ‘‘the complex whole which includes knowledge, belief, art, morals, custom
and any other capabilities and habit acquired by man as a member of society.’’
Subsequent contributions emphasize the all-inclusive nature of culture as affecting
aspects of human life in a society (Soares, Faranghmehr, & Shoham, 2007).
However, the relevant literature on culture’s consequences on consumption behavior
(see Soares et al., 2006, for a review) rarely includes examinations of how the
‘‘complex whole’’ affects purchasing and consuming behavior; rather, this body of
work focuses on reporting the influences of single dimensions of culture on consumer
behavior. For example, using case (national averages) data on tipping practices for
each of 38 different service professions in 34 different countries for Star (1988) and
culture value scores from Hofstede (1983) for 50 different countries of a large,
multinational industrial corporation and its subsidiaries, Lynn, Zinkan, & Harris
(1993, p. 483) report that tipping to be more prevalent in countries with a high
tolerance of status and power differences between people than in countries whose
tolerance of such differences is lower, ‘‘in this analysis, the power distance index
correlated at .46 ( po.02) with the prevalence of tipping.’’
The present chapter offers examples of taking steps toward proposing and
empirically testing gestalt perspectives of cultures’ consequences on consumption
behavior. The approach is unique in applying QCA (Ragin, 2008) of cultural
multiple-dimensional statements (i.e., conjunctive recipes) for testing cultures’
consequences on purchasing and consuming — working from the complex whole
rather than single-dimension perspective of culture. This study adopts Ragin’s (2008)
view that the analysis of asymmetric set relations is critically important to
social research rather than calculating the net effects of independent variables in
linear models from a symmetric (correlation and multiple regression) perspective.
Fuzzy Set Social Science and Qualitative Comparative Analysis 361
‘‘In set-theoretic work, the idea of a causal recipe is straightforward, for the notion
of combined causes is directly captured by the principle of set intersection’’ (Ragin,
2008, p. 9).
Figure 1 includes visualizations of symmetric (Panel A) and asymmetric
relationships (Panel B). Statistical analyses that apply matrix algebra (e.g.,
correlation and multiple regression analysis) are tests how well data analyses
support theoretical statements of symmetric relationships (i.e., low values of an
‘‘independent variable statement’’ relate to low values of a dependent variable and
high values of the independent variable relate to high values of a dependent variable.
An ‘‘independent variable statement’’ can by a single dependent variable or a
complex statement (model) of several independent variables. Statistical analysis
examines whether or not the data analysis of an independent statement of
one or more independent terms supports a necessary statement of a relationship
(see Panel A).
QCA applies Boolean algebra to test how well data analyses support theoretical
statements of asymmetric relationships (i.e., whether or not high values of an
antecedent condition relate to high values of an outcome condition). QCA does not
make a prediction about the relationship of low values of an independent
condition and an outcome condition because QCA recognizes that multiple routes
can result in high (membership) value of an outcome condition. Thus, QCA examines
whether or not the data analysis supports a statement of sufficiency but not necessity.
Reality in social behavior of individuals and organizations of individuals frequently
support statements of sufficiency but not necessity — a fancy way of stating that
multiple routes occur in getting to a high outcome condition such as new product
success, heavy casino gambling behavior, or very frequent road rage. See Panel B
in Table 1.
This chapter is unique and valuable both in theory and method in proposing and
testing complex-whole proposals of national cultural influences on purchasing and
consuming behavior. For example, using case (national level) data from visitors to
Australia from 14 Asian, European, and North American countries, the study in this
report includes examining the impact on consumer behavior of the American (USA)
icon, the cowboy, symbolizing the cultural recipe of high independence, low
uncertainty avoidance, high masculinity, and low power distance (i.e., all persons
(men) are equal) as well as additional cultural recipes. The findings support the
theoretical and practical importance of examining the complex whole of cultures’
consequences on purchasing and consuming behavior.
The chapter has the following structure. Following this introduction, section two
provides a brief review of research on cultures’ influence on consumption behavior
in which the dominant view in developing and testing theory is the net effects
perspective; section two includes a general theory of cultures’ consequences on
purchasing and consuming products ands services relating to international tourism.
Section three proposes adopting a set-theoretic perspective for advancing theory
and empirical research on cultures’ consequences on consumer behavior; section
three includes a set of specific complex-whole cultural propositions for examination.
Section four describes the method for examining the propositions. Section five
362
Panel A: Panel B:
Dependent Statistical Analysis Assumption Outcome Qualitative Comparative Analysis Assumption
Condition
Variable
High x
High x x x
x x
x x
x
x
x Medium
Medium x x
x
x
x x
Low x Low
Low Medium High Low Medium High

Model Statement of Independent Variable(s) Model Statement of Simple or Complex
Antecedent Condition
Figure 1: Symmetric and asymmetric statements of theoretical relationships.

Table 1: Comparison of iconic and fuzzy set membership scores for simple and
complex antecedent conditions.
Values American Hofstede Hofstede Fuzzy set

iconic world USA membership
cowboy averages value USA
Individualism (I) 1.00 .43 .91 1.00

Masculinity (M) 1.00 .50 .62 0.75
Power distance (P) 0.00 .55 .40 0.25
Uncertainty (U) 0.00 .64 .46 0.20
Not PD (BP) 1.00 0.75
Not Uncertainty (BU) 1.00 0.80
(I M BP BU) 1.00 0.75
Note: Fuzzy set scores are transformations of osted values for the USA. Hofstede’s national value scores
are available at http://www.geert-hofstede.com/. The range possible for fuzzy set scores is from 0.00 to
1.00. The .75 is equal to the lowest value among I M BP BU and represents the causal space that the
four statements share together.
present the findings of the study. Section six presents limitations of the study. Section
seven covers implications for theory and international management/marketing
practice. Section eight provides conclusions and suggestions for future research.
Researh on Cultures’ Influence on Consumption Behavior
Several reviews of the literature of cultures’ influence on consumption behavior

are available. Sojka and Tansuhaj (1995, p. 4) conclude that researchers follow
three approaches in operationalizing culture: through language, through material
goods/artifacts, and through beliefs/value systems. Soares (2005) provides a
thorough discussion on both direct and indirect measures of cultural values in
research on the influence of cultural values on consumer behavior.
Culture reflects ‘‘general tendencies of persistent preference for particular states
of affairs over others, persistent preferences for specific social processes over others,
and general rules for selective attention, interpretation of environmental cues,
and responses’’ (Tse, Lee, Vertinsky, & Wehrung, 1988). Culture is the collective
programming of the mind that distinguishes members of one society form another —
the accumulation of shared meaning, rituals, norms, and traditions among members
of a society (Solomon, 1996). Culture is both a product of actions and includes
conditioning elements of future action (Kroeber & Kluckhohn, 1952). This definition
is somewhat broader than Clark’s (1990, p. 70) statement, ‘‘the working definition of
national character speaks of enduring personality characteristics among the
populations of particular states.’’ Consequently, while the two terms may be used
interchangeably with clarity, national culture is a broader conceptualization than
embraces national character. National culture is inclusive of the dominant value
system, personality, social relationships, and public and private behaviors of a

country’s population. Certainly the following observations about national character
apply as well to national culture:
Common sense requires rejection of the notion that traits are

monolithically persuasive in all individuals in any nation. Rather,
theory suggests a model distribution of traits or a pattern of tendencies
in each nation. (Italics original in Clark, 1990, p. 70)
A substantial body of empirical work supports a core tenant of research on

national cultures: systematic variation across countries exists on the national-cultural
level. For example, Hofstede (1980) reports that, even for countries that are less well
culturally integrated, the different ethnic and/or linguistic groups have important
commonality in culture in comparison to the populations of other countries.
Smith and Schwartz (1997, p. 112) report that cultural differences among samples
from three regions in China, three in Japan, and five in the USA were dwarfed by the
much larger differences between nations. Schwartz and Ros (1995) find across a
sample of 13 countries that nation accounts for about three times more variance in
the ratings of the items used to measure national culture than any within-national
variable examined, such as gender, education, age, and marital status. However,
after reviewing the above studies, Steenkamp (2001, p. 38) still emphasizes, ‘‘The
mutual influences of micro cultures and national culture y needs more attention.’’
This chapter illustrates useful methods of attending to this need.
Nation are useful as a proxy for culture since members of a nation tend to share
a similar language, history, religion, understanding of institutional systems, and a
sense of identity (Dawar & Parker, 1994), making its use a common approach to
operationalize culture (e.g., Dawar & Parker, 1994; Steenkamp et al., 1999; Yeniyurt
& Townsend, 2003).
Triandis (1989) proposes that a country’s culture is a powerful force shaping its
people’s perceptions, dispositions, and behaviors due to strong forces towards
integration in nations, including a single dominant language, history, educational
system, political and legal system, and shared mass media, market, services, and
national symbols (e.g., flags and sports teams). Hofstede (1980, 2001) identifies four
(later, five) core dimensions of national culture that may influence consumer
predispositions (i.e., unconscious thinking, see Bargh, 1989, or ‘‘mental program-
ming,’’ see Hofstede, 2001), strategic thinking/deciding processes, and behaviors:
individualism/collectivism, uncertainty avoidance, masculinity/femininity, power
distance (and a fifth, the Confucian dynamic of long-term versus short-term
orientation). Schwartz (1994, 1997) proposes three dimensions of universal human
values (conservatism versus autonomy, hierarchy versus egalitarianism, and mastery
versus harmony).
Despite a conceptual and empirical development in national culture literature, the
delimitation of culture to the nation-state, and the deconstruction of culture as three
or four universal value dimensions may raise some potential issues as follows. First,
researchers are fully aware of the existence of multiple levels of cultures. This view
clearly shows that culture and country are not meant to imply the same concepts:
Culture can be conceptualized at different levels, including the national

level (Dawar & Parker, 1994). Thus, studying the effects of national
culture requires that there exist some meaningful degree of within-
country commonality and between-country differences in national
cultures (Dawar & Parker, 1994). Conceptual (Hofstede, 1991, 2001;
Schwartz, 1994) and empirical (Hofstede, 1980; Smith & Schwartz,
1997) evidence indicates that this is indeed the case. (Steenkamp et al.,
1999, p. 56)
Second, most people belong to different levels of culture at the same time, and
such multiple memberships complicate the identification and understanding of
national-cultural influence on consumer behaviors. While stereotyping is commonly
seen as a means to categorizing information to bring order to our environments (e.g.,
Barna, 1994; Fiske, 1993), a study of national cultures must recognize this overall
limitation on the capacity to generalize the results and findings, as well as possibly
criticism for ethnocentrism or discrimination (Barna, 1994; Clark, 1990; Peabody,
1985). The current study contributes knowledge enhancement in national culture
through exploring micro-cultural impact on national culture, and moderating role
of micro culture on national culture and international consumer behaviors.
Lastly, overemphasis on the universal or dominant national-cultural dimensions
(e.g., Hofstede, 1980, 1991; Schwartz, 1994, 1997) may result in not only
oversimplification of complex or multifaceted nature of culture but also support
ignorance of variant values across nations. However, while recognizing the limitation
of dominant-value-based national-cultural study, universally valid dimensions of
national culture plays a significant role in advancement of theoretical rigor in
national culture and cross-national research.
Values represent central goals that relate to all aspects of attitudes and behaviors
(Schwartz & Bilsky, 1987); Values not only transcend specific actions and situation,
but also serve as standards to guide the selection or evaluation of behavior, people,
and events (Smith & Schwartz, 1999). The two major virtues of the concept of
value-its relative abstraction and generality- well reflect the key feature of national
culture/character- mental programming (Hofstede, 1980, 1991) and enduring
personality (Clark, 1990).
Valid frameworks delineating dimensions of national cultural variations are
also crucial in creating a nomological framework that is capable of integrating
diverse attitudinal and behavioral phenomena and provides a basis for developing
hypotheses explaining systematic variation between cultures in attitudes and
behaviors (Smith, Dugan, & Trompenaars, 1996). Such frameworks are necessary
for international marketing researchers to the formulation of cross-cultural general-
ization beyond exploratory, qualitative comparisons, which are difficult to validate
or replicate (Steenkamp, 2001). Additionally, national cultural dimensions show a
significant relevance on cross-national consumer behaviors (e.g., Dawar & Parker,
1994; Lynn, Zinkan, & Harris, 1993; Pizam & Sussmann, 1995; Roth, 1995;
Steenkamp et al., 1999).
Reviewing National Culture Theoretical Frameworks
Relevant literature pays substantial attention to developing universally occurring

national cultural frameworks (see Hall, 1976; Hofstede, 1991, 2001; Schwartz, 1994,
1997). More significantly, although unified dimensions have not been reached,
converging domains of national culture are examined by both conceptual (e.g., Inkeles
& Levinson, 1969; Kluckhohn & Strodbeck, 1961; Schwartz, 1994, 1997), and empirical
approaches (Hofstede, 1980, 1991; 2001; Peabody, 1985; Schwartz, 1994, 1997).
Hofstede (1980, 1991, 2001) and Schwartz (1994, 1997) provide two rigorous
comprehensive frameworks that are used for cross-national theorizing and for designing
studies. In particular, the Hofstede’s framework has been noted as a ‘‘watershed
conceptual foundation for many subsequent cross-national research endeavors’’
(Fernandez, Carlson, Stepina, & Nicholson, 1997, pp. 43–44). In addition, Sivakumar
and Nakata (2001) report 1,101 citations to Hofstede’s work in during 1987–1997.
Hofstede’s framework has been criticized due to lack of correspondence between
the measurement items and conceptual definitions in the cultural dimensions. In
comparison, Schwartz framework’s provides a close match between the national-
cultural domains and their strong theoretical foundations (Kagitc- ibasi, 1997;
Steenkamp, 2001). Hofstede derives and defines four dimensions of national
culture in particular, based on the combination of empirical and eclectic analyses.
However, Clark (1990) highlights that the core four dimensions of the Hofstede
(1990, 1991)’s national culture framework show considerable similarities with other
studies, especially Inkeles and Levinsion’s (1967) theory driven dimensions.
Hofstede’s dimensions are identified based on the country-level approach. Schwartz’s
dimensions are, on the other hand, the derivation of the cultural domains/
dimensions, and subsequent possibilities of limited conceptual contrast at the
country-level (Kagitc- ibasi, 1997; Steenkamp, 2001).
With respect to sampling issues, Hofstede provides indexes and rankings for the
four dimensions of national culture, based on the matched samples of 117,000 IBM
employees in 50 different nations (Hofstede, 1980, 1991). Country scores and
rankings based on matched samples question possibility of samples’ influence on
findings. While the IBM sample is not representative of the various countries’
populations and the employees studied were from the countries’ middle class rather
than from their upper or lower classes, the focus here is on differences between
countries and not absolute scores:
Hofstede’s sample is good for this purpose because the respondents

from different countries were well matched on variables such as
corporate culture, occupation, education, sex, and age; their only
systematic differences were in nationality. Moreover, the validity of
these data is attested by the fact that they are significantly related to a
variety of other theoretically relevant, country-level data (see Hofstede,

1980) for a review. (Lynn et al., 1993, p. 482)
Schwartz (1994) also highlights the necessity of a sample matching strategy in the
absence of representative random samples across various nations. National scores on
the Schwartz’s dimensions are largely based on teacher and student samples across
32 nations.
Another area of concerns from Hofstede’s framework is questioning of stability of
country score and index, as data collection took place in 1967–1973. Hoppe (1990)
conducted an update for 19 countries in 1984 and found a reasonable stability
(correlation varying between .56 and .69). This finding suggests that Hofstede’s
ordering of nations was not determined by the IBM sample, and that this ordering
remained fairly stable over several decades. Hoppe’s findings also support the
stability of value systems. Furthermore, Hofstede’s four cultural dimensions are
accepted widely and have been used by many marketing researchers to locate and
compare countries (e.g., see Dawar & Parker, 1994; Iverson, 1997; Lynn et al., 1993;
Money & Crotts, 2003; Nakata & Sivakumar, 1996; Pizam & Sussmann, 1995; Roth,
1995; Steenkamp et al., 1999).
The conceptual framework of this study that this chapter reports is based on four
dimensions of Hofstede’s framework due to its conceptual merit and empirical
support. Kagitc- ibasi (1997, p. 11) notes, ‘‘Hofstede’s framework is still the most
comprehensive comparative study in terms of both the range of countries and
the number of respondents involved. More importantly, the four dimensions of
Hofstede’s framework are well suited for the development of integrative theory
for international consumer behavior research as they parallel areas of traditional
concerns in the explanation of consumer behavior (see Clark, 1980, p. 73, for details).
Additionally, the usefulness of the Hofstede’s framework in international marketing
is well established, whereas Schwartz’s framework has yet to be applied widely.
However, the use of Hofstede’s framework and cultural value indexes for nations
does have limitations. While prior research reports examine the independent effects
of national cultural values in management and consumer behavior using Hofstede’s
country value indexes and do so with some success (e.g., Lynn, Zinkan, & Harris,
1993), other research criticizes the accuracy and usefulness of Hofstede’s scoring and
country index scores (Bearden, Money, & Nevins, 2006; McSweeney, 2002).
General Theory of Cultures’ Consequences on International Tourism Behavior
Figure 2 summarizes a general theory of cultures’ consequences on purchasing and

consuming products and services relating to international tourism behavior. While
the theory includes propositions of substantial direct and indirect influences of each
of four dimensions of national culture on international consumer behavior, the
theory focuses on proposing the influence of configurations (i.e., causal recipes)
of cultural dimensions on international tourism behavior. Figure 2 serves as a
368
1. National Cultural Dimensions 3. Consumption Moderating Variables

and Some Causal Recipes • Prime Motive for Trip: Holiday Only versus VFR
• Prior Visit Experience
I: Individualism/Collectivism
U: Uncertainty Avoidance 3
P: Power Distance 2 5. Consumption Behaviors
M: Masculinity 6 (i.e., Behaviors in Australia
· = and During Current Visit)
~ = not 4 1
• Travel Party Composition
I I·U U
5 • Search and Use of
4. Pre-Consumption
Information while in Oz
I·P I·U·P·M U·M
Cognitions and Behavior
9 • Nights Stayed in Oz
• Total Number of Distinct Trip
P·M
• Decide on a group tour versus
P M Activities
FIT (freely independent travel)
• Number of Oz States Visited
~(I·U·P·M) • Amount of Pre-Trip Planning
• Trip to the “Outback”
• Information Search:
during Stay
total sources used • Oz Wine-related Touring
• Information Search:
• Golfing in Oz
types of sources used
• Visiting local attractions/
7
“tourist activities”
2. Subculture Dimension • Arts/Heritage (e.g., Theatre)
• Total Expenditures per Day
• Age • Total Expenditures for
8 Shopping for Goods (e.g., Gifts)
to Take Home
Figure 2: General theory of cultures’ consequences on international tourism.

conceptual framework for the study and includes a macro summary of the direct
(shown by arrows labeled 1 and 2) and indirect influences (arrow 9) of four
dimensions of national culture. Causal recipes of national culture are hypothesized
to be substantial only for first-time consumers visiting an international destination
(e.g., visiting Australia on a holiday only trip — not visiting friends and/or relatives
(VFR). (‘‘Holiday’’ refers to leisure travel that includes traveling during national
holiday periods and all other vacation periods. The survey instrument used includes
both ‘‘holiday/vacation’’ to avoid possible confusion between the British and
American use of ‘‘holiday.’’) Repeat and VFR consumption behaviors are likely to
reduce the impact of national culture on consumption plans and behaviors — arrows
3 and 4 in Figure 2 serve to indicate these propositions; repeat visitors and visitors
seeing friends and relatives are likely to be more acclimated to local contexts than
first-time visitors.
Individualism/collectivism Individualism/collectivism (IC) pertains to the degree to

which people in a country prefer to act as individuals rather than as members of a
group. In collectivist countries, close-knit social structures form, whereby people
expect their group to care for them in exchange for unwavering loyalty. In
individualistic societies, the social fabric and group norms are much looser: people
tend not to follow social norms, but rather make decisions and initiate behaviors
independently of others (Roth, 1995). Triandis declares that IC as ‘‘perhaps the most
important dimension of cultural difference in social behavior across diverse cultures of
the world.’’ While IC may be worthy of independent examination, the dimension has
been found to relate with other dimensions of national culture (e.g., r ¼ .67, po.05,
with power distance, for a review see Kagitc- ibasi, 1997). Thus, given the limited
number of nation states in the world, locating countries in all theoretically possible,
high–low, combinations is unfeasible. However, considering all possible combinations
is useful for both theory and empirical work because each combination of high–low
levels of attributes is a specific location in theoretical ‘‘property space’’ (Lazarsfeld,
1937); each of these locations, in turn, may constitute a different type.
If a relatively small number of combinations exist empirically, then the researcher
will be able to reduce a multidimensional property space in a handful of categories.
Lazarsfeld (1937, pp. 127–128) calls this simplification a functional reduction.
Stinchombe’s (1968, p. 47) remarks on typologies echo the idea of functional
reduction; he states that a ‘‘typology is a statement that a large number of variables
have only a small number of combinations of values which actually occur, with all
other combinations being rare or nonexistent. This results in a radical improvement
in social scientific theory’’ (Ragin, 2000, p. 79).
Applying the main tenants of IC leads to the following propositions. Deciding on
visiting via an organized group tour occurs less frequently among people from
nations high in individualism; contracting travel/visit arrangements using a group
tour is more popular among people from nations high in collectivism. Rationale:
group touring to and in a destination is one means to maintain social approval within
one’s own primary group (i.e., the group decided and made the trip); such group
touring may trigger reactance (Bem, 1970) as a violation of personal freedom among
people from individualistic nations.
Pre-trip information search using impersonal, commercial, sources is higher
among consumers from nations high in individualism; pre-trip information search
using personal, noncommercial, sources is higher among consumers from nations
high in collectivism; the number of different external information sources used
is greater in nations high versus low in collectivism. Rationale: in individualistic
societies people tend to make decisions and initiate behaviors independently of
others; seeking help and approval in making decisions is part of the social fabric of
collectivist, but less so of individualistic nations.
The composition of the immediate travel party more frequently includes two or
more persons among people from nations high in collectivism and more frequently
one person among people from nations high in individualism. Rationale: gaining
social approval for the trip may be realized by bringing additional family members
and friends on the trip and such approval is likely valued more in nations high in
collectivism versus individualism.
The time away from a consumer’s home nation is less among visitors
from countries characterized by high collectivism versus countries characterized
by high individualism. Rationale: loyalty to close-knit groups requires frequent
presence in the relevant groups, as such, lengthy overseas travel experiences are
valued less in nations high in collectivism. Greater shares of visitors from
nations high in collectivism shop for products to take home and their total
expenditures for goods to take home are greater compared to visitors from
nations high in individualism. Rationale: gift-giving to primary group members
remaining at home is one way to maintain/gain social approval, a need more likely
activated among people from nations high in collectivism but not high in
individualism.
Uncertainly avoidance Uncertainty avoidance (UA) refers to the degree that a

society feels threatened by uncertain, risky, ambiguous, or undefined situations
and the extent to which they try to avoid such situations by adopting strict
codes of behavior. In countries where uncertainty avoidance is high, a feeling of
‘‘what is different is dangerous prevails’’ (Hofstede, 1991, p. 119). Conversely, in low
uncertainty avoidance societies, the dominant feeling is more likely to be, ‘‘what is
different is curious’’ and worth exploring (Hofstede, 1991, p. 119). Consequently, the
general theory includes the following propositions.
Deciding on visiting via an organized group tour occurs less frequently among
people from nations low in UA; contracting travel/visit using a group tour is more
popular among people from nations high in UA. Rationale: group touring to and
in a destination is one means to reduce unforeseen risks (i.e., physical and
psychological) mainly due to the pre-planned trip itinerary to and in a destination by
profession organizations (i.e., tour operators and travel agents).
The amount of pre-trip planning is higher among visitors from high UA nations.
Rationale: through trip planning behavior serves reducing discomfort or risks with
upcoming leisure travel. The average number of different information sources
searched is higher among visitors from nations high in UA compared to nations

low in UA. Search and use of tourism information during visit is more frequent
among visitors from high UA versus low UA. Visitors from nations low in UA visit
more places, including more often visiting places infrequently visited, such as the
Australian Outback, compared to visitors from nations high in UA. Rationale:
experiencing places where comparatively few people have been is often highly valued
by people from nations low in UA while avoiding experientially unknown, and
possibly dangerous, places is highly desired by people from nations high in UA.
The frequency of hotel accommodations versus campgrounds, motels, and other
types of accommodations is higher among visitors from high versus low UA nations.
Rationale: experiencing professional services during a leisure trip increases security
and comfort at holiday destination.
Power distance Power distance (PD) is the extent to which a society accepts the fact
that power in institutions and organizations is distributed unequally. Among other
things, high PD reflects acceptance of hierarchical power structures, a perception
of differences between superior and subordinate, and a belief that power holders
are entitled to privileges; low PD reflects the opposite mental programming toward
power (see Hofstede, 2001).
Upon reflection about how certain leisure touring behaviors may associate
with PD lead to the following propositions. Shorter visits are more prevalent in
countries high versus low in PD. Rationale: successfully occupying a status rank in a
hierarchical power structure requires frequent physical presence to continually affirm
such distance while societal ties are weaker in countries low in PD and consequently,
trips longer in length are more acceptable. Participating in shopping behavior and
total expenditures on shopping are higher among visitors from countries high versus
low in PD. Rationale: Gift buying, a behavior included in shopping, is more highly
valued in countries high versus low in PD because gift giving-receiving helps to
acknowledge/reinforce hierarchical status/power positions — a consequence valued
highly in nations high versus low in PD.
Visiting well-known local attractions, engaging in organized tourist activities
(e.g., guided tours and wine-country tours), and golfing during the visit occur
more frequently among visitors from high versus low PD nations. Rationale: such
behaviors enable visitors to share with others that they actually engaged in well-
known, unique and valued, destination-linked activities that most people back home
may never get to experience — such reporting ability and practice serves to reinforce
a high status ranking, a consequence particularly valued in high versus low PD
nations. For example, golf outings to uniquely experienced, overseas, golf courses
may serve to enhance a golfer’s status in discussions with fellow golfers back home.
The frequency of hotel accommodations versus campgrounds, motels, and other
types of accommodations is higher among visitors from high versus low PD nations.
Rationale: experiencing professional services during a leisure trip, such as hotel
services, serves to demonstrate status ranking — a consequence valued higher among
visitors from high versus low PD.
Masculinity Masculinity (M) is the extent that the dominant values in society are
‘‘masculine’’ (i.e., assertiveness, the acquisition of money and things, and not caring
for others, the quality of life or people). High scores on Hofstede’s M index associate
with male dominance, and an emphasis on achievement, independence, and money.
Low scores associate with fluid sex roles, equality between the sexes, and an emphasis
on service, interdependence, and people. With the exclusion of Japan as a statistical
outlier in their analysis, Lynn et al. (1993) report a positive association between
service tipping behavior and M (r ¼ .47, po.02).
Several leisure travel activities are likely to have traits that characterize M. The use
of external information sources (versus unaided judgments) in planning the trip is less
among visitors from high versus low M nations. Rationale: independence is exhibited
to oneself and others by not relying on information from external sources — personal
independence is highly valued in high versus low M nations. Travel party size is
smaller among visitors from nations high versus low in M. Rationale: less caring for
others may be expected to translate into less need to travel with others (e.g., with
friends and family members).
Total number of leisure activities and destinations/states visited are likely to be
less among visitors from nations high versus low in M. Rationale: acquisition of
money and things (e.g., materialism) may be relate negatively with engaging leisure
activities and destinations and states visited during a leisure travel due to experiential
and intangible nature of overseas leisure travel.
Engaging in active outdoors and sports activities occurs more frequently among
visitors from high versus low M nations. Rationale: such activities demonstrate
personal achievements (e.g., white water rafting and scuba diving) that are
consequences particularly highly valued by visitors from high versus low M nations.
Two Behavioral Moderating Variables: VFR Travel and Repeat Visitation
Figure 2 illustrates (arrows 3 and 4) two trip-related variables moderating the

influence of national culture on overseas, leisure-travel, consumption behaviors:
VFR travel and repeat visitation. Visitors traveling primary to visit friends and
relatives are likely to receive suggestions on what to do and not do, and places and
events to avoid, often along with free accommodations during their visits that
serve to reduce the impact of national culture on length of stays. Consequently, (H5a)
the effect sizes of hypothesized associations among national cultural dimensions and
leisure trip behaviors are reduced for VFR compared to holiday only trips among
first-time visitors.
Because prior visits serve to acculturate visitors as to socially acceptable and
unacceptable, as well as financially rewarding and punishing, behaviors, the influence
of national cultural dimensions on consumption behaviors may be expected to be less
detectable among repeat versus first-time visitors. Consequently, the effect sizes
relationships among national cultural dimensions and leisure trip behaviors are likely
to be low for repeat versus first-time holiday-only travelers.
Thus, the theory predicts substantial effect sizes of national cultural dimensions
on travel behaviors among first-time leisure visitors who are not visiting friends and
relatives. The theory predicts non-substantial effect sizes for the national cultural
dimensions for the other three groups: first-time leisure travelers primarily VFR, and
repeat VFR, and repeat holiday-only visitors.
The Direct and Moderating Influences of the Youth and Senior Micro Cultures
Marketing strategists sometime view older consumers (e.g., over-50s) as consumers

set in their ways and a market segment best to ignore. However, the share of older
consumers in developed nations is growing from 12% in 1950 to over 20% in 2002.
In developed nations the over-50s own three-quarters of all financial assets and
account for half of all discretionary spending power. ‘‘Free time and [good] health,
combined with relative financial comfort and a greater readiness for self-indulgence,
are creating a mature market eager to consume and explore’’ (‘‘Marketing to the
Old,’’ 2002). Which view is more accurate among older consumers who do
some exploring (i.e., overseas traveling): do consumption behaviors among older
consumers reflect national culture theory-based predictions more or less closely than
younger and middle-aged consumers?
The proposition that ‘‘as consumers move through their life cycle, they undergo
predictable changes in values, lifestyles, and consumption patterns’’ (Mowen &
Minor, 1998) complements the view that youth is a time of exploration and to rebel
against the views and behaviors of parents (scenes from the movie, Rebel Without a
Cause may come-to-mind here). Is it true that in later stage in life that we become our
parents in the ways we see things and in our behaviors? Are members of older versus
younger generations more likely to match behavior patterns theoretically predicted
from their respective national cultures? If age is an important moderating variable on
the influence of national culture on overseas, leisure travel behavior, then the
differences among young travelers from nations varying in their national cultures
may be expected to be less compared to older travelers.
As Figure 2 depicts (arrows 5 and 6) a substantial age effect likely exists in
international leisure travel, for example, young travelers (1) prefer FIT (freely
independent travel) versus group tour, (2) stay more nights in the destination
country, (3) do more activities, (4) visit more destinations and states, and (4) spend
less per day compared to middle-aged and senior travelers. Rationale: younger
versus older travelers are likely to perceive more freedom to be away from home
and they often have less family and occupational obligations compared to older
travelers. Energy levels and desire to do more varied activities are likely to be higher
among younger versus older travelers while their incomes and available spending
power are lower.
A substantial age by national culture interaction effect may exist in international
leisure travel; specifically, differences in behaviors due to national cultures are less
among young versus older travelers. The findings for the visitors to Australia data do
High
Leisure Travel
Behavior
(e.g., Length of
Destination Stay) High Individualistic
Nations
Moderate
High Collectivist
Nations
Low
Youth (< 30) Middle (30−50) Seniors (> 50)
Figure 3: Age subculture influences (Theory). Note: Figure 2 depicts a significant

lopsided ‘‘U’’ shaped age main effect that includes the highest average length of
destination stays among young leisure travelers and the lowest among middle-aged
leisure travelers. Figure 2 depicts a significant age by national culture interaction
effect, for example, the interaction prediction includes an increase in length of
destination stay for seniors compared to middle-age travelers greater among nations
high in individualism versus nations high in collectivism. Figure 2 also shows a
significant main effect for the individualism/collectivism national culture dimension.
not support this proposition. In Figure 2 arrows 7 and 8 summarize the macro
proposal that age moderates the national culture impact on travel behavior.
Figure 3 depicts the more specific proposal that age moderates the impact of
national culture on number of nights spent in Australia among leisure, pure holiday
(not visiting friends and relatives) travelers. The prediction reflects Levitt’s (1983)
widely debated view that markets are becoming more similar globally in customers’
purchasing behavior toward specific brands. If so, the purchase behavior regarding
Australia as a travel destination brand should be more similar among young versus
older visitors across many national cultures.
Set-Theoretic Perspective for Advancing Theory and Empirical

Research on Cultures’ Consequences
Combinations of national culture dimensions, micro cultural, and prior behavior
contribute a sense of the whole in understanding national culture influences on
consumer behavior; such knowledge improves understanding of behavior that relates
to national cultures above the understanding of single cultural-dimension studies.
Nations consists of combinations of dimensions of national culture —’’ type
concepts’’ (Lazarsfeld, 1937) involve sets of attributes that make sense together
as a unitary concept. Social scientists often use type concepts without analyzing
them and instead focus on analyzing individual dimensions across types (e.g., Lynn
et al., 1993).
Considering three level (high, medium, and low) for each of four value
dimensions, Figure 4 presents an example of a full typology of cultural types that
builds from Hofstede’s first four dimensions. In Figure 3 each of the 81 combinations
of attributes constitutes a ‘‘property space’’ (Lazarsfeld, 1937).
Figure 4 summarizes 81 theoretically possible nations that reflect unique
combinations of levels of national cultural dimensions. In the behavioral sciences,
because systematic versus random differences in dependent measures due to
psychological and social independent variables often become clear only for
respondents scoring very high versus very low on the independent variables
(related to this point, see Bass, Tigert, & Lonsdale, 1968; Gladwell, 1996;
McClelland, 1998; Ragin, 1987), Figure 3 reflects the suggestion of performing
qualitative comparative analyses of nations using the following levels for each of the
four national cultures: the 20% ‘‘very high’’ versus the 60% high-moderate-low
versus the 20% ‘‘very low.
Lazarsfeld (1937, pp. 127–128) notes that if a relatively small number of combina-
tions exists empirically, then the researcher is able to reduce a multidimensional
property space to a handful of categories — he names this simplification a functional
reduction. Stinchcombe (1968, p. 47) stresses that functional reductionism results in
a radical improvement in understanding social phenomenon and in developing social
scientific theory.
Sivakumar and Nakata (2001) note the difficulty in controlling for possible
confounding influences in a cross-cultural study that examines only one dimension of
national culture. The most extreme analysis would be comparing measures for
dependent variables for nations scoring very high in individualism (cell 1 in Figure 4)
versus nations scoring very high in collectivism (cell 81 in Figure 4). Assuming that
substantial differences do occur in the dependent measures, the finding observed
cannot be attributed to anyone focal dimension of national culture since nations
in cell 1 are very high and nations in cell 81 are very low in all four dimensions.
Estimating differences using dependent responses for nations in cell 1 versus cell 55 is
one theoretical example of a comparison of nations scoring very high versus very low
on individualism that controls for the other three major dimensions of national
culture. Because national culture index scores are available only for a limited number
of the total nations (50 of approximately 175 recognized nations) and because
significant associations across nations occur for individualism, uncertainty avoid-
ance, and power distance, in practice less than half of the cells in Figure 4 categorize
one or more nations; the rest may be empty empirically but are useful for reflection
and for theory development.
Sivakumar and Nakata (2001) identify sets of nations for single culture and two
cultural dimension comparisons that provide some control for non-focal cultural
dimensions. However, the attempt to achieve both high difference scores for a focal
dimension as well as low difference score for non-focal dimensions moves
the researcher away from nations ranked first or last for any one cultural dimension.
For example, the country ranked first in masculinity (i.e., Japan) is not found in
376
Core Dimensions Uncertainty Avoidance

Very High (1) High/Medium/Low (2) Very Low (3)
Individualism Power Distance Masculinity = 1 2 3 1 2 3 1 2 3
1 7 9
Very High (1) Animus
High/
CAN NET US
Very High Medium/ (2)
Low
AUS,
NZ
UK
Very low(3)
Very High 28 MAL 36

High/
Medium/ High/
HK,
Low Medium/ JP SG
Low
Very low G, SW
Very High 55 IND 63

Very Low High/

Medium/ KOR,
TW
Low
75
Very low 73 Anima 81
Key. CAN = Canada, G = Germany, HK = Hong Kong, IND = Indonesia, KOR = Korea, JP = Japan, MAL = Malaysia, NET = Netherlands,
NZ = New Zealand, SG = Singapore, SW = Switzerland, TW = Taiwan, UK = United Kingdom, US = United States
Figure 4: Classifying 14 nations by four dimensions of national culture.

Sivakumar and Nakata’s country pair comparisons for masculinity controlling for
non-focal dimensions.
Possible Confounding Effects and a Quasi-Experimental Design
Pondering the need to compare nations very high versus very low in dimensions of
national culture and the high likelihood of confounding possibilities leads to
applying Cook and Campbell’s (1979) suggestion to examine multiple comparisons in
theory testing for ruling out alternative explanations, as well as to compare any sets
of nations whereby a maximum difference occurs for one dimension only. Figure 3
depicts the classification of nations using the described 20-60-20 segmenting rule
for Hofstede’s (1980) index scores for 14 nations having substantial numbers
of respondents visiting Australia in the study described below. Note that one set of
nations does differ for one dimension only: Canada being very high and Korea and
Taiwan being very low on individualism with all three nations scoring in the large
mid-range for the other three cultural dimensions.
From a dichotomous perspective (rather than a identifying three levels for each
dimension), a total of 3k – 1 logically possible groupings occur for causal conditions.
Thus, for four national culture dimensions, a total of 80 groups are possible; these
grouping include 8 single aspect groupings (e.g., high versus low in individualism),
24 two aspect groupings, 32 three aspects groupings, and 16 four aspect combina-
tions. Ragin (2000, p. 125) notes that by evaluating the comparability of the cases
conforming to each configuration, a researcher can make a preliminary assessment of
the adequacy of the aspects selected for investigation. For example, the configuration
I U P BM indicates the combination of high individualism (I), high uncertainty
avoidance (U), high power distance (P), and low masculinity (M). Note that midlevel
dots indicate combinations of characteristics and the ‘‘B’’ preceding the attribute’s
name indicates negation.
A fuzzy set perspective recognizes that an individual case (e.g., country) is not
fully in or fully out of any of the sets that make up a property space — the case has
partial membership in all crisply defined property-space locations. Given Hofstede’s
scores (transformed to fall between 0 and 1) for four dimensions of national
cultures by country, a very close correspondence can be established between fuzzy
membership scores (ranging from 0 to 1) rather than using dichotomous scales.
Using Boolean algebra the maximum score across the dimensions defines a country’s
membership score for a specific property space in two or more dimensions. For
example, a country with fuzzy sets scores of .5, .7, .2, and .9 for I, U, P, and M,
respectively, has a score of .1 for I U P BM, since BM ¼ 1.0 0.9 ¼ .1 and .1 is
the maximum score for this country for any one dimension in combination with the
scores on the other three dimensions.
Note that Figure 4 identifies two property spaces as animus (cell 7) and anima
(cell 75) locations. Our conceptualization of Jung’s concept of animus here refers to
the largely unconscious cultural position of high masculinity, high individualism,
Animus:
I·M·P·~U
Idealized American I I·P P

(Cowboy) Configuration:
I·M·~U·~P I·P·U·~M
I· P ·M ·U P·U
M·~(I·U·P) M·U Anima:

U·~(I·M·P)
Figure 5: Summary of three value systems using a Venn diagram.
high power distance, and low uncertainty avoidance (‘‘I am a man who does need
anyone, I am powerful physically, and I am afraid of nothing’’). Anima here refers to
the largely unconscious cultural position of low masculinity, low individualism, low
power distance, and high uncertainty avoid (‘‘I am woman with children who need
me, I am weak physically, and I wish to avoid being in harm’s way’’). The following
conjunctive statements represent the causal recipes for these two cells:
Property space 7, animus: M I P BU

Property space 75, anima:BM BI BP U.
The American cowboy icon comes close to the animus recipe though the argument
may have merit that this cowboy represents low power distance as a reflection of the
American ideal, ‘‘all men are equal.’’ In support of the view that the characteristics of
the American cowboy reflect the animus recipe, the property space of the United
States is very close to cell 7 in Figure 3. Figure 5 visualizes the ideal American
cowboy, animus, and anima property spaces.
The causal recipe statement I M BU BP represents the iconic American value
system — the configuration of total membership in individuality and masculinity and
total non-membership in uncertainty avoidance and power distance. This complex
antecedent condition equals 1.00. Table 2 includes the estimates for this iconic view
of the American value system based on a fuzzy set transformation of Hofstede’s
index values for the United States. The fuzzy set value for the USA equals .75 for the
causal recipe, somewhat lower than value of 1.00 but closer to the iconic estimate in
Table 2: Fuzzy membership in the set of long-visit behavior.
Number of nights in Australia Membership (M) Verbal label
45-365 M ¼ 1.0 Clearly a long visit

35-44 .5oMo1.0 More or less a long visit
26-34 M ¼ .5 In between
11-25 .1oMo.4 More or less not a long visit
0-10 M¼0 Clearly not a long visit
comparison to other countries. Hofstede’s national value scores are available at

http://www.geert-hofstede.com/.
The United States has the highest individualism index value (.91) among the
550 countries in Hofstede’s set of cultural values. Given that .91 is the highest value,
the fuzzy set for individualism includes a transformation of this value to 1.00. The
reader might wish to convert Hofstede’s index scores into fuzzy set scores; in doing
so, fuzzy set value for a country should be greater than .50 when a country’s score is
greater than the average score for a given cultural value. The fuzzy set value for a
country should be less than .50 when a country’s Hofstede score is below the average
score for a given cultural value. The fuzzy set value assignments in Table 2 are
applications of these heuristics.
Using fuzzy set scores that range from 0.00 to 1.00 enables the testing of the
relevancy of animus, anima, and the full theoretical range of causal recipes/
combinations of influence on international tourism behavior.
Method
Data and Procedure
The data for examining the general theory propositions was purchased from the
Australian Bureau of Tourism Research as part of an on-going annual study of
international visitors to Australia (i.e., the ‘‘International Visitor Survey’’ and
‘‘Supplementary Survey’’). The Australian Commonwealth and State/Territory
governments sponsor this survey annually. Data acquisition was supported for
the study by a grant awarded by the Cooperative Research Centre for Sustainable
Tourism in Australia.
Data were collected using exit interviews of visitors to Australia leaving on
commercial flights. A quota-sampling plan was followed to achieve a representative
sample of all flights from all Australian international airports. Thus, the greatest
share of interviews was completed at the Sydney Airport; however, representative
samples of interviews proportionate to traffic volume were completed at all seven
Australian international airports. The findings in this chapter are based on data
(n ¼ 3,651) from the first quarter of 2000.
All interviews were completed face-to-face with an interviewer and an adult

member of each travel party in the departure lounges for specific flights.
A professional marketing research firm (A C Nielsen) was hired to collect the data
for the study. The interviewers were native speakers in the language used during the
interviews; the interviewers work full-time for A C Nielsen.
The cooperation and completion rates were above 90 percent for respondents
requested to participate in the survey. Several additional factors were considered
to insure data collection representative of visiting travelers; for example, the
substantial majority of travelers arrived in the lounge area more than one hour
before the departure of their flight and care was taken to interview early, middle, and
late arrivals; interview quotas were used for travelers by travel party size that were
representative of flight population data.
Country Selection
A two-step procedure was used to classify respondents by national culture. First,

early in the interview each respondent was asked to name his or her country of
residence. Second, the following question was asked near the end of the interview,
‘‘What is your first language, the language that you speak most at home?’’ To be
included within a specific national culture, for example, Korean, the respondent had
to be residing currently in the same nation of the language the respondent spoke most
at home, for example, a ‘‘Korean’’ was defined to be someone living in Korea who
spoke the Korean language most often at home. Using this screening requirement
resulted in more than 90 percent useable completed surveys for each of the 14 nations
examined.
Survey responses from the 14 nations with the largest shares of visitors to
Australia were selected for analysis. The 14 include seven Asian nations: Korea,
Hong Kong (a unique economic/political city-state within the PRC), Indonesia,
Japan, Malaysia, Singapore, and Taiwan. Survey responses from four European
countries were analyzed: Germany, the Netherlands, Switzerland, and the United
Kingdom. Surveys were analyzed from two North American nations: Canada and
the USA. Finally, responses from New Zealand were included in the study. Visitors
from these 14 nations constitute more than 75 percent of the total overseas visits and
close to two-thirds of the total visitors to Australia.
Questionnaire
The survey instrument included 88 questions. Trip-related planning activities were

included in the questionnaire. The questions included asking, ‘‘Before you left
[country of residence] did you get any information about Australia for this
visit?’’ Respondents who did get information were asked to identify information
sources from a list of 11 specific information sources, plus ‘‘somewhere else.’’
Detailed questions on cities and places visited as well as participating in 29 specific

tourism-related activities were asked.
Detailed questions were asked about expenditures in 16 expense categories,
including pre-paid expenditures, airfares within Australia during the visit, car rentals,
petrol and oil costs, food, drink, and accommodations, shopping, gambling, and
entertainment (e.g., theaters, movies, zoos, museums, nightclubs, recreation, and
entry fees). The respondents were not asked to report a total amount of expenditures
for their current visit in Australia; this amount was calculated based on their answers
to the individual expenditure items. ‘‘Total visit expenditures’’ includes prepaid
expenses related to activities done in Australia during the visit and expenditures while
in Australia for the travel party, but not the overseas air fare expenses for travel to
and from Australia.
All questions were examined for clarity and ease of response in each of six major
languages (English, German, Indonesian/Malay, Japanese, Korean, and Mandarin)
used to gather data. Thus six versions of the questionnaire were prepared and the
professional full-time interviewers completed the face-to-face interviews in the native
tongues of the respondents; this procedure was used to encourage top-of-mind
responses to open-ended questions among visitors from countries of origins with
the largest shares of visitors to Australia. Thus, for example, native Germans
participating as respondents, who were capable of being interviewed in English, were
requested to respond during the survey in their native language.
In pre-testing the six versions of the survey, ten-to-twelve respondents completed
each round of draft survey instruments. Two to three rounds of revisions were
completed before the research team conducting the study was satisfied that the
questions were understood clearly. The process of clarifying the questionnaires
included two-rounds of back-translations, for example, from Japanese to Australian-
English and back again to Japanese and Australian-English.
Analyses
For testing the propositions cross-country (i.e., group) level statistical analyses
was used for testing net effects (e.g., Bass et al., 1968; Lynn et al., 1993) as well as
qualitative comparative analyses (see McClelland, 1998; Ragin, 1987) using fuzzy
set QCA software (available at fsQCA.com) for testing configuration effects. The
analysis and findings in the chapter focus on the association of cultural causal recipes
on international tourism behavior; the analysis and findings using net effects analysis
(e.g., correlation analysis) is available elsewhere (Woodside & Ahn, 2007).
To compare apples to apples for similar segments of visitors across countries, four
sets of data groups were created for both the statistical and comparative analyses:
A-consumers: first-time visitors to Australia on a purely holiday trip (i.e., no part

of the visit was motivated by business reasons and the visit did not include VFR in
Australia),
B-consumers: first-time visitors to Australia on a holiday only visit who also are
VFR in Australia,
C-consumers: repeat visitors to Australia on a purely holiday trip,
D-consumers: repeat visitors to Australia on a holiday only visit that also are VFR
in Australia.
Theory suggests that the hypothesized impacts of the four described dimensions of
national culture should vary systematically: highest for A’s, moderate for B’s, low for
C’s; and lowest for D consumers. A’s have the least direct and indirect experience
with using the brand, Australia, and thus, their reliance unconscious reliance on their
relevant national cultures should be highest. D’s are likely to adjust their behaviors
during their current visits to Australia based on: (1) their prior visit (dis)satisfactory
experiences and consumption learning that came to their working memories during
their current visits and (2) particularly frank suggestions of things to do and how to
act offered by friends and family members living in Australia. The data analyses do
not confirm these views.
Both the estimated average and median responses for each of the dependent
measures were used to estimate the country-level association with the four
dimensions of national culture. Because the pattern and size of the associations
are very similar for associations based on mean versus median values, the findings
reported below are based on averages.
Further, fine-grained, comparative analyses, were performed by examining
data for the three age segments within each of the A’s, B’s, C’s, and C’s. If the
hypothesized impacts of specific dimensions of national culture are found
consistently across all three-age segments, such a comparative analysis serves as
meta-analyses (sometimes at nearly a case-by-case micro level since the sample sizes
become small for some countries). The observed findings support consistent impacts
of national culture across the three age segments.
Data analysis includes transforming Hofstede’s index scores for country value
dimensions into fuzzy set scores. The fuzzy set scores range from 0.00 to 1.00 and the
calibration (see Ragin, 2008) of index to fuzzy set scores results in very similar values
for the two sets of data (e.g., USA country value index for individualism equals 91,
the highest individualism score among the countries in Hofstede’s data set and the
fuzzy set score for the USA is set at 1.00).
The QCA data analyses included transforming mean scores for the outcome
conditions (analogous to dependent measures in statistical analysis into fuzzy set
scores). Table 1 provides an example of calibrating values representing a long visit to
Australia. One point to fuzzy set analysis is a range of values do not represent any
theoretically useful differences and the calibration transforms the values within such
a range into the same value. For example, the range of nights in Australia of 45 to
365 days constitutes full membership (1.00) into the long visit fuzzy set value.
All visits less than 11 days are transformed into full nonmembership in the long visit
outcome condition. The data analyses include full use of fuzzy set scores ranging
from 0.00 to 1.00.
1.0
Membership
in the Set of 0.9
Long Visit
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 10 20 30 40 50 60+
Number of Nights in Australia
Figure 6: Plot of fuzzy-set memberhsip in long visits.
Figure 6 indicates the specific fuzzy set values for number of nights of visiting
Australia. The share of visitors visiting Australia in a tour group was also
transformed into fuzzy set scores as well as dollars spent shopping gifts, and number
of activities and regions in Australia visited.
QCA provides measures for consistency and coverage. Consistency values are
analogous to correlation estimates in statistical hypothesis testing. Coverage values
are analogous to effect size estimates in statistical hypothesis testing. Consistency
gauges the degree to which the cases sharing a given combination of conditions
(M I P BU) agree in displaying the outcome in question (e.g., not a long visit
(BV)). That is, consistency indicates how closely the pairing of antecedent and
outcome scores approximates a perfect subset relation.
Coverage assesses the degree to which a cause or causal recipe accounts for
instances of the outcome. When several paths to the same outcome exist, the
coverage of a given causal combination may be small. Thus, coverage gauges
empirical relevance or importance.
The calculation of the fuzzy set-theoretical consistency value is at as follows:
X minðX i ; Y i Þ
ConsistencyðX i Y i Þ ¼ P ,
Xi
where Xi equals a fuzzy set score for the antecedent condition (e.g., a specific causal
recipe) and Yi equals a fuzzy set score a specific outcome condition (e.g., long visit).
The measure of fuzzy-set coverage is simply the overlap expression for the
proportion of the sum of the membership scores in the outcome:
X minðX i ; Y i Þ
Coverage ðX i Y i Þ ¼ P :
Yi
P P
The formula for coverage of Y by X substitutes (Yi) for (Xi) in the
denominator of the formula for consistency (Ragin, 2008, p. 57).
In the XY plot output of fsqca.com software program, the value in the upper
left box shows the degree to which the data are consistent with XrY (X is a subset
of Y ). The number in the lower right box indicates coverage. For an indication of
usefulness, Ragin recommends consistency scores should be higher than 0.75 with the
observed consistency value being greater than the corresponding coverage value.
Findings
Applying QCA, the presentation of findings here are only to probe some of the
causal recipe antecedent associations with outcome conditions and to present a full
examination of the general theory of cultures’ consequences on tourism behavior.
Figure 7 presents the entire set of data for the 14 countries. To eliminate the
impact of one travel party with an extreme outcome condition having a
disproportionate impact on the findings, the data analysis in this chapter focuses
only on sets of visitors having more than 4 travel parties per category of visitors,
for example, the sample size is 41 first-time visitor parties from the USA visiting
Australia for holiday only and younger than 30 years old. See Figure 7 for this and
other sample sizes for specific contingency statements.
Anima (BM BI BP U) as a Causal Recipe Influencing Tourism Behavior
Figure 8 shows the findings for anima as a causal recipe. These findings indicate
modest support for the proposition that anima is a causal recipe that relates to not-
long-visits to Australia. The consistency scores for the X (anima) and Y (not-long
visits) are .85, 1.00, and 1.00 for young, middle-aged, and old visitor segments. Korea
and Taiwan are countries with high anima scores.
Note that European countries have low fuzzy set scores for not-long-visits in
Figure 7 as well as low scores for anima. The farther distance of the European
countries from Australia versus the Asian countries is an equally plausible explana-
tion to the high consistency values in Figure 8. However, distance alone is not a
plausible explanation for other patterns in the findings.
While this report does not show plots of anima and not-long-visit for repeat
holiday visitors, the findings are very similar for repeat visitors as first-time holiday
visitors. The consistency values are above .88 for the four XY plots for repeat holiday
visitors and coverage values are less than .35 (for the 3 age groups and the total
sample of cases).
The analysis for the older segment is not useful since the analysis is absent of most
of the Asian countries — Asian countries do not provide many over 50 aged visitors
to Australia. The analyses do indicate similar patterns for young and middle age
visitors for animal and the outcome condition.
National Culture
Japan USA Germany

(n=575) (n=281) (n=129)
First-time Repeat Visitor First-time Repeat Visitor First-time Repeat Visitor

Visitor (n=407) (n=168) Visitor (n=163) (n=118) Visitor (n=74) (n=55)
Holiday VFR Holiday VFR Holiday VFR Holiday VFR Holiday VFR Holiday VFR
(n=387) (n=20) (n=132) (n=36) (n=131) (n=32) (n=62) (n=56) (n=65) (n=9) (n=37) (n=18)
y m o y m o y m o y m o y m o y m o y m o y m o y m o y m o y m o y m o
Dependent
Variable
n of P/NC .16 .14 .23 .69 .20 0 .12 .16 .22 .24 .20 0 .42 .20 .15 .60 .73 .46 .21. 19 .09 .07 .10 .17 .43 .80 .56 0 .75 1.0 0 .18 .30 .60 .13 .40
n of P/C .71 .68 .54 .08 .20 .50 .39 .50 .33 .12 .13 0 .32 .54 .70 .20 .27 .18 .07 .19 .18 0 .06 .13 .33 .54 .67 0 0 0 .80 .50 .30 .20 0 .20
n of I/C .92 .81 .85 .54 .60 1.0 .72 .71 .33 29 .40 0 .98 .74 .92 .70 .91 .73 .29 .42 .50 .07 .15 .09 1.29 1.34 .89 .75 .15 0 1.40 .82 .40 0 .13 0
% Group Tour 37 8 0 0 25 25 0 0 7 0 5 7 32 0 0 0 0 0 9 0 0 0 5 3 22 0 0 0 0 0 0 0 0 0
. 48 36
Mean Nights 14 6 8 14 9 5 11 8 13 14 13 10 28 15 12 25 19 23 16 15 19 24 19 22 42 33 24 55 27 25 30 30 37 13 24 42
Mean Total 6.8 6.2 6.0 7.5 6.0 5.5 6.0 5.4 4.8 5.7 5.5 5.3 9.5 9.7 9.7 10.8 10.1 6.1 10.4 7.9 7.4 7.4 7.1 6.2 9.4 10. 4 10.9 7.0 7.0 12.0 7.2 9.2 10.1 4.2 7.5 6.2
Activities
n of regions 2.5 1.8 1.7 2.2 1.6 1.0 2.5 1.5 1.4 1.9 1.5 2.5 4.0 3.9 4.3 3.6 4.1 3.9 4.4 2.7 3.4 2.6 1.8 2.5 10.1 11.7 9.9 5.0 8.0 5.0 9.6 8.3 8.0 2.4 3.1 3.8
visited
Mean $ per day 297 368 253 133 322 286 256 381 219 140 144 128 149 257 203 86 165 274 187 186 230 54 125 86 112 159 136 45 71 236 125 182 157 38 112 89
$ Shopping for 1075 1201 585 874 616 200 757 797 677 219 451 443 241 710 767 703 281 509 171 342 733 188 222 333 287 438 389 280 538 545 353 278 377 106 166 399
taking home
n 280 94 13 13 5 2 67 56 9 17 15 4 41 43 47 10 11 11 14 26 22 15 18 23 21 35 9 4 4 1 5 22 10 5 8 5
Key: P/NC: Personal, Non-commercial P/C: Personal, commercial I/C: Impersonal, commercial source; Y = age < 30, M = 30-49, O = 50+
Figure 7: Comparative analysis for national cultures with key findings underlined.
Fuzzy Set Social Science and Qualitative Comparative Analysis
385
386
National Culture
Canada Korea Taiwan

(n=91) (n=105) (n=116)

(n=38) (n=14) (n=18) (n=21) (n=65) (n=5) (n=22) (n=13) (n=60) (n=11) (n=27) (n=18)
Dependent
Variable
N of P/NC .37 .67 .60 33 .00 .60 .60 .33 .00 .25 .01 .00 .15 . 35 .00 .33 .00 1.00 .01 .17 .00 .00 .50 .00 .31 .15 .00 .88 .33 .00 .23 .08 .00 .09 .50 .00
N of P/C .42 .33 .40 .33 .33 .60 .20 .17 .14 .25 .01 .00 .50 .71 1.50 .00 .00 .00 .30 .33 .00 .00 .00 .00 .78 .92 .50 .00 .67 .00 .46 .50 1.00 .09 .17 .00
N of I/NC .16 .11 .10 .17 .33 .00 .00 .00 .00 .00 .00 .00 .02 .06 .00 .00 .00 .00 .00 .08 .00 .00 .00 .00 .19 .12 .00 .00 .00 .00 .00 .08 .00 .09 .17 .00
N of I/C .79 .89 .60 .67 .00 .40 .80 .33 .71 .00 .01 .00 .50 .24 1.00 .33 1.00 .00 .50 .50 .00 .00 .50 .00 .66 .00 .50 .25 .00 .00 .54 .75 .00 .18 .17 .00
.
% Group Tour 0 0 0 0 0 0 0 0 0 0 0 0 39 82 100 0 0 100 30 42 00 0 0 0 88 81 50 0 0 0 23 50 50 0 0 0
Ave. Nights 48 40 27 77 17 14 92 21 35 11 15 37 25 06 03 11 45 51 06 14 00 31 19 17 07 06 05 09 23 00 35 06 07 11 12 112
Ave. Total & Daily 13.5 14.2 12.5 10.6 6.7 7.8 9.6 9.3 7.7 6.3 5.0 7.3 9.2 8.6 6.0 7.3 12 11 7.2 5.6 00 6.8 4.8 4.0 9.1 8.7 8.0 7.3 9.0 00 9.5 8.1 3.0 7.7 5.0 8.0
Activities (.7) (.7) (.5) (.5) (.4) (.8) (.7) (.6) (.4) (.8) (.5) (.3) (1.3) (1.8) (2) (.7) (.3) (.2) (1.9) (1.3)(0) (.4) (.5) (.2) (1.6)(1.5)(1.7) (1.0) (.6) (0) (.9 ) (1.6) (.4) (.8) (.4) (.1)
N of Oz Regions 9.1 7.7 4.2 5.0 1.7 2.8 7.8 3.5 5.1 2.5 1.2 3.0 2.8 2.1 2.0 1.3 6.0 5.0 1.4 1.1 .00 1.3 1.8 1.0 3.2 3.0 2.5 1.1 1.0 .00 2.4 2.0 3.5 1.4 1.7 1.0
& States visited (2.6)(3.4)(1.9) (2.2)(1.0)(1.2) (2.6)(1.7)(1.7) (1.5)(1.2)(1.6) (1.9) (1.8 )(2) (1.3) (2) (1) (1.4) (.9) (.0) (1) (1.8) (1) (2.5)(2.5) (2.5) (1.1) (1) (0) (1.9) (1.6) (2) (1.3) (1.5) (1)
$ per day 113 307 153 125 78 171 50 134 160 103 76 38 216 303 50 51 18 45 314 451 00 177 154 08 123 327 272 164 96 00 150 369 213 135 177 94
$ Shopping 360 727 354 270 300 175 194 590 379 225 235 25 658 1039 138 150 300 400 1036 1071 0 750 808 20 542 1705 1029 316 745 00 558 1100 850 589 842 6000
N 19 09 10 06 03 05 05 06 07 04 10 07 46 17 02 03 01 01 10 12 00 08 04 01 32 26 02 8 3 0 13 12 02 11 06 01
Figure 7: (Continued)
National Culture
UK Switzerland Hong Kong

(n=508) (n=56) (n=124)

(n=181) (n=62) (n=102) (n=163) (n=29) (n=5) (n=14) (n=8) (n=46) (n=7) (n=32) (n=39)
Dependent
Variable
N of P/NC .63 .40 .37 .65 .36 .53 .21 .13 .16 .09 .14 .09 .65 .86 .00 1.00 .00 1.00 .83 .20 .33 .00 .00 .00 .33 .11 .00 .50 .00 .00 .20 .08 .00 .14 .17 .0 0
N of P/C .42 .40 .50 .26 .09 .29 .15 .06 .19 .04 .11 .17 .45 .71 .50 .33 .00 .00 .33 .20 .67 .00 .00 .00 .46 .53 1.00 .00 .00 .00 .00 .12 .00 .14 .04 .00
N of I/NC .26 .11 .15 .17 .09 .12 .15 .06 .03 .04 .03 .05 .20 .57 .00 .00 .00 .00 .17 .00 .33 .00 .00 .00 .13 .11 .00 .17 .00 .00 .00 .00 .00 .00 .0 .00
N of I/C 1.07 .81 .44 .70 .41 .29 .39 .34 .27 .17 .08 .07 1.10 1.57 1.00 .67 .00 1.00 1.33 1.40 . 00 .67 1.00 .00 .46 .58 .33 .50 .00 .00 .20 .20 .00 .14 .0 4 .00
.
% Group Tour 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 0 0 38 42 33 0 0 0 20 24 0 0 0 0
Ave. Nights 63 34 20 43 23 33 38 31 29 26 26 45 69 38 19 29 14 46 58 24 26 4 24 15 10 8 6 9 0 0 7 12 17 1 3 14 13
Ave. Total & Daily 12.4 10.2 8.9 10.7 8.2 9.6 7.7 8.4 8.2 8.4 6 .8 8.0 11.2 7.9 11.0 10.7 4.0 13.0 11.7 8.8 4.3 4.7 10. 5 4.3 8.3 8.0 8.7 8.7 .0 .0 6.2 6.2 4.5 4.5 5.5 5.3
Activities (.4) (.5) (.6) (.4) (.5) (.4) (.3) (.4) (.4) (.5) (.4) (.3) (.3) (.3) (.6) (.7) (.3) (.3) (.2) (4) (.3) (.2) (.5) (.3) (1.2) (1.1) (1.6) (1.1) (0) (0) (1.0) (1.0) (0.2) (.7) (.7) (..5)
N of Oz Regions 10.3 6.4 4.8 3.6 3.8 4.5 5.3 6.1 5.0 3.5 2 .2 3.6 12.8 17.4 7.5 5.7 2.0 4.0 22.7 9.8 4.7 1.0 3.0 1.7 2.5 2.6 2.0 1.2 .0 1.0 1.6 1.8 1.5 2.0 1 .7 1.0
&States visited 2.9 2.6 2.3 1.5 1.6 2.0 1.8 2.4 2.2 1.6 1 .4 1.7 3.9 3.7 2.5 2.0 2.0 1.0 3.2 2.4 1.0 1. 0 2.5 1.0 1.8 1.7 1.0 1.0 .0 1.0 1.2 1.5 1.0 1.3 1 .2 1.0
$ per day 97 180 162 76 125 104 97 143 190 92 113 68 104 140 130 92 186 35 88 124 120 164 73 83 243 348 400 151 0 398 129 395 224 124 246 1320
$ Shopping 330 397 416 368 504 265 161 400 426 436 432 344 461 700 150 133 350 300 232 270 217 200 150 233 654 862 978 536 0 1500 200 897 35 286 850 267
N 82 53 46 23 22 17 33 32 37 23 36 104 20 7 2 3 1 1 6 5 3 3 2 3 24 19 3 6 0 1 5 25 2 7 29 3
387
388
National Culture
Netherlands Singapore Malaysia

(n=79) (n=117) (n=67)

(n=40) (n=10) (n=17) (n=12) (n=36) (n=4) (n=50) (n=27) (n=27) (n=4) (n=19) (n=17)
Dependent
Variable
N of P/NC .65 .42 .80 2.0 .83 .00 .17 .38 .33 .50 .00 .13 .42 .25 .00 1.00 .00 2.00 .64 .15 .08 .00 . 17 .00 .17 .42 .67 .67 .00 1.00 .50 .33 .00 .33 .00 .29
N of P/C .65 .33 .80 .00 .50 .00 .17 .25 1.00 .00 .00 .25 .53 .38 1.00 .00 .00 1.00 .55 .31 .15 .00 .08 . 00 .25 .42 .67 .33 .00 1.00 .00 .20 .50 .00 .00 .14
N of I/NC .26 .17 .20 .00 .33 .00 .33 .00 .33 .00 .00 .00 .00 .13 .00 .00 .00 1.00 .0 9 .15 .00 .00 .08 .00 .17 .00 .00 .33 .00 .00 .00 .00 .00 .00 .00 .00
N of I/C .04 .92 .60 2.00 .50 .00 .33 .75 1.00 2.00 .00 .1 3 .74 .56 1.00 1.00 .00 2.00 .64 .73 .39 .00 .08 .00 .58 .58 .33 .00 .00 1.00 1.00 .13 .50 .67 .00 .00
.
% Group Tour 9 0 20 0 0 0 0 13 0 0 0 0 21 6 0 0 0 0 0 12 0 0 0 0 42 33 0 0 0 0 0 7 0 0 0 0
Ave. Nights 81 23 32 32 21 16 46 31 24 .20 22 51 10 9 7 6 7 6 7 8 7 11 13 15 14 08 16 13 00 56 06 09 05 19 2 7 28
Ave. Total & Daily 12.1 8.8 8.0 20.0 9.0 7.0 11.2 9.5 11.0 4.5 5.0 7.1 7.3 5.0 9.0 6.0 3.5 8.0 7.3 5.4 4.3 6.3 4.7 6.0 7.6 5.8 7.7 10 00 12 6.5 5.1 3.0 8.0 4.3 3.9
Activities (.4) (.5) (.3) (.6) (.6) (.4) (.4) (.4) (.7) (.3) (.2) (.2) (.9) (.7) (1.3) (1 .0) (.6) (1.3) (1.1) (.8) (.8) (.7) (.6) (.6) (.8) (.9) (.7) (.8) (.0) (.2) (1.1) (.7) (.6) (.5) (.5) (.4)
N of Oz regions 11.2 9.1 10.2 6.0 5.2 3.3 14.3 12.0 6.0 4.0 3.0 2.5 2.8 2.3 2.0 1.0 2. 0 2.0 1.7 2.0 1.3 1.2 1.3 1.6 2.3 1.4 3.7 2.0 .00 1.0 1.0 1.5 1.0 1.7 1.4 2.1
&States visited (2.7)(2.6)(3.0) (2.0)(1.7)(1.3) (3.2)(2.9)(3.7) (1.0)(1.0)(1.5) (1.6)(1.4)(2.0) (1.0)(1.5)(1.0) (1.4)(1.2)(1.1) (1.2)(1.2)(1.0) (1.8)(1.2)(1.7) (1.7) (.0)(1.0) (1 .0)(1.2)(1.0) (1.0)(1.1)(1.3)
$ per day 105 145 139 42 60 63 78 169 336 79 185 29 212 348 362 174 184 220 304 304 709 106 94 298 150 197 59 101 00 32 268 481 1434 760 151 307
$ Shopping 310 302 167 120 160 300 167 425 967 225 100 217 432 523 1200 1000 450 300 9 91 592 384 475 353 1561 489 223 684 567 00 500 20 0 1034 225 283 380 279
N 23 12 05 01 06 03 06 08 03 02 02 08 19 16 01 01 02 01 11 26 13 06 12 09 12 12 03 03 00 01 02 15 02 03 07 07
National Culture
New Zealand Indonesia

(n=332) (n=50)
First-time Repeat Visitor First-time Repeat Visitor

Visitor (n=49) (n=283) Visitor (n=13) (n=37)
Holiday VFR Holiday VFR Holiday VFR Holiday VFR

(n=29) (n=20) (n=136) (n=147) (n=6) (n=7) (n=11) (n=26)
y m o y m o y m o y m o y m o y m o y m o y m o
Dependent
Variable
N of P/NC .64 .25 .00 .14 .00 .00 .20 .14 .06 .24 .05 .01 .00 .50 .00 .33 .33 .00 .00 .29 .00 .00 .00 .00
N of P/C .57 .75 .33 .14 .00 .00 .20 .30 .31 .10 .05 .12 .00 .50 1.00 .00 .00 .00 .00 .00 .00 .20 .00 .22
N of I/NC .07 .00 .00 .00 .00 .00 .07 .00 .04 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00
N of I/C .57 .50 .33 .07 .00 .00 .17 .37 .16 .07 02 .07 .00 .75 .00 .00 .00 .00 .00 .29 .00 .00 .17 .11
% Group Tour 7 0 0 0 0 0 7 2 4 0 0 0 0 0 100 0 0 0 0 0 0 0 0 11
Ave. Nights 9 7 12 46 33 18 19 13 12 14 8 17 14 11 03 14 22 21 15 15 06 73 14 14
Ave. Total/Daily 5.6 6.4 6.7 6.9 7.0 8.0 6.2 6.2 6.1 6.3 4.7 5.0 5.0 5.0 5.0 7.3 6.0 6.0 8.0 7.1 0.7 3.2 4.7 5.9
Activities (.9) (.9) (1.0) (.5) (.3) (.5) (.6) (.7) (.7) (.8) (.8) (.4) (.4) (.7) (1.7) (.8) (.5) (.3) (.5) (.6) (.1) (.3) (.5) (.5)
N of Oz Regions 1.3 1.5 1.0 2.0 1.0 2.0 1.4 2.3 2.5 1.4 1.6 1.7 1.0 3.0 1.0 1.7 1.0 5.0 2.0 2.4 1.0 1.2 1.3 1.4
& States visited (1.1)(1.1)(1.0) (1.0)(1.0)(1.0) (1.1)(1.3)(1.3) (1.1)(1.1)(1.2) (1.0)(1.3)(1.0) (1.0)(1.0)(4.0) (2.0)(1.3)(1.0) (1.2)(1.0)(1.3)
$ per day 181 253 21 79 32 148 141 243 203 83 144 63 14 272 533 37 22 55 35 407 535 36 549 321
$ Shopping 649 482 183 973 390 60 266 381 594 309 346 206 100 150 1000 150 383 0 100 1764 458 264 614 679
N 14 12 03 14 04 02 30 57 49 29 41 77 01 04 01 03 03 01 01 07 03 05 12 09
389
390
Figure 8: Anima (BM BI BP U) as a causal recipe affecting not long visits to Australia. Note: Findings indicate high
consistency and moderate coverage for anima as an explanation for not long visits to Australia.
Animus and Not Shopping for Gifts among Young, Middle-Aged, and Older Visitors
Figure 9 includes the findings for animus as a causal recipe and not shopping for gifts
as the outcome condition. High fuzzy set scores for not shopping for gifts to take
home indicate low dollar expenditures for such purchases. The USA youth has
high values for both animus and not shopping for gifts. The Japanese youth has low
values for both animus and not shopping for gifts.
The XY plot for young samples of country visitors provides the highest
consistency value (1.00) and modest coverage (.36). For the middle-age pairings,
the findings indicate that most middle-aged Koreans, Japanese, and Taiwanese
visitors to Australia do shop for gifts to take home while most middle-age Dutch,
Malaysians, Germans, and British are not shopping for gifts to take home.
Animus and Not Shopping for Gifts for First-Time and Repeat Holiday Visitors
Figure 9 includes a replication of findings for two distinct samples: first-time and
repeat holiday visitors. The findings include high consistency and modest coverage
for both first-time and repeat holiday XY plots for animus as a causal recipe.
Note in Figure 9 that a number of sample cases of countries from Western cultures
are high in both animus and not shopping for gifts while the opposite pattern occurs
for a number of sample cases of countries from Eastern cultures.
Limitations
This chapter offers a limited introduction to QCA for analyzing conjunctive causal
recipes of cultures’ consequences on outcome conditions relating to international
tourism. Attempting to meet the objective of keeping this report to reasonable length,
the analyses and findings do not offer a comprehensive examination of the
propositions in the general theory of cultures’ consequences on international tourism.
The focus of the findings is on examining conjunctural influences rather than net
effects on two outcome conditions (length of visit and expenditures for gifts).
A full report using QCA that includes analyzing simple causal conditions (e.g.,
collectivism only) on outcome conditions would be beneficial. (Such analysis do
support and extend findings using net effects statistical test results that Woodside,
2007) reports.
The findings in this chapter are representative of visitors to one country in one
year using only Hofstede’s value indexes and replications to other countries in other
continents using alternative value scores for countries are necessary before
concluding that conjunctive causal recipes of country values are useful for explaining
outcome conditions relating to international tourism behavior.
392
Figure 9: Animus (M I P BU) and not shopping for gifts to take home. Note: Findings indicate high consistency and
moderate coverage for animus as a causal recipe for not shopping for gifts.
Implications for Theory and International Management

and Markeitng Practice
Cultures represent a complex whole (a conjunction) of value systems. Such value

systems potentially are causal recipes that affect international tourism behavior.
The present chapter demonstrates that such complex causal recipes are helpful
in explaining not visiting an overseas destination for a lengthy stay and for not
shopping for gifts to take home.
The findings in this report do not support the view that young versus middle-age
or older visitors exhibit counter-cultural outcome behaviors. The findings do not
support the view that complex culture recipes are less influential among repeat
visitors versus first-time visitors.
The view that the pairing of animus as a causal recipe and not shopping for gifts
relates to international visitors from some Western countries receives support. Such a
pairing of relatively high animus and not shopping for gifts excludes most visitors
from Eastern countries.
Anima as a complex causal recipe does relate to not-long-visit. This finding
supports the view that being home is necessary for practicing forms of nurturing
behavior that anima reflects.
Firms in tourism related industries may benefit by designing experiences that
match with the visitors’ cultural orientations. For example most visitors from several
Eastern cultures will not engage in long visits and will engage in shopping for gifts
during their brief visits to Australia. The opposite pattern is more likely to occur for
visitors from Western cultures. Broadly summarizing, product marketing for gifts for
loved ones and friends at home relates more closely to Eastern cultures while
experience marketing during longer visits relates more closely to Western cultures.
Exceptions do occur from this strategy proposition and the proposition may not
appear to be a new proposal; however, the use of QCA permits the identification of
what countries and visitors reflect this proposition and where the exceptions occur.
The continuing tourism marketing designs that attempt to convert short visits into
long visits or long visits into longer visits may be effective for only a limited number
of origin markets. For Australia, given the dominance of short visits and complex
cultural recipes that call residents to be home, most Asian visitors are highly
unlikely to shift to long visits in comparison to visitors from Western nations.
Complex causal recipes (such as animus) permit visitors from Western cultures to
consider lengthening their visit to participate in additional ‘‘once-in-a-lifetime’’
experience. The identification of Australian-own unique experiences may be essential
for implementing such strategies effectively. Similar insights likely are applicable for
other country destinations as well.
Conclusion and Implications for Future Research
Examinations and refinements of the general theory of cultures’ consequences on

international tourism are worth of additional research attention. Culture’s influence
on purchasing and consumption tourism products and services is likely to be

contingent on age, first-time versus repeat visits, and holiday versus visiting friends
and relatives. However, the findings in this chapter do not offer much support for
this view.
The main conclusion is that QCA is useful for identifying important asymmetrical
relationships between complex causal recipes of values statements and international
tourism behavioral outcomes. The testing of such causal recipes using QCA is worth
pursuing in research on explaining and modeling international tourism behavior.
Problems and Questions for Discussion
1. A. Summarize the major findings in Figure 10 in 300 words or less. B. What

tourism-marketing strategy implications follow from these findings?
2. A. Summarize the major findings in Figure 11 in 300 words or less. B. What
tourism-marketing strategy implications follow from these findings?
3. A. Construct a table listing Hofstede index scores for each culture value appearing
in Figures 10 and 11 for each of the 14 nations. (Retrieve these values on the
internet at http://www.geert-hofstede.com/). B. Transform Hofstede’s index
1.00
.90 NET
SW
Y=Long-Visit Fuzzy-Set Membership
.80 UK
.70
.60 CAN
G
.50 US
.40 KOR
.30
.20 IND MAL

JP
.10 HK
SG
TW NZ
.00
.00 .10 .20 .30 .40 .50 .60 .70 .80 .90 1.00
I·U·P Intersection of Individualism (I), Uncertainty Avoidance (U),
and Power Distance (P)
Key. CAN = Canada, G = Germany, HK = Hong Kong, IND = Indonesia, KOR = Korea, JP = Japan,
MAL = Malaysia, NET = Netherlands, NZ = New Zealand, SG = Singapore, SW = Switzerland,
TW = Taiwan, UK = United Kingdom, US = United States
Figure 10: Sufficiency test long-visit fuzzy-set membership among young adults by
the intersection of individualism, uncertainty avoidance, and power distance.
1.00
.90 NET
SW
Y=Long-Visit Fuzzy-Set Membership
.80 UK
.70
.60 CAN
G
.50
US
.40 KOR
.30
.20 JP IND MAL
.10 HK
SG
TW NZ
.00
.00 .10 .20 .30 .40 .50 .60 .70 .80 .90 1.00
(I·U·P~M)Intersection of Individualism (I),

Uncertainty Avoidance (U), Power Distance (P) and not Masculinity (~M)
Key. CAN = Canada, G = Germany, HK = Hong Kong, IND = Indonesia, KOR = Korea, JP = Japan,
MAL = Malaysia, NET = Netherlands, NZ = New Zealand, SG = Singapore, SW = Switzerland,
TW = Taiwan, UK = United Kingdom, US = United States
Figure 11: Sufficiency test long-visit membership among young adults by the Union
of Individualism, Uncertainty Avoidance, Power Distance, and not Masculinity
(I A PBM).
scores for each nation for each value after creating a fuzzy set scale for each of the
four cultural values.
4. A. After completing problem 3 above, use your fuzzy set scales to compute the
causal-recipe fuzzy-set value for I M BPBU (the American iconic recipe) for
each nation. B. Create a figure that plots the country scores for I M BPBU and
the long visit membership scores that appear in Figures 10 and 11 for the 14
nations.
5. A. Summarize the major findings that appear in your diagram in answering
problem 4. B. What tourism-marketing strategy implications follow from these
findings?
Chapter 18
Conclusions: Principles for Doing

Case Study Research
Synopsis
Chapter 18 closes the book with twelve principles relevant for doing case
study research. The chapter includes brief discussions of specific must-read
literature for each principle. The discussion also emphasizes that accuracy
(validity) comes first, not generality. The chapter emphasizes that the
dominant logic in seeking generality by using surveys whereby informants
write-out answers, tick boxes, and never have the opportunity to answer
questions that they themselves frame fails to deliver accuracy except possibly
when informants are describing evaluating their own recent experiences
(see Chapter 2 for further details). The following key thoughts signify the
twelve principles:
Configural not net effects

Unconscious not conscious thinking
Dynamic not cross sectional designs
Multiple routes not one model only
Predictive validity not only a best fitting model
Context not context free
Conjunctive-disjunctive not compensatory decision-making
Systems thinking not independent versus dependent conditions
Multi-person not one-person
Satisfy not optimize decisions
Unobtrusive evidence not just obtrusive interviews or observations
Visual not just verbal data collection and interpretation.
If we are concerned about the imprecision of case studies as research data,

we can console ourselves by noting that a man named Darwin was able to
write about a study of the Galapagos Islands and a few other cases. To the
best of my recollection, there are not statistics in Darwin’s book (Simon, 1991,
p. 128).
Introduction
Chapter 18 closes the book with principles to live-by in doing case study research.
These principles serve as rationales supporting the usefulness for theory creation,
doing valid research, and implementing successful practice (i.e., workable
implemented strategies).
Good news in response to Simon’s (1991) expression of concern about the
imprecision of case studies! The discussion and research examples in several chapters
in this book describe how precision is possible in case study research. Methods
providing for precision include qualitative (configural) comparative analysis (QCA,
see Chapter 2), building-in degrees of freedom, ethnographic decision-tree modeling,
and confirmatory personal introspections.
Chapter 18 presents 12 principles relevant for doing case study research. The
principles serve both as rationales for doing case study research versus using empirical
positivistic methods and as guides on how to do case study research. Nothing is
sacrosanct about the number 12; including 20 principles for case study research is the
goal for the second edition.
The quality in doing and writing-up case study research improves with practice.
Big mistakes usually occur by novices in doing case study research. All case study
researchers make mistakes but usually do improve with practice; studying case-study
research methods before going into the field to collect data can be helpful. Mistakes
include asking blunt questions at a first meeting with informants and asking ‘‘why’’
questions. The second edition will elaborate on this thirteenth principle — practice to
improve quality; ask informants what you should ask; ask indirect questions and ask
questions indirectly before asking blunt questions; consider not asking questions at
all until the third or fourth meeting with informants. The next 12 sections describe
12 principles relevant for case study research theory development, designing
methods, and interpreting data.
Configural Efects Not Net Effects
Outcomes are viewable better as results of causal recipes and not combinations of
independent variables and interaction terms. ‘‘In conventional quantitative research,
independent variables are seen as analytically separable causes of the outcomes under
investigation. Typically, each causal variable is thought to have an autonomous or
independent capacity to influence the level, intensity, or probability of the dependent
variable’’ (Ragin, 2008, p. 112).
Achieving accuracy requires applying methods that account for contingency and
complex antecedent conditions. For example, at least a few substantial changes in
contexts relevant to an outcome always change the outcome — antecedents are
viewable best as recipes and not linear combinations of independent influences.
The addition of complex interaction terms (e.g., four-way interactions of independent
variables) presents severe problems of interpretations. Expecting techniques designed
Conclusions: Principles for Doing Case Study Research 399
specifically to estimate net effects (e.g., correlation and structural equation modeling)
in linear-additive models to do a good job of assessing causal recipes, especially in
situations that include multiple recipes, is unreasonable (Ragin, 2008, p. 113).
Homework reading to assign yourself: Ragin (2008). Ragin is the leading authority
on qualitative comparative analysis. Start with his 2008 book because the Boolean
algebra tools in this book are more advanced and useful than his earlier books.
Unconscious Not Conscious
Humans (all of us) have biases favoring the belief that we plan what we do and know
why we have done whatever we have done. A double-whammy over-confidence bias
occurs. An over-confidence bias occurs when someone’s subjective confidence in her
judgments is reliably greater than her objective accuracy, especially when confidence
is relatively high. Failure to recognize and agree that such a bias is relevant in our
descriptions of causes of our own behavior is the second whammy.
When outcomes are good, we usually conclude, ‘‘I did it!’’ Humans tend to report
that their decisions and behaviors were the principal (or only) causes of good
outcomes. When outcomes are bad, we conclude, ‘‘They changed the rules on me!’’
Humans tend to report that conditions and actions of others were the principal (often
the only) cause of bad outcomes. Humans fail to (consciously) notice how context
and manipulations by behavioral scientists and marketers influence their behavior
(see Bargh & Chartrand, 1999).
Consequently, collecting conscious-based data via self-reports (e.g., subjective
personal introspections and survey responses) is insufficient for achieving high
accuracy. Humans need to reflect on their own behavior and collect unconscious-
thinking data (e.g., latency-response tests at implicit.harvard.edu) to achieve high
accuracy in describing and understanding their own thinking and behavior.
Homework reading to assign yourself: Strangers to Ourselves: Discovering the
Adaptive Unconscious by Timothy D. Wilson (2002). Wilson (2002) provides
examples of how and why ‘‘buyers lie’’ unconsciously usually when buying houses
and in other buying and using contexts. His book’s title really is food for thought.
Dynamic Not Cross-Sectional

The dominant logic for most organizational studies, research in marketing, and
strategic management research is to collect all the data in the main study in one
survey and then to test a conceptual model using structural equation modeling.
The conceptual models include no lagged relationships (and the write-ups of these
studies include only best-validity estimates and almost never include tests for
predictive validity). While the conceptual models almost always imply lagged
relationships between the independent variables and the dependent variables, the
empirical models fail to test for such relationships.
Not following the dominant logic is best. Collecting data from the same sources
for several days, weeks, years, or decades is best. Doing cross-lagged analysis
(e.g., Woodside, 1999), field experiments where manipulations precede outcomes
(List, 2006), system dynamics modeling in case study research (e.g., Hall, 1984), and
qualitative comparative analysis (e.g., Ragin, 2008) are useful alternatives to avoid
the failure to recognize empirically for lagged relationships.
Go back into the field twice, thrice, and several times more if possible to see if
informants’ predictions become reality. Participant observations (e.g., van Maanen,
1978) that include long periods in the field (i.e., weeks, months, and years) alleviate
the lack of information on lagged relationships that cross-sectional data analysis
ignores.
Homework to assign yourself: Hall (1984). Hall (1984) remains a brilliant gem
in examining lagged relationships unobtrusively of management decisions and
performance outcomes of the firm in a case study context. Roger Hall’s 1984 article is
remarkably good — an intellectual feast! This 1984 paper won first-place in an
international competition sponsored by the Institute for the Study of Management
Science (TIMS) for the ‘‘Most Original New Contribution to the field of
Organizational Analysis and Design’’ in 1983. The paper is available in full at
http://www.systemdynamics.org/conferences/1983/proceed/parallel-vol2/hall533.pdf.
Multiple Routes Not One Model Fits All
Useful rules-of-thumb include the expectations that if a study has five antecedent
conditions with each having low, medium, and high values, about five-to-ten
percent of the resulting 243 possible causal recipes (35 ¼ 243) results in a high value
for the outcome condition in the investigation; 30–50 percent of the causal recipes
lead to low-to-medium values in the outcome condition, and the rest of the complex
antecedent conditions (i.e., causal recipes) are ‘‘remainders’’, that is, they include
no cases.
Several models in a multiple regression analysis will achieve close to the same
maximum level that is reachable for the coefficient of determination (adjusted R2) in
an empirical positivistic study; yet most of these studies only report one or two
empirical models. Similar reports are made for structural equation models. Case
study researchers using qualitative comparative analysis (see the software program at
fsqca.com) know better — they report all relevant routes to high outcome levels for
the outcome condition.
Homework to assign yourself: Perfetto and Woodside (2009). Perfetto and
Woodside (2009) describe several routes leading to extremely frequent casino
gambling using relatively complex recipes of demographic antecedent conditions.
This study reports unique recipes for whales (wealthy), big fish (modest income) and
jumbo shrimp (very low income) extremely frequent casino gamblers.
Predictive Validity Not Just Best Fit Validity
Gigerenzer (2008) shows that multiple regression models do extremely well at

creating a best fitting model for a given dataset but perform less well in comparison
to more simple heuristics at predictive validity — testing the model’s accuracy for
a separate set of data. Presenting only best-fit model findings is the dominant logic
in the management, organizational studies, and marketing literatures but doing so is
bad practice.
Testing for predictive validity with hold out samples is always possible and doing
so substantially increases the added value for both empirical positivistic and
interpretative case studies. Yes, case studies can apply empirical positivistic analysis
methods (e.g., latency measurement and building-in degrees-of-freedom analysis).
Homework to assign yourself: [223]Gigerenzer (2008). A brilliant scholar who is
the direct link to the present from Herbert Simon’s work in case study research
(e.g., Cyert et al., 1956) for creating formal models of bounded rationally. Gigerenzer
(2008) is essential reading.
Context Not Context-Free
Simon (1990, p. 7) most famously reports, ‘‘Human rational behavior y is shaped

by a scissors whose two blades are the structure of the task environment and the
computational capabilities of the actor.’’
Simon’s earlier development of the concept of ‘‘bounded rationality’’

illustrates well the scissors of the mind-environment interface. In a
boundedly rational world, environments are much more complex than
any single actor can ever fully comprehend due to the computational
limitations of the human mind. Hence, actors must construct local
cognitive models that enable them to manage their environments
effectively. The cogency of this insight, that action takes place in
cognitively tractable local spaces, was a needed antidote to conceptions
of the panoramic managerial mind promoted by neoclassical econom-
ics. (Porac & Sharpira, 2007, p. 2007)
Because multiple-actors directly or indirectly participate in the immediate or

prior contexts relevant to the main focus of research on most human behaviors,
collecting data from multiple informants and for multiple context both are useful for
confirming/refuting the accuracy of the findings and expand on understanding of
the contingencies affecting outcomes uniquely for each context. The big mistake
in the dominant logic in management research is not planning a research study to
examine for context effects.
Homework to assign yourself: Woodside (2003). Woodside (2003) shows how

thinking, deciding, and actions in business-to-business supplier–customer relation-
ships are contingent on how the supplier categorizes customers and sizes of
customers’ available business. The study illustrates context–contingency middle-
range theory building on of how alternative contexts affect marketer–customer
decision-making.
Conjunctive-Disjunctive Not Compensatory Decision Making
For deciding on actions in new contexts, an individual or group usually asks two
questions. First, does a given action meet the minimum levels necessary on three to
nine prerequisites? Second, is the given action likely to provide high performance for
a critical outcome?
The first question expresses the conjunctive decision rule: set minimum levels
necessary for each prerequisite and discard candidate actions failing to meet all
minimums requirements.
The second question expresses the disjunctive decision rule: select the action that is
superior for the most critical prerequisite — among the remaining options following
learning the answer to the first questions.
For new contexts, the rules actually in use include several definable (this discussion
includes six) phases that occur several times before a choice is made. For example:
Define the problem/opportunity

Search a little to find a few (three to nine) potential options
Apply the conjunctive rule
Redefine the problem/opportunity
Apply the conjunctive rule
Redefine the problem/opportunity
Apply the disjunctive rule
Apply the disjunctive rule
Implement the decision
Evaluate the performance realized after implementing the decision.
Witte (1972) emphasizes that phases are identifiable in complex (new or major
problem/opportunity) management decision making but several of these phases
repeat during the processes. Research by Mintzberg et al. (1976) confirms empirically
and elaborates theoretically on how phases repeat in complex decision making.
Earlier empirical findings by Cyert et al. (1956) and recent findings by Woodside and
Wilson (2000) provide additional support.
What decision makers do not do is a Ben Franklin when making decisions.
Sometimes decision makers tell you that they perform Ben Franklins but if the
researcher conducts ‘‘long interviews’’ (McCracken, 1988) of two-plus hours with

two to five participants in the decision process, she learns that the actual decision
process was an application of the conjunctive-disjunctive rules.
What is doing a Ben Franklin? Ben Franklin was one of the Founding Fathers of
the United States. His writings include recommendations to apply a compensatory
decision rule for major decisions with two or more options (e.g., go versus no go;
option A, B, or C). A compensatory decision rule includes assigning pluses and
minuses to all features relevant in the decision for each alternative; adding up the
pluses and subtracting the minuses; the final values are compared; the option with the
highest score is selected.
A variant on this fully rational model for decision making is to apply varying
weights of importance to the features. The decision maker may assign different
weights of importance to the features and multiple the weights after assigning
positive or negative values to an option for each feature (e.g., 3 to + 3) and then
sum the weighted scores for each option. Afterward, the decision maker selects the
option with the highest summed score.
While a few decision makers do take the time to apply such compensatory
decision rules, they will then refer to non-compensatory rules when actually deciding
(Woodside & Wilson, 2000). Decision makers apply non-compensatory decision rules
(such as the conjunctive-disjunctive rule) for two reasons. First, such rules are easier
to apply than compensatory rules. Second, the non-compensatory rules actually
provide higher predictive validity (though lower best fit validity) in accuracy versus
compensatory rules (Gigerenzer, 2008) — this second reason was recently proposed
theoretically and confirmed empirically by Gigerenzer (2008).
Consequently, while the theory of reasoned action (Ajzen & Fishbein, 1980;
Fishbein & Ajzen, 1975) — a compensatory model on decision making — is
theoretically elegant and pleasing but it is not relevant sufficiently for describing or
understanding actual decision making. Also, asking informants to respond to
measures of importance using 1–5 or 1–7 scales and then multiplying informants’
responses to feature evaluations by importance responses is unsound if the objectives
include describing and understanding real-life decision making.
Homework to assign yourself: Gladwell (2005), Gigerenzer (2007), and Gigerenzer
(2008). Gladwell (2005) and Gigerenzer (2007) are useful, popular, airport-reading
translations of scientific research on how thinking and deciding are actually
done. Gigerenzer (2008) is a heavy-duty scientific read. (Someone needs to nominate
Gigerenzer for the Nobel Prize in Economics before 2020. He deserves both the
nomination and the award.)
Systems Thinking Not Linear Thinking
The Fifth Discipline: The Art & Practice of the Learning Organization (1990) describes
systems thinking. Systems’ thinking includes the assumption that relationships
include feedback loops. The view that decisions are never final but lead to actions
that lead to further decisions that lead to further actions is systems thinking.
Systems’ thinking is necessary for system dynamics modeling. Jay W. Forrester is
the founder of system dynamics as a field of study. Hall (1984) applies many of the
principles of system dynamic modeling in his study that he reports in his award
winning case study research article.
Systems’ thinking recognizes that layers of relationships occur among antecedent
conditions (using QCA language) and variables (using empirical positivistic
language). For example consider antecedent conditions A, B, C, and D: where A(t)
affects B(t+1) and B(t+2) affects C(t+3) and C(t+3) affects D(t+4) and D(t+4) affects
A(t+5), where t ¼ days, weeks, or years. Note that among these relationships that all
conditions are both antecedent and outcome conditions — systems’ thinking includes
the proposition that all conditions or variables are both independent and dependent.
‘‘Hidden demons’’ are found in systems. Hidden demons are seemingly
unimportant relationships that can drive the system to failure. Executives frequently
fail to recognize the impact of these relationships before the system breakdowns and
dies. Hall (1976, 1999) provides details on how hidden demons cause systems’ failures.
System Dynamics Review is the leading journal in the field. Consider becoming
a member of the System Dynamics Society (go to http://www.systemdynamics.org/),
if you wish to become a hands-on researcher in this subfield of case study research.
Homework to give yourself: Hall (1976) and Sterman (2000). Read Hall for
a highly readable empirical example of system dynamics research. Sterman (2000)
is a first-rate full treatise on how to do system dynamics research.
Multiple Party Not Single Person
Von Hippel (1988) reports the discovery in his ethnographic study of new
manufacturing processes that manufacturers initially fail to report that ‘‘lead
customers’’ came up with details on what to invent, what the invention should look
like, and how the invention should work. When good things happen, most humans
tend to try to take all the credit for themselves. But with long interviews individually
among several persons on site (i.e., in contexts), a more accurate picture emerges:
several persons have involvement in the innovation process. The Sources of
Innovation, the book resulting from his Ph.D. dissertation is downloadable for free
at http://web.mit.edu/evhippel/www/sources.htm.
Look for multiple-party participation and build theory to include the participa-
tion by multiple parties in case-study research projects. For example, friends, family
members, and co-workers have direct and indirect influences in initial purchases
of most products, services, and brands. Suppliers and customer firms influence
manufacturers in new product development. Woodside and Biemans (2005) provide
a set of propositions of the roles of multiple parties in IMDAR processes
(i.e., innovation, manufacturing, diffusion, acceptance and rejection of innovations).
Homework to give yourself: von Hippel (2008). Also, in a paper awarded best
article in the volume of its publication, Woodside (1996b) provides an insightful
description of the roles multiple parties play in rejecting superior new technologies —
it’s a worthy backup homework to assign yourself.
Satisfy Not Optimize
Because humans face many tasks everyday and seek to make tradeoffs between
accuracy and effort when making decisions (Payne, Bettman, & Johnson, 1993),
decision makers rarely seek to find an optimal option to a problem or opportunity
even if such an option might be available. Simon (1990) points out that humans
lack the cognitive resources to maximize: they usually do not know the relevant
probabilities of outcomes, they can rarely evaluate all outcomes with sufficient
precision, and their memories are weak and unreliable. Also, an optimal decision
may not exist for a given decision-making context. Simon proposes the bounded
rationality theory of decision making.
A question that might arise here is how might decision makers improve their
decisions when in the heat of a context requiring that a decision be made? That is,
how can decision makers decide on taking an action somewhat better than one that
results in disaster and that many persons implement in the heat of a given context?
Weick (2007) provides several articles on this issue. He uses the following account to
stress the importance of improving decision making in high stress contexts:
At least 23 wildland firefighters have died in four separate incidents

since 1990 with their tools beside them. In every case, they died within
sight of safety zones that could have been reached if they had been
lighter and moved faster. For example, at the South Canyon disaster
outside Glenwood Springs, Colorado, 14 firefighters were killed on
July 6, 1994, when they failed to outrun a fire that exploded through a
stand of oak trees just below them. One firefighter, whose body was
found a mere 250 ft from safety at the top of the ridge, was still wearing
a backpack and still had a chain saw in his hand. (Weick, 2007, p. 6)
Weick (2007) proposes six solutions to improving the quality (i.e., effectiveness
not efficiency) of decisions (please refer to his article for a full discussion of these
possible solutions). The point here is that designing decision making that
incorporates his training proposals may be possible. Also case study researcher
may wish to undertake historical analysis (Golder, 2000) of cases to find additional
instances of very wise decision making that incorporates Weick’s proposals and
others versus typical decision making. Using QCA, what are the complex antecedent
conditions resulting in wise versus typical solutions? These issues may represent a
worthwhile Ph.D. dissertation topic and method.
Homework to assign yourself: Weick (2007). Note-to-self: be sure to include a
chapter on case study research on normative decision-making issues in the second
edition.
Unobtrusive Not Just Obtrusive Observations and Interviews
Humans and other animals are likely to report beliefs, attitudes, and behaviors about
themselves that do not match with their beliefs, attitudes, and behaviors when acting
unobserved in natural contexts.
Several reasons support this phenomenon. Humans have limited conscious access
to their own unconscious thinking (Wegner, 2002; Wilson, 2002). Humans edit their
thoughts that surface from their memories to protect their egos. Humans edit their
responses to be socially correct. Humans frequently change their answers depending
on the framing context and wording of the question (Levin, Schneider, & Gaeth,
1998). Placing the same human in different contexts frequently changes their beliefs,
attitudes, and behaviors dramatically. Field studies confirm that humans act
differently in real-life versus laboratory settings (e.g., List, 2006).
Collecting data unobtrusively in natural field settings is a powerful method for
learning how humans think and act in real-life. Avoid asking leading questions.
Avoid asking any questions at all!
This view might seem obvious but implementing research methods from such a
base requires more creativity and different tools that the use of fixed-point surveys.
Fixed-point surveys represent the main data collection method in most masters’
theses and Ph.D. dissertations in marketing, strategic management, and organiza-
tional studies — possibly from 1960 to 2010.
A key suggestion here is to use methods other than fixed-point surveys.
One context that is appropriate for fixed-point surveys is for assessing customer
evaluations of recent product–service experiences — especially when such surveys
reduce bias by not identifying the sponsor of the study and achieve response
rates above 50 percent for consumers selected for the study (e.g., Woodside, Frey, &
Daly, 1999).
Homework to assign yourself: Webb et al. (2000). Webb et al. (2000) cover 20+
methods for unobtrusive data collection and provide sources in the literature
with details on the implementation of each one. This recommendation goes hand-in-
hand with Mintzberg’s (1979) recommendation to do ‘‘direct research’’ by going
into real-life contexts that you are studying (see also Mintzberg, 1979) in strategic
management — an insightful view that applies to all fields of management, social
psychology, and sociology.
Visual Not Just Verbal Data Collection and Interpretation
Humans first communicate visually. Kaufman and Kaufman (1980) demonstrates

that babies of all ages (four days to six months) have preferences among visual
alternatives. Verbal languages developed thousands of years after Homo sapiens and
their ancestors were using visual communications. Humans’ report visual scenes,
actions, and locations in their dreams (e.g. Jung, 2009).
Mu
ltip
Mu le P
ltip art
Mu le P ies
Visual ltip M arty
le Pult
i
artple
Un y Pa
rti
ob es
tru
·C l siv Sy
· B ra e ste
A figu m B
n sT
Co hi
nk
A king
in
g
Thin
Unconscious ys t ems
S
outes
Sy
ste
m C
Sa
y
sT
iple R
tis
hi
fy
Con
nk
lidit
in
Dynamic
Mult
Con
g
junc
e Va
junc
tive-
ictiv
tive-
Pred
Disju
Disju
nctiv
e
nctiv
e
C o n t e x t
Figure 1: The case study research flowering learning tree: communicating visually 12 case study research principles.
Conclusions: Principles for Doing Case Study Research
407
Because most thinking occurs unconsciously and individuals have great difficulty
in retrieving unconscious thoughts, Zaltman (2003) and Zaltman and Zaltman (2008)
create and apply visual data collection methods that aid individuals to retrieve
and interpret information from memories. Also refer to Chapters 4 and 7 in the book
in your hands or on the screen.
Homework to assign yourself: Reed (2010). Reed (2010) provides excellent
training in how to think visually as well as a review of much literature of research on
visual thinking and communicating.
Conclusion
Figure 1 concludes this final chapter with a happy-face flowering learning tree as a
metaphor for the principles for case study research theory, methods, and practice.
Figure 1 illustrates a key feature in a unique visual format for each of the
12 principles. Let us nurture this tree’s future growth.
References
Adaval, R., & Wyer, R. S. (1998). The role of narratives in consumer information processing.
Journal of Consumer Psychology, 7(3), 207–245.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior.
Englewood Cliffs, NJ: Prentice Hall.
Alexis, M., Haines, G. H., & Simon, L. (1968). Consumer information processing: The case of
women’s clothing (pp. 197–205). In: R. L. King (Ed.). Proceedings of the fall conference,
Marketing and the New Science of Planning, American Marketing Association, Chicago.
Allan, J., Fairclough, G., & Heinzen, B. (2002). The power of the tale: Using narratives for
organisational success. Chichester/West Sussex: Wiley.
Allen, D. E. (2002). Toward a theory of consumer choice as sociohistorically shaped practical
experience: The fits-like-a-glove (FLAG) framework. Journal of Consumer Research, 28(4),
515–532.
Allison, G. (1971). Essence of decision: Explaining the Cuban missile crisis. Boston, MA: Little,
Brown and Company.
Allport, G. W. (1985). The historical background of social psychology. In: G. Lindzey &
E. Aronson (Eds), Handbook of social psychology (Vol. 1, pp. 1–46). New York: Random
House.
Altman, I. (1992). A transactional perspective on transitions to new environments.
Environment and Behavior, 24(2), 268–280.
Ambrosini, V., & Bowman, C. (2002). Mapping successful organizational routines. In:
S. A. Huff & M. Jenkins (Eds), Mapping strategic knowledge (pp. 19–45). London: Sage.
Anderson, J. C., & Narus, J. A. (1990). A model of distributor firm and manufacturer firm
working partnerships. Journal of Marketing, 54, 42–58.
Anderson, J. R. (1982). Acquisition of cognitive skill. Psychological Review, 89, 369–406.
Anderson, S. M., & Glassman, N. S. (1996). Responding to significant others when they are
not there: Effects on interpersonal inference, motivation, and affect. In: R. M. Sorrentino &
E. T. Higgins (Eds), Handbook of motivation and cognition (Vol. 3, pp. 262–321). New York:
Guildford.
Andreasen, A. R. (1985). Backward market research. Harvard Business Review, 63(May–June),
176.
Angelmar, R., & Stern, L. W. (1978). Development of a content analytic system for analysis of
bargaining communication. Journal of Marketing Research, 15(February), 93–102.
Argyris, C. (1982). Reasoning, learning and action. San Francisco, CA: Jossey-Bass.
Armstrong, J. S. (2003). Discovery and communication of important marketing findings:
Evidence and proposals. Journal of Business Research, 56(1), 69–84.
Arnold, S. J., & Fischer, E. (1994). Hermeneutics and consumer research. Journal of Consumer
Research, 21(1), 55–70.
Arnould, E., & Price, L. L. (1993). River magic: Extraordinary experience and the extended
service encounter. Journal of Consumer Research, 20(June), 24–45.
Arnould, E. J., & Wallendorf, M. (1994). Market-oriented ethnography: Interpretation

building and marketing strategy formulation. Journal of Marketing Research, 31(November),
484–503.
Bagozzi, R., & Nataraajan, R. (2000). The year 2000: Looking forward. Psychology &
Marketing, 17, 1–11.
Bagozzi, R. P., & Dabholkar, P. A. (2000). Discursive psychology: An alternative conceptual
foundation to means-end chain theory. Psychology & Marketing, 17(7), 535–586.
Bales, R. F. (1968). Interaction process analysis. In: D. L. Sills (Ed.), International encyclopedia
of the social sciences (pp. 465–471). New York: Crowell-Collier and Macmillan.
Barcus, F. E. (1961). A content analysis of trends in Sunday comics 1900–1959. Journalism
Quarterly, 38, 171–180.
Bargeman, B., & van der Poel, H. (2006). The role of routines in the vacation decision-making
process of Dutch vacationers. Tourism Management, 27(4), 707–720.
Bargh, J. (2002). Losing consciousness: Automatic influences on consumer judgment,
behavior, and motivation. Journal of Consumer Research, 29(September), 280–285.
Bargh, J. A., & Chartrand, T. L. (1999). The unbearable automaticity of being. American
Psychologist, 54(7), 462–479.
Bargh, J. A. (1989). Conditional automaticity: Varieties of automatic influence in social
perception and cognition. In: J. S. Uleman & J. A. Bargh (Eds), Unintended thought (pp. 3–51).
New York: Guilford Press.
Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects
of trait construct and stereotype activation on action. Journal of Personality & Social
Psychology, 71(2), 230–244.
Barna, L. M. (1994). Stumbling blocks in inter cultural communication. In: L. Samovar &
R. Porter (Eds), Inter cultural communication: A reader. Belmont, CA: Wadsworth
Publishing Company.
Bartunek, J. M. (1984). Changing interpretative schemes and organizational restructuring:
The example of a religious order. Administrative Science Quarterly, 29, 355–372.
Bartunek, J. M. (1993). Rummaging behind the scenes of organizational change — and finding
role transitions, illness, and physical space. In: R. W. Woodman & W. A. Pasmore (Eds),
Research in organizational change and development (Vol. 7, pp. 41–76). Greenwich, CT: JAI.
Bass, F. M., Tigert, D. J., & Lonsdale, R. T. (1968). Market segmentation: Group versus
individual behavior. Journal of Marketing Research, 5(August), 264–270.
Bauer, R. A. (1967). Consumer behavior as risk taking. In: D. F. Cox (Ed.), Risk taking and
information handling in consumer behavior. Harvard University: Boston, MA.
Bearden, W. O., Money, R. B., & Nevins, J. L. (2006). Multidimensional versus
unidimensional measures in assessing national culture values: The Hofstede VSM94
example. Journal of Business Research, 59(2), 195–203.
Bearden, W. O., & Woodside, A. (1978). Consumption occasion influence on consumer brand
choice. Decision Sciences, 9(April), 274–281.
Bearden, W. O., & Woodside, A. G. (1976). Interactions of consumption situations and brand
attitudes. Journal of Applied Psychology, 61, 764–769.
Becken, S., & Gnoth, J. (2004). Tourist consumption systems among overseas visitors:
Reporting on American, German, and Australian visitors to New Zealand. Tourism
Management, 25(3), 375–385.
Becker, H. S. (1998). Tricks of the trade: How to think about your research while you’re doing it.
Chicago: University of Chicago Press.
References 411
Belk, R. (1974). An exploratory assessment of situational effects in buyer behavior. Journal of

Marketing Research, 11(May), 156–163.
Belk, R., Güliz, G., & Askegaard, S. (2003). The fire of desire: A multisited inquiry into
consumer passion. Journal of Consumer Research, 30(December), 326–351.
Belk, R. W. (1975). Situational variables and consumer behavior. Journal of Consumer
Research, 2, 157–164.
Belk, R. W. (1987). Material values in the comics: A content analysis of comic books featuring
themes of wealth. Journal of Consumer Research, 14, 26–42.
Bem, D. J. (1970). Beliefs, attitudes, and human affairs. Belmont, CA: Brooks/Cole.
Berger, A. (1971). Comics and culture. Journal of Popular Culture, 5(Summer), 164–177.
Bettman, J. K. (1979). An information processing theory of consumer choice. Reading, MA:
Addison-Wesley.
Bettman, J. R. (1970). Information processing models of consumer behavior. Journal of
Marketing Research, 7(3), 370–376.
Bettman, J. R. (1971a). A graph theory approach to comparing consumer information
processing models. Management Science, 18(Part 2), 114–128.
Bettman, J. R. (1971b). The structure of consumer choice processes. Journal of Marketing
Research, 8(November), 465S4.
Bettman, J. R., & Park, C. W. (1980). Effects of prior knowledge, experience, and phase of the
choice process on consumer decision processes: A protocol analysis. Journal of Consumer
Research, 7, 234–248.
Bettman, J. R., & Weitz, B. A. (1983). Attributions in the board room: Causal reasoning in
corporate annual reports. Administrative Science Quarterly, 28, 165–183.
Bettman, J. R., & Zins, M. A. (1978). Constructive processes in consumer choice. Journal of
Consumer Research, 4(September), 75–85.
Biemans, W. G. (1989). Developing innovations within networks. Groningen, The Netherlands:
University of Groningen.
Boller, G. W. (1990). The vicissitudes of product experience: Songs of our consuming selves.
Advances in Consumer Research, 17, 621–626.
Boller, G. W., Babakus, E., & Olson, J. C. (1989). Viewer empathy in response to drama ads:
Development of the VEDA scale. Unpublished Working Paper Number 402–489. Fogleman
College of Business and Economics, Memphis State University.
Bonoma, T. V., & Rosenberg, H. (1975). Theory-based content analysis: A social influence
perspective for evaluating group process. Chicago: Institute for Juvenile Research.
Booth, W. C. (1961). The rhetoric of fiction. Chicago: University of Chicago Press.
Bougon, M. G., & Komocar, J. (1990). Directing strategic change: A dynamic wholistic
approach. In: A. S. Huff (Ed.), Mapping strategic thought. Chichester, UK: Wiley.
Bourdieu, P., & Johnson, R. (1993). The field of cultural production. New York: Columbia
University Press.
Briassoulis, H. (2002). Sustainable tourism and the question of the commons. Annals of
Tourism Research, 29(4), 1065–1085.
Bronfenbrenner, U. (1986). Recent advances in research on the ecology of human
development. In: R. K. Silbereisen, K. Eyferth & G. Rudinger (Eds), Development as
action in context: Problem behavior and normal youth development (pp. 287–309). New York:
Springer.
Bronfenbrenner, U. (1992). Ecological systems theory. In: R. Vasta (Ed.), Six theories of child
development (pp. 187–250). Philadelphia: Jessica Kingsley Publishers.
Brown, A. D. (1994). Authoritative sensemaking in a public inquiry report. Organization

Studies, 25, 95–112.
Brown, A. D., & Jones, M. (1998). Doomed to failure: Narratives of inevitability and
conspiracy in a failed IS project. Organization Studies, 19, 73–88.
Brown, A. D., Stacey, P., & Nandhakumar, J. (2008). Making sense of sensemaking narratives.
Human Relations, 61(8), 1035–1062.
Brown, S. (1998). The wind in the wallows: Literary theory, autobiographical criticism and
subjective personal introspection. Advances in Consumer Research, 25, 25–30.
Brown, T. J., Dacin, P. A., Pratt, M. G., & Whetten, D. A. (2006). Identity, intended image,
construed image, and reputation: An interdisciplinary framework and suggested terminology.
Journal of the Academy of Marketing Science, 34(2), 99–106.
Browne, J. (1973). The used car game. Lexington, MA: D.C. Heath.
Browne, J. (1976). The used car game. In: M. P. Golden (Ed.), The research experience. Itasca,
IL: Peacock.
Brunel, F. F., Tietje, B. C., & Greenwald, A. G. (2004). Is the implicit association test a
valid and valuable measure of implicit social cognition? Journal of Consumer Psychology, 14,
385–404.
Bruner, J. (1990a). Acts of meaning. Cambridge, MA: Harvard Business School Press.
Bruner, J. (1991). The narrative construction of reality. Critical Inquiry, 18(1), 1–21.
Bruner, J. S. (1990b). Culture and human development: A new look. Human Development, 33,
344–355.
Bruning, J. L., & Kintz, B. L. (1968). Computational handbook of statistics. Glenview, IL:
Scott, Foreman and Company.
Burmann, C., & Zeplin, S. (2005). Building brand commitment: A behavioural approach to
internal brand management. Journal of Brand Management, 12(4), 279–300.
Busch, P., & Wilson, D. T. (1976). An experimental analysis of a salesman’s expert and
referent bases of social power in the buyer–seller dyad. Journal of Marketing Research,
13(February), 3–11.
Butcher, S. H. (1961). Aristotle’s poetics (Trans.). New York: Hill and Wang.
Cabeza, R., & Nyberg, L. (2000). Imaging cognition II: An empirical review of 275 PET and
fMRI Studies. Journal of Cognitive Neuroscience, 12(1), 1–47.
Calder, B. J. (1977). Structural role analysis of organizational buying: A preliminary
investigation. In: A. G. Woodside, J. N. Sheth & P. D. Bennett (Eds), Consumer and
industrial buying behavior. New York: Elsevier.
Campbell, D. T. (1965). Variation and selective retention in socio-cultural evolution. In:
H. R. Barringer, G. I. Blanksten & R. Mack (Eds), Social change in developing areas
(pp. 19–49). Cambridge, MA: Schenkman.
Campbell, D. T. (1975). Degrees of freedom in the case study. Comparative Political Studies,
8(July), 178–193.
Campbell, D. T. (1979). A tribal model of the social system vehicle carrying scientific
knowledge. Knowledge: Creation, Diffusion, Utilization, 1(3), 181–201.
Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the
multi-trait multi-method matrix. Psychological Bulletin, 56, 81–105.
Campbell, J. (1968). Creative mythology: The masks of god. New York: Penguin.
Campbell, J. (1973). The hero with a thousand faces. Princeton: Princeton University Press.
Campbell, J. (1974). The mythic image. Princeton: Princeton University Press.
Capon, N., & Hulbert, J. (1975). Decision systems analysis in industrial marketing. Industrial
Marketing Management, 4, 143–160.
References 413
Carlsmith, J. M., Ellsworth, P. C., & Aronson, E. (1976). Methods of research in social
psychology. Reading, MA: Addison-Wesley.
Celsi, R. L., Rose, R. L., & Leigh, T. W. (1993). An exploration of high risk leisure
consumption through skydiving. Journal of Consumer Research, 20(June), 1–23.
Chamberlin, T. C. (1890). The method of multiple working hypotheses. Science, 15(366),
92–96.
Chase, W. G., & Simon, H. A. (1973). Perceptions in chess. Cognitive Psychology, 4, 55–81.
Chebat, J. C., Vercollier, S. D., & Gelinas-Chebat, C. (2003). Drama advertisements:
Moderating effects of self-relevance on the relations among empathy, information
processing, and attitudes. Psychological Reports, 92, 997–1104.
Chen, S., & Anderson, S. (1999). Relationships from the past in the present: Significant-other
representations and transference in interpersonal life. In: M. P. Zanna (Ed.), Advances in
experimental social psychology (Vol. 31, pp. 123–190). San Diego: Academic Press.
Choffray, J.-M., & Lilien, G. L. (1978). Assessing response to industrial marketing strategy.
Journal of Marketing, 42(2), 20–31.
Choi, I., Nisbett, R. E., & Norenzayan, A. (1999). Causal attribution across cultures:
Variation and universality. Psychological Bulletin, 125, 47–63.
Christensen, C. M., & Bower, J. L. (1996). Customer power, strategic investment, and the
failure of leading firms. Strategic Management Journal, 17(2), 197–218.
Christensen, G. L., & Olson, J. C. (2002). Mapping consumers’ mental models with ZMET.
Psychology & Marketing, 19(6), 477–502.
Clark, T. (1990). International marketing and national character: A review and proposal for
an integrative theory. Journal of Marketing Research, 54(4), 69–79.
Cohen, M., March, J., & Olsen, J. (1972). A garbage can model of organizational choice.
Administrative Science Quarterly, 17, 1–24.
Coulter, R. A., Zaltman, G., & Coulter, K. S. (2001). Interpreting consumer perceptions
of advertising: An application of the Zaltman metaphor elicitation technique. Journal of
Advertising, 30(4), 1–21.
Christensen, C. M. (1997). The innovator’s dilemma. Cambridge, MA: Harvard Business
School Press.
Churchill, G. A., Jr. (1991). Marketing research: Methodological foundations. Chicago: Dryden
Press.
Chute, H. (2008). Comics as literature? Reading graphic narrative. PMLA, 123(March), 452–465.
Clark, H. H., & Schober, M. F. (1992). Asking questions and influencing answers. In:
J. M. Tanur (Ed.), Questions about questions (pp. 15–48). New York: Sage.
Cohen, M., March, J., & Olsen, J. (1972). A garbage can model of organizational choice.
Conklin, H. (1968). Ethnography. In: D. L. Sills (Ed.), International encyclopedia of social
science (pp. 115–208). New York: Free Press.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation. Boston: Holt Mifflin.
Cooper-Thomas, H., van Vianen, A., & Anderson, N. (2004). Changes in person —
Organization fit: The impact of socialization tactics on perceived and actual P–O fit.
European Journal of Work and Organizational Psychology, 13(1), 52–78.
Coopey, J., Keegan, O., & Emler, N. (1997). Managers’ innovations as ‘‘sense-making’’.
British Journal of Management, 8(3), 301–315.
Cox, D. (1967). An in depth study of consumer buying behavior. In: D. Cox (Ed.),
Risk taking and information handling in consumer behaviour. Boston, MA: Harvard
University Press.
Crotts, J. C. (1999). Consumer decision making and prepurchase information search.

In: A. Pizam & Y. Mansfeld (Eds), Consumer behavior in travel and tourism (pp. 149–168).
New York/London/Oxford: The Haworth Press.
Cyert, R. M., & March, J. G. (1963). A behavioral theory of the firm. Englewood Cliffs, NJ:
Prentice Hall.
Cyert, R. M., Simon, H. A., & Trow, D. B. (1956). Observation of a business decision. Journal
of Business, 29(October), 237–238.
Dawar, N., & Parker, P. (1994). Marketing universals: Consumers’ use of brand name, price,
physical appearance, and retailer reputation as signals of product quality. Journal of
Marketing, 58(2), 81–95.
Dean, J. W. (1986). Decision processes in the adoption of advanced technology. Working Paper.
College of Business Administration, Pennsylvania State University, University Park, PA.
Dearborn, D. C., & Simon, H. A. (1958). Selective perception: A note on the departmental
identification of executives. Sociometry, 21, 140–144.
de Chernatony, L. (1999). Brand management through narrowing the gap between brand
identity and brand reputation. Journal of Marketing Management, 15(1/3), 157–179.
de Chernatony, L., Cottam, S., & Segal-Horn, S. (2006). Communicating services brands’
values internally and externally. The Services Industries Journal, 26(8), 819–836.
Decrop, A. (1999a). Tourists’ decision-making and behavior processes. In: A. Pizam &
Y. Mansfeld (Eds), Consumer behavior in travel and tourism (pp. 103–133). New York/
London/Oxford: The Haworth Press.
Decrop, A. (1999b). Qualitative research methods for the study of tourist behavior. In:
A. Pizam & Y. Mansfeld (Eds), Consumer behavior in travel and tourism (pp. 335–365).
New York/London/Oxford: The Haworth Press.
Delgadillo, Y., & Escalas, J. E. (2004). Narrative word of mouth communication: Exploring
memory and attitude effects of consumer storytelling. Advances in Consumer Research, 31,
186–192.
Denzin, N. (1978). Sociological methods: A sourcebook (2nd ed.). New York: McGraw-Hill.
Denzin, N. K., & Lincoln, Y. S. (1994). Introduction: Entering the field of qualitative research.
In: N. K. Denzin & Y. S. Lincoln (Eds), Handbook of qualitative research. Thousand Oaks,
CA: Sage.
Dichter, E. (1960). The strategy of desire. New York: McGraw-Hill.
Dichter, E. (1964). Handbook of consumer motivations: The psychology of the world of objects.
New York: McGraw-Hill.
Dichter, E. (1985). Dichter on consumer motivation. New York: Elsevier/North-Holland.
Dillman, D. A. (1978). Mail and telephone surfers: The total design method. New York: Wiley-
Interscience.
Doyle, J. R., & Sims, D. (2002). Enabling strategic metaphor in conversation: A technique
for cognitive sculpting for explicating knowledge. In: A. S. Huff & M. Jenkins (Eds),
Mapping strategic thought (pp. 63–85). Thousand Oaks, CA: Sage.
Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organizational images and member
identification. Administrative Science Quarterly, 39(2), 239–263.
Dwyer, R., Schurr, P. H., & Oh, S. (1987). Developing buyer–seller relationships. Journal of
Marketing, 51, 11–27.
Dyer, W. G., Jr., & Wilkins, A. L. (1991). Better stories, not better constructs, to generate
better stories, a rejoinder to Eisenhardt. Academy of Management Review, 16(3), 613–619.
Eden, C., Jones, S., Sims, D., & Smithin, T. (1979). The intersubjectivity of issues and issues
of intersubjectivity. Working paper, and appears in (1981) Journal of Management Studies,
18, 37–47.
References 415
Eichenwald, K. (2000). The informant. New York: Broadway Books, Inc.

Eisenhardt, K. (1989). Building theories from case study research. Academy of Management
Review, 14(4), 532–550.
Ellis, C. (1991). Sociological introspection and emotional institution. Symbolic Interaction,
14(1), 23–50.
Ericsson, K. A., & Simon, H. A. (1993). Protocol analysis: Verbal reports as data. Cambridge,
MA: MIT Press.
Escalas, J. E. (1998). Advertising narratives: What are they and how do they work? In: B. Stern
(Ed.), Representing consumers: Voices, views, and visions (pp. 267–289). New York:
Routledge Press.
Escalas, J. E. (2004). Narrative processing: Building consumer connections to brands. Journal
of Consumer Psychology, 14, 168–180.
Escalas, J. E., & Stern, B. B. (2003). Sympathy and empathy: Emotional responses to
advertising dramas. Journal of Consumer Research, 29, 566–578.
Esch, F.-R., Rutenberg, J., Strödter, K., & Vallaster, C. (2005). Verankerung der Markenidentität
durch behavioral branding. In: F. R. Esch (Ed.), Moderne markenführung. Grundlagen —
innovative ansätze — praktische umsetzungen (pp. 985–1008). Wiesbaden: Gabler.
Evans, F. B. (1967). Selling as a dyadic relationship—A new approach. Canadian Review of
Sociology and Anthropology, 4(February), 24–40.
Ewert, J. C. (2000). Reading visual narrative: Art Spiegelman’s Maus. Narrative, 8, 87–103.
Evans, J. S. (2008). Dual-processing accounts of reasoning, judgment and social recognition.
Annual Review of Psychology, 59, 255–278.
Farley, J. U., & Howard, J. A. (1975). Control of ‘error’ in market research data. Lexington,
MA: Lexington Books.
Fauconnier, G. (1997). Mapping in thought and language. Cambridge, UK: Cambridge
University Press.
Feinson, R. (1998). The animal in you. New York: St. Martin’s Griffin.
Feldman, J. M., & Lynch, J. G. (1988). Self-generated validity and other effects of
measurement on belief, attitude, intention, and behavior. Journal of Applied Psychology, 73,
421–435.
Felman, S., & Laub, D. (1992). Foreword. Testimony: Crises of witnessing in literature,
psychoanalysis, and history (pp. xiii–xx). New York: Routledge.
Ferguson, E. S. (1977). The mind’s eye: Nonverbal thought in technology. Science, 197,
827–836.
Fernandez, D. R., Carlson, D. S., Stepina, L. P., & Nicholson, J. D. (1997). Hofstede’s country
classification 25 years later. Journal of Social Psychology, 137(1), 43–54.
Fielding, N. G., & Lee, R. M. (1991). Using computers in qualitative research. London: Sage.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to
theory and research. Reading, MA: Addison-Wesley.
Fisher, R. (1993). Social desirability bias and the validity of indirect questioning. Journal of
Consumer Research, 20(September), 303–315.
Fiske, S. T. (1993). Social cognition and social perception. In: L. Porter & M. Rosenzweig
(Eds), Annual review of psychology. Palo Alto, CA: Annual Reviews Inc.
Fodness, D., & Murray, B. (1997). Tourist information search. Annals of Tourism Research,
24(3), 503–523.
Fodness, D., & Murray, B. (1998). A typology of tourist information search strategies. Journal
of Travel Research, 37(2), 108–119.
Fodness, D., & Murray, B. (1999). A model of tourist information search behavior. Journal of
Travel Research, 37(3), 220–230.
Fontenot, R. J., & Wilson, E. J. (1997). Relational exchange: A review of selected models
for a prediction matrix of relationship activities. Journal of Business Research, 39, 5–12.
Ford, A. (1999). Modeling the environment. Washington, DC: Island Press.
Fournier, S. (1998). Consumers and their brands: Developing relationship theory in consumer
research. Journal of Consumer Research, 24(4), 343–374.
Freeman, D. (1983). Margaret Mead and Samoa: The making and unmaking of an
anthropological myth. Cambridge: Harvard University Press.
French, J. R. P., & Raven, B. (1959). The bases of social behavior. In: D. Cartwright (Ed.),
Studies in social power. Ann Arbor, MI: University of Michigan.
Freud, S. ([1908] 1965). The interpretation of dreams. New York: Avon.
Freud, S. (1923). The ego and the id. New York: W.W. Norton.
Gadel, M. S. (1964). Concentration by salesman on congenial prospects. Journal of Marketing,
28(April), 64–66.
Gaither, C. (2002). How much do doctors really know? The Boston Globe Magazine,
pp. 12–14ff.
Gardner, W. (1990). CONTIME: Continuous-time analysis of parallel streams of behavior.
Multivariate Behavioral Research, 25(2), 205–206.
Geertz, C. (1973a). The interpretation of cultures. New York, NY: Basic Books.
Geertz, C. (1973b). Thick description: Toward an interpretive theory of culture. In: C. Geertz
(Ed.), The interpretation of cultures: Selected essays (pp. 3–30). New York: Basic Books.
Gergen, K. J., & Gergen, M. M. (1988). Narrative and the self as relationship. Advances in
Experimental Social Psychology, 21, 17–56.
Gharajedaghi, J. (2006). Systems thinking: Managing chaos and complexity. Amsterdam:
Butterworth-Heinemann.
Gibbert, M. (2006). Generalizing about uniqueness. Journal of Management Inquiry, 15(2),
145–151.
Giddens, A. (1979). Central problems in social theory: Action, structure, and contradiction in
social analysis. London: Macmillan.
Giddens, A. (1984). The constitution of society: Outline of the theory of structuration.
Cambridge: Polity.
Gigerenzer, G. (2000). Adaptive thinking: Rationality in the real world. New York: Oxford
University Press.
Gigerenzer, G. (2001). The adaptive toolbox. Bounded rationality: The adaptive toolbox, Gerd
Gigerenzer & Reinhard Selten (pp. 37–50). Cambridge, MA: MIT Press.
Gigerenzer, G. (2002). Adaptive thinking. Oxford, UK: Oxford University Press.
Gigerenzer, G. (2007). Gut feeling: The intelligence of the unconscious. New York: Viking.
Gigerenzer, G. (2008). Rationality for mortals: How people cope with uncertainty. Oxford:
Oxford University Press.
Gigerenzer, G., & Todd, P. M.The ABC Research Group. (1999). Simple heuristics that make
us smart. New York: Oxford University Press.
Gilmore, M. J., & Murphy, J. (1991). Understanding classroom environments: An organiza-
tional sensemaking approach. Educational Administration Quarterly, 27(3), 392–429.
Gilovich, T. (1991). How we know what isn’t so. New York: Free Press.
Gilovich, T., Griffin, D., & Kahneman, D. (Eds). (2002). Heuristics and biases. New York:
Cambridge University Press.
Givhan, R. (2005). Wookie at me! Clothes make the myth. Washington Post, May 10,
p. CO1. Available at http://www.washingtonpost.com/wp-dyn/content/article/2005/05/19/
AR2005051901877.html
References 417
Gladwell, C. (1989). Ethnographic decision tree modeling. Thousand Oaks, CA: Sage.
Gladwell, M. (1996). The tipping point. The New Yorker, 72, 32–39.
Gladwell, M. (2002). Tipping point: How little things can make a big difference. Boston: Back
Bay Books.
Gladwell, M. (2005). Blink: The power of thinking without thinking. Boston: Back Bay.
Gladwin, C. H. (1989). Ethnographic decision tree modeling. Thousand Oaks, CA: Sage.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory. Chicago: Aldine.
Goatly, A. (1997). The language of metaphors. London: Routledge.
Goffman, E. (1959). The presentation of self in everyday life. Garden City, NJ: Doubleday.
Golder, P., & Tellis, G. J. (1993). Pioneer advantage: Marketing logic or marketing legend?
Journal of Marketing Research, 30(2), 158–170.
Golder, P. N. (2000). Historical method in marketing research with new evidence on long-term
market share stability. Journal of Marketing Research, 37(2), 156–172.
Goldsmith, E. (1971). The limits of growth of natural systems. General Systems, 16, 69–75.
Gould, G. (2005). Comparing, not confirming personal introspection: A comment on
Woodside. Psychology & Marketing, 23, 253–256.
Gould, S. (1991). The self-manipulation of my pervasive, perceived vital energy through
product use: An introspective-praxis perspective. Journal of Consumer Research, 18,
194–207.
Gould, S. J. (1995). Researcher introspection as a method in consumer research: Applications,
issues, and implications. Journal of Consumer Research, 21(4), 719–722.
Green, K. C., & Armstrong, J. S. (2007a). Structured analogies for forecasting. International
Journal of Forecasting, 23(3), 365–376.
Green, K. C., & Armstrong, J. S. (2007b). The value of expertise for forecasting decisions in
conflicts. Interfaces, 37(3), 287–299.
Green, K. C., & Tashman, L. (2008). Should we define forecast error as e ¼ F A or
e ¼ A F? Foresight, 10(1), 38–40.
Grice, H. P. (1975). Logic and conversation. In: P. Cole & J. L. Morgan (Eds), Syntax and
semantics, Volume 3: Speech acts (pp. 41–58). New York: Academic Press.
Grove, S. J., & Fisk, R. P. (1992). Observational data collection methods for services
marketing: An overview. Journal of the Academy of Marketing Science, 20, 217–240.
Gutman, J. (1997). Means-end chains as goal hierarchies. Psychology & Marketing, 14(6),
545–560.
Gutman, J., & Reynolds, T. J. (1978). An investigation of the levels of cognitive abstraction
utilized by consumers in product differentiation. In: J. Eighmey (Ed.), Attitude research
under the sun (pp. 57–64). Chicago: American Marketing Association.
Hakansson, H. (Ed.) (1982). International marketing and purchasing of industrial goods.
New York/Chichester: Wiley.
Hall, R. I. (1976). A system pathology on an organization: The rise and fall of the old Saturday
evening post. Administrative Science Quarterly, 21(2), 185–211.
Hall, R. I. (1984). The natural logic of management policy making: Its implications for the
survival of an organization. Management Science, 30, 905–927.
Hall, R. I. (1991). The policy maps of managers. Working Paper. Drake Centre for
Management Studies, University of Manitoba, Winnipeg.
Hall, R. I. (1999). A study of policy formation in complex organizations: Emulating group
decision-making with a simple artificial intelligence and a system model of corporate
operations. Journal of Business Research, 45, 157–171.
Hardin, G. (1968). The tragedy of the commons. Science, 162, 1243–1248.
Hedberg, B. (1981). How organizations learn and unlearn. In: P. C. Nystrom &
W. H. Starbuck (Eds), Handbook of organizational design (Vol. 1, pp. 3–27). New York:
Oxford University Press.
Heider, F. (1958). The psychology of interpersonal relations. Wiley.
Heimann, P. (1956). Dynamics of transference interpretations. International Journal of Psycho-
Analysis, 37, 303–310.
Heisley, D. D., & Levy, S. J. (1991). Autodriving: A photoelicitation technique. Journal of
Consumer Research, 18(December), 257–272.
Hersen, M., & Barlow, D. H. (1976). Single-case experimental designs: Strategies for studying
behavior change. New York: Pergamon.
High Performance Systems. (1995). STELLA: High performance systems. Available at http://
www.hps-inc.com/
Hilton, D. J. (1990). Conversational processes and causal explanation. Psychological Bulletin,
107, 81–86.
Hilton, D. J. (1995). The social context of reasoning: Conversational inference and rational
judgment. Psychological Bulletin, 118(2), 248–271.
Hiltunen, A. (2002). Aristotle in Hollywood: The anatomy of successful storytelling. Bristol,
UK: Intellect Books.
Hirschman, E. (1986). Humanistic inquiry in marketing research: Philosophy, methods, and
criteria. Journal of Marketing Research, 13(August), 237–250.
Hirschman, E. (2000a). Consumers’ use of intertextuality and archetypes. Advances in
Consumer Research, 27, 57–63.
Hirschman, E. (2000b). Heroes, monsters, and messiahs. New York: Andrews McMeel.
Hirschman, E. C. (1992). The consciousness of addiction: Toward a general theory of
compulsive consumption. Journal of Consumer Research, 19(September), 155–179.
Hofstede, G. (1980). Culture’s consequences: International differences in work-related values.
Beverly Hills, CA: Sage.
Hofstede, G. (1983). National cultures in four dimensions: A research-based theory of
cultural differences among nations. International Studies of Management and Organization,
8, 46–74.
Hofstede, G. (1991). Culture and organizations: Software of the mind. London: McGraw-Hill.
Hofstede, G. (2001). Culture’s consequences. Thousand Oaks, CA: Sage.
Hofstede, G. (2003). Culture’s consequences: International differences in work-related values
(2nd ed.). Beverly Hills, CA: Sage.
Hogarth, R. (1987). Judgment and choice (2nd ed.). Chichester: Wiley.
Hogarth, R. M. (2001). Educating intuition. Chicago: University of Chicago Press.
Holbrook, M. B. (1986). I’m hip: An autobiographical account of some consumption
experiences. In: R. J. Lutz (Ed.), Advances in consumer research (Vol. 13, pp. 614–618).
Provo, UT: Association for Consumer Research.
Holbrook, M. B. (1988). The psychoanalytic interpretation of consumer behavior: I am an
animal. Research in Consumer Behavior, 3, 149–178.
Holbrook, M. B. (1995). Consumer research: Introspective essays on the study of consumption.
Sage: Thousand Oaks, CA.
Holbrook, M. B. (1999). Consumer value: A framework for analysis and research. London:
Routledge.
Holbrook, M. B. (2003). Time travels in retrospace: Unpacking my grandfather’s trunk —
Some introspective recollections of Brule River. In: S. Brown & J. Sherry (Eds), Retroscapes
(pp. 218–241). Thousand Oaks, CA: Sage.
References 419
Holbrook, M. B. (2005a). Consumption experience, customer value, and subjective personal

introspection: An illustrative photographic essay. Journal of Business Research, 59, 714–725.
Holbrook, M. B. (2005b). Customer value and autoethnography: Subjective personal intro-
spection and the meanings of a photograph collection. Journal of Business Research, 58, 45–61.
Holbrook, M. B., & Kuwahara, T. (1998). Collective stereographic photo essays: An
integrated approach to probing consumption experiences in depth. International Journal of
Research in Marketing, 15(2), 201–221.
Holt, D. B. (1991). Rashomon visits consumer behavior: An interpretive critique of naturalistic
inquiry. Advances in Consumer Research, 18, 57–62.
Holt, D. B. (2002). Why do brands cause trouble? A dialectical theory of consumer culture and
branding. Journal of Consumer Research, 29, 70–90.
Holt, D. B. (2003). What becomes an icon most? Harvard Business Review, 3, 43–49.
Holt, D. B. (2004). How brands become icons. Cambridge, MA: Harvard University Business
School Press.
Holt, D. B., & Thompson, C. J. (2004). Man-of-action heroes: The pursuit of heroic
masculinity in everyday consumption. Journal of Consumer Research, 31, 425–440.
Homans, G. C. (1950). The human group. New York: Harcourt, Brace & Company.
Homans, G. C. (1974). Social behavior: Its elementary forms. New York: Harcourt, Brace, and
Jovanovich.
Hoppe, M. H. (1990). A comparative study of country elites: International differences in
work related values and learning and their implications for international management training
and development. Unpublished doctoral dissertation. University of North Carolina, Chapel
Hill, CA.
Howard, J. A., Hulbert, J., & Farley, J. U. (1975). Organizational analysis and information-
systems design: A decision-process perspective. Journal of Business Research, 3(April),
133–148.
Howard, J. A., & Morgenroth, W. M. (1968). Information processing model of executive
decision. Management Science, 14(3), 416–428.
Huff, A. S. (1990). Mapping strategic thought. Chichester, UK: Wiley.
Huff, A. S., Huff, J. O., & Barr, P. (2001). When firms change direction. New York: Oxford
University Press.
Huff, S. A., & Jenkins, M. (2002). Mapping strategic knowledge. London: Sage.
Hulbert, J. (2003). Organizational analysis and information system design: A road revisited.
Journal of Business and Industrial Marketing, 18(6/7), 509–513.
Humphreys, L. (1970). Tearoom trade: Impersonal sex in public places. Society, 7(3), 10–25.
Hunt, S. D. (1993). Objectivity in marketing theory and research. Journal of Marketing,
57(April), 76–91.
Hunter, J. E., Schmidt, F. L., & Jackson, G. B. (1982). Meta-analysis: Cumulating Research
findings across studies. Thousand Oaks, CA: Sage.
Inkeles, A., & Levinson, D. J. (1969). National character: The study of model personality and
sociocultural system. In: G. Lindzey, & E. Aronson (Eds), The handbook of social
psychology (2nd ed., Vol. 4). Cambridge, MA: Addison-Wesley.
Iser, W. (1976). The act of reading: A theory of aesthetic response. Baltimore, MD: John
Hopkins University Press.
Iverson, T. J. (1997). Decision timing: A comparison of Korean and Japanese travelers.
International Journal of Hospitality Management, 16(2), 209–219.
Jain, P. C., & Daljeet, D. (2006). Narration: The essence of Buddhist art. Available at http://
www.exoticindiaart.com/acrobat/buddhistart.pdf
Johnston, R. J., & Tyrrell, T. J. (2005). A dynamic model of sustainable tourism. Journal of
Travel Research, 44(November), 124–134.
Jung, C. (1969). Archetypes and the collective unconscious (The collected works of C. G. Jung,
Vol. 9, Part 1). Princeton, NJ: Princeton University Press.
Jung, C. G. (1916/1959). The archetypes and the collective unconscious. In: H. Read,
M. Fordham, & G. Adler (Eds), Collective works (Vol. 9, Part 1). Princeton, NJ: Princeton
University Press.
Jung, C. G. (2009). The red book. New York: Norton.
Kagitc- ibasi, C. (1997). Individualism and collectivism. In: J. W. Berry, M. Segall &
C. Kagitcibasi (Eds), Handbook of cross-cultural psychology: Social behavior and application
(2nd ed., pp. 1–49). Boston: Allyn and Bacon.
Kahneman, D., & Frederick, S. (2002). Representativeness revisited: Attribute substitution in
intuitive judgment. In: T. Gilovich, D. Griffin & D. Kahneman (Eds), Heuristics and biases:
The psychology of intuitive judgment (pp. 49–81). Cambridge: Cambridge University Press.
Kaplan, A. (1964). The logic of inquiry. San Francisco: Chandler.
Kassarjian, H. (1983). Social values and the Sunday comics: A content analysis. In:
R. P. Bagozzi & A. M. Tybout (Eds), Advances in consumer research (Vol. 10, pp. 434–438).
Ann Arbor, MI: Association for Consumer Research.
Kassarjian, H. (1984). Males and females in the funnies: A content analysis. In: R. E. Pitts, Jr.
& A. G. Woodside (Eds), Personal values and consumer psychology (pp. 87–109). Lexington,
MA: Lexington Books.
Kassarjian, H. H. (1977). Content analysis in consumer research. Journal of Consumer
Research, 4, 8–18.
Kaufman, R., & Kaufman, F. (1980). The face schema in 3- and 4-month-old infants: The role
of dynamic properties of the face. Infant Behavior and Development, 3, 331–339.
Kelley, H. H. (1967). Attribution theory in social psychology. In: D. Levine (Ed.),
Nebraska symposium on motivation (Vol. 15, pp. 192–240). Lincoln: University of Nebraska
Press.
Kelley, H. H. (1973). The process of causal explanations in interaction: Some dialogues on
dialogue. In: J. H. Harvey, W. J. Ickes & R. F. Kidd (Eds), New directions in attribution
research (Vol. 3, pp. 307–328). Hillsdale, NJ: Lawrence Erlbaum.
Kelvin, P. (1971). The bases of social behaviour — An approach in term of order and value.
London: Holt, Rinehart & Winston Ltd.
Kepner, C., & Tregoe, B. (1965). The rational manager. New York: McGraw-Hill.
King, R. L., & Woodside, A. G. (2000). Qualitative comparative analysis of travel and tourism
consumption systems. Tourism Analysis, 5(2–4), 105–111.
Klein, M. (1952). The origins of transference. International Journal of Psycho-Analysis, 33,
433–438.
Kluckhohn, F., & Strodtbeck, R. (1961). Variations in value orientations. Evanston, IL: Row,
Peterson and Company.
Kluth, A. (2006). Among the audience: A survey of new media. The Economist, April 22,
pp. 3–20.
Kotler, P., & Pfoertsch, W. (2007). Being known or being one of many: The need for brand
management for business-to-business (B2B) companies. Journal of Business & Industrial
Marketing, 22(6), 357–362.
Kozak, A. (1996). Local clinicians need knowledge tools. American Psychologist, 51, 1335–1336.
Kozinets, R. V. (2002). The field behind the screen: Using netnography for marketing research
in online communities. Journal of Marketing Research, 39, 61–72.
References 421
Krauss, A., & Union Station. (2008). When you say nothing at all. Available at http://
www.youtube.com/watch?v ¼ QjsjZWlRVvo
Kroeber, A. L., & Kluckhohn, C. (1952/1994). Culture: A critical review of concepts and
definitions. Peabody Business Review, 35(July–August), 113–121.
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago Press.
Langer, E. J. (1975). The illusion of control. Journal of Personality & Social Psychology, 32,
311–328.
Langley, A. (1999). Strategies for theorizing from process data. Academy of Management
Review, 24(4), 691–710.
Langley, A., Mintzberg, H., Pitcher, P., Posada, E., & Saint-Macary, J. (1995). Opening up
decision-making: The view from the black stool. Organization Science, 6(3), 260–279.
Laughlin, H. P. (1970). The ego and its defenses. New York: Appleton-Century-Crofts.
Lazarsfeld, P. F. (1937). Some remarks on the typological procedures in social research.
Festschrift fur Sozialforschung, 6(1), 119–139.
Leigh, T. W., & McGraw, P. F. (1989). Mapping the procedural knowledge of industrial sales
personnel: A script-theoretic investigation. Journal of Marketing, 53, 16–34.
LePla, J. F., & Parker, L. M. (1999). Integrated branding: Becoming brand-driven through
company-wide action. London: Quorum Books.
Levin, I. P., Schneider, S. L., & Gaeth, G. J. (1998). All frames are not created equal:
A typology and critical analysis of framing effects. Organizational Behavior and Human
Decision Processes, 76(2), 149–188.
Levine, G. M., Halberstadt, J. B., & Goldstone, R. L. (1996). Reasoning and the weighting of
attributes in attitude judgments. Journal of Personality & Social Psychology, 70, 230–240.
Levitt, T. (1983). The globalization of markets. Harvard Business Review, 61(3), 92–102.
Levy, S. J. (1996). Stalking the amphisbaena. Journal of Consumer Research, 23, 163–176.
Lichenstein, S., Fischoff, B., & Phillips, L. (1982). Calibration and probabilities: The state
of the art in 1980. In: D. Kahneman, P. Slovic & A. Tversky (Eds), Judgment under
uncertainty: Heuristics and biases (pp. 306–334). Cambridge: Cambridge University Press.
Lindemans, M. F. (2009). Medusa. Encyclopedia Mythica. Available at http://www.pantheon.
org/articles/m/medusa.html
Lindquist, J. D. (1975). Meaning of image. Journal of Retailing, 50(4), 29–38.
Lippman, P. C. (2007). Practice theory, pedagogy, and the design of learning environments:
Knowledge communities. Retrieved October 24, 2007. Available at http://www.aia.org/
cae_a_20031101_justathought
List, J. A. (2006). The behavioralist meets the market: Measuring social preferences and
reputation effects in actual transactions. Journal of Political Economy, 114(1), 1–37.
Loewenstein, G. (2001). The creative destruction of decision research. Journal of Consumer
Research, 28(December), 499–505.
Lombard, G. F. F. (1955). Behavior in a selling group. Boston: Division of Research, Graduate
School of Business Administration, Harvard University.
Lynn, M., Zinkhan, G. M., & Harris, J. (1993). Consumer tipping: A cross-country study.
Journal of Consumer Research, 20(3), 478–488.
Maani, K. E., & Majaraj, V. (2004). Links between systems thinking and complex decision
making. System Dynamics Review, 20(1), 21–48.
Malle, B. F. (1999). How people explain behavior: A new theoretical framework. Personality
and Social Psychology Review, 3(1), 23–48.
Malle, B. F. (2004). How the mind explains behavior. Cambridge, MA: Bradford Book/MIT
Press.
Malle, B. F., & Bennett, R. E. (1998). People’s praise and blame for intentions and actions:
Implications of the folk concept of intentionality. Unpublished manuscript.
Malle, B. F., & Knobe, J. (1997). The folk concept of intentionality. Journal of Experimental
Social Psychology, 33, 101–121.
March, J. G., Sproull, L. S., & Tamuz, M. (1991). Learning from samples of one or fewer.
Organization Science, 2(1), 1–13.
March, J. G., & Sutton, R. I. (1997). Organizational performance as a dependent variable.
Organizational Science, 8(6), 698–706.
Mark, M., & Pearson, C. S. (2001). The hero and the outlaw: Building extraordinary brands
through the power of archetypes. New York: McGraw-Hill.
Marketing to the Old. (2002). Economist, 10(August), 51–52.
Martin, R. (2007). The opposable mind. Boston: Harvard Business School Press.
Martin, W. (1986). Recent theories of narrative. Ithaca, NY: Cornell University Press.
Martin, W. (2005). Real perspectival selves. Theory & Psychology, 15, 207–224.
Masion, D., Greenwald, A. G., & Bruin, R. H. (2004). Predictive validity of the implicit
association test in studies of brands, consumer attitudes, and behavior. Journal of Consumer
Psychology, 14, 405–415.
McCarthy, J. E., & Perreault, W. D., Jr. (1993). Basic marketing (11th ed.). Boston, MA:
Richard D. Irwin, Inc.
McClelland, D. C. (1961). The achieving society. Princeton, NJ: D. Van Norstrand Company, Inc.
McClelland, D. C. (1988). Human motivation. Cambridge: Cambridge University Press.
McClelland, D. C. (1998). Identifying competencies with behavioral-event interviews.
Psychological Science, 9(5), 331–339.
McClelland, D. C., Koestner, R., & Weinberger, J. (1992). How do self-attributed and implicit
motives differ? In: C. P. Smith (Ed.), Motivation and personality: Handbook of thematic
content analysis (pp. 49–72). Cambridge, UK: Cambridge University Press.
McClure, J., & Hilton, D. J. (1997). Because you can’t always get what you want: When
preconditions are better explanations than goals. British Journal of Social Psychology, 36,
223–240.
McClure, J. L. (2002). Goal-based explanations of actions and outcomes. European Review of
Social Psychology, 12, 201–235.
McCort, D. J., & Malhotra, N. K. (1993). Culture and consumer behaviour: Toward an
understanding of cross-culture consumer behaviour in international marketing. Journal of
International Consumer Marketing, 6(2), 91–127.
McCracken, G. (1987). Culture and consumption among the elderly: Three research objectives
in an emerging field. Aging and Society, 7, 203–224.
McCracken, G. (1988). The long interview. Thousand Oaks, CA: Sage.
McKee, R. (2003). Storytelling that moves people: A conversation with screenwriting coach,
Robert McKee. Harvard Business Review, 6, 51–55.
McKercher, B. (2003). Adopting a marketing approach to achieve sustainable cultural
tourism. International Journal of Tourism Sciences, 3(1), 129–142.
McSweeney, B. (2002). Hofstede’s model of national cultural differences and their
consequences: A triumph of faith – A failure of analysis. Human Relations, 55(1), 89–118.
Mead, M. (1928). Coming of age in Samoa: A psychological study of primitive youth for Western
civilization. New York: William Morrow.
Merton, R. K. (1957). Social theory and social structure (2nd ed.). Glencoe, IL: Free Press.
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook
(2nd ed.). Thousand Oaks, CA: Sage.
References 423
Miles, S. J., & Mangold, W. G. (2005). Positioning southwest airlines through employee
branding. Business Horizons, 48(6), 535–545.
Miller, D., & Friesen, P. H. (1982). Structural change and performance: Quantum versus
piecemeal-incremental approaches. Academy of Management Journal, 25, 867–892.
Miller, D. C. (1991). Handbook of research design and social measurement. Newbury Park, CA:
Sage.
Mink, L. O. (1978). Narrative form as a cognitive instrument. In: R. H. Canary (Ed.),
The writing of history (pp. 129–149). Madison, WI: University of Wisconsin Press.
Mintzberg, H. (1979). An emerging strategy of ‘direct’ research. Administrative Science
Quarterly, 24(4), 582–589.
Mintzberg, H., Raisinghani, D., & Theoret, A. (1976). The structure of ‘‘unstructured’’
decision processes. Administrative Science Quarterly, 21(2), 246–275.
Mitroff, I. I., & Kilmann, R. H. (1976). On organization stories: An approach to the design
and analysis of organizations through myths and stories. In: R. H. Kilman, L. R. Pondy &
D. P. Slevin (Eds), The management of organization design (Vol. 1, pp. 189–207). New York:
North-Holland.
Mohr, J., & Spekman, R. B. (1994). Characteristics of partnership success: Partnership
attributes, communication behavior, and conflict resolution techniques. Strategic Management
Journal, 15, 135–152.
Moisio, R., & Arnould, E. J. (2005). Extending the dramaturgical framework in marketing:
Drama structure, drama interaction and drama content in shopping experiences. Journal of
Consumer Behaviour, 4, 1–11.
Money, R. B., & Crotts, J. C. (2003). The effect of uncertainty avoidance on information
search, planning, and purchases of international travel vacation. Tourism Management,
24(2), 191–202.
Montgomery, D. F. (1975). New product distribution: An analysis of supermarket buyer
decisions. Journal of Marketing Research, 12(August), 255–264.
Moore, A. (1985). Rosanzerusu is Los Angeles: An anthropological inquiry of Japanese
tourists. Annals of Tourism Research, 12, 619–643.
Morgan, R. M., & Hunt, S. (1994). The commitment-trust theory of relationship marketing.
Journal of Marketing, 58, 20–38.
Morgenroth, W. M. (1964). A method for understanding price determinants. Journal of
Marketing Research, 1(3), 17–26.
Morley, I. E. (1978). Bargaining and negotiation: The character of experimental studies. In:
H. Brandstatter, J. E. Davis & H. Schuler (Eds), Dynamics of group decisions (pp. 1975–2206).
Beverly Hills: Sage.
Morley, I. E., & Stephenson, G. N. (1977). The social psychology of bargaining. London:
George Allen and Unwin.
Moss, M. W. (2002). Why management theory needs Popper: The relevance of falsification. In:
Reason in Practice: The Developing Philosophy of Management Conference 26-29 June.
Unpublished paper. Oxford, UK.
Mowen, J. C., & Minor, M. (1998). Consumer behavior (5th ed.). Upper Saddle River, NJ:
Prentice-Hall.
Munby, H. (1986). Metaphor in the thinking of teachers: An exploratory study. Journal of
Curriculum Studies, 18(2), 197–209.
Na, W. B., Marshall, R., & Woodside, A. G. (2009). Decision system analysis of
advertising agency decisions. Qualitative Market Research: An International Journal,
12(2), 153–170.
Nakata, C., & Sivakumar, K. (1996). National culture and new product development: An
integrative review. Journal of Marketing, 60(1), 61–72.
Nataraajan, R., & Bagozzi, R. (1999). The year 2000: Looking back. Psychology & Marketing,
16, 631–642.
Newell, A., & Simon, H. A. (1972). Human problem solving. Englewood Cliffs, NJ: Prentice
Hall.
Nicosia, F. M., & Wind, Y. (1977). Behaviour models of organizational buying processes.
In: F. M. Nicosia & Y. Wind (Eds), Behavioural models for market analysis: Foundations for
marketing action (pp. 96–120). Hinsdale, IL: Dryden.
Nielsen, RL. (1986). Personal interview. Lincoln, Nebraska: SRI Research Center.
Nittoli, S. (2003). A confirmatory personal introspection of my university choice. Working Paper.
Department of Marketing, Boston College, 28 pp.
Nutt, P. C. (1984). Types of organizational decision processes. Administrative Science
Quarterly, 29(3), 414–450.
Nutt, P. C. (1993). The formulation processes and tactics used in organizational decision
making. Organization Science, 4(2), 226–251.
Nutt, P. C. (1998). Framing strategic decisions. Organization Science, 9(2), 195–216.
Office of Fair Trade. (2003). Queensland Government. Retrieved October 24, 2007. Available
at http://www.fairtrading.qld.gov.au
Olshavsky, R. W. (1973). Customer–salesman interaction in appliance retailing. Journal of
Marketing Research, 10(May), 208–212.
Orr, J. E. (1990). Sharing knowledge, celebrating identity: Community memory in a service
culture. In: D. Middleton & D. Edwards (Eds), Collective remembering (pp. 169–189).
London: Sage.
Pace, W. R. (1962). Oral communication and sales effectiveness. Journal of Applied
Psychology, 44(October), 487–488.
Padgett, D., & Allen, D. (1997). Communicating experiences: A narrative approach to creating
service brand image. Journal of Advertising, 26, 49–62.
Park, C. W., Hughes, R. W., Thurkal, V., & Friedman, R. (1981). Consumers’ decision plans
and subsequent choice behavior. Journal of Marketing, 45(Spring), 33–47.
Pattinson, H. M., & Woodside, A. G. (2007). Innovation and diffusion of software technology:
Mapping strategies. Oxford, UK: Elsevier.
Paun, D. (1997). A study of ‘‘best’’ versus ‘‘average’’ buyer–seller relationships. Journal of
Business Research, 39, 13–21.
Payne, J. W., Bettman, J., & Johnston, E. (1993). The adaptive decision maker. New York:
Cambridge Press.
Payne, J. W., Bettman, J. R., & Johnson, E. J. (1993). The adaptive decision maker. Cambridge,
UK: Cambridge University Press.
Payne, S. L. (1951). The art of asking questions. Princeton, NJ: Princeton University Press.
Peabody, D. (1985). National characteristics. Cambridge, UK: Cambridge University Press.
Pennington, A. L. (1968). Customer–salesman bargaining behavior in retail transactions.
Journal of Marketing Research, 5, 255–262.
Perfetto, R., & Woodside, A. G. (2009). Extremely frequent behavior in consumer research:
Theory and empirical evidence for chronic casino gambling. Journal of Gambling Studies,
25, 297–316.
Pettigrew, A. M. (1973). The politics of organizational decision making. London: Tavistock.
Pettigrew, A. M. (1975). The industrial purchasing decision as a political process. European
Journal of Marketing, 9(March), 4–19.
References 425
Pettigrew, A. M. (1995). Longitudinal field research on change: Theory and practice. In:
G. P. Huber & A. H. Van De Ven (Eds), Longitudinal field research methods (pp. 91–125).
Thousand Oaks, CA: Sage.
Pfeffer, J. (1981). Power in organizations. Boston, MA: Pittman Publishing.
Pieters, R., Baumgartner, H., & Allen, D. (1995). A means-end chain approach to consumer
goal structures. International Journal of Research in Marketing, 12(3), 227–244.
Pizam, A., & Sussmann, S. (1995). Does nationality affect tourist behavior. Annals of Tourism
Research, 22(4), 901–917.
Porac, J., & Shapira, Z. (2007). On mind, environment, and Simon’s scissors of rational
behavior. Journal of Management and Governance, 5, 206–212.
Posner, G. (1998). VW Day, sometimes buying a car means burying the past. New York Times
Magazine, October 4, p. 128.
Puccinelli, N. M., Braun, K., & Mast, F. (2002). Implicit predictors of consumer behavior
(Case No. 502-043). Cambridge, MA: Harvard Business School.
Quinn, J. B. (1980). Strategies for change: Logical incrementalism. Georgetown, Canada:
Irwin-Dorsey.
Ragin, C., Drass, K., & Davey, S. (2007). Fuzzy-Set/Qualitative Comparative Analysis 2.0.
Available at http://www.u.arizona.edu/Bcragin/fsQCA/software.shtml
Ragin, C. C. (1987). The comparative method: Moving beyond qualitative and quantitative
strategies. Berkeley: University of California Press.
Ragin, C. C. (1989). The comparative method: Moving beyond qualitative and quantitative
strategies. Berkeley, CA: University of California Press.
Ragin, C. C. (1997). Turning the tables: How case-oriented methods challenge variable
oriented methods. Comparative Social Research, 16(1), 27–42.
Ragin, C. C. (2000). Fuzzy-set social science. Chicago: University of Chicago Press.
Ragin, C. C. (2008). Redesigning social inquiry: Fuzzy sets and beyond. Chicago: Chicago
University Press.
Rapaille, C. (2004). Breaking the code of luxury (Archetype Discoveries). New York:
Worldwide.
Rapaille, C. (2006). The culture code: An ingenious way to understand why people around the
world live and buy as they do. New York: Broadway.
Raymore, L. (2002). Facilitators to leisure. Journal of Leisure Research, 34(1), 37–51.
Rebel without a Cause. (1955). Hollywood, CA: Warner Bros Pictures.
Reed, S. K. (2010). Thinking visually. New York: Psychology Press.
Repenning, N. P. (2001). Understanding fire fighting in new product development. Journal of
Product Innovation Management, 18(5), 285–300.
Rihoux, B., & Ragin, C. C. (2009). Configurational comparative methods. Los Angeles:
Sage.
Robinson, P. C., Faris, C. W., & Wind, Y. (1967). Industrial buying and creative marketing.
Boston: Allyn and Bacon.
Rosenbaum, M. S., & Spears, D. L. (2006). Who buys what? Who does that? The case of
golden week in Hawaii. Journal of Vacation Marketing, 12(3), 246–255.
Ross, L., & Nisbett, R. E. (1991). The person and the situation. New York: McGraw-Hill.
Roth, M. S. (1995). The effects of culture and socioeconomics on the performance of global
brand image strategies. Journal of Marketing Research, 32(2), 163–175.
Rsingh.net. (2005). Gender representation, February 6. Available at http://www.rsingh.net
Ryle, G. (1949). The concept of mind (New University of Chicago Press edition 2002). Chicago:
University of Chicago Press.
Sachs, H., Schegloff, E., & Jefferson, G. (1978). A simplest systematic for the organization of
turn-taking in conversation. In: J. Schenkein (Ed.), Studies in the organization of
conversational interaction (pp. 1–54). New York: Academic Press.
Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for the organization
of turn-taking in conversation. Language, 50, 696–735.
Sanday, P. R. (1979). The ethnographic paradigm(s). Administrative Science Quarterly,
24(December), 527–550.
Schank, R. C. (1990). Tell me a story: A new look at real and artificial memory. Cambridge,
UK: Cambridge University Press.
Schank, R. C. (1999). Dynamic memory revisited. Cambridge, UK: Cambridge University Press.
Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals and understanding: An inquiry into
human knowledge. Hillsdale, NJ: Erlbaum.
Schegloff, E. A. (1996). Turn organization: One intersection of grammar and interaction. In:
E. Ochs, E. A. Schegloff & S. Thompson (Eds), Interaction and grammar (pp. 52–133).
Cambridge, UK: Cambridge University Press.
Schegloff, E. A., Koshik, I., Jacoby, S., & Olsher, D. (2002). Conversation analysis and applied
linguistics. American Review of Applied Linguistics, 22, 3–31.
Schein, E. (1971). Organizational socialization. In: B. Hinton & H. Reitz (Eds), Groups and
organizations: Integrated readings in the analysis of social behavior (pp. 210–215). Belmont:
Wadsworth Publishing Company.
Schenkein, J. (1978). Studies in the organization of conversational interaction. New York:
Academic Press.
Schindler, R. M., & Holbrook, M. B. (2003). Nostalgia for early experiences as a determinant
of consumer preferences. Psychology & Marketing, 20, 275–302.
Schlenker, B. R. (1980). Impression management: The self-concept, social identity, and
interpersonal relations. Monterey, CA: Brooks/Cole.
Schwarz, N. (1999). Self-reports: How the questions shape the answers. American Psychologist,
54(2), 93–105.
Schwartz, S. H. (1994). Beyond individualism/collectivism: New cultural dimensions of values.
In: U. Kim, H. C. Triandis, C. Kagitcibasi, S. C. Choi & G. Yoon (Eds), Individualism and
collectivism: Theory, method and applications (pp. 85–119). Thousand Oaks, CA: Sage.
Schwartz, S. H. (1997). Values and culture. In: D. Munro, S. Carr & J. Schumaker (Eds),
Motivation and culture (pp. 69–84). New York: Rutledge.
Schwartz, S. H., & Bilsky, W. (1987). Towards a psychological structure of human values.
Journal of Personality and Social Psychology, 53(3), 550–562.
Schwartz, S. H., & Ros, M. (1995). Values in the west: A theoretical and empirical challenge to
the individual-collectivism cultural dimension. World Psychology, 1(2), 91–122.
Senge, P. M. (1990). The fifth discipline. New York: Doubleday.
Shank, R. C. (1990). Tell me a story: A new look at real and artificial intelligence. New York:
Scribner’s Sons.
Shank, R. C. (1999). Dynamic memory revisited. Cambridge, UK: Cambridge University Press.
Shimp, T., Hyatt, E., & Snyder, D. (1991). A critical appraisal of demand artifacts in consumer
research. Journal of Consumer Research, 18, 273–283.
Siegel, S. (1956). Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill.
Simon, H. A. (1960). The new science of management decision. New York: Harper & Row.
Simon, H. A. (1990). Invariants of human behavior. Annual Review of Psychology, 41, 1–19.
Simon, H. A. (1991). Bounded rationality and organizational learning. Organization Science,
2(1), 125–134.
References 427
Simonson, I., & Nowlis, S. M. (2000). The role of explanations and need for uniqueness in
consumer decision making: Unconventional choices based on reasons. Journal of Consumer
Sims, D. B. P., & Doyle, J. R. (1995). Cognitive sculpting as a means of working with
managers’ metaphors. OMEGA, 23(2), 117–124.
Sirsi, A. K., Ward, J. C., & Reingen, P. H. (1996). Microcultural analysis of variation in
sharing of causal reasoning about Behavior. Journal of Consumer Research, 22(4), 345–372.
Sivakumar, K., & Nakata, C. (2001). The stampede toward Hofstede’s framework: Avoiding
the sample design pit in cross-cultural research. Journal of International Business Studies,
32(3), 555–574.
Skinner, B. F. (1996). Operant behavior. In: W. K. Honig (Ed.), Operant behavior: Areas for
research and application (pp. 12–32). New York: Appleton-Century-Crofts.
Sloan, A. P., Jr. (1964). My years at general motors. New York: A Macfadden Book.
Smith, A. (2000). Leximancer: The document mapping system. Key Centre for Human
Factors and Applied Cognitive Psychology, University of Queensland. Available at www.
leximancer.com
Smith, P. B., Dugan, S., & Trompenaars, F. (1996). National culture and the values of
organizational employees: A dimensional analysis across 43 countries. Journal of Cross-
Cultural Psychology, 24(2), 231–264.
Smith, P. B., & Schwartz, S. H. (1997). Values. In: J. W. Berry, H. Segall & C. Kagitcibasi
(Eds), Handbook of cross-cultural psychology: Social behavior and application. (2nd ed., pp.
77–118). Boston: Allyn and Bacon.
Smith, R. A., & Lux, D. S. (1993). Historical method in consumer research: Developing causal
explanations of change. Journal of Consumer Research, 19(March), 595–610.
Soares, A. (2005). The influence of culture on consumers: Exploratory and risk taking behavior.
Unpublished doctoral dissertation. Escola de Economia e Gestão, University of Minho.
Soares, A., Farhangmehr, M., & Shoham, A. (2007). Hofstede’s dimensions of culture in
international marketing studies. Journal of Business Research, 60(3), 277–284.
Sojka, J. Z., & Tansuhaj, P. S. (1995). Cross-cultural consumer research: A twenty-year review.
Advances in Consumer Research, 22(1), 461–474.
Solomon, M. (1996). Consumer behavior (3rd ed.). Upper River Saddle, NJ: Prentice Hall.
Sood, S., Woodside, A. G., & Miller, K. (2007). Analysing iconic consumer brand weblogs.
American Association for Artificial Intelligence. Available at http://www.aaai.org/Papers/
Symposia/Spring/2006/SS-06-03/SS06-03-041.pdf
Spiegleman, A. (1986). Maus I: A survivor’s tale — My father bleeds history. New York:
Pantheon.
Spiegleman, A. (1991). Maus II: A survivor’s tale — And here my troubles began. New York:
Pantheon.
Spiggle, S. (1985). 7-up, Pepsi Art, and Sunkist Art. In: E. C. Hirschman & M. B. Holbrook
(Eds), Advances in consumer research (Vol. 12, pp. 11–16). Provo, UT: Association for
Consumer Research.
Spiggle, S. (1986). Measuring Social values: A content analysis of Sunday comics and
underground comix. Journal of Consumer Research, 13(June), 100–113.
Stanovich, K. E., & West, R. F. (1998). Cognitive ability and variation in selection task
performance. Think Reasoning, 4, 193–230.
Star, N. (1988). The international guide to tipping. New York: Berkeley.
Steenkamp, J. B. (2001). The role of national culture in international marketing research.
International Marketing Review, 18(1), 30–44.
Steenkamp, J. B., ter Hofstede, F., & Wedel, M. (1999). A cross-national investigation into the
individual and national cultural antecedents of consumer innovativeness. Journal of
Marketing, 63(2), 55–69.
Sterman, J. (2000). Business dynamics: Systems thinking and modeling for a complex world.
New York: McGraw-Hill/Irwin.
Stern, B. (1994). Classical and vignette television advertising dramas: Structural models,
formal analysis, and consumer effects. Journal of Consumer Research, 20, 601–615.
Sternthal, B., Tybout, A. M., & Calder, B. J. (1987). Confirmatory versus comparative
approaches to judging theory tests. Journal of Consumer Research, 14, 114–125.
Stevenson, W. B., Bartunek, J. M., & Borgatti, S. P. (2003). Front and backstage
processes of an organizational restructuring effort. Journal of Applied Behavioral Science,
39(3), 243–258.
Stinchcombe, A. L. (1968). Constructing social theories. New York: Harcourt, Brace,
Jovanovitch.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton.
Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Combining qualitative and
quantitative approaches. Thousand Oaks, CA: Sage.
Taylor, J. L. (1977). Exchange behavior among life insurance selling and buying centers in
naturalistic settings. Unpublished doctoral dissertation, College of Business Administration,
University of South Carolina, Columbia.
Taylor, J. L., & Woodside, A. G. (1979). Exchange behavior among salesman and customers in
natural settings. Working paper. University of South Carolina, Columbia.
Taylor, J. L., & Woodside, A. G. (1980). An examination of the structure of buying-selling
interactions among insurance agents and prospective customers. In: J. Olson (Ed.),
Advances in consumer research (Vol. 7). Ann Arbor, MI: Association for Consumer
Research.
Taylor, J. L., & Woodside, A. G. (1981). Exchange behavior among salesmen and customers
in natural settings. In: P. Reingen & A. G. Woodside (Eds), Buyer interactions: Empirical
research and normative issues (pp. 45–63). Chicago: American Marketing Association.
Tellis, G. J. (1988). The price sensitivity of competitive demand: A meta-analysis of sales
response models. Journal of Marketing Research, 15(3), 331–341.
Tellis, G. J., & Franses, P. H. (2006). Optimal data interval for advertising response models.
Marketing Science, 25(3).
Tellis, G. J., & Golder, P. N. (1996). First to market, first to fail? The real causes of enduring
market leadership. Sloan Management Review, 37(2), 65–75.
Tellis, G. J., Niraj, R., & Yin, E. (2009). Does quality win? Network effects versus quality in
high-tech markets. Journal of Marketing Research. Available at http://www-rcf.usc.edu/
Btellis/qualitywin.doc
Tellis, G. J., & Wernerfelt, B. (1987). Competitive price and quality under asymmetric
information. Marketing Science, 6(3), 240–253.
Terkel, S. (1974). Working: People talk about what they do all day and how they feel about what
they do. New York: Pantheon Books.
Thelen, E. M., & Woodside, A. G. (1997). What evokes the brand or store? International
Journal of Research in Marketing, 14(November), 125–143.
Thompson, C., Rindfleisch, A., & Arsel, Z. T. (2006). Emotional branding and the strategic
value of the Doppelgänger brand image. Journal of Marketing, 70(1), 50–64.
Thompson, C. J. (1997). Interpreting consumers: A hermeneutical framework for deriving
marketing insights from the texts of consumers’ consumption stories. Journal of Marketing
Research, 34(4), 438–455.
References 429
Thompson, C. J., Locander, W. B., & Pollio, H. R. (1989). Putting consumer experience back
into consumer research: The philosophy and method of existential-phenomenology. Journal
of Consumer Research, 16, 133–147.
Thompson, C. J., Pollio, R., & Locander, W. B. (1994). The spoken and the unspoken:
A hermeneutic approach to understanding the cultural viewpoints that underlie consumers’
expressed meanings. Journal of Consumer Research, 21(3), 432–452.
Thomson, K., de Chernatony, L., Arganbright, L., & Sajid, K. (1999). The buy-in benchmark:
How staff understanding and commitment impact brand and business performance. Journal
of Marketing Management, 15(8), 819–835.
Thorngate, W. (1976). ‘In general’ vs. ‘It depends’: Some comments on the Gergen–Schlenker
debate. Personality and Social Psychology Bulletin, 2(3), 404–410.
Tosti, D. T., & Stotz, R. D. (2001). Brand: Building your brand from the inside out. Marketing
Management, 10(2), 28–33.
Triandis, H. C. (1989). The self and social behavior in differing cultural contexts. Psychological
Review, 96(3), 506–520.
Tse, D. K., Lee, K.-H., Vertinsky, I., & Wehrung, D. A. (1988). Does culture matter? A cross-
cultural study of executives’ choice, decisiveness and risk adjustment in international
marketing. Journal of Marketing, 52(4), 81–95.
Tucker, W. T. (1967). Foundations for a theory of consumer behavior. New York: Holt,
Rinehart & Winston.
Tufte, E. R. (2001). The visual display of quantitative data. Providence, RI: Graphics Press.
Tulving, E. (1972). Episodic and semantic memory. In: E. Tulving & W. Donaldson (Eds),
Organization of memory (pp. 382–402). New York: Academic Press.
Tulving, E. (1985). How many memory systems are there? American Psychologist, 40, 385–398.
Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology, 53,
1–25.
Turnbull, W. (1986). Everyday explanation: The pragmatics of puzzle resolution. Journal for
the Theory of Social Behavior, 16, 141–160.
Urde, M. (2003). Core-value based corporate brand building. European Journal of Marketing,
37(7/8), 1017–1040.
Vallaster, C., & de Chernatony, L. (2005). Internationalisation of services brands: The role of
leadership during the internal brand building process. Journal of Marketing Management,
21(1–2), 181–203.
Vallaster, C., de Chernatony, L., & von Wallpach, S. (2006). Weblogs: How they contribute
to co-construct a corporate brand: The case of Gate Gourmet. In: L. de Chernatony &
S. Cottam (Eds), Proceedings of the thought leaders international conference on brand
management 2006, Birmingham.
Van Maanen, J (1978). The asshole. In: P. K. Manning & J. Van Maanen (Eds), Policing:
A view from the street. Santa Monica, CA: Goodyear Publishing Available at http://
petermoskos.com/readings/Van_Maanen_1978.pdf
Van Maanen, J. (1979). The fact of fiction in organizational ethnography. Administrative
Science Quarterly, 24(December)34–41, 539–550.
Van Someren, M. W., Barnard, Y. F., & Sandberg, J. A. C. (1994). The think aloud method.
London: Academic Press.
Varela, J. A. (1971). Psychological solutions to social problems: An introduction to social
technology. New York: Academic Press.
Vlosky, R. P. (1995). Sources of competitive advantage for wood products distribution suppliers.
Louisiana Forest Products Laboratory, Working Paper 12, Louisiana State University
Agricultural Center, Baton Rouge, LA.
Vlosky, R. P., & Wilson, DT. (1994). Technology adoption in channels. In: A. Parvatiyar &
J. Sheth (Eds), Proceedings of the 1994 research conference on relationship marketing.
Atlanta, GA: Emory University.
Von Hippel, E. (1988). Sources of innovation. Oxford: Oxford University Press.
Von Wallpach, S., & Woodside, A. (2009). Theory and practice of enacted internal branding:
Theory, practice, and an experiential learning case study of an Austrian B2B company.
In: M. Glynn & A. G. Woodside (Eds), Advances in business marketing and purchasing.
Volume 15: Business-to-business brand management (pp. 389–425). London: Emerald
Publishing.
Vyas, N., & Woodside, A. G. (1984). An inductive model of industrial supplier choice
processes. Journal of Marketing, 48(Winter), 30–45.
Wallendorf, M., & Brucks, M. (1993). Introspection in consumer research: Implementation
and implications. Journal of Consumer Research, 20, 339–359.
Wang, L. C., Baker, J., Wagner, J. A., & Wakefield, K. (2007). Can a retail Web site be social?
Journal of Marketing, 71, 143–157.
Warwick, D. P. (1975). A theory of public bureaucracy: Politics, personality, and organization in
the state department. Cambridge, MA: Harvard University Press.
Webb, E., & Weick, K. E. (1979). Unobtrusive measures in organizational theory.
Administrative Science Quarterly, 24(December), 650–659.
Webb, E. J., Campbell, D. T., Schwartz, R. D., & Sechrest, L. (1966). Unobtrusive measures.
Chicago: Rand McNally.
Webb, E. J., Campbell, D. T., Schwartz, R. D., & Sechrest, L. (2000). Unobtrusive measures
(Comment: revised ed.). Thousand Oaks, CA: Sage.
Weber, P. S., & Manning, M. R. (2001). Cause maps, sensemaking, and planned
organizational change. Journal of Applied Behavioral Science, 37(2), 227–251.
Webster, F. (1992). The changing role of marketing in the corporation. Journal of Marketing,
56, 1–17.
Weed, L. L. (1991). Knowledge coupling: New premises and new tools for medical care and
education. New York: Springer-Verlag.
Wegner, D. M. (2002). The illusion of conscious will. Cambridge, MA: Bradford Books, MIT
Press.
Weick, K. (2007). Drop your tools on reconfiguring management education. Journal of
Management Education, 31(1), 5–16.
Weick, K. E. (1969). The social psychology of organizing. Reading, MA: Addison-Wesley.
Weick, K. E. (1979). The social psychology of organizing (2nd ed.). Reading, MA: Addison-Wesley.
Weick, K. E. (1993). The collapse of sensemaking in organizations: The Mann Gulch disaster.
Weick, K. E. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage.
Weick, K. E. (1996). Drop your tools: An allegory for organizational studies. Administrative
Science Quarterly, 41(3), 301–313.
Weick, K. E., & Bougon, M. G. (1986). Organizations as cognitive maps. In: H. P. Sims, Jr. &
D. A. Gioia (Eds), The thinking organization (pp. 102–135). San Francisco: Jossey-Bass.
Weick, K. E., & Sutcliffe, K. M. (2001). Managing the unexpected: Assuring high performance
in an age of complexity. San Francisco, CA: Jossey-Bass.
Weiner, B. (1995). Judgments of responsibility. New York: Guilford.
Weiser, J. (1988). See what I mean? Photography as nonverbal communication in cross-
cultural psychology. In: F. Poyatos (Ed.), Cross-cultural perspectives in nonverbal
communication (pp. 240–290). New York: Hogrefe Publishers.
References 431
Weitz, B. A., Sujan, H., & Sujan, M. (1986). Knowledge, motivation, and adaptive behavior:
A framework for improving selling effectiveness. Journal of Marketing, 50(4), 174–191.
Wells, W. (1989). Lectures and dramas. In: P. Cafferata & A. M. Tybout (Eds), Cognitive and
affective responses to advertising (pp. 13–20). Lexington, MA: Lexington Books.
Wells, W., & Losciuto, L. (1966). Direct observation of purchasing behavior. Journal of
Marketing Research, 3, 227–229.
Wells, W. D. (1993). Discovery-oriented consumer research. Journal of Consumer Research,
19(March), 489–504.
Wertime, K. (2002). Building brands & believers: How to connect with consumers using
archetypes. Singapore: Wiley.
Whittington, R. (1992). Putting Giddens into action-social-systems and managerial agency.
Journal of Management Studies, 29(6), 693–712.
Whyte, W. F. (1943). Street corner society. Chicago: University of Chicago Press.
Whyte, W. F. (1990). Participatory action research. Newbury Park, CA: Sage.
Whyte, W. F. (1991). Social theory for action. Newbury Park, CA: Sage.
Whyte, W. F. (1943/1993). Street corner society. Chicago, IL: Chicago University Press.
Willett, R. P., & Pennington A. (1966). Customer and salesman: The anatomy of choice in a
retail setting. In: R. M. Haas (Ed.), Proceedings of the fall educators’ conference, American
Marketing Association, Chicago (pp. 598–616).
Williams, J. D. (1992). Reflections on a black middle-class consumer: Caught between two
worlds or getting the best of both? In: J. F. Sherry, Jr. & B. Sternthal (Eds), Advances in
consumer research (19, pp. 850–856). Provo, UT: Association for Consumer Research.
Wilson, D. T. (1977). Dyadic interactions. In: A. G. Woodside, J. N. Sheth & P. D. Bennett
(Eds), Consumer and industrial buying behavior (pp. 355–365). New York: North Holland.
Wilson, E. J., & Vlosky, R. P. (1997). Partnering relationship activities: Building theory from
case study research. Journal of Business Research, 39, 59–70.
Wilson, E. J., & Wilson, D. T. (1988). ‘Degrees of freedom’ in case study research of
behavioral theories of group buying. Advances in Consumer Research, 15, 587–594.
Wilson, E. J., & Woodside, A. G. (1999). ‘Degrees-of-freedom analysis’ of case study data in
business marketing research. Industrial Marketing Management, 28(3), 215–231.
Wilson, T., & Nisbett, R. E. (1978). The accuracy of verbal reports about the effects of stimuli
on evaluations and behavior. Social Psychology, 41(June), 118–130.
Wilson, T. D. (2002). Strangers to ourselves: Discovering the adaptive unconscious. Cambridge,
MA: Belknap.
Wilson, T. D., Hodges, S. D., & LaFleur, S. J. (1995). Effects of introspecting abut reasons: Inferring
attitudes from accessible thoughts. Journal of Personality & Social Psychology, 69, 16–28.
Wind, Y. (1966). Industrial buying behavior: Source loyalty in the purchase of industrial
components. Ph.D. dissertation, Stanford University, Stanford, CA.
Wind, Y., & Cardozo, R. N. (1974). Industrial market segmentation. Industrial Marketing
Management, 3(May), 153–165.
Witte, E. (1972). Field research on complex decision-making processes — The Phase theorem.
International Studies of Management and Organization, 2(October), 156–182.
Wolf, M. (1990). China notes: Engendering anthropology. In: R. Sanjek (Ed.), Field notes
(pp. 343–355). Ithaca, NY: Cornell University Press.
Woodside, A. G. (1977). Ethnographic research in marketing performance auditing. Working
Paper. University of South Carolina, Columbia.
Woodside, A. G. (Ed.) (1992). Advances in business marketing. Volume 5: Mapping how
industry buys. , Greenwich, CT: JAI Press.
Woodside, A. G. (Ed.) (1996a). Case studies for industrial and business marketing. Stamford,
CT: JAI Press.
Woodside, A. G. (1996b). Theory of rejecting superior, new technologies. Journal of Business &
Industrial Marketing, 11(3/4), 25–43.
Woodside, A. G. (1999). Advertising and consumption of alcoholic beverages. Journal of
Consumer Psychology, 8(2), 167–186.
Woodside, A. G. (2003). Middle-range theory construction of the dynamics of organizational
marketing-buying behavior. Journal of Business & Industrial Marketing, 18(4/5), 309–335.
Woodside, A. G. (2004a). Advancing means–end chains by incorporating Heider’s balance
theory and Fournier’s consumer-brand relationship typology. Psychology & Marketing, 21,
279–294.
Woodside, A. G. (2004b). Advancing from subjective to confirmatory personal introspection
in consumer research. Psychology & Marketing, 21(12), 987–1010.
Woodside, A. G. (2005). Market-driven thinking: Achieving contextual intelligence. London:
Butterworth-Heinemann.
Woodside, A. G. (2006). Overcoming the illusion of will and self-fabrication: Going beyond
Naı̈ve subjective personal introspection to an unconscious/conscious theory of behavior
explanation. Psychology & Marketing, 23(3), 257–272.
Woodside, A. G. (2007). Collectivism-individualism’s influence on international travel behavior.
Working paper. Boston College, Carroll School of Management.
Woodside, A. G. (2008a). Using the forced metaphor-elicitation technique (FMET) to meet
animal companions within self. Journal of Business Research, 61, 280–287.
Woodside, A. G. (2008b). Using the forced metaphor-elicitation technique (FMET) to meet
animal companions within self. Journal of Business Research, 61(3), 480–487.
Woodside, A. G. (2008c). Anti-social behaviour: Profiling the lives behind road rage.
Marketing Intelligence & Planning, 26(5), 459–480.
Woodside, A. G., & Ahn, I. (2007). Culture’s consequences on experiencing international
tourism services and products. In: A. G. Woodside & D. Martin (Eds), Tourism
management: Analysis, behavior and strategy (pp. 28–61). Wallingford, UK: CABI.
Woodside, A. G., & Biemans, W. G. (2005). Modeling IMDAR processes. Journal of Business
& Industrial Marketing, 20(7), 380–393.
Woodside, A. G., Caldwell, M., & Spurr, R. (2006). Advancing ecological systems theory in
lifestyle, leisure, and travel research. Journal of Travel Research, 44(3), 259–272.
Woodside, A. G., & Chebat, J.-C. (2001). Updating Heider’s balance theory in consumer
behavior. Psychology & Marketing, 18, 475–496.
Woodside, A. G., & Davenport, W. (1974). The effect of salesman similarity and expertise on
consumer purchasing behavior. Journal of Marketing Research, 11, 198–202.
Woodside, A. G., & Fleck, R. A., Jr. (1979). The case approach to understanding brand
choice. Journal of Advertising Research, 19(2), 23–30.
Woodside, A. G., Floyd-Finch, N., & Wilson, E. J. (1986). Psychodrama and beverage
consumption. In: M. Laroche & A. Strazzieri (Eds), Proceedings of the 13th international
research seminar in marketing, University of Marseille, Institute d’Adminstration des
Enterprises, Aix-en-Provence (pp. 478–511).
Woodside, A. G., MacDonald, R., & Trappey, R. J., III. (1997). Measuring linkage-
advertising effects on customer behavior and net revenue. Canadian Journal of Adminis-
trative Sciences, 14(2), 214–228.
Woodside, A. G., & Megehee, C. M. (2009). Travel storytelling theory and practice. Anatolia,
20(1), 86–99.
References 433
Woodside, A. G., & Moore, E. M. (1983). Direct response marketing: Understanding

patronage behavior and testing competing strategies. In: W. R. Darden & R. F. Lusch
(Eds), Patronage behavior and retail management (pp. 179–205). New York: North-Holland.
Woodside, A. G., Pattinson, H. H., & Miller, K. E. (2005). Advancing hermeneutic research
for interpreting interfirm new product development. Journal of Business & Industrial
Marketing, 20, 364–379.
Woodside, A. G., & Samuel, D. M. (1981). Observation of centralized corporate procurement.
Industrial Marketing Management, 10(3), 191–205.
Woodside, A. G., & Soni, P. K. (1991). Direct response advertising information: Profiling
heavy, light and nonusers. Journal of Advertising Research, 31, 26–36.
Woodside, A. G., Sood, S., & Miller, K. (2007). When consumers and brands talk:
Storytelling theory and research in psychology and marketing. Psychology & Marketing,
25(2), 145–197.
Woodside, A. G., & Uncles, M. D. (2005). How behavioral primacy interacts with short-term
marketing tactics to influence subsequent long-term brand choice. Journal of Advertising
Research, 45, 229–240.
Woodside, A. G., & Wilson, E. J. (2000). Constructing thick descriptions of marketers’ and
buyers’ decision processes in business-to-business relationships. Journal of Business &
Industrial Marketing, 15(5), 354–369.
Woodside, A. G., & Wilson, E. J. (2002a). Respondent inaccuracy. Journal of Advertising
Research, 42(6), 1–13.
Woodside, A. G., & Wilson, E. J. (2002b). Respondent inaccuracy. Journal of Advertising
Woodside, A. G., & Wilson, E. J. (2003). Case study research for theory-building. Journal of
Business & Industrial Marketing, 18, 493–508.
Wall Street Journal (WSJ). (2008). Versace presents a collage of a Collection, February
22. Available at http://blogs.wsj.com/runway/2008/02/22/versace-presents-a-collage-of-a-
collection/
Yeniyurt, S., & Townsend, J. D. (2003). Does culture explain acceptance of new products in a
country? An empirical investigation. International Marketing Review, 20(4), 377–396.
Yin, R. K. (1989). Case study research: Design and methods. Newbury Park, CA: Sage.
Yin, R. K. (1994). Case study research design and methods (2nd ed.). Thousand Oaks, CA:
Sage.
Zadeh, L. (1965). Fuzzy sets. Information and Control, 8, 338–353.
Zaltman, G. (2003). How customers think: Essential insights into the mind of the markets.
Boston: Harvard Business School Press.
Zaltman, G., & Coulter, R. H. (1995). Seeing the voice of the customer: Metaphor-based
advertising research. Journal of Advertising Research, 35, 35–51.
Zaltman, G., LeMasters, K., & Heffring, M. (1982). Theory construction in marketing.
New York: Wiley.
Zaltman, G., & Zaltman, L. (2008). Marketing metaphoria: What deep metaphors reveal about
the minds of consumers. Boston: Harvard Business Press.
Ziller, R. C. (1990). Photographing the self: Methods for observing personal orientation.
Newbury Park, CA: Sage.
Zuikier, H. (1986). The paradigmatic and narrative modes in goal-guided inference. In:
R. M. Sorrentino & E. T. Higgins (Eds), Handbook of motivation and cognition (pp. 465–502).
New York: Guilford.
Subject Index
AAE, 132 case study research, 1–42, 44–46, 48, 50,

adaptive unconscious, 132, 144, 52, 54, 56, 58, 60, 62, 64, 66, 68,
146–147, 155–156, 399 70, 72, 74, 76, 78, 80, 82, 86–88,
advanced attribution error, 130, 132 90, 92, 94, 96, 98, 100, 102, 104,
advanced overconfidence bias, 131–132 107–108, 110, 112, 114–116, 118,
advertising, 45, 58, 61–62, 70–71, 87, 120, 122, 124, 126, 130, 132, 134,
104, 148, 194, 255, 257, 274, 276, 136, 138, 140, 144, 146, 148, 150,
280, 309, 323 152, 154, 156, 158, 160, 162, 164,
alcohol, 236–239 166, 168, 170, 172, 174, 176, 178,
anecdote, 21, 23, 177 180, 182, 184, 186, 188–192, 194,
anima, 376–379, 384, 390, 393 196, 198, 200, 202, 204, 208, 210,
animus, 359, 376–379, 391–393 212, 214, 216, 218, 220, 222,
antagonist, 44–45, 92 224–261, 263–289, 291–319, 322,
anthropomorphic, 41–42, 47 324, 326, 328, 330, 332, 334, 336,
AOB, 132 338, 340, 342–344, 346, 348, 350,
archetype, 42–44, 47–48, 53–54, 56–58, 352, 354, 356, 360, 362, 364, 366,
63, 85, 88–92, 96, 100–103 368, 370, 372, 374, 376, 378, 380,
asymmetric, 26, 360–362 382, 384, 386, 388, 390, 392, 394,
auto-driving, 5, 226, 240, 265, 267, 397–408
270–271, 275, 278, 281 catharsis, 43, 88, 91–92, 99–100
automatic thought-retrieval, 139 causal recipe, 19, 25–26, 28, 30, 33, 359,
autonoetic, 46 361, 367–369, 378–379, 381,
awakening, 88, 91–92, 96–98 383–384, 390–394, 398–400
cause mapping, 343–344, 347–348, 352,
backstage, 158, 163, 173–174, 176, 354
187–188 Clark Kent, 101–102
balance theory, 63, 70, 102, 225, climax, 93
227–228
cohort auditor, 118
Barq’s, 144
collectivism, 364, 368–370, 374–375, 391
Blog, 43–44, 58–60, 62–64, 69–71, 77, 91,
96, 100–101 commodity purchase planning, 333
Boone, 268, 272–277, 280–289 commons, 354
Buddhist, 91, 130 configural, 397–398, 407
building-in degrees-of-freedom, 12, 268, confirmation, 30, 33, 105, 114, 126,
401 130–131, 133, 164, 195, 199, 203,
buying center, 246–247, 250, 252, 322 211, 261, 265, 277
436 Subject Index
confirmatory personal introspection, 24, dampening, 347–348, 352

87, 107, 109, 111, 113–115, 117, decision systems analysis, 15, 23, 33,
119, 121, 123, 125, 127, 129, 164, 38–39, 215
398 decision, 1–2, 4, 8, 12–13, 15, 20, 23,
conjunctive-disjunctive, 397, 402–403, 31–33, 35, 38–39, 107, 113–117,
407 120–122, 124–126, 143, 169,
conscious, 3–5, 15, 42–44, 46, 49–51, 54, 207–211, 213–215, 217, 219–221,
59, 68, 70, 85–88, 92, 96, 102–103, 223–224, 232, 241, 243–253,
105, 107–108, 113, 115–116, 260–261, 270, 275, 280, 293,
121–123, 125–126, 129–133, 300–302, 304, 306–307, 322–326,
135–137, 139, 141, 147, 157, 160, 329, 334, 339–342, 397–398,
162, 164, 166, 175–176, 179, 184, 402–403, 405
226, 238, 291–293, 296, 302–304, deliberate, 23–24, 33, 36, 157–158,
307–308, 310, 397, 399, 406 183–184
consistency, 29–30, 158, 186, 360, dependent variable, 10, 13, 19, 31, 295,
383–384, 390–392 310, 345, 360–362, 375, 385–389,
context, 1–2, 5, 9, 20–21, 33, 35, 40, 86, 398–399
99, 104, 109, 116, 145, 159, 162, description, 6, 8, 11–13, 16, 32, 39, 62,
164, 166, 189, 191–193, 229, 241, 67, 113, 119, 136, 149, 154–155,
244–246, 252–253, 259–260, 264, 175, 177, 184, 208–209, 214, 222,
291, 293, 295–296, 321, 343, 397, 229, 231–232, 264, 283, 292, 301,
399–402, 405–406 322, 329, 333, 405
contingency, 15, 207–211, 213, 220, deviation amplifying, 348
222, 291, 304, 308, 384, DOF, 38–39, 67
398, 402 drama, 41, 43–45, 49, 51, 53, 55–56, 58,
control, 6, 11–13, 16, 42, 49, 60, 88, 160, 61–63, 68–71, 139–140
238, 248, 338, 352–353, 357, 375 dual processing theory, 86, 104
conversation analysis, 189–190 dynamic causal modeling, 12–13
cool being, 68 dynamic, 3, 12–13, 30, 33–34, 183–184,
corporate purchase agreement, 324, 194, 344, 351–352, 354–355, 364,
330–332, 340 397, 399, 404, 407
cost avoidance, 26–27, 214, 217, 221,
338 ecological systems theory, 293, 297
coverage, 17–18, 22–25, 29–33, 37, 39, emic, 1, 8, 16, 39, 44, 49, 51, 54, 56, 58,
134, 204, 242, 360, 383–384, 63, 67, 69, 105, 117, 125–126, 140,
390–392 158–161, 164, 166–167, 169, 172,
CPA, 325 179, 182–185, 188, 301
CPP, 333 empirical positivism, 18, 20
creative destructive, 107 enactment, 53–54, 56, 67, 103, 162, 167,
critical test, 241, 243–244, 263, 268, 169–170, 183, 186
282 environment, 32, 62, 71, 118, 161,
culture, 1, 16, 48, 87, 108, 116, 133, 143, 164, 166, 170, 172, 177, 179, 181,
145, 169–170, 193, 242, 297, 325, 185, 188, 193, 213, 264, 281,
359–360, 363–367, 369, 372–377, 293, 296, 321–322, 343, 348, 354,
380, 382, 385–389, 393–394 357, 401
Subject Index 437
Epsilon Corporation, 322, 325–328, Hawaii, 44, 295, 297–314

339 help, 32, 35, 39, 43–44, 47–48, 78, 92,
Epsilon, 322, 325–329, 333–334, 98, 113–114, 120, 125, 132, 136,
337–339, 341–342 144, 154, 166, 184, 195, 197,
ethnographic, 9, 12–13, 31, 113, 209, 201, 203, 226, 234, 240, 254,
293, 301, 321, 325, 342, 398, 404 273–274, 289, 306, 324, 340,
etic, 8, 16, 39, 54, 56, 58, 63, 105, 343, 351–352, 370
125–126, 140, 143, 155, 184–185 hero, 43–44, 48, 55, 57, 88–89, 145
exchange behavior, 194, 205 hidden demon, 33, 343–344, 404
explanation, 8, 11–13, 16, 24, 101, hierarchy, 46, 225, 364
129–130, 134–136, 141, 172, 220, historical method, 25, 30, 115–116
226, 251, 259–260, 295, 297, 345, Holocaust, 49, 86, 227
347, 367, 384, 390 humanistic, 115, 117–118, 126
explicit, 5, 8–9, 16, 18, 20, 32, 35, 37, 51,
104, 107, 114, 125, 131–133, icon, 50, 361, 378
138–141, 143–145, 147, 156, 184, IKEA, 169
208, 223, 243, 291, 294, 297, 348, illusion, 129–133, 135, 137, 139, 141
355 imbalance, 49, 52, 63, 69, 102, 225,
228
feedback, 8, 10, 23, 33, 116, 122, 136, IMDAR, 34, 404
244, 322, 336, 343–345, 351–352, implicit, 18, 24, 32, 35–36, 51, 104, 110,
404 114, 130–133, 138–141, 143–145,
FIT- freely independent traveler, 300, 147–148, 156, 166, 208, 223, 291,
368, 373 294, 297, 354, 399
fit-like-a-glove, 291 independent variable, 10, 19, 31,
FLAG, 134, 291, 293, 295–297, 304 295, 310, 345, 360–362, 375,
flow, 31, 45, 51, 63, 67, 156, 169, 300, 398–399
321, 326, 329, 342 individualism, 359, 363–364, 368–370,
FMET, 23–24, 105, 118–120, 125, 374–379, 382, 394–395
138–140, 143–145, 147–149, 151, inertia, 33, 80
153–155, 166, 251 informant, 5, 24–25, 31–32, 35–39,
folk, 24, 129–130, 134, 226 70–71, 104–105, 107–110, 114,
forced metaphor elicitation technique, 117, 119–120, 122–126, 129–130,
23–24, 105, 123, 139 136–141, 143, 145, 148–149,
front stage, 158–163, 167–168, 172, 154–155, 185, 194, 229–230, 236,
176–179, 181, 183–184, 186, 188 249, 254, 291–292, 296–298, 301,
FSSS, 25, 30, 32, 38, 40 303, 319, 321, 341
fundamental attribution error, 129–130, inside auditor, 117–118, 125
132, 136, 141 inter-judge reliability, 250
fuzzy set, 19, 21, 23, 25–28, 359, 361, internal branding, 157–158, 160,
363, 365, 367, 369, 371, 373, 375, 162–163, 167, 183–184
377–379, 381–385, 387, 389, 391,
393, 395 ‘‘Joe the Plumber’’, 23
journey, 57, 66, 77, 80, 82, 89, 91–92,
golf, 314, 343–357, 371 97–101
438 Subject Index
laddering, 148, 225–231, 233, 235, multiple methods, 1, 33, 38, 107, 115,
237–240 121, 125, 129, 354–356
law of the instrument, 21, 40 multiple routes not one model only, 397
lead user, 34 myth, 41, 44, 47, 50, 54, 59, 67–69, 71,
level 1, 3, 238, 240 88, 97, 145
level 2, 3–4, 238, 240
level 3, 4, 238 naı̈ve observation, 12
level 4, 4, 238 narrative, 41, 44–46, 50–51, 63, 67–70,
level 5, 4–5, 238 85–89, 91, 93, 95–97, 99, 101, 103,
little trickster, 43, 55, 57, 90 105, 110, 242, 303
logical positivistic, 241–242 ‘‘new wave theory’’, 272, 282
long interview method, 24, 264, 268–269, nonconscious, 132, 144–145, 147, 155
292, 297, 310
loops, 8, 10, 13, 23, 33, 79, 122, 345, operational data, 6, 8, 16
347–349, 351–352, 404 opposable mind, 345, 351–352, 355
lying, 3, 11, 97 optimality, 354
over-confidence bias, 399
market prices, 26, 211, 216, 220
Paris, 51, 56, 62, 64, 67, 70–71, 73–76
masculinity, 48, 361, 363–364, 368, 372,
participant, 8–9, 12–13, 16, 32, 35, 113,
375–378, 395
117–118, 124, 132, 182, 193, 195,
means-end laddering, 225, 227, 229, 231,
198–199, 205, 209, 244, 264–265,
233, 235, 237, 239
321–323, 325–327, 329, 331, 333,
MEC, 225–228, 230–232, 238, 240
335, 337–339, 341–342, 354, 400
member check, 20, 32, 58, 70, 114, 127,
participant observation, 9, 12–13, 32, 35,
130–131, 141
113, 193, 205, 244, 264–265,
mental model, 1, 8–9, 16, 114, 223
321–323, 325, 327, 329, 331, 333,
meta-analysis, 252 335, 337, 339, 341–342, 400
metaphor, 5, 22–24, 67, 104–105, 107, phase theory, 121
110–113, 118–120, 123, 130, polyscenic, 91, 93
138–139, 143–144, 148, 164, 166, postulate of commensurate complexity,
170, 172, 176–177, 179, 226, 230, 17–18, 22–23, 39
251, 408 postulate of disproportionate
meta-sensemaking, 6, 11 achievement, 17–18, 23, 25, 40
micro-tipping point, 291, 293 power distance, 360–361, 363–364,
mixed-methods, 23, 33, 35–36, 130, 368–369, 371, 375–378, 394–395
132–133 prediction, 6, 11–13, 16, 19, 39, 209,
Monet, 64, 76 243–244, 246–247, 250–251,
monoscenic, 93, 97, 102–103 253–255, 257–259, 261, 348,
Moonstruck, 55 351–352, 361, 374
mother-of-goodness, 43, 56 predictive validity, 397, 399, 401, 403,
motivation, 100, 115, 120, 156, 253 407
Mountain Dew, 41, 44, 48, 88 prequel, 49, 52, 86, 91–92, 96, 228–229,
MTP, 291, 293–295, 304, 309–311 231, 233, 235, 237, 239
multi-person, 397 presentational data, 6, 8
Subject Index 439
‘‘proper pleasure’’, 92 Street Corner Society, 32, 39, 193,

property space, 17, 22, 24, 37–38, 369, 342
375, 377–378 structural equation modeling, 19, 25, 37,
protagonist, 23, 44–46, 49–52, 59–60, 360, 399
68, 70–71, 85, 87–88, 91–92, structure, 32, 34, 45, 50, 52–53, 55,
96–101 58, 61, 91, 103, 143, 169, 172,
puzzle, 195, 198–199, 201–204, 243 178–179, 181, 228, 323, 361,
371, 401
QCA, 294–295, 310–311, 359–361, subjective personal introspection, 21, 23,
381–384, 391, 393–394, 398, 108, 117, 126–127, 129–130,
404–405 137–138, 143–145, 147, 149, 151,
qualitative comparative analysis, 294, 153, 155, 399
359, 361–363, 365, 367, 369, 371, symmetric, 360–362
373, 375, 377, 379, 381, 383, system 1, 85–87, 104–105
385, 387, 389, 391, 393, 395, system 2, 85–86, 104–105
399–400 system dynamics, 23, 38, 343, 345,
347–349, 351–357, 400, 404
rebel, 41, 44, 48, 88, 373 systems thinking, 6, 38, 343, 345, 347,
rejection inducing dimension, 122 349, 351, 353–355, 357, 397,
relative preference dimension, 122 403–404, 407
request for quotation, 249
RFQ, 218, 329, 331, 334, 336, 338–339 tacit knowledge, 24, 166
role-playing, 24 talk, 15, 45, 65, 81, 104, 124, 168, 186,
189–191, 193–197, 199, 201, 203,
salesman, 80, 194–195, 197–199, 205, 254, 306, 313–314
201–205 Tape-recording, 191
satisfy, 175, 246, 397, 405, 407 TAT, 138–140
self-fabrication, 129, 131, 133, 135, 137, Technorati, 43–44, 71
139, 141 ‘‘The asshole’’, 30
self-report, 18, 25, 50, 87, 96, 102, 105, Thematic Apperception Test, 139
138, 141, 143–144, 154–155, 205, thick description, 6, 13, 20, 23–24, 31,
311, 399 39, 107, 139, 179, 207–211, 213,
SEM, 19–20, 25, 37 215, 217, 219, 221–223, 263–264,
sensemaking, 6, 8, 11, 52, 130, 158, 160, 269, 292, 298, 303, 310
166, 183, 223 tipping point, 96, 291, 293, 298, 309,
sentiments, 6, 14–16, 228, 230 351
solution, 13, 22, 168, 195, 198–199, 201, Tommy Hilfiger, 62, 65, 68, 70, 78, 81
203–204, 246, 248, 299, 322, 338, Trade-off dimension, 121–123
345, 354 tragedy of the commons, 354
‘‘stay-the-course’’ theory, 268, 282 transference theory, 143, 155
storytelling, 41–53, 55–59, 61–63, 65, triangulation, 6–9, 16, 22–23, 33, 35–36,
67–71, 73, 75, 77, 79, 81, 83, 86, 38–39, 48, 113, 131, 261, 264, 325,
88, 91–93, 100, 103, 105, 111–112, 342
120, 141, 143, 148–149, 155, 164, typology, 134–135, 138, 230–231, 369,
166, 170, 179, 185–186, 303, 309 375
440 Subject Index
uncertainty avoidance, 359, 361, 364, Versace, 62, 66, 69–71, 81–83, 87,
368, 370, 375–378, 394–395 93–103, 105
unconscious, 3–5, 15–16, 24, 42–44, VFR, 368–369, 372–373, 381–382,
46–48, 50–51, 54, 56, 59, 70, 385–389
85–88, 91, 96–98, 102–105, Visual narrative art, 85–87, 89, 91, 93,
107–110, 113–116, 118–119, 95, 97, 99, 101, 103, 105
121–126, 129–130, 132–133, Volkswagen, 41, 56, 88, 102
136–137, 143–147, 155–157, 160,
162, 164, 166, 170, 179, 184, 226, ‘‘Weick’s clock’’, 22–23
232, 238, 240, 291–293, 296–297, will, 3, 12, 21–22, 35, 37, 48–49, 63–64,
303–304, 310, 359, 364, 377–378, 67, 72, 75, 77, 82, 98–99, 101, 104,
382, 397, 399, 406–408 108, 110, 121–122, 126, 129–135,
unintentional, 134–136, 145 137, 139, 141, 145, 154–155, 161,
unobtrusive evidence, 397 168–169, 176, 179, 182, 184, 196,
unstructured, 32, 178 198, 200, 205, 207, 215, 217–219,
unsustainable, 344, 348–349, 354–355 221, 228, 234, 236, 238, 240, 246,
259, 272–273, 275, 284–285, 289,
value, 11, 20, 26–28, 121, 134, 139–140, 291–292, 302–303, 309, 311–312,
148, 159–160, 162–164, 168, 314, 319, 323, 327, 332, 335–337,
170, 176–182, 205, 207, 214–215, 342, 345, 347–348, 355, 369, 393,
223, 226, 229–230, 233–238, 398, 400, 403
241, 244, 292, 306, 321, 328, 330, ‘‘Woodside’s box’’, 23, 32
333, 345, 347, 351–352, 354–357,
359–361, 363–365, 367, 375, ZMET, 5, 118–119, 139–140, 148, 166
378–379, 382–384, 391, 393–395, zoomorphistic, 24, 143–144, 148–149,
400–401 154–155, 167

Case Study

Uploaded by

Copyright:

Available Formats

Case Study

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Case Study

Uploaded by

Copyright:

Available Formats

What is the book about based on the contents page?

What is the book about based on the contents page?

What are some of the research methods discussed in the book?

What are some of the research methods discussed in the book?

CASE STUDY RESEARCH:

Theory Methods Practice

United Kingdom North America Japan

First edition 2010

Reprints and permission service

British Library Cataloguing in Publication Data

1. Building Theory from Case Study Research 1

16. Systems Thinking and System Dynamics Modeling 343

Most thinking occurs unconsciously

organization) to be too low-grade in accuracy to be acceptable. Too much nuance is

Time: micro-seconds, minutes, hours, days, weeks, months, decades, centuries

the complicated individual can sense variations in a larger environ-

System Dynamics (Chapter 16)

Figure 1: Time and thinking property spaces.

Building Theory from Case Study Research

Introduction: Achieving a Broad Perspective When

CSR is an inquiry that focuses on describing, understanding, predicting, and/or

A case study is an empirical inquiry that investigates a contemporary

Why Case Study Research is Useful

‘‘I Hate Lying Like That’’

The operational constructs using closed-ended responses developed by researchers

Three-Person and Five-Way Mental Processing in Industrial Marketing

Thinker, Decision-Maker, Thinker, Decision-Maker,

Figure 1: The multiple mental processes in research on industrial marketing–buying

Deep Understanding: The Principal Objective of Case Study Research

Research Steps Required to Achieve Deep Understanding

Data in this category [presentational] are often ideological, normative,

Observe Document Observe Document Observe Document

Senge (1990) makes a number of telling points about decision-makers’ mental

researchers seek a deep understanding by directly observing in ‘‘real time’’

The Core Criticisms of Large Sample Survey Research

Performance information itself colors subjective memories, percep-

collection methods across several time periods. Meta-sensemaking — the researcher’s

Theory Building and Theory Testing Using Case Study Research

The Objectives of Case Study Research

includes forecasting near term and/or long-term psychological states, behaviors, or

9. Naïve 10. 11. 12.

13. Participant 14. Action 15. Ethno- 16. Clinical

Figure 3: Classifying CSR by research objectives.

explaining, predicting, and controlling case behavior without starting with a

A social system’s identity nodes and loops are typically over

Cell 9 is a description without explanation, prediction, or control: a naı̈ve report

Suggestions Regarding Selection of Case Study Research Objectives

Core Propositions in Case Study Research

CSR is an inquiry focusing on describing, understanding, predicting, and/or

Bridging the Chasm between Survey

Chapter 2 describes how behavioral science research methods that management

three objectives to accomplish. The chapter proposes replacing Thorngate’s (1976)

Case Study Researchers’ Principal Criticisms of Fixed-Point Surveys

Third, correlation analysis, multiple regression, and structural equation modeling

Criticisms of Case Study Research by Empirical Positivists

Researchers applying classical empirical positivistic methods (e.g., correlation,

Figure 1: Research methods’ attainment of objectives.

theory development — highly accurate portrayals of real-word processes and little to

Building Bridges across the Chasm

Members of both research camps appear to be guilty of the ‘‘law of the