Kristina Herold A4
Kristina Herold A4
Kristina Herold A4
Acta Universitatis
Lappeenrantaensis 665
Supervisor Professor Sanna-Katriina Asikainen
LUT School of Business and Management
Lappeenranta University of Technology
Finland
ISBN 978-952-265-860-9
ISBN 978-952-265-861-6 (PDF)
ISSN-L 1456-4491
ISSN 1456-4491
Lappeenranta 2015
75 pages
In responding to the calls for more research on what happens after WOM information is
received, and how it affects marketing-relevant outcomes, this dissertation extends prior
WOM literature by investigating how consumers process information in a high-
involvement service domain, in particular higher-education. Further, the dissertation
studies how the form of WOM influences consumer choice. The research contributes to
WOM and services marketing literature by developing and empirically testing a
framework for information processing and studying the long-term effects of WOM.
The results of the dissertation are presented in five research publications. The
publications are based on longitudinal data. The research leads to the development of a
proposed theoretical framework for the processing of WOM, based on theories from
social psychology. The framework is specifically focused on service decisions, as it
takes into account evaluation difficulty through the complex nature of choice criteria
associated with service purchase decisions. Further, other gaps in current WOM
literature are taken into account by, for example, examining how the source of WOM
and service values affects the processing mechanism.
The research also provides implications for managers aiming to trigger favorable WOM
through marketing efforts, such as advertising and testimonials. The results provide
suggestions on how to design these marketing efforts by taking into account the
mechanism through which information is processed, or the form of social influence.
I have been honored to receive feedback from Professor Robert East and PhD Bodo
Lang, who have served as my preliminary examiners. Your insights and
encouragements have been invaluable and inspiring for me to continue my research in
the area of Word-of-Mouth.
I would also like to thank Professor Richard Petty for the feedback on my work during a
challenging phase in my dissertation process. Your support provided the courage and
strength to take my ideas forward. Also a special thank you to Professor Jennifer Argo,
your inspiring words made me realize my goals in academia. I also value the insights
and support from Professor Jane Hemsley-Brown, and the feedback from anonymous
reviewers during both journal and conference review processes. This feedback has
improved and strengthened my research to a great extent.
I gratefully acknowledge the financial support, which I have received from the Research
Foundation of Lappeenranta University of Technology, Foundation for Economic
Education and Emil Aaltonen Foundation.
I have been privileged to spend five months at the Scandinavian Consortium for
Organizational Research (SCANCOR), Stanford University. During this time I found a
new passion for research, and for this I owe a special thank you to many individuals,
Henrika Franck, Paul Savage, Ulrik Hvidman, David Lallemant, Andrea Carafa, Yavuz
Akaln, Christian Wheeler, Sarah Soule, Annette Eldredge, Markus Paukku and
Kirsimarja Blomqvist, to name a few. You all have inspired me to broaden my horizons
and challenge myself. The positive energy at SCANCOR is something I will always
cherish. Further, I want to thank Professor Sanjit Sengupta for the opportunity to present
my research to the faculty of San Francisco State University.
I would also like to express my sincere gratitude for all the support and help from
Minna Ranta, Terttu Hynynen, and Johanna Jauhiainen, you have always made things
happen. Thank you also to Joan Nordlund for helping me with my academic writing and
proofreading my work.
Also the faculty at LUT School of Business and Management have provided a strong
support network of knowledge and inspiration. Thank you Mikael Collan, Hanna-Kaisa
Ellonen, Maija Hujala, Anne Jalkala, Olli Kuivalainen, Kalevi Kylheiko, Pivi
Maijanen-Kylheiko, Sami Saarenketo, Liisa-Maija Sainio, Hanna Salojrvi, and Sanna
Sintonen.
Also the peer support from other doctoral students has been a great source of energy, for
this I owe especially a thank you to Jenni Sipil, Heini Vanninen, and Mohamadali Ahi.
A special thank you to Daria Volchek and Lasse Torkkeli for your insights, guidance,
chocolate bars, and multiple laughs. Also, thank you Minna Oinonen for our venting
lunches, and to Elina and Jussi Reponen for our treasure hunt, these have been great
times of unwinding.
My friends have been a crucial part of this process. Thank you for the encouragement,
inspiration and especially taking my mind away from my dissertation through the many
experiences and memories we have shared throughout the years. Julle has my special
thank you, as you were a vital supporter who encouraged me to start this journey.
Finally, my family, iti, Isi, Nanna, Cam, Mika and Anna, you are my backbone. The
warmth, your positive attitude, inspiration, and the incredible support that you bring to
my life is amazing! There have been many ups and downs during the past four years
and you have always been the source of my energy and smiles. Thank you for believing,
supporting and loving. I want to end my acknowledgements with words, which have
echoed in my head multiple times during this process. They are from an important
person who I wish could have been here, my grandfather: Periksi ei anneta - Birger
Herold.
Kristiina Herold
To my family in Finland, South-Africa, and Sweden
Contents
Abstract
Acknowledgements
Contents
List of publications 11
1
Introduction 13
1.1
Background.............................................................................................. 14
1.2
Research objectives ................................................................................. 17
1.2.1
Research gap addressed by the dissertation................................. 17
1.2.2
Scope and objectives of the study ............................................... 19
1.3
Structure .................................................................................................. 21
1.4
Definitions of the key concepts ............................................................... 22
1.4.1
Word-of-Mouth ........................................................................... 22
1.4.2
Attitude ........................................................................................ 23
1.4.3
Information diagnosticity ............................................................ 24
1.4.4
Forms of WOM ........................................................................... 24
1.5
Research context: HE applicants decision-making ................................ 24
5
Conclusions 51
5.1
Theoretical contribution of the study ...................................................... 51
5.1.1
Broadening the concept of information processing in WOM
research ....................................................................................... 51
5.1.2
Bringing forth the context: the nature of choice criteria in
complex service decisions ........................................................... 52
5.1.3
Introducing a survey approach to the multiple roles of
variables postulate ....................................................................... 53
5.1.4
Extending the understanding of the behavioral processes of
WOM through two different forms ............................................. 53
5.2
Managerial implications .......................................................................... 54
5.3
Limitations and suggestions for future research ...................................... 56
References 59
Publications
11
List of publications
The dissertation consists of the introductory part, which provides an overview of the
dissertation and the following publications. The publications included in the dissertation
are summarizing the contribution of the author of this dissertation. All publications were
written in cooperation with other co-authors. However, the author of the dissertation
was the first and corresponding author in four cases.
PUBLICATION I
Evolahti, R., Herold, K., and Sundqvist, S. (2011). International Students Personal
Values, Educational Motives and University Choice Criteria. Proceedings of the
ANZMAC Conference, 28th -30th December, Perth, Australia.
The author was partly responsible for the development of the theoretical framework of
the study, setting the research questions, and discussion of the implications.
PUBLICATION II
Herold, K., Tarkiainen, A., and Sundqvist, S. (2013). The Impact of Word-of-Mouth on
Attitude Formation: an Information Processing Perspective. Proceedings of the
American Marketing Association Winter Marketing Educators Conference, 15th-17th
February, Las Vegas, NV, US. (This article is revised and further submitted)
The author was mainly responsible for the collection of the data, development of the
theoretical framework of the study, setting the research questions, and discussion of the
implications.
PUBLICATION III
Herold, K., Sipil, J., Tarkiainen, A. and Sundqvist, S. (2015). How Service Values
Influence the Processing of Word-of-Mouth in the Evaluation of Credence Beliefs.
(This article is submitted)
The author was responsible for the development of the research plan, collection and
analysis of the data, and writing most of the manuscript.
PUBLICATION IV
Herold, K., Tarkiainen, A., and Sundqvist, S. (2016). How the Source of Word-of-
Mouth Influences Information Processing in the Formation of Brand Attitudes.
Forthcoming in Journal of Marketing for Higher Education, 26(1).
The author was mainly responsible for the development of the research plan, data
collection, development of the theoretical framework of the study, setting of the
hypothesis, data analysis, and reporting of the results and implications of the study.
12 List of publications
PUBLICATION V
Herold, K., Sipil, J., Tarkiainen, A., and Sundqvist, S. (2014) Facts or Opinions -
Which Make a Difference? Word-of-Mouth and Attitude Change in a High-involvement
Service Context. Proceedings of the ANZMAC Conference, 1st-3rd December, Brisbane,
Australia. (This article is revised and further submitted)
The author was mainly responsible for the development of the research plan, data
collection, setting of the hypothesis, data analysis, and reporting of the results and
implications of the study.
13
1 Introduction
Consumers are known to discuss products, brands, and experiences with each other
(Berger, 2013). These discussions are probably the most prevalent element of consumer
behavior (Burnkrant & Cousineau, 1975). In marketing, these discussions are known as
Word-of-Mouth (WOM), and refer to any information about the target object or brand
transferred from one individual to another (Brown, Barry, Dacin & Gunst, 2005, p.
125). As WOM has an impact on decision-making, it has been of interest to academia
and practitioners for decades (East, Hammond & Lomax, 2008; Liu, 2006; Sweeney,
Soutar & Mazzarol, 2012). In fact, previous research suggests that WOM is the primary
factor behind 20 to 50 percent of all purchasing decisions (Boughin, Doogan & Vetvik,
2010), and is more impactful than advertising (East et al., 2005).
The importance of WOM is further highlighted in the services domain (Mangold, Miller
& Brockway, 1999; Murray, 1991; Sweeney, Soutar & Mazzarol 2008), since services
(compared to goods) are more difficult to evaluate prior to purchase given that they are
intangible, heterogeneous, and perishable in nature (Murray, 1992). Further, when a
decision is important to consumers, they rely on WOM for information, as it is
perceived as a trustworthy source of information that reduces risk, simplifies
complexity, and increases the confidence of consumers in decision-making (Berger,
2014). Thus, high-involvement services provide an interesting context to study WOM.
Therefore, this dissertation will take part in the discussion concerning the impact of
WOM on consumer evaluation during a high-involvement decision, namely regarding
international Masters level education.
Even though WOM has been of interest to academia and practitioners, there are still
many gaps in the current literature related to WOM. For example, research on the
receiver side has been conducted to a lesser extent (Sweeney, Soutar & Mazzarol,
2008), which may be partly due to the difficulty in studying received WOM. According
to East and Uncles (2008), this difficulty arises from the informal, unplanned, and
unpredictable nature of WOM, which also makes managing WOM a complex issue.
Prior research on WOM effectiveness has generally focused on source, message, and
receiver characteristics (e.g., Bansal & Voyer, 2000; Gilly, Graham, Wolfinbarger &
Yale, 1998). This line of research has provided important insight into the influence of
WOM and its boundary conditions; however, for WOM to have an impact on
marketing-relevant outcomes, the manner in which consumers use WOM in decision-
making must be understood as well (Yang, 2012). As important information processing
mechanisms occur prior to marketing-relevant outcomes, their understanding brings
new insights into consumer behavior (Plassmann, Venkatraman, Huettel & Yoon,
2015). Thus, focusing on the processing of WOM information may provide an
interesting perspective not only to understanding how the receivers of WOM make use
of the information in decision-making, but also to understanding the long-term effects
of the information on marketing-relevant outcomes (Petty, Wheeler & Tormala, 2013).
14 1 Introduction
In the study, a theoretical model was developed to understand how consumers process
WOM information in forming marketing-relevant outcomes, such as attitude formation
and change in a high involvement service context. The aim of the model was to
contribute to both WOM and services marketing literature. In addition, managerial
guidelines for designing marketing campaigns for WOM are provided.
1.1 Background
Due to its importance to decision-making, WOM has attracted interest from academia
and practitioners for decades (e.g., East, Hammond & Lomax, 2008; Liu, 2006;
Sweeney et al., 2012). The fields of marketing and social psychology have researched
WOM and they form the background to this dissertation. The impact of WOM on a
multitude of marketing-related outcomes such as attitudes, purchase intentions, and
choice behavior is well known from prior research (Berger, 2014). This line of research
has highlighted important variables for increasing the effectiveness of WOM, for
example, source characteristics such as source expertise and tie strength (e.g., Bansal &
Voyer, 2000; Gilly et al., 1998; Voyer & Ranaweera, 2015; Wangenheim & Bayn,
2004), message characteristics such as valence and richness of the message (Eisend,
2013; Laczniak, DeCarlo & Ramaswami, 2001; Sweeney et al., 2012; Sweeney, Soutar
& Mazzarol, 2014), and situational characteristics such as involvement with the
purchase (Voyer & Ranaweera, 2015; Wangenheim & Bayn, 2007). Further,
consumers who actively seek WOM or, in other words, who engage in a process of
vigorously seeking and ultimately obtaining information are more affected by WOM
(Bansal & Voyer, 2000), especially during high-risk decisions (Fang, Lin, Liu & Lin,
2011).
However, there have also been some varying results and boundary conditions found in
this line of research, which suggests that there are underlying information processes that
affect the impact of WOM on marketing-relevant outcomes. For example, Charlett,
Garland, and Marr (1995) found that WOM impacted consumer attitudes and purchase
probability to the same extent regardless of WOM valence (i.e., how positive or
negative WOM is). Moreover, East et al. (2008) reported that positive WOM was more
influential on brand choice than negative WOM, and found some evidence that actively
1.1 Background 15
sought WOM was more influential in the case of positive as opposed to negative WOM.
As a more recent example, Jun, Cha, and Aggarwal (2011) noted that source
characteristics source expertise and tie strength mattered in the impact of WOM only
when the information was tensile as opposed to more specific in details.
Thus, more research is needed to investigate what happens after WOM information has
been received and how it affects marketing-relevant outcomes such as attitudes and
purchase intentions (Martin, 2014; Martin & Lueg, 2013; Yang et al., 2012). This
knowledge provides insights into the likelihood of WOM effectiveness, and thus leads
to a better understanding of the impact of WOM on decision-making (Marting & Lueg,
2013). Perceived information diagnosticity has been a central construct in prior
literature focusing on the processing of WOM, since perceived diagnosticity
determines the likelihood of information utilization (Herr, Kardes & Kim, 1991, p.
457). Diagnosticity is a broad cognitive concept (Feldman & Lynch, 1988; Lynch,
2006) that refers to the perceived relevance or usefulness of the information related to
decision-making (Ahluwalia, Unnava & Burnkrant, 2001). Thus, information
diagnosticity represents a relatively thoughtful, cognitive, and deliberate manner of
information processing, and from this viewpoint, WOM is effective if it is perceived
relevant and useful for the decision.
Mangold et al. (1999) highlights the receivers felt need for information and advice as
the most frequent trigger of WOM in the services domain. However, WOM is not
necessarily always perceived as diagnostic even though the receiver is in need of
information, as some details may be processed more as simple cues of information, as
suggested by Voyer and Ranaweera (2015). For example, on some occasions, the
evaluation of a brand may consist of difficult and complex choice criteria such as
advanced technical details, which the consumer may not be able to evaluate but are
relevant criteria for decision-making. In this case, diagnostic information, may become
less relevant since the consumer is not able to process the central arguments in the
information but still wants to form an opinion of the brand. Additionally, on some
occasions, the discussions may be about irrelevant issues for the decision maker as they
may concern issues relating to the decision but not about the reasons as to why the
decision maker is actually making the choice, also known as service values. Thus, the
discussion may be perceived as less diagnostic. Further, certain sources of WOM may
not provide the most diagnostic information, which is rich in facts and arguments;
nevertheless, the support and approval of these social sources may be important for the
decision maker. On these occasions (i.e., the complex nature of choice criteria, service
values, the source of WOM) consumers may, for example, focus more on the general
tone of the discussions, try to understand whether the sources views are positive about
the issues, and not perceive the information as diagnostic. Therefore, an interesting
question can be raised regarding the processing of WOM: is only diagnostic information
utilized, or might there be other forms of WOM additionally relevant for decision-
making?
16 1 Introduction
There have been some suggestions based on significant findings that WOM might have
other less cognitive forms of impact. For example, Voyer and Ranaweera (2015)
proposed that the closeness of the sender and receiver (i.e., tie strength) could be
considered as a simple cue of information, since their results implied that the closeness
of the source weakens the effect of product involvement on WOM influence. Further,
with similar results concerning source credibility, Van Hoye and Lievens (2005)
suggest that future research could study the elaboration likelihood model (ELM) and
WOM, which according to the authors, may provide more insight in order to understand
the different information processing mechanisms of WOM (Van Hoye & Lievens,
2005).
In the persuasion literature, deriving from social psychology, this variety of different
information processing mechanisms has been acknowledged by dual-processing models
of influence such as the ELM (Petty & Cacioppo, 1986). The models have
acknowledged that there are two distinct processing routes through which information is
used in decision-making. These two routes differ in the extent of cognitive processing
of information, that is, the level of analysis of the information for its central merits
(Petty, 1994). An additional literature stream, also arising from social psychology, is
social influence literature. Social influence literature has also raised the issues of two
distinct forms of influence: informational and normative (Deutsch & Gerard, 1955).
Informational influence is based on information and facts, thus focusing more on the
message. Normative social influence, in turn, implies conformation with the
expectations of a social source and is responsive to the sources preferences and values
(Kaplan, 1989). Both persuasion and social influence literature streams may provide
interesting theoretical frameworks for understanding how the receiver uses WOM in
decision-making, as the central construct arising from WOM literature, perceived
diagnosticity, can be seen to have many similarities with these theories. First, with
cognitive processing arising from persuasion literature and, second, with informational
influence arising from social influence literature. As both forms of influence
(Wangenheim & Bayn, 2004) and information processing routes are suggested to
appear in WOM discussions (Voyer & Ranaweera, 2015), understanding the role of the
less cognitive processing mechanisms, in addition to information diagnosticity, may
provide insights into WOM literature.
Therefore, since extant research on WOM has acknowledged the relevance of perceived
diagnosticity, and thus the more thoughtful processing of WOM, this research will aim
to additionally understand how the more simple processing mechanisms operate during
a high-involvement service decision. The ELM will form the theoretical perspective of
this study and additionally will apply social influence theories to provide a broader
perspective on the underlying mechanisms of information processing. Thus, by focusing
on the receiver of WOM during a high-involvement service decision, taking a
longitudinal perspective, this study aims to contribute to the understanding of the
processing of WOM. Taken together, the usage of WOM will be addressed through its
impact on beliefs, attitude formation, attitude change, and consumer choice, thus aiming
to narrow the gap of what happens after WOM is received and positioning this study at
1.2 Research objectives 17
the intersection of WOM, services marketing, social influence, and persuasion literature
streams. Figure 1 presents the positioning of the dissertation in relevant literature fields
and central concepts arising from each intersection.
Thus, when it comes to understanding how WOM is processed, not only its
diagnosticity but also less cognitive processing routes and forms of WOM may provide
interesting insights into the impact of WOM on marketing-related outcomes. Less
cognitive processing of information, such as information processed as simple cues or
18 1 Introduction
heuristics, is important to understand, since its impact on attitude strength differs from
information processed through a more cognitive route (Petty et al., 2013). Strong
attitudes are those that are persistent over time and thus understanding attitude
formation and change is interesting for researchers and practitioners. Therefore, the
processes by which WOM impacts attitude formation is important to understand not
only for the initial effects in attitude formation but also because these processes help to
clarify the long-term consequences of WOM and the behavioral processes of WOM.
All this suggests that the extant literature on WOM identifies the importance of the
diagnosticity of WOM and its relevance in consumer decision-making, which is
emphasized when considering services. However, it leaves different processing
mechanisms, especially the more simple and less cognitive ones yet unexplored. Thus,
following the lead of social psychology and incorporating a theoretical background
from social psychology with prior literature in WOM and services marketing domains,
the present studys author (and the fellow researchers) engaged in an investigation of
the current state of how WOM is processed by consumers using an information
diagnosticity perspective.
First, a gap in current knowledge was identified in the processing of WOM through
dual-processing models of persuasion. The more cognitive processing of WOM is well
acknowledged through prior research focusing on WOM diagnosticity. These studies
have mainly focused on the accessibility-diagnosticity framework (Bone, 1995; Herr et
al., 1991; Van Hoye & Lievens, 2005) and provided novel insights into how consumers
process WOM. Correspondingly, this research will focus on the diagnosticity of WOM
information, but with a similar approach as to Andrews (2013), thus leaving out the
accessibility of information from memory. However, the less thoughtful information
processing mechanisms derived from the ELM and social influence theories will be
taken into account. With the direction of these theories, the aim is to understand the
manner in which the nature of choice criteria and service values affect how consumers
utilize WOM information.
Second, deriving from social influence literature, two forms of social influences have
been identified to affect consumer behavior (normative and informational). Prior
research has noted that both forms of influence are highly relevant in complex and high-
involvement choices. Additionally, prior research has acknowledged that different
social sources (e.g., family, friends) appear to differ in the underlying motivational
1.2 Research objectives 19
effect, or form, of social influence (Childers & Rao, 1992). Further, the very source
itself is also noted in WOM literature as an avenue of future research (Chen, 2008;
Berger, 2014; Zhao & Xie, 2011). While informational influence has a close
resemblance to information diagnosticity, studying normative WOM influence in
parallel with WOM information diagnosticity, taking into account the source of WOM,
will provide further insights into how consumers process WOM.
Third, in addition to the mechanisms of information processing and forms of WOM, this
study will take a longitudinal perspective to consumer decision-making and investigate
the forms of WOM in two distinct phases, evaluation and choice. Berger (2014) calls
for more attention to the behavioral processes of WOM, and since information
processing is generally seen as a dynamic process that evolves over time as more
information is acquired (Briol & Petty, 2012), it is essential to understand how WOM
affects consumer behavior from an attitude change perspective. Further, the different
forms of social influences, informational and normative, are suggested to have distinct
impacts on consumer choice (Kaplan, 1989). Therefore, taking an attitude change
perspective, and understanding the impact of normative and informational
(diagnosticity) influences, will provide further evidence of the behavioral mechanisms
underlying WOM.
Even though the extant literature has recognized the importance of different processing
mechanisms (Voyer & Ranaweera, 2015; Van Hoye & Lievens, 2005) affecting the
impact of WOM in attitude formation and attitude change, there is an empirical gap in
terms of the studies focusing on these processing mechanisms and different forms of
social influence in a high-involvement service context. Thus, investigating different
forms of WOM and processing mechanisms operating in parallel, taking higher
education (HE) as a specific high-involvement service context, creates a relevant
research frontier and addresses theoretical and empirical gaps in the existing knowledge
on WOM.
The study addresses the following research question: How is WOM information
processed in a high-involvement service environment?
The following sub-questions help to provide the answers to the main research question,
through the five research papers included in the structure of the dissertation:
The research process of the dissertation follows four steps. First, the decision-making
process of an international HE applicant is presented. Second, an analysis of the existing
literature at the intersection between WOM and the ELM is conducted. Based on the
results, a conceptual framework is developed for the processing of WOM information.
Third, an analysis of the existing literature at the intersection between WOM and
different forms of social influences is conducted. Fourth, a synthesis of the theoretical
perspectives (ELM and social influences) is presented through the nature of choice
criteria, service values, source effects, and behavioral process of WOM.
The study makes theoretical contributions within the field of WOM by addressing the
study objectives and providing managerial implications. From a theoretical perspective,
the dissertation advances the understanding by forming a conceptual framework of the
processing of WOM information and providing insights into how consumers process
WOM during a high-involvement decision. From a practical point of view, the study
provides insights into the formation of attitudes and long-term effects of WOM, thus
helping marketers in designing more effective WOM campaigns. The author believes
that the study could be a step towards a deeper understanding of how consumers make
use of WOM information in decision-making, and consequently towards revealing what
happens after WOM has been received. Figure 2 outlines the positioning of the
publications within the theoretical framework of this dissertation.
1.3 Structure 21
1.3 Structure
The first chapter of the Introduction of the dissertation provides a background to the
study and introduces the research objectives and context of the study. The second
chapter starts with the theoretical foundations of information processing and social
influences and presents a conceptual model of the processing of WOM information.
Additionally, it presents a synthesis of the theoretical applications for the processing of
WOM information by contextualizing the framework to services marketing through the
nature of choice criteria and then providing models for how service values and the
source of WOM affect this conceptualization. Finally, the chapter provides a model of
the behavioral mechanisms of WOM through an attitude change and consumer choice
perspective. The third chapter summarizes the methodological issues of the study and
describes the research design and data collection applied. The fourth chapter describes
the objectives and results of the publications included. The final chapter concludes by
answering the research question of the study and providing a description of the
dissertations contribution to the existing body of knowledge. Table 1 provides a
description of the publications included in the dissertation.
22 1 Introduction
1.4.1 Word-of-Mouth
There is some variety in how WOM is defined and conceptualized in the literature. It is
referred to as informal communications between consumers concerning the ownership,
usage or characteristics of particular goods, services and/or their sellers (De Matos &
Rossi, 2008, p. 578), and in its broadest sense includes any information about the
target object or brand transferred from one individual to another (Brown et al., 2005, p.
1.4 Definitions of the key concepts 23
1.4.2 Attitude
Attitudes refer to the general and relatively enduring evaluations people have of all
kinds of objects, including products, brands, ads, and ideas (Horcajo, Briol & Petty,
2010, p. 239). The importance of attitudes is acknowledged in marketing literature due
to their central role in behavioral outcomes, such as purchase intention (e.g., Martin &
Lueg, 2013; Sundaram & Webster, 1999) and probability of purchase (e.g., Charlett et
al., 1995). Therefore, the process of attitude formation has gained the interest of
marketing researchers and practitioners aiming to persuade favorable consumer attitudes
(Argyriou & Melewar, 2011). Additionally, Kumar (2015) raised the importance of
consumer attitudes in marketing research and emphasized that marketers should
identify, manage, and understand consumer attitudes. A brand attitude is the overall
evaluation of the brand and forms the basis of consumer behavior such as brand choice
(Keller, 1993). Further, prior research has noted that in consumer behavior research it is
typically assumed that WOM functions through attitude change (Wangenheim &
Bayn, 2007), this view is also adopted in the dual-processing models of persuasion
(Briol & Petty, 2012). Therefore, in this dissertation, the processing of WOM and its
impact on attitudes is studied from a broad perspective, taking into account the
formation of brand attitudes and beliefs. Additionally, attitude change and consumer
choice are assessed. Understanding how WOM impacts the extent of positive brand
attitudes and changes them will provide a broader perspective of the processing of
WOM in decision-making.
24 1 Introduction
Service attributes also known, as choice criteria, are the desired outcomes from the
product or service, which consumers use to evaluate and compare different brands to
1.5 Research context: HE applicants decision-making 25
one another (Engel, Blackwell & Miniard, 1986). In HE research, there exists rich
evidence that applicants have several choice criteria, which further increases the
complexity of evaluation (e.g., Cubillo-Pinilla et al., 2009; Joseph & Joseph, 1998;
Kallio, 1995; Maringe, 2006; Soutar & Turner, 2002). Typical choice criteria identified
by previous research include tuition fees and other costs, reasonable entry requirements,
academic reputation of the university and country, good career prospects, campus
atmosphere, friends choice of university, and family opinion. Among the most
important criteria in the evaluation of an institution, also known as pull factors, are
reputation, career prospects, and academic value (e.g., Mazzarol & Soutar, 2002, Soutar
& Turner, 2002).
Further, service values have been shown to impact HE students behavior and provide
insights into what is driving students or applicants towards a choice (Durvasula et al.,
2011; Lages & Fernandes, 2005). Durvasula et al. (2011) identified three dimensions of
personal (service) values related to HE: peaceful life (SVPL), social recognition
(SVSR), and social integration (SVSI). SVPL applies when consumers experience the
service as supporting pleasurable living, tranquility, security and/or harmony SVSI
relates to the potential of the service to facilitate the development of stronger and better
relationships both professionally and socially, thus fostering SVSI (Lages & Fernandes,
2005). Finally, SVSR relates to the individuals perception of how the service fosters
respect from others and status improvement (Lages & Fernandes, 2005).
Thus, prior research on HE has highlighted the high-involvement nature of the decision-
making process and identified the relevant choice criteria, service values, and the focal
sources of WOM. Therefore, as HE is a complex service, it is interesting to study the
processing of WOM information.
27
2.1.1 ELM
ELM is a dual-processing theory of influence. The key notion in the dual-processing
theories is that some processes of attitude formation and attitude change require
relatively high amounts of cognitive effort, whereas other processes require relatively
little cognitive effort (Petty et al., 2013). The ELM was originally developed to
understand why some attitude changes endured over time, whereas others are more
persistent, and it is based on the premise that the extent of thinking about a persuasive
message determines attitude persistence (Briol & Petty, 2012). Elaboration, the focal
concept in the model, refers to the amount of thinking and is determined by the
motivation and ability of the consumer to process information (Rucker & Petty, 2006).
It is assumed that consumers generally want to understand information and develop
accurate views of the world, thus highlighting the more cognitive processing of
information (Petty, 1994). However, if either the motivation or ability to process
information is lower, then less cognitive mechanism may be used for information
processing. A central assumption is that the same variable (e.g., praise or activity) can
have different effects on attitudes, depending on the situation and context (Petty et al.,
2013).
In the ELM, information processing occurs through two separate routes: the central and
the peripheral. The central route is related to high levels of elaboration in which the
28 2 Theoretical points of departure
receiver focuses on the central merits of issue-relevant information (Petty et al., 2013).
When information is mainly processed through the central route, attitude formation and
change result from extensive thinking of information related to the evaluative topic
(Briol & Petty, 2012). When consumers effortfully scrutinize available information,
with high levels of ability and motivation to do so, they will form a strong and reasoned
(but not necessarily unbiased) attitude (Petty et al., 2013). Thus, attitudes formed
through the central route are more stable and predictive of behavior (Petty, 1994). The
peripheral route, on the other hand, requires less elaboration, and simple cues or
heuristics can change attitudes (Rucker & Petty, 2006). In this route, attitudes can be
affected without extensive issue-relevant thinking or relevant arguments. Thus, the
receiver does not deliberately process the information for its central merits (Petty,
1994). Attitudes that are changed by information processed through the peripheral route
are suggested to be weaker and thus more susceptible to change (Petty et al., 2013).
Generally, when the decision is of high personal relevance to the consumer (e.g., HE),
information is processed through the central route, but if either the ability or motivation
to process the information is decreased, peripheral route processing may occur in
parallel to the central route (Petty, 1994).
Another aspect of the ELM is the multiple roles of variables postulate, meaning that the
same variable such as activity and praise can affect attitudes for different reasons in
different ways (Rucker & Petty, 2006). The process through which a certain variable
will affect persuasion depends on the level of elaboration and the context (Petty, 1994).
It could, for example, serve as an argument or a simple cue or affect the receivers
thoughts through the amount, valence, or degree of attitude confidence (Rucker & Petty,
2006). Simple mechanisms are at play when the level of elaboration is low, because
consumers do not analyze the information for its relevance and do not require extensive
cognitive effort, and thus the mere level of positivity can affect attitudes. When the
level of elaboration is high, on the other hand, consumers who are highly motivated and
have the ability to process it carefully analyze all the information (Rucker & Petty,
2006).
efforts of its relevance and usefulness, and therefore referring to the peripheral route (c).
Additionally, both processing routes can be activated simultaneously, occurring in
parallel processing of information (c, a + b).
Since the ELM holds that any variable can be processed as a simple cue or an argument
depending on different circumstances (Petty, 1994), this perspective should still be
explored in the WOM domain, and the conceptual framework presented in Figure 1 may
provide a good tool for this. However, there are also other forms of WOM (normative
and information) identified in the social influence literature that may provide additional
insights and are thus presented in the following section.
source. Compliance, in turn, occurs because the consumer hopes to attract a favorable
reaction from the social source (Kelman, 1961).
Both normative and informational influences have been identified in complex and high-
involvement consumer decisions (Lord et al., 2001), and the strength of WOM is
suggested to depend on the level of both forms of influence (Wangenheim & Bayn,
2004). However, the normative form of influence has attracted less attention in the
WOM literature, especially with regards to high-involvement service choice. Normative
influence has been described as highly impactful on consumer choice especially when
the evaluation is public or visible to others (Kaplan, 1989) and is highly impactful in
changing behavior (Nolan, Schultz, Cialdini, Goldstein & Griskevicius, 2008).
In the electronic WOM domain (eWOM), certain variables such as argument quality
and type of reviews are considered to reflect central route processing. Other variables
such as source credibility and number of reviews are suggested to reflect peripheral
processing. (For an extensive review on the topic, please see Cheung and Thadani,
2012.) Since the ELM holds that any variable can be processed as a simple cue or an
argument depending on different circumstances (Petty, 1994), this perspective should be
explored in the WOM domain, and the conceptual framework presented in Figure 3 may
provide a good tool for this. However, there are also other forms of WOM (normative
and information) identified in the social influence literature, and the processing of these
different forms is more understood and can be treated more similarly in prior eWOM
studies as a direct relationship of these influences and the marketing-relevant outcome.
Figure 4 represents the conceptual model for the different types of WOM influencing
brand attitudes.
terms of four issues: first, a deeper understanding of how the nature of choice criteria
impacts information processing routes in presented. Second, the impact of service
values on the processing of WOM is assessed. Third, a framework of how different
sources of WOM affect the formation of brand attitudes is created. Finally, the impact
of different forms of WOM on attitude change is evaluated.
According to the ELM, in some situations, the peripheral route may affect attitudes
when the level of elaboration is high. Weakening the motivation to process information,
or the ability to evaluate its central merits, may result in higher levels of peripheral route
processing (Petty, 1994). Consumers seem to be motivated to process issue-relevant
information in evaluating high-involvement services, but as noted previously, the
criteria may be difficult to evaluate, and thus the ability to do so may vary depending on
what is being evaluated. Consumers must have the resources and the capacity to
understand and process the information in order to be able to fully evaluate its central
merits (Rucker & Petty, 2006). The way the message is presented may negatively affect
this ability. It may pass too quickly, for example, or be perceived as ambiguous (Petty,
1994). Hence, the nature of choice criteria is suggested to have a central role in
understanding why information is processed through a certain route in the formation of
attitudes, especially in the service context, given the high incidence of complex criteria.
Various models of brand attitudes have been proposed in the literature, of which a
multiattribute approach has been widely accepted (Keller, 1993). In this approach,
brand attitudes are viewed as a function of the associated attributes and benefits that
are salient for the brand (Keller, 1993, p. 4). The multiattribute approach seems
32 2 Theoretical points of departure
appropriate for the HE context to conceptualize brand attitudes, since prior research has
identified the choice criteria applicants use to evaluate and compare HE institutions
(e.g., Cubillo-Pinilla et al., 2009; Joseph & Joseph, 1998; Maringe, 2006; Mazzarol &
Soutar, 2002; Soutar & Turner, 2002). Further, prior research on WOM has noted that
positive WOM tends to be more effective when considering credence than search
criteria (Lim & Chung, 2011), but what remains unclear is how WOM information is
processed in the formation of the choice criteria. Given the increased complexity in
evaluating experience and credence criteria, the consumer may be forced to process
simple cues due to the ambiguity or difficulty of the information. Consumers who
assume that they cannot form a veridical attitude based on the information, but who are
highly motivated to process it, can according to Petty (1994) either delay their decision
or generate their own relevant information. If the consumer cannot delay the decision
and is unable to generate information, he/she may rely on accessible simple cues in
order to form an evaluative attitude (Petty, 1994). Therefore, it is suggested that
peripheral processing of WOM information as simple cues and heuristics will increase
and the evaluation will become more difficult. Figure 5 outlines this theorizing and
adapts it to the conceptual framework of the processing of WOM information.
Previous studies on values and consumer susceptibility to social influences (i.e., WOM)
such as Hoffman and Broekhuizen (2009) and Kropp et al. (2005) tend to categorize
values broadly as internal, external, and fun/excitement-related. Internal values, such as
self-fulfillment and self-respect, are less dependent on social others, whereas external
values relate more closely to the opinion of others, such as being well respected, secure,
and fulfilled in ones relationships (Homer & Kahle, 1988). Thus, the fulfillment of
external values may be more dependent on others (Batra, Homer & Kahle, 2001). This
distinction is highly relevant in the WOM context given the evidence that external
values positively influence susceptibility to social influences (i.e., WOM) (Hoffman &
Broekhuizen, 2009). Given that values are context-specific, and that Durvasula et al.
(2011) studied service values in the HE domain, it can be argued that the influential
values (presented previously in this dissertation: SVPL, SVSI, SVSR) represent more
external values in that all three are mainly dependent on people other than the applicant.
However, even if these three dimensions of service values tend to represent external
values, they differ in their underlying orientation. SVPL operates on a self-oriented
level, whereas SVSI and SVSR are more socially oriented (Durvasula et al., 2011).
According to Batra et al. (2001), security could be conceived of as an internal value,
even if its fulfillment does not depend solely on the consumer. They use peace of mind
as an example, which may relate to an external threat, or to an internal state of mind;
thus, SVPL could be said to have an internal origin even if it is generally perceived as
an external value (Batra et al., 2001). Given the differences in the underlying orientation
between the service-value dimensions, it is likely that their impact on the processing of
WOM may vary, since internal and external values are suggested to have opposite
effects on consumers tendency to use social influences (i.e. WOM) in their evaluation.
Figure 6 represents the conceptual framework for how service values impact the
processing of WOM in the formation of a credence criteria-based belief.
34 2 Theoretical points of departure
Goodrich and Mangleburg (2010) studied parental and peer influences in young
consumers purchasing behavior. They found that parental social power over the choice
of products was mainly coercive, referent or legitimate, based on social power theory
(French & Raven, 1959). Coercive power is used to threaten non-compliers with
negative consequences, referent power reflects the receivers need to identify with the
sender, and legitimate power reflects the senders right or authority to impose
behavioral requirements on the receiver (Raven, 2008). Thus, what these social powers
have in common is the notion that social influence comes from the perceived qualities
2.2 Theoretical approaches to WOM utilizationA synthesis 35
of the source and is not reflective of the message, and thus could be seen as a form of
normative influence (Kaplan, 1989). Childers and Rao (1992) also investigated familiar
and peer-based reference groups, and found that when consumers were considering
purchasing more luxury products, the strongest peer influence was informational rather
than utilitarian or value-expressive (i.e., normative). The same effect occurred for
specific brand decisions among consumers (Childers & Rao, 1992). As there are some
references to combined normative and informative influence when the source is a
friend, it is suggested that both forms of WOM influences may be active through this
source. Further, according to the social power theory, informational influence generates
two separate forms of social power: expert and informational power (Raven, 2008).
Expert power reflects the receivers belief that the source is knowledgeable and has
some superior insight into the topic at hand, whereas informational power leads to
cognitive change, meaning the giver of the information argues and explains certain
points to the receiver (Raven, 2008). Wilson and Sherrell (1993) find that source
expertise tends to have a strong effect on persuasion and attitudes, suggesting that this
could be attributable to the cognitive nature of the shared information. Thus, this would
suggest that information from the teacher source is mainly processed through the central
route.
public, as it is discussed extensively with family and friends (Binsardi & Ekwulugo,
2003; Kallio, 1995). Kaplan (1989) argues than when a choice is public, normative
influence may be more effective than informative influence on choice, suggesting that
the normative form of WOM may have a stronger impact on HE applicants choice than
informational WOM in later phases of decision-making.
Further, as stated previously, the ELM assumes that attitudes formed through the central
route are generally more persistent over time (Rucker & Petty, 2006). In the context of
HE, applicants are highly involved with the choice and tend to process information
extensively (Chapman, 1986), it is therefore suggested that they are likely to process
information through the central route as much as their ability enables. Since search
criteria-based attitudes are simple to evaluate, this research suggests that the search-
based attitude dimension is already formed mainly through the central route in previous
phases of the decision-making process. Thus, due to the stability of attitudes formed
through the central route, a change in the search-based dimension may be marginal, and
the effect of change is thus smaller on consumer choice. Additionally, the credence-
based attitude dimension may be more susceptible to change than the search-based
dimension, due to the increased amount of peripheral processing of WOM arising from
the difficulty of evaluating credence-based attitudes. Attitudes formed through the
peripheral route are acknowledged to be more susceptible to change, because they are
not as accessible and are more uncertain than attitudes processed through the central
route (Petty & Briol, 2012).
Since, normative influence was reflecting more peripheral processing, and the role of
normative influence is suggested to be stronger in the final stages, it is an interesting
setting for understanding the underlying behavioral mechanisms of WOM. Figure 8
depicts information processing routes and the forms of social influences affecting
attitude change and consumer choice.
The sample consisted of international applicants for Masters degree programs of four
Finnish universities who applied during the spring of 2012. An email with a link to an
online questionnaire (Qualtrics software) was sent to all applicants. The data
represented multiple Masters-level programs. The first data-collection phase (T1)
followed the search phase, but was carried out before the applicants had received any
information from the universities concerning their acceptance or otherwise. The second
data set (T2) followed the choice phase, when the applicants had received letters of
acceptance from different universities, and had made their final choice but had not yet
38 3 Research design and methods
started their studies. In other words, they did not yet have any study-related experience
at the universities. Thus, the data represented real choices and incorporates data points,
which are relatively rarely discussed in the literature. The sample for the search phase
(T1) comprised 1.718 respondents, a response rate of 43.8 percent. Of these, 1.110
respondents proceeded to the second point of data collection, the choice phase (T2),
which took place in August 2012. This second phase yielded 481 responses (response
rate 43.3 percent). Of these, 149 were usable for the longitudinal data, which included
students who had been accepted into one of the original four universities (T1).
Retrospective surveys are subject to bias, such as sampling bias and measurement bias
(East & Uncles, 2008). Sampling bias is associated with the representativeness and
relevance of the sample to the research at hand (East & Uncles, 2008). As the aim of
this dissertation is to understand the impact and processing of WOM during a high-
involvement service choice, an important objective relating to the representativeness of
the sample is to assess whether the sample represents the aim. The sample was collected
at a time when the applicants had already applied to one of the four universities, and the
extent of WOM discussions was assessed in the survey prior to measures of WOM by a
single item slider scale ranging from 0 to 100, anchored from not at all to a great
extent (see Appendix 1). This was conducted in T1 for each source of WOM (family,
teachers, and friends): This part of the questionnaire concerns the discussions you have
had with other people about university x. Please indicate below the extent to which you
have had such discussions, and in T2 for each source of WOM: This part of the
questionnaire concerns the discussions you have had with other people about university
x when you were making your choice on which university to start your studies at,
AFTER you have received your acceptance letter from the university. Please indicate
below the extent to which you have had such discussions. If the responded indicated
that they had had discussions with the social source to some extent, then the following
WOM scales (activity, praise, and diagnosticity) were presented. Thus, the sample can
be seen as representative from this perspective, but a limitation to the sample is that
only one high-involvement service is represented, which also relates to external validity.
In order to reduce measurement bias, both surveys were distributed as close as possible
to the WOM discussions and subsequent behavior, in order to minimize the time span
for recall. Additionally, pretesting was conducted prior to the launch of the surveys to
increase precision in wording and clarity of statements, thus complying with the
compatibility of measures principles (East & Uncles, 2008).
error and less bias than results yielded by, for example, multiple regression, and thus
can be seen as an important tool for marketing researchers (Iacobucci, 2009).
Additionally, SEM modeling takes into account measurement error and incorporates
unobservable variables, and thus can be considered as a strong method for theory testing
and empirical model building in marketing (Fornell & Larcker, 1981).
SEM refers to a general method for estimating the unknown coefficients in a set of
linear structural equations (Jreskog, 1970, p. 1) and is used to specify the
phenomenon under study in terms of tentative cause and effect variables and various
causal effects (Jreskog & Srbom, 1982, p. 404). In this dissertation SEM with
LISREL 8.80 software is conducted for hypothesis testing, as the aim is to test theory on
a phenomenon level. This method is generally adopted in confirmatory studies, (i.e.,
confirming hypothesis based on theory); however, it can also be useful for exploratory
analysis by a sequence of confirmatory analyses, which can be based on more tentative
theory or hypothesis (Jreskog & Srbom, 1993). Confirmatory factor analysis (CFA) is
based on models, which are assumed to describe, explain, or account for empirical data
in terms of relatively few parameters. The CFA is based on a priori information about
the data structure and builds on at least tentative theory, and thus is more confirmatory
than exploratory in nature (Jreskog & Srbom, 1993). In this dissertation, CFA is
conducted in order to determine overall model fit and provide insights into construct
validity, that is, analyze the extent to which observed variables reflect the theoretical
constructs they are hypothesized to reflect (Fornell & Larcker, 1981). These validity
issues are elaborated more in the next section, which focuses on validity and reliability
issues of the dissertation work.
Content validity or face validity refers to the extent in which items are in line with their
conceptual definition (Ping, 2004). Measures for the focal concepts in the different
models appearing in Publications 1-5 in this dissertation were adapted from prior
literature, thus increasing their face validity. The main reason for the adaptation of the
measures was to adapt them into the HE context (except for the measure of service
values, which was originally developed by Durvasula et al. (2011) for HE). After the
adaptation of the measures, an outside expert performed a language revision to the
40 3 Research design and methods
survey. The survey measures were examined by two other researchers prior to data
collection, in order to increase content and face validity. Criterion validity refers to
measure correspondence with other known valid and reliable measures of the same
construct (Ping, 2004, p. 130). Since the survey was rather lengthy (mean response
time for T1: 31 minutes, and for T2: 33 minutes), additional measures for the constructs
could not be added to the instrument. However, since the measures were from previous
research, their validation has been assessed previously. Through the pretesting of both
surveys, each measure was discussed with the pretesters in order to comprehend their
view and understand the measured concepts based on the survey instrument, thus
contributing to face and criterion validities.
After data collection, data can be analyzed for its construct validity through convergent
and discriminant validities (Fornell and Larcker, 1981). Convergent validity refers to
the common variance the variables in a construct share, and is commonly assessed
through average variance extracted (AVE) and constructs reliability (CR). AVE
measures the amount on variance captured by a construct in relation to measurement
error, and CR measures the reliability of the construct (Fornell & Larcker, 1981). AVE
values above 0.50 are usually considered to reflect sufficient construct validity (Hair et
al., 1998), and CR values should be above 0.60 to reflect sufficient internal consistency
(Bagozzi & Yi, 1988). The reliability of this dissertation work will be further discussed
3.3 Validity and reliability of the study 41
in the next paragraph. Discriminant validity refers to the uniqueness of a construct and
can be assessed in different ways: by squaring correlations and comparing them with the
AVEs (Fornell & Larcker, 1981) or by means of nested modeling, as recommended by
Anderson and Gerbing (1988). In the nested modeling approach the correlation between
a pair of constructs is assessed and should be significantly different from 1.0, which
confirms sufficient discriminant validity. Discriminant validity and convergent validity
criteria are reported in the dissertation for each model in Publications 1-5.
Reliability is defined broadly as the degree to which measures are free from error and
therefore yield consistent results (Peter, 1979, p. 6). In order to increase the reliability
of marketing research, Peter (1979) suggests that surveys should use multi-item scales,
improve the clarity of survey instructions, and report construct reliabilities. In this
dissertation research, all measures were multi-item measures and preformed soundly in
both waves of the data collections. Construct reliabilities were reported for all measures
in the different models of Publications 1-5. Additionally, as stated previously, the
survey was pretested and language checked in order to increase clarity and avoid
ambiguous statements. Further, increasing the reliability of this dissertation work was
the two-wave data collection, since the same measures for WOM and brand attitudes
yielded consistent construct reliabilities and results throughout the publications.
43
The first publication provides insights into the context of the research, and studies the
linkage between personal values, educational motives, and choice criteria. The results
indicate that personal values have an important role in forming educational motives,
which in turn, affects university choice criteria. The publication therefore provides
insights into the complexity of HE decision-making.
The second publication focuses on the processing of WOM information during the
search phase of HE decision-making. The paper presents a conceptual framework
through which the processing of WOM can be understood and takes into account the
nature of choice criteria in the formation of brand attitudes. Further, the publication
provides empirical evidence that WOM is processed through both information
processing routes, and suggests that the nature of choice criteria explains the occurrence
of this parallel processing.
The third and fourth publications adopt the same model as the second study.
Additionally, the third publication focuses on the effects service values have on the
processing of WOM information in the formation of a complex credence-based belief,
while the fourth publication investigates how the very source of WOM (family, friends,
teachers) affects the processing routes of WOM.
The fifth study uses a longitudinal model to investigate the behavioral processes of
WOM, since it focuses on how the form of WOM (normative of informational) affects
attitude change, and impacts subsequent choice behavior.
Taken together, the Publications II, III, and IV focus on brand attitude and belief
formation in the search stage, in which the final choice has not yet been made. Through
these publications, the results suggest that the processing of WOM occurs mainly
through the central route, in which information is analyzed for its central merits.
However, the nature of choice criteria, service values, and the source of WOM impact
this process. As the nature of choice criteria becomes more complex, suggesting that
evaluation is more difficult the extent of peripheral route processing increases. The
service values have distinct effects on information processing, depending on their
orientation. Moreover, all three values affect central route processing, but not peripheral
route processing, which might suggest that there are no trade-offs between the
processes. Further, the different sources of WOM are associated with different forms of
influence, and thus the processing of WOM information from these social sources tends
to vary.
Through the conceptual model developed in Publication II, and further validated in
Publications III and IV, it also became evident that WOM activity it processed through
both processing routes, depending on the nature of choice criteria. However, WOM
44 4 Summary of the publications and review of the results
praise was in all publications processed through the peripheral route, which suggests
that WOM praise is processed more as a simple cue of information and not evaluated
for its diagnostic properties. Thus, the publications provide insights into the multiple
roles of variables postulate in the case of WOM in complex and high-involvement
service evaluations.
Publication V focuses on the choice phase and highlights the behavioral processes of
WOM. The publication emphasized the central role of normative WOM in attitude
chance and consumer choice. Further, the central role of normative WOM was
highlighted in the case of a positive decision. Additionally, the results provided novel
insights into the role of informational WOM during a high-involvement service choice,
since informational WOM seemed to be more active when the choice was negative.
Thus, the objective of this study was to provide insights into the linkage between
personal values, educational motives, and choice criteria. The results thus increased
understanding of the decision-making process of a HE applicant. The first paper had a
central role in the structure of this dissertation, since it provides understanding of the
high-involvement and complex nature of the context of this study.
Main contribution
This study extended prior research on HE marketing by applying the goal structure
approach to HE decision-making. By providing empirical evidence of how personal
values affect motives, and how motives affect relevant choice criteria, this study
provides unique insights into linkages between these central variables in the HE
domain.
Based on the results, the SVSR seems to have an important role in explaining
educational motives, especially in the case of instrumental motives. However,
instrumental motives had only a small effect on choice criteria, and the effects that were
significant were negative. Social and scholastic motives seemed to play a larger and
positive role in explaining choice criteria.
4.2 Publication II: The Impact of Word-of-Mouth on Attitude Formation: an 45
Information-processing Perspective
The study thus addressed the objective of how do international HE applicants evaluate
institutions. By examining the linkages between the central concepts, the study provided
further evidence of the complexities and multifaceted nature of HE applicants decision-
making.
Prior research on the processing of WOM has studied WOM diagnosticity and aimed to
understand the factors that contribute to the overall perceived diagnosticity of
communication. This paper aimed to understand the processing of WOM through the
ELM, by adopting the concept of information diagnosticity. In addition to the diagnostic
information, the publication aimed to understand the simple processes of WOM
operating in parallel to the cognitive and diagnostic information.
Thus, the objective of this study is to understand the processing of WOM information,
and its effects on brand attitude formation in a high-involvement service context. The
first paper has an important role in the structure of this dissertation, since it conceptually
builds and empirically tests a model for the processing of WOM for complex service
evaluations, which is further adopted and developed in Publications II and III.
Main contribution
This study expanded the prior research on WOM and services marketing by developing
and empirically testing a conceptual framework of WOM information processing, and
applying it to a high-involvement service context by taking into account the complex
nature of choice criteria.
Adopting the SEC framework into the evaluation of brand attitudes, allows division of
the attitude construct into three distinct levels, which vary in terms of evaluation
difficulty. Through this distinction, the processing of WOM was shown to vary, and
different processing routes were active depending on the difficulty of the nature of the
evaluative attitude. This contributed to the understanding of how consumers utilize
WOM information in their evaluations of complex services.
The study thus addressed the objective of how consumers utilize WOM information, by
providing the theoretical background of social psychology and forming a conceptual
framework in which the context of complex service is taken into account through the
nature of choice criteria.
This study aimed to take the framework developed in Publication II further, by studying
how service values impact the processing of WOM. Since, Publication II showed
interesting findings of parallel processing regarding the most difficult and important
credence criteria, this study focused on beliefs based on the specific criteria and aimed
to gain a deeper understanding of their formation.
Therefore, the objective of this paper was to assess the effect of service values on the
processing of WOM information and the impact of WOM on the evaluation of complex
beliefs. The role of the second study is relevant since it focuses theoretically on an
important topic related to information processing, the service values, which reflect the
underlying reasons as to why a consumer is making the decision in the first place.
Main contribution
The study further validates the proposed model of information processing of WOM,
since the results are in line with the parallel processing of WOM in the formation of
complex beliefs. Thus, the study provided further insights into how consumers utilize
WOM by tapping into the values that matter while making important evaluations.
First, based on the results, service values were found to have distinct effects on the
processing of WOM information. Self-oriented values, which reflect values that are less
dependent on others, were noted to increase the extent of cognitive processing of WOM
in the formation of complex beliefs. However, the opposite effect occurred when
focusing on more socially oriented values.
Second, the study did not find any trade-offs between more cognitive and simple cue
processing of information, meaning that when self-oriented values increased more
cognitive processing of WOM, they did not simultaneously decrease the extent of the
simpler processing of information. The finding encourages further research on the area
but also suggest that when WOM is considered, the tradeoffs are not relevant in forming
complex beliefs through more simple processing. This result is more in line with
theorizing based on the Heuristic Systematic Model (HSM) (Chaiken et al., 1989),
which is also a dual-processing model.
4.4 Publication IV: How the Source of Word-of-Mouth Influences Information 47
Processing in the Formation of Brand Attitudes
4.4 Publication IV: How the Source of Word-of-Mouth Influences
Information Processing in the Formation of Brand Attitudes
Overall objective
In prior WOM research relatively little was understood about how the source of WOM
(e.g., friends, family) affects the impact of WOM on attitude formation. The objective
of this study, with its focus on information processing, was therefore to enhance the
understanding of how consumers utilize WOM information when the source of
information is taken into account. Additionally, the research field of HE found some
varying results regarding WOM effectiveness. Some researchers highlighted the crucial
role of WOM in evaluating HE institutions, whereas others found weaker support,
especially in the case of Masters degree or graduate students.
The role of this study in the structure of the dissertation was to further develop the
conceptualization of information processing and take into account identified research
gaps in WOM literature by focusing on the source of WOM.
Main contribution
Through the ELM and social influence theories, this study was able to provide further
insights into the complexity of the WOM phenomena. Each source (family, friends and
teachers) had its own unique effect on attitude formation. Additionally, as in Publication
I, this study adopted the SEC framework and further verified the focal role of the
complex nature of choice criteria in understanding the processing of WOM.
First, the family source was found highly impactful in the formation of brand attitudes,
and this impact was strongest for the most complex attitudes of the SEC framework.
However, interestingly, the family source was found to provide information mainly
through the peripheral route. Therefore, information diagnosticity (more cognitive
processing) was found irrelevant for the overall model.
Second, the friend source provided information, which was processed through both
routes. Friends also had a notable impact in attitude formation, and WOM praise was
largely influential in the formation of the more complex criteria. Additionally, the
processing in the more simple attitudes was impacted by simple cues but to a lesser
extent than the more difficult criteria.
Third, the teacher source also provided information that was analyzed through both
processing routes. Again, the more complex criteria were formed mainly through simple
cues of WOM information, and this effect grew stronger as the difficulty in evaluating
brand attitudes increased. However, the more simple attitude was formed only through
simple WOM information, which contradicted the prevalent view that cognitive
processing is most active when the information is easy to process. However, this may be
due to the very source of information and the content of the attitude measure.
48 4 Summary of the publications and review of the results
Finally, interaction effects of WOM praise and activity were also studied. Significant
interactions were found in some situations, especially concerning the formation of
simple attitudes. Additionally, a negative interaction between WOM praise and activity
was found affecting information diagnosticity, which would warrant further research to
negative WOM concerning the teacher source of information. The cognitive processing
route has a unique pattern in each source, which may be related to the specific brand
attitude criteria which were adapted by this research. Thus, further research could
investigate more neutral criteria, which vary in their level of difficulty but are not
related content wise to a specific source.
The role of this study in the structure of the dissertation was to provide new insights
into the use of WOM information by adapting two distinct forms of WOM influence,
normative and informational. And further, to investigate their role in choice phase of
decision-making. Thus, this study aimed to respond to the objective in this dissertation
regarding the underlying behavioral process of WOM.
Main contribution
The main results indicate that different forms (normative and informational) of WOM
affect consumers in the final choice phase. The results of the study imply that normative
WOM from close relevant others seems to play a more significant role in attitude
change (credence and search from the SEC framework) and consumer choice, especially
when focusing on highly complex credence-based attitude.
Additionally, post hoc analysis was conducted in order to understand the role of
informational influence in the final choice phase. In the analysis, the focus was on the
formation of beliefs for when a positive choice was made compared to results for when
the choice was negative. The results of the post hoc analysis indicated that
informational WOM had a significant effect on both search- and credence-based beliefs
4.5 Publication V: Facts or Opinions - Which Make a Difference? Word-of-Mouth
49
and Attitude Change in a High-involvement Service Context
when the choice was negative and that normative influence played a more important
role in credence-based belief formation when the choice was positive. Thus, the main
contribution of this study lay in demonstrating the highly central role of normative
influence in consumer behavior, but additionally pointing out the central role on
informational influence during a negative choice.
Throughout this dissertation, the aim was to understand how the receiver utilizes WOM
in decision-making by focusing on the processing routes of WOM and different forms
of WOM. The combined results provide evidence of the complexity of WOM and
highlight the important role of processing WOM information as simple cues and of the
normative role of WOM, which has not been paid attention to in prior research to a
great extent. Finally, the results provided evidence that WOM has a central role in
forming attitudes, beliefs, and consumer choice.
51
5 Conclusions
Taken together, the purpose of this dissertation was to enhance scientific knowledge in
the field of marketing by narrowing the gap identified in WOM literature concerning the
processing of WOM by consumers. In order to achieve this goal, the two following
research objectives were set:
The second research objective was addressed by collecting longitudinal data, as high-
involvement choices are associated with long and extensive information search phases.
The contextualization of knowledge of the relationships between WOM and related
outcomes was additionally addressed by adopting the SEC framework in Publications
II, III, IV and V. Thus, focusing on the nature of choice criteria through attitudes, which
are associated with complex service decisions, provided knowledge on why certain
processing routes of WOM are active and effective during the evaluation of a service.
mainly focused on its more cognitive nature, through information diagnosticity (e.g.,
Herr et al., 1991). However, this research suggests, based on the assumptions of the
ELM that more simple processing mechanisms are active and can operate in parallel
with diagnostic information. To the authors best knowledge, this is the first study to
investigate both processing routes of WOM information in the services domain on an
empirical level.
Through Publications III and IV the developed framework in Publication I for WOM
information processing was further verified, with additional gaps identified in the WOM
literature. First, by incorporating services values, and second, by focusing on specific
sources of WOM. Through, the results of Publications 2, 3, and 4 it becomes evident
that simple processing mechanisms are important to identify and acknowledge, since
they have an additional impact in the formation of brand attitudes and beliefs and
further are mostly present in the information provided by each source of WOM.
More specifically, the findings of the different publications suggest that the different
processing routes are complimentary in nature. In Publications II, III, and IV the results
indicated a higher impact on the outcome when parallel processing occurs. The only
exception for this is information from the family source, which seemed to have
compatible explanatory power on brand attitudes compared to the other sources for the
more complex criteria, even though the cognitive processing route seemed to be
irrelevant for the processing of information through the source. This finding interests
more investigation into the relevance of normative influence and information processed
as simple cues. However, taken together, the results imply that parallel processing of
information occurs especially in the processing of complex criteria which suggests that
more cognitive processing is active as far as the ability to evaluate difficult criteria
enables, and the peripheral route provides additional information for the formation of a
veridical attitude.
5.1.2 Bringing forth the context: the nature of choice criteria in complex service
decisions
In addition to recognizing both processing routes, this study provides novel insights into
the reasons that certain processing routes are active by applying the SEC framework.
Through the SEC framework, the author was able to demonstrate how difficulty in
evaluating, arising from the nature of choice criteria, affects the processing routes of
WOM information. Prior research has acknowledged that this difficulty affects
consumers ability to meaningfully evaluate the information (Murray, 1991). This
research thus aimed to consider this notion and empirically study its effects on
information processing. Prior research on social psychology has acknowledged that the
ability to evaluate information affects information processing routes (Petty, 1994), but
this had not been evaluated in the WOM domain through relevant choice criteria.
Through the results of Publications II, III, and IV, it can be noted that the parallel
processing of both routes increases as the complex nature of choice criteria increases.
5.1 Theoretical contribution of the study 53
This implies that more complex and important attitudes are not as stable as the more
simple attitudes. To provide insights into this proposition, Publication V studied how
WOM changes attitudes, and based on the results, it seems that the more complex
criteria indeed change more than simple attitudes do based on WOM, and that this
change affects consumers choice to a greater extent. This provides additional support
for the importance of understanding the processing on WOM through both information
processing routes.
However, prior WOM research, especially that focusing on complex services, has not
investigated the roles of WOM constructs (praise and activity). Thus, the results of
Publications I, III, and, IV provide insights into this issue. In all the three publications,
WOM praise was not mediated in any model, which would suggest that WOM praise is
not analyzed for its diagnostic properties but is considered more as a simple cue of
information. Further, WOM activity was mediated in multiple occasions, but it was also
processed as simple cues. This implies the dual role of WOM activity. Therefore, this
study contributes to WOM research by providing insights into how different constructs
of WOM are processed. These results may also shed more light into the findings
reported by Liu (2006) and Wu (2013) concerning the focal role of WOM activity (i.e.,
volume). As Wu (2013) notes, when the effect of WOM activity is controlled for, the
valence of the message (i.e., WOM praise) becomes less relevant in consumer
evaluation. Based on the results of this dissertation, it seems that WOM activity is
indeed impactful since it is processed for its diagnostic properties to a greater extent
than WOM praise. Prior research may have therefore found the focal role of WOM
volume (compared to valence) in impacting outcomes, such as awareness (Liu, 2006).
(Nolan et al., 2008). This has also been noted in the WOM domain, and it has been
suggested that the strength of WOM depends on the level of both forms of influence
(Wangenheim & Bayn, 2004). However, to the authors best understanding, this has
not been empirically studied in the WOM domain in relation to consumer choice.
This research provides empirical evidence that the normative form of WOM is, in fact,
highly impactful in a complex service choice. The results of Publication V suggest that
normative influence plays a crucial role in changing complex attitudes and affecting
consumer choice. Additionally, the results provide evidence that normative influence is
highly relevant in the formation of beliefs when a positive decision is made. However,
the results also highlight the role of informational WOM when the decision is negative,
thus extending the understanding of how these two forms of WOM impact behavioral
processes.
Marketers can attempt to trigger WOM through indirect and direct marketing efforts.
Indirect efforts include general advertising attempts, such as, testimonials, teaser
campaigns, and celebrity endorsements (Lang & Hyde, 2013). Direct efforts include
incentivized WOM (i.e., the company gives a reward for spreading WOM), or targeting
influencers (Lang & Hyde, 2013). The results of dissertation show that both processing
routes are active when potential customers of high-involvement services engage in the
evaluation process, and this effect is dependent on differences in the nature of the
choice criteria. The results, therefore, provide insights into how information in terms of
choice criteria conveyed by advertisements or testimonials could be portrayed in order
to support both processing mechanisms. Additionally, influencers or incentivized WOM
senders can be educated beforehand on how they could portray the message of the target
brand based on the same knowledge.
A useful starting point here is that managers should identify the relevant choice criteria
associated with the service and acknowledge that different activities should be targeted
in an attempt to communicate these criteria effectively. Identifying the complex criteria
is the first step in understanding how their potential consumers may utilize WOM in
5.2 Managerial implications 55
Many marketers attempt to trigger positive WOM of their brand, and the findings of this
study provide guidelines for those engaged in high-involvement services. Given that the
more simple search-based criteria are processed mainly through the central route
(wherein issue-relevant information is scrutinized and other information may be
disregarded as irrelevant), marketers should do their best to provide potential consumers
with clear information of search criteria through marketing efforts that gives them
relevant content to discuss, thereby harnessing the practical value of WOM. Practical
value has been shown to be a motivation for consumers to engage in WOM discussions
(Berger, 2013), given that they ask for advice and share information. If a marketing
manager invests in a large buzz campaign, celebrity endorsement, or arousal-generating
viral video that highlights search-based information, he or she may not be spending
wisely, as consumers may consider the extra effort irrelevant to their judgment (Petty,
1994). On the other hand, such efforts may be highly effective in conveying experience
and credence criteria based information. It should be remembered that both information
processing routes are active when consumers evaluate these difficult criteria. Thus, a
viral campaign must in addition to creating arousal, convey credible information for
scrutinizing the information and harnessing practical value.
Additionally, marketers can attempt to promote the ability of the consumer to process
information in their attempt to create strong brand attitudes. This can be done by
designing marketing efforts, which promote message repetition, prevent the message
from being presented too fast, and control for distractions (Petty et al., 2013). These are
important issues a marketer should take into account when designing marketing
campaigns facilitating WOM, especially when credence criteria are highly important in
evaluation, such as in the case of complex services. Since WOM activity is processed
for its diagnostic properties to a greater extent that WOM praise, marketers could devote
more attention to increasing WOM activity. However, marketers should ensure that the
active WOM concerning their service is not negative in valence, and that the valence of
the communication is also relevant.
In terms of service values, the results imply that consumers with high levels of values
related to peace and contentment are more likely to process information as arguments
than consumers with low levels of such values, whereas the opposite is true for
consumers who highly value social integration and, given the direction of our results,
for those who value social recognition. Thus, given the presence of simple information
cues in the formation of complex and focal beliefs, marketing managers should aim to
support the level of cognitive processing among consumers. These supporting efforts
should be stronger when targeting consumers who value social integration and social
recognition, and tend to scrutinize the informational aspects of messages to a lesser
extent. One suggestion is that marketers can reflect service values in their marketing
efforts: in the case of social integration the marketer could provide information about
56 5 Conclusions
Additionally, a marketer should identify the focal sources of WOM information, since
the very source impacts the processing routes of WOM. A marketer can attempt to
target families by means of different communication mediums, such as direct mail and
advertising. These communication mediums incorporated with relevant information
presented clearly and understandably to applicants parents in their own language may
also facilitate more information rich discussions. Since some sources and dimensions of
brand attitude were more associated with peripheral route processing, a marketer could
aim to target these sources and discussions concerning the certain criteria to a greater
extent by aiming to increase WOM activity. More complex criteria (credence) could be
conveyed with a high level of positivity in order to increase overall impact not only via
largely fact-based communication but in combination. Further, more simple information
(search) could be encouraged in a clear manner, without devoting extensive effort to
creating arousal and excitement. Marketers should be cautious in targeting these sources
given the evidence of some negative effects in the evaluation of search-based criteria.
Further, due to the important role of normative WOM in the choice phase, a marketer
should place extra effort in increasing this form of WOM in this phase. Marketing
efforts targeted on normative or informational WOM differ in nature in that the latter
requires arguments and facts, whereas normative influence may require the arousal of
positive emotions in particular. According to the results of the dissertation, marketing
efforts should aim to strengthen the influence of normative WOM from close relevant
others, especially in the final phase of the choice process. Given the focal role of
normative WOM identified in our analyses, it could be concluded that marketers need to
sell their services not only to the decision maker but also to close relevant others such
as family and friends, whose approval and support seem to be strongly influential.
This dissertation focused on the ELM as the primary theoretical background. However,
the focus on ELM may entail the exclusion of some potentially applicable theories such
as the Heuristic Systematic Model (HSM), which also examines dual-processing of
persuasion (Chaiken et al., 1989) and provides a useful framework for understanding
information processing. However, the ELM acknowledges that there are multiple forms
of peripheral processes and does not limit them to retrieving an accessible and stored
decision rule, as HSM does. ELM enables the processing of new information for low-
level conditions, since in some occasions consumers receive information about products
or services that they do not have prior experience with, such as in the case of this study.
However, other important assumptions (such as dual-processing, ability and motivation
to process information) affecting this research based on the ELM are shared with the
HSM. Further investigations with experimental designs are also recommended to
establish causality and investigate individual differences of the consumers ability and
motivation to process WOM information.
Interestingly, none of the simple cues (direct effects of WOM praise or WOM activity
on credence belief) seemed to be moderated by service values, which is a somewhat
surprising finding, as the ELM postulates a tradeoff between the impacts of high- and
low-effort processes on judgments, such as that the former strengthens and the latter
weakens (Petty et al., 2013). The fact that none of the simple cues were influenced by
service values even if the central-route processing was could imply that there are no
tradeoffs. This calls for more research in the area. A richer approach to measuring
WOM, taking into account content and negative WOM for instance, may provide
insight on the issue.
Further, cost-related issues, which are significant as evaluation criteria (e.g., Soutar &
Turner, 2002), were excluded from the study for contextual reasons (i.e., only two of the
participating universities charged tuition fees). Within the HE context, it would be
interesting to find out if the relevant criteria are the same for the actual decision maker
and their close relevant others. Earlier research has identified critical choice criteria for
the decision maker, but given the implication in our results that HE applicants seek the
approval and support of family and friends, their evaluation criteria should also be
investigated.
The author believes that the conceptualization of WOM as activity and praise is
sufficient to capture the phenomenon of WOM, but acknowledges the limitation of
additional variables of WOM, and therefore suggests this for further research. Future
58 5 Conclusions
research could adopt a richer approach to the study of WOM by including additional
elements such as negative WOM (Goyette, Line, Bergeron & Marticotte, 2010;
Sweeney et al., 2014), cognitive content, and content richness (Sweeney et al., 2012).
Negative WOM may provide an interesting perspective, since negative WOM attributed
toward the brand affects consumer evaluations and has been proposed to have higher
diagnostic properties (Laczniak et al., 2001) therefore increasing the role of information
processing through the central route. Additionally, investigating whether these other
concepts of WOM are processed as simple cues of information or evaluated for their
diagnostic properties in a similar manner as in the conceptual framework presented in
this dissertation will lead to further insights into how consumers process WOM.
59
References
Ahluwalia, R., Unnava, H.R. and Burnkrant, R.E. (2001). The moderating role of
commitment on the spillover effect of marketing communications. Journal of
Marketing Research 38(4), pp. 458-470.
Ajzen, I., Brown, T.C. and Rosenthal, L.H. (1996). Information Bias in Contingent
Valuation: Effects of Personal Relevance, Quality of Information, and Motivational
Orientation. Journal of Environmental Economics and Management, 30(1), pp. 43-
57.
Alba, J., Lynch, J., Weitz, B., Janiszewski, C., Lutz, R., Sawyer, A. and Wood, S.
(1997). Interactive home shopping: Consumer, retailer, and manufacturer incentives
to participate in electronic marketplaces. Journal of Marketing, 61(3), pp. 38-53.
Anderson, J.C. and Gerbing, D.W. (1988). Structural equation modeling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103(3), pp.
411-423.
Andrews, D. (2013). The interplay of information diagnosticity and need for cognitive
closure in determining choice confidence. Psychology & Marketing, 30(9), pp. 749-
764.
Bagozzi, R.P. and Yi, Y. (1988). On the evaluation of structural equation models.
Journal of the Academy of Marketing Science, 16(1), pp. 74-94.
Bansal, H.S. and Voyer, P.A. (2000). Word-of-mouth Processes within a Service
Purchase Decision Context. Journal of Service Research, 3(2), pp. 166-177.
Batra, R., Homer, P.M. and Kahle, L.R. (2001). Values, susceptibility to normative
influence, and attribute importance weights: A nomological analysis. Journal of
Consumer Psychology, 11(2), pp. 115-128.
Bearden, W.O. and Etzel, M.J. (1982). Reference group influence on product and brand
purchase decisions. Journal of Consumer Research, 9(2), pp. 183-194.
Berger J. (2013). Contagious: Why Things Catch on. Simon & Simon: New York, NY.
Boughin, J., Doogan, J. and Vetvik, O.J. (2010). A new way to measure word-of-mouth
marketing. McKinsey Quarterly, 2, pp. 113-116.
Briol, P. and Petty, R.E. (2012). A history of attitudes and persuasion research, in
Kruglanski A. and Stroebe, W. (Eds.), Handbook of the History of Social Psychology
(pp.285-320), Psychology Press, New York.
Brown, J.J. and Reingen, P.H. (1987). Social ties and Word-of-Mouth referral behavior.
Journal of Consumer Research, 14(3), pp. 350-362.
Brown, T.J., Barry, T.E., Dacin, P.A. and Gunst, R.F. (2005). Spreading the word:
Investigating antecedents of consumers positive Word-of-Mouth intentions and
behaviors in aretailing context. Journal of the Academy of Marketing Science, 33(2),
pp. 123-138.
Bryman, A. and Bell, E. (2015). Business research methods. 4th edition. New York:
Oxford University Press.
Chaiken, S., Liberman, A. and Eagly, A.H. (1989). Heuristic and systematic processing
with and beyond the persuasion context. In J. S. Uleman and J. A. Bargh (Eds.),
United Thought (pp. 212-252). New York, NY: Guilford Press.
Chapman, D.W. (1981). A model of student college choice. The Journal of Higher
Education, 52(5), pp. 490-505.
Charlett, D., Garland, R. and Marr, N. (1995). How damaging is negative word of
mouth. Marketing Bulletin, 6(1), pp. 42-50.
Cheung, C.M.K. and Thadani, D.R. (2012). The impact of electronic word-of-mouth
communication: A literature analysis and integrative model. Decision Support
Systems, 52(1), pp. 461-470.
Childers, T.L. and Rao, A.R. (1992). The influence of familial and peer-based reference
groups on consumer decisions. Journal of Consumer Research, 19(2), pp. 198-211.
Cubillo-Pinilla, J.M., Zuniga, J., Losantos, I.S. and Sanchez J. (2009). Factors
influencing international students evaluations of higher education programs. The
Journal of American Academy of Business, 15(1), pp. 270-278.
Darby, M.R. and Karni, E. (1973). Free competition and the optimal amount of fraud.
Journal of Law and Economics, 6(1), pp. 67-88.
Deutsch, M. and Gerard, H.B. (1955). A study of normative and informational social
influences upon individual judgment. Journal of Abnormal and Social Psychology,
51(3), pp. 629-636.
Durvasula, S., Lysonski, S. and Madhavi, A.D. (2011). Beyond service attributes: Do
personal values matter? Journal of Services Marketing, 25(1), pp. 33-46.
East, R., Hammond, K., Lomax, W. and Robinson, H. (2005). What is the effect of a
recommendation? The Marketing Review, 5(2), pp. 145-157.
East, R., Hammond, K. and Lomax, W. (2008). Measuring the impact of positive and
negative word of mouth on brand purchase probability. International Journal of
Research in Marketing, 25(3), pp. 215-224.
Engel, J.F., Blackwell, R.D. and Miniard, P.W. (1986). Consumer Behavior. Dryden
Press: New York.
62 References
Fabrigar, L.R., Porter, R.D. and Norris, M.E. (2010). Some things you should know
about structural equation modeling but never thought to ask. Journal of Consumer
Psychology, 20(2), pp. 221-225.
Fang, C.H., Lin, T.M., Liu, F. and Lin, Y.H. (2011). Product type and word of mouth: a
dyadic perspective. Journal of Research in Interactive Marketing, 5(2/3), pp. 189-
202.
Feldman, J.M. and Lynch, J.G. (1988). Self-generated validity and other effects of
measurement on belief, attitude, intention and behavior. Journal of Applied
Psychology, 73(3), pp. 421-435.
Fornell, C. and Larcker, D.F. (1981). Evaluating structural equation models with
unobservable variables and measurement error. Journal of Marketing Research,
18(1), pp. 39-50.
French, J.R.P. Jr., and Raven, B.H. (1959). The bases of social power. In D. Cartwright
(Ed.), Studies in Social Power (150-167). Ann Arbor, MI: Institute of Social
Research.
Gilly, M.C., Graham, J.L., Wolfinbarger, M.F. and Yale, L.J. (1998). A dyadic study of
interpersonal information search. Journal of the Academy of Marketing Science,
26(2), pp. 83-100.
Goodrich, K. and Mangleburg, T.F. (2010). Adolescent perceptions of parent and peer
influences on teen purchase: an application of social power theory. Journal of
Business Research, 63(12), pp. 1328-1335.
Goyette, I., Line, R., Bergeron, J. and Marticotte, F. (2010). e-WOM scale: Word-of-
mouth measurement scale for e-services context. Canadian Journal of Administrative
Sciences, 27(1), pp. 5-23.
Hair, J.F.Jr., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998). Multivariate data
analysis, Fifth edition. New Jersey: Prentice-Hall.
Hartline, M.D. and Jones, K.C. (1996). Employee performance cues in a hotel service
environment: Influence on perceived service quality, value, and word-of-mouth
intentions. Journal of Business Research, 35(3), pp. 207-215.
Herr, P.M., Kardes, F.R. and Kim, J. (1991). Effects of word-of-mouth and product
attribute information on persuasion: An accessibility-diagnosticity perspective.
Journal of Consumer Research, 17(4), pp. 454-462.
References 63
Homer, P.M. and Kahle, L.R. (1988). A structural equation test of the values-attitude-
behavior- hierarchy. Journal of Personality and Social Psychology, 54(4), pp. 638-
646.
Horcajo, J., Briol, P. and Petty, R.E. (2010). Consumer persuasion: Indirect change
and implicit balance. Psychology & Marketing, 27(10), pp. 938-963.
Hunt, S.D. (1990). The truth in marketing theory and research. Journal of Marketing,
40(3), pp. 17-28.
Iacobucci, D. (2008). Mediation analysis (No. 156). Thousand Oaks, CA: Sage.
Iacobucci, D. (2009). Everything you always wanted to know about SEM (structural
equations modeling) but were afraid to ask. Journal of Consumer Psychology, 19(4),
pp. 673-680.
Johnston, T.C. (2010). Who and what influences choice of university? Student and
university perceptions. American Journal of Business Education, 3(10), pp. 15-23.
Jun, S.Y., Cha, T. and Aggarwal, P. (2011). How much better? The effect of tensile
information on word-of-mouth effectiveness. Journal of Marketing Theory and
Practice, 19(3), pp. 263-276.
Jreskog, K.G. (1970). A general method for estimating a linear structural equation
system. ETS Research Bulletin Series, 1970(2), pp. 1-41.
Jreskog, K.G. and Srbom, D. (1993). LISREL 8: Structural equation modeling with
the SIMPLIS command language. Lincolnwood, IL: Scientific Software
International.
Kahle, L.R. (1983). Social values and social change: Adaptation to life in America.
New York: Praeger.
Kahle, L.R., and Kennedy, P. (1989). Using the list of values (LOV) to understand
consumers. Journal of Consumer Marketing, 6(3), pp. 5-12.
Kallio, R.E. (1995). Factors influencing the college choice decision of graduate
students. Research in Higher Education, 36(1), pp. 109-124.
Kelman, H.C. (1961). Process of opinion change. The Public Opinion Quarterly, 25(1),
pp. 57-78.
Kropp, F., Lavack, A.M. and Silvera, D.H. (2005). Values and collective self-esteem as
predictors of consumer susceptibility to interpersonal influence among university
students. International Marketing Review, 22(1), pp. 7-33.
Kumar, V. (2015). Evolution of marketing as a discipline: What has happened and what
to look out for. Journal of Marketing, 79(1), pp. 1-9.
Laczniak, R.N., DeCarlo, T.E. and Ramaswami, S.N. (2001). Consumers responses to
negative word-of-mouth communication: An attribution theory perspective. Journal
of Consumer Psychology, 11(1), pp. 57-73.
Lages, L.F. and Fernandes, J.C. (2005). The SERPVAL scale: A multi-item instrument
for measuring service personal values. Journal of Business Research, 58(11), pp.
1562-1572.
Lang, B. and Hyde, K.F. (2013). Word of mouth: What we know and what we have yet
to learn. Journal of Consumer Satisfaction, Dissatisfaction & Complaining Behavior,
26, pp. 3-18.
Lim, B.C. and Chung, C.M.Y. (2011). The impact of word-of-mouth communication on
attribute evaluation. Journal of Business Research, 64(1), pp. 18-23.
Liu, Y. (2006). Word of mouth for movies: Its dynamics and impact on box office
revenue. Journal of Marketing, 70(3), pp. 74-89.
Lord, K.R., Lee, M. and Choong, P. (2001). Differences in normative and informational
social influence. Advances in Consumer Research, 28, pp. 280-285.
MacKenzie, S.B. and Podsakoff, P.M. (2012). Common method bias in marketing:
causes, mechanisms, and procedural remedies. Journal of Retailing, 88(4), pp. 542-
555
Martin, W.C. & Lueg, J.E. (2013). Modeling word-of-mouth usage. Journal of Business
Research, 66(7), 801-808.
Nolan, J.M., Schultz, P.W., Cialdini, R.B., Goldstein, N.J. and Griskevicius, V. (2008).
Normative social influence is underdetected. Personality and Social Psychology
Bulletin, 34(7), pp. 913-923.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1985). A conceptual model of service
quality and its implications for future research. Journal of Marketing, 49(4), pp. 41-
50.
Peter, J.P. (1979). Reliability: A review of psychometric basics and recent marketing
practices. Journal of Marketing Research, 16(1), pp. 6-17.
Petty, R.E. (1994). Two routes to persuasion: State of the art. In Eelen, P., Pichelle, M.,
d'Ydewalle, G. and Bertelson, P. (Eds.), 25th International Congress of Psychology:
State of the art lectures, Erlbaum, Hillsdale, NJ.
66 References
Petty, R.E. and Cacioppo, J.T. (1986). Communication and persuasion: Central and
peripheral routes to attitude change. New York, NY Springer-Verlag.
Petty, R.E., Wheeler, S.C. and Tormala, Z.T. (2013). Persuasion and attitude change. In
Weiner, I. B. and Lerner, M. J. (Eds.), Comprehensive Handbook of Psychology:
Vol. 5, Personality and Social Psychology (2nd Ed.), pp. 369-390. John Wiley &
Sons, New York.
Ping, R.A.Jr. (2004). On assuring valid measures for theoretical models using survey
data. Journal of Business Research, 57(2), pp. 125-141.
Raven, B.H. (2008). The bases of power and the power/interaction model of
interpersonal influence. Analyses of Social Issues and Public Policy, 8(1), pp. 1-22.
Rokeach, M. (1973). The Nature of Human Values. New York: Free Press.
Rucker, D.D. and Petty, R.E. (2006). Increasing the effectiveness of communications to
consumers: Recommendations based on elaboration likelihood and attitude certainty
perspectives. Journal of Public Policy & Marketing, 25(1), pp. 39-52.
Soutar G.N. and Turner, J.P. (2002). Students preferences for university: A conjoint
analysis. International Journal of Education Management, 16(1), 40-45
Sundaram, D.S. and Webster, C. (1999). The role of brand familiarity on the impact of
word-of-mouth communication on brand evaluations. Advances in Consumer
Research, 26, pp. 664-670.
Sweeney, J.C., Soutar, G.N. and Mazzarol, T. (2008). Factors influencing word of
mouth effectiveness: receiver perspectives. European Journal of Marketing, 42(3/4),
pp. 344-364.
Sweeney, J.C., Soutar, G.N. and Mazzarol, T. (2012). Word of mouth: Measuring the
power of individual messages. European Journal of Marketing, 46(1/2), pp. 237-257.
Sweeney, J.C., Soutar, G.N. and Mazzarol, T. (2014). Factors enhancing word-of-mouth
influence: positive and negative service-related messages. European Journal of
Marketing, 48(1/2), pp. 336-359.
References 67
Voyer, P.A. and Ranaweera, C. (2015). The impact of word of mouth on service
purchase decisions: examining risk and the interaction of tie strength and
involvement. Journal of Service Theory and Practice, 25(5), pp. 636-656.
Wangenheim, F.V. and Bayn, T. (2004). The effect of word of mouth on services
switching measurements and moderating variables. European Journal of Marketing,
38(9/10), pp. 1173-1185.
Wangenheim, F.V. and Bayn, T. (2007). The chain from customer satisfaction via
word-of-mouth referrals to new customer acquisition. Journal of the Academy of
Marketing Science, 35(2), pp. 233-249.
Wilson, E.J. and Sherrell, D.L. (1993). Source effects in communication and persuasion
research: A meta-analysis of effect size. Journal of the Academy of Marketing
Science, 21(2), pp. 101-112.
Yang, S., Hu, M., Winer, R.S. Assael, H. and Chen, X. (2012). An empirical study of
word-of-mouth generation and consumption. Marketing Science, 31(6), pp. 952-963.
Zhao, M. and Xie, J. (2011). Effects of social and temporal distance on consumers
responses to peer recommendations. Journal of Marketing Research, 48(3), pp. 486-
496.
69
Please give your opinion about the information you received from the discussions you
had about University of Technology (LUT). The left-hand column indicates which
discussions are of concern.1
Please indicate the extent to which you agree or disagree with the statements in the left-
hand column on information received from discussions with your FRIENDS2 about
Lappeenranta University of Technology (LUT)1
1
The name of the university that appears in the survey was dependent on which
university the applicant had applied.
2
Each source of WOM was asked separately
70
Normative Influence
Please indicate below the extent you feel people close to you think about your choice to
study at Lappeenranta University of technology (LUT)1
1
The name of the university that appears in the survey was dependent on the applicants
positive decision.
Personal/service values
Please indicate the extent to which you agree or disagree on what a Master's degree
allows you to achieve
71
Educational motives
How important were the following reasons motivating you to study a higher education
degree
72
Choice Criteria
Now please, go back in time when you were making a decision to take a full university
degree in Finland. To what extent the following factors had an effect on your decision-
making
73
Please indicate below the importance of the following factors to you in your choice of a
university for Master's degree studies. Please respond to all the questions.
74
Beliefs
To what extent do you agree or disagree that the following factors are associated with
the university1. Please respond to all the questions.
1
The name of the university that appears in the survey was dependent on the applicants
positive and negative decision.
75
Question formatting (T1): This part of the questionnaire concerns the discussions you
have had with other people about Lappeenranta University of Technology (LUT).
Please indicate below the extent to which you have had such discussions. (Publications
2, 3, and 4)
Question formatting (T2): This part of the questionnaire concerns the discussions you
have had with other people about Lappeenranta University of Technology (LUT), when
you were making your choice on which university to start your studies at, AFTER you
have received your acceptance letter from the university. (Publication V)
Appeared in both T1 and T2: Please indicate below the extent to which you have had
such discussions
ACTA UNIVERSITATIS LAPPEENRANTAENSIS
630. LUOSTARINEN, LAURI. Novel virtual environment and real-time simulation based methods
for improving life-cycle efficiency of non-road mobile machinery. 2015. Diss.
632. KOLOSENI, DAVID. Differential evolution based classification with pool of distances and
aggregation operators. 2015. Diss.
634. KIVYIRO, PENDO. Foreign direct investment, clean development mechanism, and
environmental management: a case of Sub-Saharan Africa. 2015. Diss.
637. ABDEL WAHED, MAHMOUD. Geochemistry and water quality of Lake Qarun, Egypt.
2015. Diss.
638. PETROV, ILYA. Cost reduction of permanent magnet synchronous machines. 2015. Diss.
639. ZHANG, YUNFAN. Modification of photocatalyst with enhanced photocalytic activity for
water treatment. 2015. Diss.
640. RATAVA, JUHO. Modelling cutting states in rough turning of 34CrNiMo6 steel. 2015. Diss.
641. MAYDANNIK, PHILIPP. Roll-to-roll atomic layer deposition process for flexible electronics
applications. 2015. Diss.
642. SETH, FRANK. Empirical studies on software quality construction: Exploring human factors
and organizational influences. 2015. Diss.
643. SMITH, AARON. New methods for controlling twin configurations and characterizing twin
boundaries in 5M Ni-Mn-Ga for the development of applications. 2015. Diss.
650. ROOZBAHANI, HAMID. Novel control, haptic and calibration methods for teleoperated
electrohydraulic servo systems. 2015. Diss.
651. SMIRNOVA, LIUDMILA. Electromagnetic and thermal design of a multilevel converter with
high power density and reliability. 2015. Diss.
653. ZUBEDA, MUSSA. Variational ensemble kalman filtering in hydrology. 2015. Diss.
655. SOKOLOV, MIKHAIL. Thick section laser beam welding of structural steels: methods for
improving welding efficiency. 2015. Diss.
656. GORE, OLGA. Impacts of capacity remunerative mechanisms on cross-border trade. 2015.
Diss.
657. AURINKO, HANNU. Risk assessment of modern landfill structures in Finland. 2015. Diss.
658. KAIJANEN, LAURA. Capillary electrophoresis: Applicability and method validation for
biorefinery analytics. 2015. Diss.
660. ANNALA, SALLA. Households willingness to engage in demand response in the Finnish
retail electricity market: an empirical study. 2015. Diss.
661. RIABCHENKO, EKATERINA. Generative part-based Gabor object detector. 2015. Diss.
662. ALKKIOMKI, VILLE. Role of service and data reuse in enterprises. 2015. Diss.
663. VNTSI, OLLI. Utilization of recycled mineral wool as filler in wood plastic composites.
2015. Diss.