Podsakoff Et Al 2016 Recommendations For Creating Better Concept Definitions in The Organizational Behavioral and

Article
Organizational Research Methods

2016, Vol. 19(2) 159-203
ª The Author(s) 2016
Recommendations Reprints and permission:
sagepub.com/journalsPermissions.nav
for Creating Better DOI: 10.1177/1094428115624965
orm.sagepub.com
Concept Definitions in the
Organizational, Behavioral,
and Social Sciences
Philip M. Podsakoff1, Scott B. MacKenzie2,

and Nathan P. Podsakoff3
Abstract
Despite the importance of establishing good, clear concept definitions in organizational research, the field
lacks a comprehensive source that explains how to effectively develop and articulate a concept’s domain.
Thus, the purpose of this article is to explain why clear conceptual definitions are essential for scientific
progress and provide a concrete set of steps that researchers can follow to improve their conceptual
definitions. First, we define what is meant by a concept, describe the functions served by concepts in
scientific endeavors, and identify problems associated with a lack of conceptual clarity. Then we explain
why it is so difficult to adequately define concepts. Next, we provide a series of recommendations for
scholars in the organizational, behavioral, and social sciences who are either trying to define a new
concept or revise the definition of one that already exists in the field. Following this, we provide some
examples that generally meet the criteria for a good conceptual definition. We conclude with a set of
questions that authors, reviewers, and editors can use as a guide for evaluating concept definitions.
Keywords
concept, construct, necessary and sufficient concept structure, family resemblance concept struc-
ture, discriminant validity, deficiency, contamination, criteria for good conceptual definitions
Concepts—scientific language—comprise the indispensable tools of science, and the irredu-

cible element in the process of scientific explanation. Precision in this fundamental area is
imperative for the production of significant theory. (DiRenzo, 1966, p. 3)
1
Warrington College of Business Administration, University of Florida, Gainesville, FL, USA
2
Department of Marketing, Kelley School of Business, Indiana University, Bloomington, IN, USA
3
Department of Management and Organizations, Eller College of Management, University of Arizona, Tucson, AZ, USA
Corresponding Author:
Philip M. Podsakoff, Warrington College of Business Administration, University of Florida, Gainesville, FL 32611, USA.
Email: [email protected]
160 Organizational Research Methods 19(2)
A good part of the work called ‘‘theorizing’’ is taken up with the clarification of concepts—and
rightly so. It is in this matter of clearly defined concepts that social science research is not
infrequently defective. (Merton, 1958, p. 114)
Concept formation lies at the heart of all social science endeavors. It is impossible to con-
duct work without using concepts. It is impossible even to conceptualize a topic, without
putting a label on it. Concepts are integral to every argument for they address the most
basic question of social science research: what are we talking about? (Gerring, 2012,
p. 112)
It is obvious from these opening quotes that social scientists have recognized the importance of
good conceptual definitions for quite some time. Indeed, discussions of the importance of provid-
ing good definitions of the concepts we use in the social sciences can be traced back to the work of
John Stuart Mill (1882), and interest in this topic has continued relatively undiminished to the
present day (DiRenzo, 1966; Gerring, 1999a, 1999b, 2012; Ghiselli, Campbell, & Zedeck,
1981; Goertz, 2006; Goertz & Mahoney, 2012; MacKenzie, 2003; MacKenzie, Podsakoff, & Pod-
sakoff, 2011; Podsakoff, Podsakoff, MacKenzie, & Klinger, 2013; Suddaby, 2010; Weber, 1949).
Unfortunately, it is also equally obvious that the problem of inadequate conceptual definitions
remains an issue for scholars in the organizational, behavioral, and social sciences. Indeed, Locke
(2003) has argued that ‘‘a major factor retarding progress in our field—and almost every intellec-
tual field today—has been the use of sloppy, careless, or subjective definitions’’ (p. 415). Similar
points have recently been made by Gerring (2012), Goertz (2006), Goertz and Mahoney (2012),
and MacKenzie et al. (2011), among others.
Not only has the problem of poor conceptual definitions been a longstanding one, but several
researchers have observed that this problem is also widespread. For example, Molloy, Chadwick,
Ployhart, and Golden (2011) found that only about 14% of the 93 studies reported between 2006 and
2009 that were designed to test the resource-based theory of the firm (RBT) identified the necessary
and sufficient conditions for a resource to be considered an intangible asset, and fewer than half
(46%) of the conceptual definitions in the RBT literature made an attempt to distinguish the focal
construct from related constructs. This is consistent with Suddaby (2010), who noted that ‘‘One
of the more commonly cited reasons for rejecting a manuscript at AMR is that reviewers feel the
submission lacks ‘construct clarity’’’ (p. 346), and also with Locke (2012), who took an even more
pessimistic view of the field when he concluded that ‘‘As someone who has been reviewing journal
articles for more than 30 years, I estimate that about 90% of the submissions I get suffer from prob-
lems of conceptual clarity’’ (p. 146).
In view of this, the purpose of this article is to explain why clear conceptual definitions are essen-
tial for scientific progress and provide a concrete set of steps that organizational and behavioral
scholars can follow to improve their conceptual definitions. First, we define what is meant by a con-
cept, describe the functions served by concepts in scientific endeavors, and identify the problems
associated with a lack of conceptual clarity. Then we explain why it is so difficult to adequately
define concepts. Next, we provide a series of recommendations that can serve as a guide for scholars
in the organizational, behavioral, and social sciences who are either trying to define a new concept or
revise the definition of one that already exists in the field. Finally, we provide some examples from
the literature that generally meet the criteria for having a good conceptual definition. As we will see,
this is not an easy task. However, it is important because, to our knowledge, there is no single paper
in the literature that addresses all of these issues despite the fact that this problem is both longstand-
ing and pervasive (cf. Locke, 2012; MacKenzie, 2003; MacKenzie et al. 2011; Molloy & Ployhart,
2012; Nunnally & Bernstein, 1994; Osigweh, 1995; Suddaby, 2010).
Podsakoff et al. 161
What Is a Concept?
Although the term concept has been used in a variety of different ways in different disciplines
(Adcock, 2005; Binning & Barrett, 1989; Cronbach & Meehl, 1955; DiRenzo, 1966; Goertz, 2006;
Locke, 2012; Riggs, 1975; Sartori, 1984; Sartori, Riggs, & Teune, 1975a, 1975b), for the purposes
of this paper, we define concepts as cognitive symbols (or abstract terms) that specify the features,
attributes, or characteristics of the phenomenon in the real or phenomenological world that they are
meant to represent and that distinguish them from other related phenomena. Thus, a concept is a
cognitive symbol that has meaning for the scientific community that uses it. Moreover, the fact that
it specifies the attributes or features that define it suggests that it is the combination of these features
that allows one to distinguish the concept from other, related concepts. This is consistent with Sartori
(1984), who argued that a concept is ‘‘The basic unit of thinking. It can be said that we have a concept
of A (or of A-ness) when we are able to distinguish A from whatever is not-A’’ (p. 74). Finally, the fact
that a concept is abstract in nature implies that it is generally not directly observable and that we can
only obtain indirect evidence of its relationships with other concepts through its operationalization.
Our definition is also consistent with Goertz (2006), who has argued that
The core attributes of a concept constitute a theory of the ontology of the phenomenon under
consideration. Concepts are about ontology. To develop a concept is more than providing a
definition: it is deciding what is important about an entity. The arguments about why attribute
X is important form part of the ontological theory of the object. (p. 27)
However, our view differs somewhat from Goertz in that he espoused an explicitly ‘‘realist’’ view of
concepts, whereas we are open to both realist and constructivist perspectives for two reasons. First,
we agree with Bagozzi’s (2007) observation that a strict realist conception ‘‘becomes less tenable
when we consider such abstract or theoretical concepts as attitudes, attributions, beliefs, desires,
emotions, goals, intentions, motivation, and personality traits’’ because it has not been possible and
it is unclear whether it will ever be possible ‘‘to relate such mental events and states closely with
physical processes in the brain or to make claims about how the mental events or states function
in a physical sense’’ (p. 230). Second, we agree with researchers (Blalock, 1968; Bledow & Frese,
2009; Bollen & Bauldry, 2011; Bollen & Lennox, 1991; Diamantopoulos & Papadopoulos, 2010;
Diamantopoulos, Riefler, & Roth, 2008; Judge & Kammeyer-Mueller, 2012; Law & Wong,
1999; Law, Wong, & Mobley, 1998; MacKenzie et al., 2011; Nye, Brummel, & Drasgow, 2014;
Williams, Vandenberg, & Edwards, 2009) who have noted that for some multidimensional concepts,
a more constructivist conception may be more appropriate (at least with respect to the ontological
relations between the subdimensions and the more general/abstract concept).
Defined in this manner, concepts are similar to hypothetical constructs, which are viewed as the-
oretical entities. However, neither of these terms should be confused with latent constructs that are
empirical entities designed to represent a hypothetical construct or concept. Our distinction is sim-
ilar to that made by Osigweh (1989), who has distinguished between theoretical concepts and
empirical or observational concepts. Equating latent constructs with the hypothetical constructs they
are intended to represent is what A. Kaplan (1964) has referred to as reification: ‘‘Reification is more
than a metaphysical sin, it is a logical one. It is the mistake of treating a notational device as though it
were a substantive term, what I have called a construct as though it were observational’’ (p. 61).
Therefore, in order to avoid confusion, when we refer to the term construct in the paper, we will
mean it to be a hypothetical variable or concept, unless otherwise indicated.
Note that our definition not only includes concepts that take a necessary and sufficient concept
structure (e.g., Sartori, 1970, 1984) but also those that take a family resemblance concept structure
(e.g., Goertz, 2006; Wittgenstein, 1953). The roots of the necessary and sufficient concept structure
date back to Aristotle, and it is considered the standard approach for defining and conceptualizing
phenomena (Goertz, 2006). According to Komatsu (1992), ‘‘The distinguishing assumption of the
classical view is that concepts are defined by sets of individually necessary and collectively suffi-
cient attributes’’ (p. 502). He further notes that this assumption implies that (a) there is no ambiguity
about whether an object is a member of the conceptual category because if it possesses the necessary
and sufficient attributes it is a member of the category and if it doesn’t it is not and (b) objects are
either a member of the category or they are not; it is not a matter of degree.
In contrast to this classical view, another way to define a concept is by means of a family resem-
blance structure. Wittgenstein (1953) is credited with calling attention to the fact that some concepts
fit better into a family resemblance concept structure than into a necessary and sufficient concept
structure. He presents the concept of ‘‘games’’ as an example:
Consider for example the proceedings that we call ‘‘games.’’ I mean board-games, card-games,
ball-games, Olympic games, and so on. What is common to them all?—Don’t say: ‘‘There must
be something common, or they would not be called ‘games’’’—but look and see whether there is
anything common to all. For if you look at them you will not see something that is common to
all, but similarities, relationships, and a whole series of them at that. To repeat: don’t think, but
look!—Look for example at board-games, with their multifarious relationships. Now pass to
card-games; here you find many correspondences with the first group, but many common fea-
tures drop out, and others appear. When we pass next to ball-games, much that is common is
retained, but much is lost.—Are they all ‘‘amusing’’? Compare chess with noughts and crosses.
Or is there always winning and losing, or competition between players? Think of patience. In
ball-games there is winning and losing; but when a child throws his ball at the wall and catches
it again, this feature has disappeared. Look at the parts played by skill and luck; and at the dif-
ference between skill in chess and skill in tennis. Think now of games like ring-a-ring-a-roses;
here is the element of amusement, but how many other characteristic features have disappeared!
And we can go through the many, many other groups of games in the same way; can see how
similarities crop up and disappear . . . . And the result of this examination is: we see a compli-
cated network of similarities overlapping and criss-crossing: sometimes overall similarities,
sometimes similarities of detail. I can think of no better expression to characterize these simila-
rities than ‘‘family resemblances’’; for the various resemblances between members of a family:
build, features, colour of eyes, gait, temperament, etc., etc. overlap and criss-cross in the same
way.—And I shall say: ‘‘games’’ form a family. (pp. 31-32)
Thus, according to Wittgenstein, some concept categories are defined by their family resemblance,
rather than by a specific set of necessary and sufficient attributes. According to this view,
A family resemblance relationship consists of a set of items of the form AB, BC, CD, DE. That
is, each item has at least one, and probably several, elements in common with one or more
other items, but no, or few, elements are common to all items. (Rosch & Mervis, 1975, p. 575)
In other words, an object’s membership in a category is determined by the extent to which it has
attributes in common with other objects in the category. Importantly, as noted by Komatsu
(1992): (a) Every attribute must be shared by more than one member of the category, (b) attributes
shared by many members are more central than attributes shared by only a few members, and (c)
category members that share many attributes with other members bear greater family resemblance
to the category (i.e., are more prototypical of it) than members sharing only a few attributes.
Table 1 summarizes some of the key differences between these two types of concept structures.
First, these two concept structures differ in terms of their underlying logical and structural rules.
Table 1. Differences Between Necessary and Sufficient and Family Resemblance Concept Structures.
Necessary and Sufficient Concepts Family Resemblance Concepts
Logical rule AND OR

Structural rule If and only if all n attributes are present If at least m of n attributes are present
Attribute substitutability No Yes
Example attribute matrix
Relation between the Intention is negatively related to extension; as the number of Intention can be positively related to extension; as the number of
intension of the defining attributes increases, the number of qualifying cases defining attributes increases, the number of qualifying cases may
concept and its decreases. increase.
extension
163
Necessary and sufficient concepts are defined in terms of a logical conjunction (AND) of attributes
and follow the structural rule that an object qualifies for membership if and only if all n attributes are
present. In contrast, family resemblance concepts are defined in terms of a logical disjunction (OR)
of common attributes shared by at least two objects in the category and follow the structural rule that
an object qualifies for membership if at least m of n attributes are present. Thus, attribute substitut-
ability is not permitted in the necessary and sufficient concept structure but is permitted in the family
resemblance concept structure. This is illustrated in the attribute matrices shown in Table 1. As
shown in the necessary and sufficient attribute matrix, qualifying cases must contain all of the nec-
essary and jointly sufficient attributes (e.g., Cases 1-4). In this example, Cases 5 and 6 fail to qualify
because they do not have all of the necessary attributes. Indeed, Case 6 fails to qualify even though it
only lacks a single necessary attribute.
In contrast, in the family resemblance attribute matrix, qualifying cases can substitute one shared
attribute for another as long as they meet the m of n criterion. In our hypothetical example, we
required a case to possess three of eight attributes in order to qualify. Based on this criterion, Cases
1 through 4 qualify, and Cases 5 and 6 do not qualify. Notice that in Cases 1 through 4, no single
attribute is shared by all cases, but each attribute is shared by at least one other qualifying case.
In addition, in the family resemblance attribute matrix, Case 3 is the most prototypical one because
it shares the most attributes with other members of the category, and attributes 2 and 6 are the most
central because they are shared by the most category members. A final difference between these two
concept structures concerns the relationship between the attributes used to define a concept (its
intension) and the number of cases to which the concept applies (its extension). For necessary and
sufficient concept structures, Sartori (1984) has noted that the number of cases to which a concept
applies decreases as the number of necessary attributes increases (i.e., extension is negatively related
to intension). However, Goertz (2006) has noted that this may not be true for family resemblance
concepts since increasing the number of attributes can sometimes increase the number of cases to
which a concept applies.
Goertz (2006) has noted that Hicks, Misra, and Ng’s (1995) concept of a welfare state is a good
example of a concept having a family resemblance structure with multiple sufficient attributes.
According to Hicks et al., a country can be considered a welfare state as long as it adopts any three
of the following four classic welfare programs: (a) old age pensions, (b) health insurance, (c) work-
man’s compensation, and (d) unemployment compensation. Thus, in this conceptualization, there
are no necessary attributes because any combination of three of the four attributes (i.e., welfare pro-
grams) is sufficient for a country to qualify as a welfare state.
Although we are not aware of any concepts in the management literature that have been explicitly
conceptualized as having a family resemblance structure, we believe that the concept of High
Performance Human Resource Practices (HPHRP) appears to fit this type of structure (cf. Delery
& Doty, 1996; Huselid, 1995; Kehoe & Wright, 2013). For example, Kehoe and Wright (2013) mea-
sured perceptions of HPHRP by asking respondents to indicate which of 15 HR practices were used
in their organization, including job applicants taking formal tests; job applicants participating in
structured interviews; employees having a fair complaint process; employees earning group bonuses
for productivity, performance, or other group performance outcomes; employees earning individual
bonuses (or commissions) for productivity, performance, or other individual performance outcomes;
employees receiving a formal evaluation of their performance at least once per year; employees reg-
ularly receiving formal communication regarding company goals and objectives; and the company
making a change in how work is completed based on the suggestion(s) of employees in the last four
months. Following this, Kehoe and Wright (2013) report that, ‘‘We created an additive index of HR
practice items, resulting in a continuous measure reflecting each employee’s perceptions that partic-
ular practices from a high-performance HR system were used to manage employees in his or her job
group’’ (p. 375). As justification for this procedure, the authors cited Youndt, Snell, Dean, and Lepak
(1996) who argued that,
Such an additive approach to combining HR practices into an index suggests that . . . perfor-
mance [can be improved] either by increasing the number of practices . . . employ[ed] within
the system or by using the practices in an HR system in a more comprehensive and widespread
manner. This approach is conceptually and empirically better than a multiplicative approach to
creating HR systems because it does not reduce the index value to zero if a single HR practice
is absent from a system. Instead, the absence of a practice only weakens the net effect of the
system (MacDuffie, 1995; Osterman, 1994). (p. 849)
Others (MacDuffie, 1995; Youndt et al., 1996) have used a similar index to represent the HPHRP
concept. This scoring procedure reveals that no individual HR practice is necessary for a firm to be
identified as having high performance human resource practices, and the HR practices are conceptua-
lized as interchangeable substitutes for each other. Of course, one may argue that the HPHRP concept
does not explicitly follow the m of n attributes criterion typical of family resemblance concepts.
Nevertheless, since none of the attributes identified are necessary requirements of a high performance
HR system and all of them are substitutable for each other, this concept appears to follow a family
resemblance concept structure in that at least one (m) of the 15 HR practices (n) is required for an orga-
nization to be classified as having a high performance HR practice system; and the greater the number
of practices that an organization possesses, the more prototypical it is of a HPHRP organization.
The Functions of Theoretical Concepts in Scientific Endeavors

Theoretical concepts serve several critical functions in the scientific enterprise. One of the primary
functions of scientific concepts is that they help organize the phenomena in the world around us into
meaningful categories and describe their attributes. A good conceptual definition should identify the
set of fundamental characteristics or key attributes that are common (and potentially unique) to the
phenomenon of interest. This precision not only clarifies the intension or meaning of the concept
(Sartori, 1984) but also prevents the same concept from being used to refer to different phenomena.
For example, based on their meta-analysis, Hulleman, Schrager, Bodmann, and Harackiewicz
(2010) concluded that achievement goal researchers have used the same label for conceptually dif-
ferent constructs. Similarly, in their review of the literature, Klein, Molloy, and Cooper (2009) note
that organizational commitment has been defined by some as an attitude, by others as a behavior, by
others as identification with the organization, and by still others as value congruence with the orga-
nization. Obviously, these phenomena are not all the same and would be measured in very different
ways, thus emphasizing the importance of clearly articulating the fundamental attributes of a con-
cept. Another function of theoretical concepts is that they provide organizational and behavioral
scientists with a common language to communicate their ideas to each other. Clear conceptual def-
initions not only identify the nature of the concept of interest, but they also help distinguish the focal
concept from other, seemingly similar concepts in the field. Finally, one of the most important func-
tions of theoretical concepts in science is that they serve as the essential building blocks of theory
(Bacharach, 1989; Dubin, 1969; Goertz, 2006; Klein & Zedeck, 2004; Mill, 1882; Shoemaker, Tan-
kard, & Larosa, 2004; Zaltman, Pinson, & Angelmar, 1973). Indeed, as noted by DiRenzo (1966),
Conceptual definition and theory formulation go hand in hand as necessary steps in one uni-
fied process of scientific research. The analysis of concepts is but one phase—a fundamental
requisite—of that complex process of scientific inquiry which culminates in theory. Concepts,
thus, are irreducible elements of theory or theoretical systems . . . . The more precise and
refined the conceptual elements, the more precise and refined the theory. (p. 6)
In summary, concepts serve as the fundamental building blocks of theory, allowing us to organize
complex phenomena with a common language that, when done well, facilitates communication
between researchers.
Problems With the Lack of Conceptual Clarity

A lack of conceptual clarity causes a number of problems—both at the conceptual and the opera-
tional levels. First, at the conceptual level, a lack of clarity often makes it difficult to distinguish the
focal concept from other similar concepts in the field, thus undermining discriminant validity. For
example, Netemeyer, Bearden, and Sharma (2003) have noted that,
In defining constructs, care must be taken as to what is included in the domain of the construct
and what is excluded from this domain, as sources of invalidity can have their origins in the
construct definition process . . . . If the domain is too broadly defined, extraneous factors or
domains of constructs other than the target construct may be included. (p. 89)
Indeed, this is why a number of authors (Churchill, 1979; MacKenzie et al., 2011; Nunnally &
Bernstein, 1994; Schwab, 1980) have argued that an essential part of the construct validation process
is distinguishing the focal construct from other related constructs. Second, as noted by several
authors (Le, Schmidt, Harter, & Lauver, 2010; Morrow, 1983; Singh, 1991; Tepper & Henle,
2011), a lack of conceptual clarity may also lead to concept ‘‘proliferation’’ (i.e., the development
of concepts with different names but overlapping conceptual domains). This is important because
several scholars (Aquino & Thau, 2009; Blalock, 1968; Le et al., 2010; Morrow, 1983; Popper,
2002; Singh, 1991; Tepper & Henle, 2011) have noted that one of the main factors that undermines
discriminant validity and inhibits conceptual progress in science is the proliferation of different
terms for the same concept. For example, Singh (1991) has argued that when concepts are not ade-
quately distinguished from each other, it is likely to: (a) lead to confusion among researchers
because some of them see the concepts as distinctive entities and build streams of research around
one of them, whereas other scientists treat them interchangeably and (b) raise questions about our
cumulative understanding of the antecedents, correlates, and consequences of the concept (or con-
cepts) of interest. Thus, it probably should not be surprising that one of the potential criticisms that
organizational and behavioral scientists encounter when they develop and propose a new concept is
that it represents nothing more than ‘‘old wine in new bottles’’ (Johnson, 1999; Kristjansson, 2012;
Spell, 2001) or that the concept is redundant with concepts that already exist in the field (Blalock,
1968; Schwab, 1980; Singh, 1991; Tesser & Krauss, 1976).
In addition, a lack of conceptual clarity can lead to difficulty in specifying and testing the nomo-
logical network of the concept, thus undermining nomological validity. If one does not have a clear
idea of what the concept means, it is difficult to identify related concepts or to specify whether they
are antecedents, consequences, or correlates of the focal concept. This is also likely to inhibit prog-
ress in the field (Aquino & Thau, 2009; Popper, 2002; Sober, 1981; Tepper & Henle, 2011) by lead-
ing to: (a) the acceptance (or rejection) of theoretical statements that are not justified (MacKenzie,
2003), (b) unwarranted modification of theories regarding the relationships between a focal concept
and other concepts in its nomological network, and/or (c) decreased confidence in the findings
obtained from research (Schwab, 1980).
At the operational level, a lack of conceptual clarity increases the likelihood of a mismatch
between the concept and measures or manipulations of it (Adcock & Collier, 2011; Goertz,
2006), thus undermining construct validity. This is consistent with Ghiselli et al. (1981), who noted
that a clear conceptual definition
directly suggests the kinds of operations that should be employed in obtaining a categorization
of individuals in terms of that property. A definition that is vague conveys only in a general
way information about the nature of the variable and is of little value in providing guides for
the development of appropriate operations. (p. 16)
Beyond this, a lack of conceptual clarity also increases the likelihood that operationalizations
of the concept (either by measures or manipulations) will be deficient and/or contaminated
(MacKenzie, 2003; MacKenzie et al., 2011; Nunnally & Bernstein, 1994; Schwab, 1980). Although
there are a number of reasons why the operationalization of a construct may be either deficient or
contaminated (e.g., poor item generation, poor selection of indicators for its measurement, focusing
on only some aspects of the conceptual definition and ignoring others, poor choice of a manipula-
tion, etc.), one of the main causes of these measurement problems is a poor definition of the concept
in the first place. For example, one reason why a measure/manipulation may be contaminated is
because the definition of the theoretical concept is ambiguous, too broadly defined, or lacks ade-
quate precision. Similarly, a measure/manipulation may be deficient because the definition of the
focal concept fails to clearly articulate all of its essential properties. The consequence of these prob-
lems is that the measures/manipulations of the focal concept may contain greater levels of systematic
and/or random measurement error (Viswanathan, 2005). Thus, it is probably fair to say that research-
ers go a long way toward solving these potential problems with their operationalizations to the extent
that they provide clear and concise definitions of the concepts they are trying to measure/manipulate
in the first place.
What Makes Defining Concepts So Difficult?

There are several possible reasons why it is difficult to develop good conceptual definitions. The
first problem is the difficulty in selecting the attributes or characteristics that define a concept. Mill
(1882) recognized this fundamental problem over a century ago when he noted that
To define, is to select from among all the properties of a thing, those which shall be understood
to be designated and declared by its name; and the properties must be well known to us before
we can be competent to determine which of them are fittest to be chosen for this purpose. (p. 8)
Unfortunately, it is not easy to specify the attributes or characteristics that capture the ‘‘essence’’ of a
concept, and developing good conceptual definitions requires a substantial amount of cognitive
effort and disciplined thinking on the part of the researcher. Indeed, as lamented by Nunnally and
Bernstein (1994), ‘‘no precise method can be stated to outline the domain of variables for a construct
properly . . . the theorizing process is necessarily intuitive’’ (p. 88). However, even though this may
be an intuitive process, we believe that there are ways to structure and guide this theorizing, which
we will discuss in later sections of the paper.
The second problem is that most scholars are not taught how to develop conceptual definitions as
part of their doctoral training. Thus, coming up with a good definition of a theoretical concept is not
a part of their research repertoire. Indeed, as noted by Goertz (2006), this problem is not restricted to
researchers in the organizational and behavioral sciences but is more broadly applicable to the social
sciences in general:
In spite of the primordial importance of concepts, they have received relatively little attention
over the years in the social sciences . . . . Yet, the contrast with the massive literature on quan-
titative measures, indicators, scales, and the like cannot be more extreme. Hence we have a
paradox: as Mill noted, concepts are a central part of our theories, yet researchers . . . have
focused very little attention on social science concepts per se. (pp. 1-2)
Goertz and Mahoney (2012) have argued that another reason for the paradox described previously is
that whereas qualitative scholars often focus their efforts and energy on developing concepts that are
nonmathematical and substantive in nature, quantitative scholars have generally focused the major-
ity of their attention on developing scales with good psychometric properties (e.g., high reliability,
good factor structures, etc.). More specifically, these authors note that
Quantitative and qualitative scholars differ systematically and often dramatically in their
approaches to concepts and measurement . . . . Some of these differences are not particularly
surprising. Qualitative scholars have long, involved, ‘‘wordy’’ discussions about the meaning
of concepts. In this respect, they resemble (political) philosophers, who also spend much time
on concept analysis. By contrast, quantitative scholars need data for their statistical models.
Accordingly, they focus attention on the nature and quality of quantitative measures. They
spend less time on the concept and more time on operationalization, aggregation and resulting
datasets. (Goertz & Mahoney, 2013, p. 205)
Indeed, a similar point was made by DiRenzo (1966) nearly 50 years ago. Goertz (2006) further
notes that the only way to improve this situation and close the gap between qualitative and quanti-
tative researchers is to focus first on the development of the definition of concepts before focusing
on their measurement.
Of course, this divide is not only limited to the political science arena. It also exists in other dis-
ciplines where the majority of research has a decided quantitative focus. For example, Bagozzi
(1984) has argued that a major impediment to theory development in the field of marketing is that
Substantively oriented researchers emphasize conceptual aspects of their research but some-
times place relatively less stress on methods and measures. Methodologically oriented
researchers tend to focus relatively more on techniques than on theory development. At the
same time, both camps spend more of their energies reading literatures and interacting with
others with similar leanings. This tends to perpetuate the split. (p. 27)
Related to the aforementioned point, it is probably fair to say that traditional research in the orga-
nizational and behavioral sciences tends to emphasize statistical conclusion and internal validity
over construct validity. For example, in their classic book on research methods, Cook and Campbell
(1979) focused first on issues of whether ‘‘conclusions about covariation are made on the basis of
statistical evidence’’ (p. 37; i.e., statistical conclusion validity) and then on the question of ‘‘whether
there is a causal relationship between the two [variables]’’ (p. 50; i.e., internal validity) before they
ask whether there is ‘‘the possibility that the operations that are meant to represent a particular cause
or effect construct can be construed in terms of more than one construct’’ (p. 59; i.e., construct valid-
ity). This emphasis may have arisen in part out of their focus on experimental research (cf. Shadish,
Cook, & Campbell, 2002, p. 98) and in part from their focus on analytical techniques that are incap-
able of simultaneously estimating both measurement and structural relationships. However, regard-
less of the reason, this set of priorities seems to have been tacitly adopted in the literature.
Figure 1. Summary of stages for developing good conceptual definitions.
Recommendations for Developing Good Conceptual Definitions

In this section, we provide a set of guidelines that we believe will help researchers to develop
better conceptual definitions and that should also benefit editors and others who are responsible
for assessing the merits of conceptual definitions proposed in the papers they review. Generally
speaking, these guidelines can be organized into the following four stages shown in Figure 1:
(a) identify potential attributes of the concept by collecting a representative set of definitions,
(b) organize the potential attributes by theme and identify any necessary and sufficient or
shared ones, (c) develop a preliminary definition of the concept, and (d) refine the conceptual
definition.
However, before describing each of these stages, it is important to keep in mind three caveats
regarding the guidelines summarized in Figure 1. First, although we describe the stages as being dis-
crete and sequential, they actually overlap to some extent, and the concept definition process is itera-
tive in nature. Second, in developing these guidelines, we have tried to strike a balance between
depth of treatment and breadth of coverage. Undoubtedly, there are other worthwhile techniques that
could be utilized during some of the steps in the process shown in Figure 1 that we are not aware of
or that we chose to exclude. However, we felt that describing every possible technique that could
potentially be useful at each stage would undermine our primary goal of outlining an organized set
of activities that set priorities for what needs to be done in order to develop a strong conceptual def-
inition. Finally, it is important to bear in mind that there may be practical limitations that prevent
researchers from implementing all of the recommendations discussed in our guidelines. Neverthe-
less, because of the critical role that concept definitions play in the development of valid theories
and operationalizations of concepts, we strongly encourage researchers in the field to use as many
of the techniques that we describe as is necessary to develop clear and concise definitions of their
theoretical concepts.
Stage 1: Identify Potential Attributes of the Concept and/or Collect a Representative

Set of Definitions
The goal of the initial phase of the concept definition process is to identify the potential attributes of
the focal concept. There are a variety of activities that may be helpful in achieving this goal, includ-
ing: searching the dictionary for synonyms or antonyms, surveying the literature, interviewing
subject-matter experts/practitioners, using focus groups, conducting case studies, comparing the
focal concept with its opposite pole, and thinking about how the concept might be operationalized.
Each of these activities will be discussed in the remainder of this section. The key point to remember
is that the objective of this stage of the concept development process is to cast a wide net and not to
try to specify the defining properties of the concept prematurely.
Search the Dictionary. Generally speaking, the process of developing the definition of a new concept is
an inductive one. One place to start might be to examine the definition of the focal concept in a dic-
tionary (Jaccard & Jacoby, 2010; Locke, 2012). There are many encyclopedias and dictionaries avail-
able for this purpose (some of which focus on specific domains within the organizational and social
sciences). Although Jaccard and Jacoby (2010) have noted that dictionary definitions often differ from
scientifically oriented definitions, they also point out that the dictionary can prove useful in specifying
the definition of the focal concept with greater clarity and insight. In addition, another useful purpose
for the dictionary is to look at antonyms of the focal concept because identifying the opposite meaning
of the concept may help clarify its underlying attributes as well as distinguish it from other concepts.
Survey the Literature. Another thing that it is helpful to do if the concept has been previously defined
by others is to survey the literature to identify the various ways in which the concept has been
defined previously and the attributes or characteristics that other researchers consider critical to its
definition. This search can prove helpful by providing the researcher with a sense of whether pre-
vious definitions contain the true essence of the concept or fail to do so because they either miss one
or more of its key attributes or include attributes that they should not. It is important during this
phase of the process to (a) keep track of the different definitions that have been used in the literature
and (b) the frequency with which these different definitions have been reported. Gerring (2012) has
noted that this helps in two ways. First, it helps identify what other scholars feel are the defining
characteristics of the concept. Second, it helps the researcher identify when he or she has conducted
an adequate survey of the literature (i.e., when the researcher reaches the point where the definitions
and usages of the concept become highly redundant, it may serve as a signal that the researcher can
terminate his or her search). The literature search is also helpful because it provides the researcher
with critical information about those concepts from which the focal concept should be distinguished.
Maynes and Podsakoff (2014) provide a good example of a table that summarizes the attributes of
several concepts identified in a literature review. The primary objective of their study was ‘‘to
develop a new, more expansive voice behavior framework; one that includes behaviors that either
challenge or support the status quo as well as behaviors that possess either positive or negative attri-
butes’’ (Maynes & Podsakoff, 2014, p. 88). To begin this process, they first reviewed the literature in
order to identify the different types of voice behavior along with their fundamental attributes. The
results of that search are reported in Table 2. This table lists: (a) the authors associated with (b) dif-
ferent conceptualizations of voice behavior and (c) the key attributes that distinguish each of these
types of voice. For example, the table indicates that among the key attributes of Hirschman’s (1970)
classic definition of voice behavior is that it challenges the status quo; it is conspicuous, visible, and
direct; and that it can have either a prohibitive or a promotive focus. Similarly, Maynes and Podsak-
off’s (2014) summary of the key attributes of Gorden’s (1988) definition of active/constructive voice
behavior is that it promotes improvement and either preserves or challenges the status quo. From this
Table 2. Summary of Attributes of Employee Voice Conceptualizations.
Author(s) Conceptualization of Voice Behavior Key Attributes
Hirschman (1970) Voice behavior

‘‘Voice is . . . [an attempt] to change, rather than to Challenges the status quo
escape from, an objectionable state of affairs.’’ (p. 30)
‘‘Voice is . . . [a] ‘messy’ concept because it can be Conspicuous/visible
graduated, all the way from faint grumbling to violent Direct
protest; it implies articulation of one’s critical opinions
rather than a private, ‘secret’ vote in the anonymity of a
supermarket; and finally, it is direct and straightforward
rather than roundabout.’’ (p. 16)
‘‘Voice is like exit in that it can be overdone: The Critical about
discontented . . . members could become so harassing organization
that their protests would . . . hinder rather than help.’’ Prohibitive or promotive
(p. 31) focus
Gorden (1988) Active/constructive voice
Expressions that actively reaffirm or correct the activities Promotes improvement
of the organization.
Behaviors include making suggestions, bolsterism Preserves or challenges
(supporting the status quo), and principled dissent. the status quo
(p. 285)
Active/destructive voice
Expressions that are actively antagonistic toward the Oriented toward doing
organization and its activities. harm
Behaviors include verbal aggression, bad-mouthing, and Challenges the status quo
antagonistic exit. (p. 285)
Passive/constructive voice
Behaviors indicating acceptance of and submission to Preserves the status quo
organizational activities.
Behaviors include nonverbal support, compliance, and
cooperation. (p. 258)
Passive/destructive voice
Behaviors that indicate resigned conformance to the Preserves the status quo
activities of the organization.
Van Dyne and Voice behavior
LePine (1998) ‘‘promotive behavior that emphasizes the expression of Promotes improvement
constructive challenge intended to improve rather than Challenges the status quo
merely criticize’’ (p. 109) Positive intentions
Van Dyne, Ang, and Prosocial voice
Botero (2003) ‘‘non-required behavior that emphasizes the expression Active, discretionary
of change-oriented comments.’’ behavior
(p. 1370) Challenges the status quo
The expression of ‘‘work related ideas, information, or Intentional
opinions based on cooperative motives. Thus, this . . . Altruistic motive
behavior is intentional, proactive, and other-oriented.’’ (p.
1371)
Behaviors include suggestions for change and Promotes improvement
recommendations for improvements.
Defensive voice
The expression of ‘‘work-related ideas, information or Self-protective
opinions—based on fear—with the goal of protecting the
self.’’ (p. 1372)
(continued)
Table 2. (continued)
Author(s) Conceptualization of Voice Behavior Key Attributes
Behaviors include expressing ideas that shift attention Motivated by fear

elsewhere based on fear and proposing ideas that focus on
others to protect the self.
Acquiescent voice
The ‘‘expression of work-related ideas, information, or Preserves/promotes the
opinions—based on feelings of resignation. Acquiescent status quo
voice is disengaged behavior that is based on feeling Motivated by resignation
unable to make a difference. Thus, it results in expressions
of agreement and support based on low self-efficacy to
affect any meaningful change.’’ (p. 1373)
Behaviors include expressions of support and agreement.
Liang, Farh, and Promotive voice
Farh (2012) ‘‘employees’ expression of new ideas or suggestions for Challenges the status quo
improving the overall functioning of their work unit or
organization . . . . Such voice is ‘promotive’ in the sense
that it is focused on a future ideal state.’’ (p. 74)
Behaviors include proposing ways to change the status Promotes change/
quo and offering suggestions for improvement. improvement
Prohibitive voice
‘‘Prohibitive voice describes employees’ expressions of Prevents harm
concern about work practices, incidents, or employee
behavior that are harmful to their organization.’’ (p. 75)
Behaviors include expressing concern over current work Preserves or challenges
practices that are problematic and expressions focused on the status quo
preventing problematic initiatives from being
implemented.
Source: Adapted from Maynes, T. D., and Podsakoff, P. M. (2014). Speaking more broadly: An examination of the nature, ante-
cedents, and consequences of an expanded set of employee voice behaviors. Journal of Applied Psychology, 99, 87-112. Copy-
right 2013 by the American Psychological Association. Reprinted with permission.
table, they developed conceptual definitions for several different types of voice behavior that we will
elaborate on in Stage 2 of the concept definition process.
One final point worth making is that although the process of conducting a survey of the literature
may go fairly smoothly in many cases, on some occasions the search may be complicated because
researchers find that some concepts with different names essentially capture the same conceptual
domain space (i.e., there is concept proliferation). As we noted earlier, the problem with construct pro-
liferation is that it may obscure the pattern of findings in the literature, result in the development of
multiple or conflicting measures of the concept, and impede theoretical progress. Therefore, when
researchers have reason to believe that concept proliferation may be a problem, we would recommend
that they consider the following. First, it is particularly important in the early stages of a literature
review for researchers to cast a wide net so that they can identify existing definitions of the focal con-
cept as well those that are related to it. For help in the process of identifying related concepts, Short,
Broberg, Cogliser, and Brigham (2010) note that source books such as Rodale’s (1978) The Synonym
Finder can be used to develop a list of synonyms to potentially help clarify the domain of a concept (or
multiple discrete word lists for each of the concept’s dimensions, if it is conceptualized as multidimen-
sional in nature). Beyond this, we have found that it is useful to refer to the keywords provided by
authors that are reported in many of the organizational and behavioral journals (typically following
the abstract of the article). Unfortunately, not all journals report this information. However, the Web
of Science provides a service that it calls ‘‘Keywords Plus’’ for some of the journals in the field. These
keywords can be found on the page that reports the article-specific information (title of the paper,
author information, article citation, abstract, etc.). Because these keywords appear to be derived from
an algorithm connecting article content with related terms, they are often more comprehensive than
those provided by the authors, and they may be particularly helpful to researchers attempting to iden-
tify concepts that are related to the one that is the focus of their research. Once these keywords have
been identified, researchers can conduct additional searches of articles identified by the keywords to
determine the extent to which these other concepts overlap with the concept of interest in terms of their
definitions and fundamental attributes. In those cases where researchers find that their focal concept is
different from a related concept (i.e., because these concepts each possess some unique, non-
overlapping attributes), this information can be used to help distinguish the focal concept from other
concepts identified in the search. In contrast, in those cases where it is difficult to discriminate the
existing definitions of the focal concept from a related concept (i.e., because the focal concept shares
all of the same attributes of the related concept, plus a few more), then researchers can use this infor-
mation in order to explain how their new or revised definition of the focal concept subsumes or incor-
porates other related concepts that already exist in the field.
Interview Subject-Matters Experts, Colleagues, and/or Practitioners. In addition, it may prove worthwhile
to consult with subject-matter experts, colleagues, and/or practitioners to help identify the defini-
tions (and attributes) of the focal concept. We believe that this is a particularly effective, but often
underutilized, technique (see Merton, Fiske, & Kendall, 1990, for helpful suggestions on how to
conduct a focused interview). The use of this technique is illustrated in Table 3, which is adapted
from a paper by Sonenshein, DeCelles, and Dutton (2014). Based on data obtained from 29 inter-
views with self-identified climate change supporters, Sonenshein et al. (2014) identified seven dif-
ferent ‘‘support challenges [that] are real or imagined difficulties that issue supporters encounter,
arising from a wide variety of contexts, including at and outside of work’’ (p. 14). Table 3 shows
several representative quotations drawn from these interviews that were used to inductively identify
the different types of support challenges. This is a good example of how interviews can be used to
identify key themes or attributes of a phenomenon.
Focus Groups and Direct (Structured) Observation. Two other techniques that may prove worthwhile in
this stage of the concept development process are focus groups (Challagalla, Venkatesh, & Kohli,
2009; Tuli, Kohli, & Bharadwaj, 2007) and direct (structured) observation (Martinko & Gardner,
1985; Mintzberg, 1975). The first of these techniques (focus groups) was used by Challagalla et al.
(2009) in their attempts to identify the various types of proactive post-sales service activities that sup-
pliers can use after a sale has been completed. Based on nine different focus group sessions with 94 man-
agers in both business-to-business and business-to-customer settings, these authors were able to identify
the key attributes and define three different dimensions of proactive post-sales service: (a) proactive pre-
vention, (b) proactive education, and (c) proactive feedback seeking. Table 4 describes the differences
between these forms of behavior in terms of their defining characteristics, provides illustrative examples
of each, and identifies related concepts previously mentioned in the literature. This table provides a good
example of how to summarize the results of a series of focus group interviews.
The second technique (direct observation) was used by Mintzberg (1975) in his oft-cited study on
the nature of managerial work. Mintzberg was interested in addressing a seemingly simple question—
‘‘What do managers do?’’ Although he was intimately familiar with the literature on leadership and
management that preceded his work, Mintzberg felt that much of this earlier literature on management
was simply wrong because it was not based on sound, empirical evidence. So, the purpose of his
research was to gather empirical data from managers to try to discover the various roles they per-
formed. He used five executives as the sample for his study and employed structured observation
Table 3. Summary of Various Types of Issue Support Challenges Identified by Sonenshein, DeCelles, and Dut-
ton (2014).
Second-Order First-Order
Themes Categories Exemplary Quotations
Relational Complacency of Patty: ‘‘It didn’t give me a lot of hope as far as middle America is
challenges others concerned. Frankly the people that I was working with were just as
middle America as you can get. I mean they got all their news from Fox.
They drove their giant pick-up trucks and complained about the price of
gas. To them it was like a white male world . . . . Trying to explain things
like why you should care about the earth is really difficult to people like
this.’’
Veronica: ‘‘Getting people to do something, so anything, like just do tiny
little things like re-replace their light bulbs and just, you know, drive less
and stuff like that.’’
Skepticism of Roger: ‘‘I think I’ve been trying to bring it up with people who aren’t
others already the converted, so to speak, and trying to open up conversations
about it, and the challenge is to take either people who are completely
apathetic or who actively disagree or who have framed it into a box to
get them more, because they are very entrenched in their . . . whoever
it is, it is kind of a dogma. It’s a religious belief almost. It’s either you are
or you aren’t, so switching people’s perspective on it is pretty hard.’’
Ashley: ‘‘A number of people that are still skeptical about it even though
that’s changing, that’s decreasing, but you always run into people that
are like, ‘Oh, you know, I don’t believe it’s true.’ You know, and to that I
say, well, you know, even if they don’t believe it’s true, they could still
do it just to save money for the company, you know, but then I guess
the other thing is it requires investment and I guess just making people
comfortable with that level of investment.’’
Cognitive Scope of issue Doug: ‘‘There’s the make-or-break issue of our generation and it’s the
challenges equivalent . . . . It’s equivalent to World War II as for our grandparents’
generation, so this will be the, this will be a defining moment in history in
humanity and relations to each other and especially the planet. You
know, if we don’t solve it in the next 20 to 30 years, then generations
will be impacted, will be detrimental . . . . It’s one of those preeminent
humanity issues that needs to be vastly understood by mass populous in
order to do the type of radical changes necessary, and it is a huge
challenge to get to that point where folks are willing to make sort of a
sacrifices and life-style choices and sacrifices in consumption patterns
necessary to get us there.’’
Lucy: ‘‘Well it is such a . . . I mean it is such a large issue. You know like if
you are by yourself as an individual and what kind of impact you can have
it is so, it appears to be so small . . . . There has to be . . . I think in order
to get more people to start to change their actions, they need to feel
like they are actually going to if others are doing it too. That it is not just,
oh, me off here by myself redoing something because what is the point
of that? I can’t make a significant enough impact on my own.’’
Complexity of Jack: ‘‘Such a complex issue, that it’s difficult to . . . be versed in all the
issue different areas that climate change impacts. So business, economics,
policy, environment, you know, energy and social implications,
developed countries versus developing countries—it’s such a wide-
ranging issue that it’s difficult to feel like you’re sort of, if not an expert,
then at least, you know, well versed in all the different topic areas.’’
(continued)
Second-Order First-Order
Themes Categories Exemplary Quotations
Bruce: ‘‘It’s the most complicated, you know, to quote Timothy Worth,
it’s probably the most difficult and complicated challenge humanity has
ever faced, more so than the spread of nuclear weapons, because it’s so
perverse. The sources of emissions are so varied in so many different
geographic locations, and they are directly tied with quality of life
increases, that I think this issue is going to be an enormously difficult one
to solve.’’
Interconnectivity Tim: ‘‘This is a problem that’s going to involve changing so many
of issue embedded infrastructures, and procedures, and mind-sets, and, you
know, it . . . it’s really going to take like just a quantum shift in . . . in
our awareness, in our processes, in the way we do things and the way
we approach the world around us . . . . Climate change is sort of a
broad-reaching issue that spans across all other disciplines and is
something that we all need to start thinking about immediately.’’
Alice: ‘‘I just think there’s so many stakeholders involved, and it’s quite a
challenge, and . . . the more stakeholders you have involved, the more
time it takes to get it sorted out, to get something that everybody can
sign up to.’’
Source: Adapted from Sonenshein, S., DeCelles, K. A., and Dutton, J. E. (2014). It’s not easy being green: The role of self-
evaluations in explaining support of environmental issues. Academy of Management Journal, 57, 7-37. Copyright 2014 by the
Academy of Management. Reprinted with permission.
as his research methodology. His research proceeded through three different stages. First, he collected
preliminary information about: (a) all of the appointments that each of the executives had conducted
the previous month, (b) the organization each executive worked in, and (c) the background, personal-
ity, managerial style, and other characteristics of each of the executives. Second, Mintzberg spent a
week with each executive and kept a chronological record of all the events that took place in the exec-
utives’ lives for that week, a record that described each piece of mail they received, and another record
that described each verbal contact. Finally, Mintzberg coded the data from the three records, and based
on this, he identified 10 different roles that managers fill. The first set of these roles is interpersonal in
nature. They include a figurehead role, a leader role, and a liaison role. The second set relates to the
informational responsibilities of the manager and includes the roles of monitor, disseminator, and
spokesperson. The final set of roles is decisional in nature and includes the manager’s role as entre-
preneur, disturbance handler, resource allocator, and negotiator. Although the definition of each of
these concepts (roles) goes beyond the scope of this paper, it is obvious that Mintzberg’s structured
observation approach was a useful technique for identifying and defining the roles played by managers
in organizational settings.
Case Studies. Another approach that may prove worthwhile in helping to identify the attributes of a
concept is the case study technique (Eisenhardt, 1989; Yin, 2014). According to Eisenhardt and
Graebner (2007),
Building theory from case studies is a research strategy that involves using one or more cases
to create theoretical constructs, propositions and/or midrange theory from case-based, empiri-
cal evidence . . . . Case studies are rich, empirical descriptions of particular instances of a phe-
nomenon that are typically based on a variety of data sources. (p. 25)
176
Table 4. Forms of Proactive Post-Sales Service Behaviors Reported by Challagalla, Venkatesh, and Kohli (2009).
Proactive Prevention Proactive Education Proactive Feedback Seeking
Defining characteristics Proactively . . . Proactively . . . Proactively . . .

Monitoring for symptoms of product- Providing product-related information Obtaining diagnostic information on
related problems through mechanical, and skills to help customers realize product performance
electronic. or human means; taking greater value Soliciting complaints
remedial action Informing customers of alternative Asking about customer preferences
Looking for product mishandling and products that might serve their needs and current gaps
taking corrective action more effectively
Illustrative examples (from B2B B2B B2B

interviews with Information technology services: ‘‘My Medical diagnostics: ‘‘We have Industrial gas supply: ‘‘We supply gases
managers) vendors have a modem-connected continuous follow-up with customers that have a shelf life—have an
system, so they get the health check . . . more to maintain a relationship.’’ expiration date. We now leave SMS
[on our machines]. They, in turn, Information technology security: ‘‘We messages alerting (customers) to the
update our managers, so we know do yearly calls to every one of our imminent expiry.’’
things will not drop off.’’ clients—executive management to Information technology consulting:
Nuclear industry: ‘‘We use executive management. [Invariably,] ‘‘We’ll call the client and say, ‘Did you
standardized report cards (nuclear there are latent dissatisfaction issues. know your contract allows for xyz?’
reactors)—whoever prepares the We would never have uncovered these (We’ll also) proactively educate new
card, it has the same format. If we without reaching out.’’ employees of the customer who may
sense a problem at a site, it goes up not know that they have a contract with
several levels in our organization.’’ [us].’’
B2C B2C B2C
Home purchase: ‘‘My real estate agent Credit cards: ‘‘AmEx calls and tells me Health care: ‘‘Our doctor’s office called
sent me a reminder to file for there is a change in my credit card to see how the patient was doing. They
homestead.’’ usage pattern.’’ didn’t need to. It was very nice of them
Courier: ‘‘This [company] proactively Auto: ‘‘I received an e-mail from my to do so.’’
provides information about product automobile manufacturer saying that Home painting: ‘‘Obtain feedback; then
status rather than me call/ask for it.’’ there were problems with my car.’’ obtain feedback again after problem is
fixed (to identify any problem with a
completed job).’’
(continued)
Proactive Prevention Proactive Education Proactive Feedback Seeking
Related literature Tax, Colgate, and Bowen (2006): Grapentine (2006): Proactive service is Crie (2003): Rather than be passive
Prevention is key to managing a latent construct, and an item for with complaints, companies should be
customer failures and recommend a measuring it includes ‘‘keeps proactively gathering information
three-step framework for preventing customers informed about new about sources of dissatisfaction.
failures. products and services.’’ Berry and Leighton (2004): Proactive
Maxham and Netemeyer (2002): Call methods are necessary to identify
for research to explore customer service problems. They suggest that
responses to proactive service senior managers must obtain feedback
recoveries initiated by the supplier (i.e., from customers.
fix problems before customers
complain).
Barker, Lane, Holbrook, Vadrevu, and
Padalina (2005): Case example of
Lucent using preventive maintenance.
Bolton, Lemon, and Verhoef (2008): A
high-level service contract could
include proactive service (e.g., repairs,
preventative maintenance).
Note: B2B ¼ business to business; B2C ¼ Business to customer.
Source: Challagalla, G., Venkatesh, R., and Kohli, A. K. (2009). Proactive postsales service: When and why does it pay off? Journal of Marketing, 73, 70-87. Copyright 2009 by the Amer-
ican Marketing Association. Reprinted with permission.
177
Table 5. Examples of the Resource Rigidity Concept from Gilbert’s (2005) Case Study.
Newspaper Evidence Examples: Financial and Operational Constraints
Beacon A Interviews, archival Sales: ‘‘Print reps could sell the online product, public documents
documents, public sponsorship but with varying degrees of success. Their margins
documents were higher on other products that were easier for them to sell.
Online was really just a novelty to them.’’ (sales representative)
Beacon B Interviews, archival Sales: Staff provided training but then dropped the program multiple
documents times owing to lack of client interest.
Press A Interviews, archival Sales: ‘‘I occasionally sell a bundled print and online package . . . it is
documents, public hard to really know what print advertisers would want.’’ (sales
documents representative)
Press B Interviews, archival Newsroom: Editors call online staff ‘‘low-brow content’’ for using
documents, public radio feeds for breaking news, personals, and unedited user-posted
documents content and refuse to work together. (newspaper editor)
Expositor A Interviews, archival Budgeting: ‘‘And in the end, the only real value is cash and cash
documents, public creation. You can’t build a business just on potential or hope.’’
documents (CEO)
Expositor B Interviews, archival Budgeting: ‘‘Look, when we roll these up into our budgets we miss our
documents, public targets.’’ (vice president, product development)
documents
Morning News A Interviews, archival Newsroom: ‘‘I will be goddamned if some online reporter is going to
documents, public call my sources and say they are from our paper.’’ (publisher)
documents
Morning News B Interviews, public Sales: ‘‘We bundled print and online, but there were clients who
documents wanted online only. These were less interesting to sales reps and
the organization wasn’t ready to deal with that reality.’’ (vice
president, marketing)
Source: Adapted from Gilbert, C. G. (2005). Unbundling the structure of inertia: Resource versus routine rigidity. Academy of
Management Journal, 48, 741-763. Copyright 2005 by the Academy of Management. Reprinted with permission.
A good example of the use of a case study involving multiple organizations to inductively develop a
concept is Gilbert’s (2005) study of the structure of organizational inertia in the face of technological
change in the newspaper industry. Based on in-depth interviews, archival documents, and direct
observations, he identified two different types of inertia. The first type, which represents a failure
to change resource investment patterns, is called resource rigidity. The second type, which repre-
sents a failure to change the organizational processes that use those resource investments, is called
routine rigidity. Table 5 shows examples of the resource rigidity concept drawn from Gilbert’s
research with the different newspapers from which he gathered his data. Based on these data, he con-
cluded that a lack of customer overlap, a different selling process, and a lower relative gross margin
combine to produce resource rigidity.
In addition to using case studies for developing the definition of new concepts, Welch, Rumyant-
seva, and Hewerdine (2015) have also noted that case studies may prove helpful for reconstructing
(or redefining) concepts that have gone through conceptual stretching. According to Sartori (1970),
conceptual stretching occurs when, in the quest for developing more generalizable concepts,
researchers extend the number of cases to which a conceptual definition applies without changing
the set of attributes used to define the concept. The result of this process is that it produces a con-
ceptual mismatch between the attributes used to define the concept and the cases the concept actu-
ally encompasses. Welch et al. have noted that when a concept has been stretched, it is important for
researchers to select and use appropriate cases (or exemplars of the concept) in order to reconstruct
its meaning. Although the criteria recommended for selecting the appropriate cases in this
reconstruction process go beyond the scope of our paper, we encourage researchers who feel that a
concept has gone through conceptual stretching to refer to the Welch et al. paper for a discussion of
this issue.
Compare the Focal Concept With Its Opposite Pole. Goertz and his colleagues (Goertz, 2006; Goertz &
Mahoney, 2012; Klein, Goertz, & Diehl, 2008) have noted the potential value of examining the
opposite (or negative) pole of the focal concept in order to help identify its underlying attributes.
According to Klein et al. (2008), ‘‘In developing a concept and its corresponding measure one needs
to think seriously about the ‘positive’ and ‘negative’ poles of the concept’’ (p. 68). Using this idea,
Klein et al. noted that although most research in international relationships define peace as the oppo-
site of war on a ‘‘conflict-cooperation’’ continuum, treating peace in this manner neglects the fact
that it does not mean the absence of conflict and that peace is not the same as cooperation. They
then go on to argue that it is more appropriate to view the peace continuum from rivalry to negative
peace to positive peace and identify the core features (e.g., the nature of the conflict, the types of
communication and issues between the parties, and the agreements, institutions, and diplomacy
between the parties) that distinguish each of these concepts.
Thus, the basic idea is that by carefully considering the ways in which opposite poles of the con-
cept differ from each other, one can gain a better understanding of the key attributes of the conti-
nuum defining the focal concept and whether it is unidimensional or bipolar. For example, if one
were to define satisfaction as a state of emotional contentment, one could ask what is the opposite
of satisfaction? Is that the same as being ‘‘dissatisfied’’ or is it merely ‘‘not satisfied’’? The former
might suggest a bipolar conceptual structure ranging from extremely content to extremely upset, and
the latter might suggest a unipolar conceptual structure ranging from extremely content to not con-
tent. This conceptual distinction has important methodological implications (cf. Green, Goldman, &
Salovey, 1993; Russell & Carroll, 1999; Tay & Drasgow, 2012).
A relatively good example of comparing the focal concept with its opposite pole has been pro-
vided by Frese and Fay (2001) in their discussion of the self-starting aspect of their personal initia-
tive (PI) concept. As noted by these authors,
It is sometimes easier to describe the other pole of PI: non-self-starting behavior. If a task is
prescribed in detail, and the person follows the prescription, there is no self-starting behavior.
Often consultants have the role of prescribing the tasks for managers in detail. Following the
consultant’s advice is the opposite of PI, particularly if the consultant coaches the manager
through the process of implementing an idea. The more a job incumbent deviates from pre-
scriptions or the less clear the prescriptions are, the more he or she is able to show PI. (p. 143)
Of course, using this technique is not without its challenges. For researchers interested in applying
this technique in their own work, one possible starting point is to examine antonyms of the focal
concept in the dictionary. Antonyms are, by definition, words that mean the opposite of another.
However, different kinds of antonyms imply different things about the nature of the continuum
between the conceptual poles. Gradable antonyms express relationships along a single continuum,
like best and worst, wise and foolish, abundant and scarce, safe and risky, big and small, simple and
challenging, experienced and inexperienced, important and unimportant, and so on. Complementary
antonyms express two opposite possibilities in which the presence of one quality or state signifies
the absence of the other (i.e., there is no intermediate state) like hired or fired, pre or post, absent
or present, push or pull, pass or fail, input or output, and so on. Relational antonyms capture situa-
tions where a relationship defines two opposite possibilities, like employer or employee, supervisor
or subordinate, franchisor or franchisee, mentor or mentee, customer or supplier, and so on. There-
fore, examining the antonyms of the focal concept may provide useful information about what
represents its opposite pole and may subsequently help researchers to better identify the core attri-
butes of their focal concept.
Another way to help clarify the core attributes of a concept by identifying its opposite pole is to ask
what the absence of the focal concept represents. For example, a researcher interested in defining orga-
nizational citizenship behavior (OCB) might ask what the absence of OCB (or ‘‘not’’ OCB) means and
then try to identify those attributes that help explain the difference between OCB and ‘‘not’’ OCB. Using
this technique may also prove useful in determining whether other concepts are the opposites of the focal
concept or represent different, related concepts. Thus, a researcher interested in determining whether
OCB is the opposite of counterproductive work behaviors (CWB) might also ask what attributes help
explain the difference between CWB and a lack of CWB (or ‘‘not’’ CWB). If the attributes associated
with a lack of OCB are virtually the same as the attributes of CWB, then the concepts can be thought as
bipolar concepts, and the researcher can use this information to help clarify what the core attributes of the
focal concept are. If, on the other hand, the attributes associated with the lack of OCB are not the same as
the attributes that define CWB, then these concepts may best be thought of as separate concepts, and the
researcher can use this information to help discriminate between these related concepts. Thus, in either
case, the information should prove useful in helping to clarify the definition of the focal concept.
Examine Existing and/or Create New Operationalizations of the Concept. Jaccard and Jacoby (2010) have
noted that since ambiguities in a conceptual definition often reveal themselves when a scientist is
trying to operationalize a concept, it may prove worthwhile for researchers to ask themselves what
kinds of questions/items/observations they might use to measure the concept. More specifically, they
argue that this is helpful because ‘‘As you formulate these questions, the meaning of the concept often
will become clearer to you’’ (Jaccard & Jacoby, 2010, p. 81). We think that this makes sense. In addi-
tion, another possible approach is to examine the content of existing measures of the concept or mea-
sures of related concepts. This can be important because (a) identifying a measure that seems to
capture the phenomenon of interest but is excluded by the preliminary definition might help to clarify
areas where the definition may be deficient and (b) evaluating whether a measure is contaminated
requires one to identify what is captured by the measure that is outside of the conceptual domain.
How Many Techniques Are Necessary to Develop a ‘‘Good’’ Conceptual Definition?. Although we have
identified several different techniques that researchers can potentially use to help in the concept
development process, it is probably fair to say that the specific number of techniques required to
develop a good conceptual definition depends somewhat on the breadth and depth of the discussion
of the concept in the extant literature. For example, in those cases where several different conceptual
definitions already exist and the goal of researchers is to provide a clearer conceptual definition of an
already existing concept, then conducting a thorough review of the literature, identifying the key
attributes based on these definitions, and organizing these attributes into meaningful conceptual
themes before developing their definition may be the most important activities. In these cases,
authors might use Gerring’s (2012) rule that they stop searching for more definitions when the def-
initions (and the attributes associated with them) become redundant. On the other hand, when a
researcher is trying to develop the definition of a new concept that has not been studied extensively
in the previous literature, the use of some inductive, qualitative techniques, such as case studies,
focus groups, and interviews with subject-matter experts or working professionals, are probably
going to be necessary to adequately flesh out the conceptual domain of the concept. Indeed, these
inductive techniques are generally underutilized in the concept development process, and even in
those cases where multiple concept definitions already exist in the literature, new insights can be
gained through the use of them. However, with that said, we believe that developing a good concep-
tual definition requires a considerable amount of critical thinking, intellectual effort, and creativity
and that simply using the techniques we have identified without also expending this type of effort is
Figure 2. Example of organizing attributes into common themes. Adapted from Sonenshein, S. DeCelles, K. A.,
& Dutton, J. E. (2014). It’s not easy being green: The role of self-evaluations in explaining support of environ-
mental issues. Academy of Management Journal, 57, 7-37. Copyright 2014 by the Academy of Management. Rep-
rinted with permission.
unlikely to result in a good definition. So, as in the case of most other things, the quality of the effort,
as opposed to its quantity, is an important factor.
Stage 2: Organize the Potential Attributes by Theme and Identify Any Necessary
and Sufficient Ones
In some cases, it is possible that researchers using one or more of the aforementioned techniques end
up identifying what they consider to be the core set of attributes necessary to define their focal con-
cept. If that is the case, then they can move to the third stage of the concept development process
(developing a preliminary definition of the concept). However, in many other cases, it is likely that
the result of Stage 1 of the process will be the identification of a fairly large number of attributes and
definitions of the focal concept. In these cases, the next steps in the process are to (a) organize the
potential attributes by their underlying themes and, if possible, (b) identify any necessary and/or suf-
ficient ones (Gerring, 2012; Sartori, 1984).
An example of organizing the attributes of a phenomenon into a smaller set of themes and aggre-
gate dimensions is provided by Sonenshein et al. (2014). As indicated in Figure 2, these authors
organized the seven primary attributes of a concept called environmental issue support challenges
into three related themes (relational challenges, cognitive challenges, and organizational chal-
lenges). Identifying these underlying themes helps to clarify the essence of the concept, and once
identified, they can be helpful in distinguishing the concept from other related concepts.
Once the potential attributes have been organized into underlying themes, it is helpful to evaluate
whether each theme is a necessary or sufficient property of the concept. Necessary (essential) prop-
erties are things that all exemplars of the concept must possess. Sufficient (unique) properties are
things that only exemplars of the concept possess. For example, to define the concept of submarine,
one might begin by identifying the essential and unique attributes of a submarine. A submarine is a
nautical vessel capable of floating, carrying people, operating below the surface of the water, and
propelling itself. It may also possess an emergency escape trunk and internal torpedo tubes. Given
these attributes, the key is to identify the essential and unique attributes that distinguish a submarine
from other nautical vessels. As shown in Table 6, some of these attributes are essential for any nau-
tical vessel (e.g., capable of floating, ability to carry people), so they are not unique to submarines.
Other attributes are helpful for distinguishing submarines from other nautical vessels. For example,
the attribute of operating below the surface of the water distinguishes a submarine from a Viking
Table 6. Identifying Necessary and Sufficient Attributes of the Concept of a ‘‘Submarine’’.
Viking Fishing
Attributes Submarine Diving Bell Longboat Boat Conclusions
A1: can float Present Present Present Present Necessary but not
sufficient
A2: can carry people Present Present Present Present Necessary but not
sufficient
A3: can operate below the surface of Present Present Absent Absent Necessary but not
the water sufficient
A4: has a self-propulsion system Present Absent Present Present Necessary but not
sufficient
A5: possesses an emergency escape Present Absent Absent Absent Sufficient but not
trunk necessary
A6: internal torpedo tubes Present Absent Absent Absent Sufficient but not
necessary
A1 and A2 and A3 and A4 Present Absent Absent Absent Necessary and jointly
sufficient
longboat or a fishing boat, but it does not distinguish it from a diving bell. However, one attribute
that does distinguish a submarine from a diving bell is its ability to propel itself. Finally, one attri-
bute that is unique to submarines but that is not necessary (unless you’re a member of the crew) is the
possession of an emergency escape trunk. Thus, as shown in the table, some individual attributes
(e.g., capable of operating below the surface, floating, carrying people, and propelling itself) are nec-
essary but not sufficient to qualify a naval vessel as a submarine, whereas other individual attributes
(e.g., emergency escape trunk, internal torpedo tubes) are sufficient but not necessary. Unfortu-
nately, no individual attribute is both necessary and sufficient for qualifying a vessel as a submarine.
However, as a set, the attributes of floating, carrying people, operating below the surface, and pro-
pelling itself are jointly sufficient to qualify a vessel as a submarine.
An illustration of the implicit use of this process found in the organizational behavior literature
has been provided by Maynes and Podsakoff (2014). In their review of the existing definitions of
employee voice, they identified the necessary (essential) attributes of voice behavior as,
an individual’s voluntary and open communication directed toward individuals within the orga-
nization that is focused on influencing the context of the work environment . . . . Furthermore,
voice behaviors share several important attributes: (a) they are exhibited by individual employ-
ees; (b) they are not silent . . . , anonymous . . . , or neutral . . . ; (c) they conspicuously stake
out an employee’s position relative to the status quo . . . . To be considered voice, the expression
must be (a) openly communicated, (b) organizationally relevant, (c) focused on influencing the
work environment, and (d) received by someone inside the organization. (p. 88)
According to their definition, each of these attributes is necessary, and the possession of all of them
is jointly sufficient for a behavior to qualify as voice behavior. For example, if an employee provides
improvement-oriented suggestions to a manager, that would qualify as an example of voice behavior
because this behavior is: exhibited by an employee; not silent, anonymous, or neutral; an expression
of the employee’s position relative to the status quo; openly communicated; organizationally rele-
vant; focused on influencing the work environment; and received by someone inside the organiza-
tion. In contrast, as noted in Table 7, if an employee anonymously notifies a regulatory agency of
misconduct in the organization (i.e., exhibits whistle-blowing), these behaviors would not qualify
Table 7. Comparing the Necessary and Sufficient Attributes for Maynes and Podsakoff’s (2014) Employee
Voice Concept to Other, Related Concepts.
Anonymous Whistle Helping

Attributes Voice Suggestions Blowing Behavior Conclusions
A1: voluntary behavior exhibited by Present Present Present Present Necessary but not
individual employees sufficient
A2: not silent, anonymous, or neutral Present Absent Absent Present Necessary but not
sufficient
A3: stakes out a position relative Present Present Present Absent Necessary but not
to the status quo sufficient
A4: openly communicated Present Absent Absent Present Necessary but not
sufficient
A5: organizationally relevant Present Present Present Present Necessary but not
sufficient
A6: focused on influencing context Present Present Present Absent Necessary but not
of the work environment sufficient
A7: Received by someone inside the Present Present Absent Present Necessary but not
organization sufficient
A1 and A2 and A3 and A4 and A5 Present Absent Absent Absent Necessary and jointly
and A6 and A7 sufficient
as employee voice because they are anonymous, not openly communicated, and not received by
someone inside of the organization. Likewise, although anonymous recommendations offered by
an employee in a suggestion box contain several of attributes associated with employee voice,
according to their definition, it is distinguishable from voice because it is anonymous, silent, and
not openly communicated.
Another example of researchers identifying a set of necessary and jointly sufficient attributes of a
concept has been provided by Judge, Locke, and Durham (1997) in their development of the defi-
nition of core self-evaluations (CSE). These authors noted that to be included as a part of CSE, a trait
must satisfy three fundamental criteria. More specifically, a trait must (a) have an evaluative
(vs. descriptive) focus, (b) be central (fundamental) to one’s self-concept, and (c) have a broad
(vs. a narrow) scope. Using these inclusion criteria, the authors were able to explain why some traits
(locus of control, generalized self-esteem, generalized self-efficacy, and emotional stability) are
included in the definition of CSE while others (e.g., agreeableness) are not. However, it is important
to note that although this is a good example of the point we are illustrating, others (e.g., R. E. John-
son, Rosen, & Levy, 2008) have argued that the set of attributes identified by Judge et al. is concep-
tually deficient and that a fourth attribute (‘‘chronic beliefs about one’s own self-regulatory and
behavioral capacities’’) should be added.
Although identifying the necessary and sufficient attributes is desirable for developing a good
conceptual definition, this is not easy to do. Indeed, Molloy et al.’s (2011) review of the
resource-based theory literature concluded that ‘‘only 14% explicated the necessary and sufficient
conditions for a resource to be considered intangible’’ (p. 1504). This is consistent with our own
reading of the organizational behavior literature. Nevertheless, we believe that if a set of jointly suf-
ficient and necessary attributes could be identified, it would greatly reduce the ambiguity and
enhance the precision of the conceptual definition being developed.
However, in some cases, it may not be possible or appropriate to identify necessary and sufficient
attributes due to the nature of the concept. As noted earlier, concepts with a family resemblance
structure do not have any necessary and sufficient attributes but instead share subsets of attributes
that may substitute for each other. For concepts with this structure, the goal is to identify the shared
attributes across subsets of cases and specify how many attributes are required for a case to qualify as
an instance of the concept. This can be done by constructing a table like the attribute matrix in the
right-hand column of Table 1. Once this is done, the central attributes and prototypical cases can be
identified. Identifying the central attributes shared by the most category members is important
because these attributes should be given greater prominence in the definition of the concept. Identi-
fying the most prototypical cases is important because they are the best examples to include in the
definition of the concept.
Although the family resemblance structure is not commonly recognized in the field of manage-
ment and organizational behavior, concepts that use checklists for measures often adopt this struc-
ture implicitly. Indeed, management researchers commonly use checklists to measure a variety of
different concepts, including the social readjustment required by stressful life events (e.g., Kinicki,
1985; Lauer & Thomas, 1976; Tharenou & Harker, 1982; Vicino & Bass, 1978; Weiss, Ilgen, &
Sharbaugh, 1982), positive work events (Ilies, Keeney, & Scott, 2011), negative life events (LoSavio
et al., 2011), daily work stress (Gable, Reis, & Elliot, 2003; Liu, Wang, Zhan, & Shi, 2009),
goal-enhancing and goal-disruptive events (Zohar, Tzischinski, & Epstein, 2003), and interpersonal
conflict and social support (Ilies, Johnson, Judge, & Keeney, 2011). For example, one of the most
commonly used checklists to measure the concept of life stress is the social readjustment rating scale
developed by Holmes and Rahe (1967). Their scale consists of a checklist of 43 stressful life events
(e.g., marriage, death of a spouse, change in residence, major personal injury or illness, retirement
from work, being fired from work) derived from clinical experience, which are weighted by the
extent to which they require ‘‘social readjustment,’’ which is defined as ‘‘the intensity and length
of time necessary to accommodate to a life event, regardless of the desirability of this event.’’ The
greater the number of stressful events experienced by a person, the greater their overall social read-
justment score. Thus, from a conceptual perspective, these stressful life events are conceived of as
multiple sufficient (but not necessary) conditions requiring social readjustment. In other words, a
person is said to have experienced social readjustment stress if they experienced 1 (m) or more of
these 43 events (n), but no particular event is required to produce this outcome. And the greater the
number of events experienced, the greater the social readjustment that is required.
Finally, although we have described the necessary and sufficient concept structure and the family
resemblance concept structure as distinct forms, it is possible to imagine concepts with a hybrid
structure such that there are one or more necessary attributes plus some additional attributes required
(Goertz, 2006). In these cases, the researcher must specify both the necessary attribute(s) as well as
other shared attributes of which at least m of n must be present to define the concept. For example,
one might define the flu as a medical condition that requires a fever plus several (m) of the following
additional symptoms (n): muscle aches, chills, nausea, diarrhea, runny nose, headache, loss of appe-
tite, or sore throat. In this example, having a fever would be necessary but not sufficient by itself to
diagnose someone as having the flu because a fever is symptomatic of many other illnesses (includ-
ing the flu), and some combination of other symptoms (i.e., muscle aches, chills, nausea, diarrhea,
runny nose, headache, loss of appetite, or sore throat) is required to identify the illness as the flu.
Stage 3: Develop a Preliminary Definition of the Concept

The first step in developing a preliminary definition of a concept is to describe the general nature of
its conceptual domain by specifying the type of property the concept represents and the entity to
which that property applies (MacKenzie et al., 2011). This is consistent with Sartori (1970,
1984), who argues that when defining a concept, researchers should specify (a) the intension of the
concept (the phenomenon to which the concept refers) and (b) the extension of the concept (the
referents to which the concept applies). By property, we mean that the definition should specify the
nature of the phenomenon to which the focal concept refers. This could include intrinsic
characteristics (e.g., cognitive ability, personality traits, etc.), thoughts (e.g., cognitions, values,
intentions, etc.), feelings (e.g., attitudes, emotions, etc.), perceptions (e.g., role perceptions, per-
ceived job or organizational characteristics, justice perceptions, etc.), actions (behaviors, activities,
etc.), or performance metrics (e.g., employee productivity, team productivity, SOE, ROI, stock
price, profitability, etc.). By entity, we mean that the definition should specify the object or event
to which the property applies, such as a person, a task, a process, a relationship, a dyad, a group
or team, an organization, or a culture.
Next, as noted by MacKenzie et al. (2011), it is important to further clarify the intension of the
focal concept by elaborating on its conceptual theme. This can be done by specifying the fundamen-
tal attributes/characteristics possessed by exemplars of it. For concepts with a necessary and suffi-
cient concept structure, each attribute/characteristic specified in the conceptual theme must be
common to all exemplars of the concept, and it helps if at least some of these attributes/character-
istics are uniquely possessed by exemplars of the concept. This is the very heart of a good necessary
and sufficient concept structure definition. Indeed, as noted by Cohen and Nagel (1934),
‘‘A definition,’’ according to Aristotle, ‘‘is a phrase signifying a thing’s essence.’’ By the
essence of a thing he understood the set of fundamental attributes which are the necessary and
sufficient conditions for any concrete thing to be a thing of that type. It approximates to what
we have called the conventional intension of a term. (p. 235)
In contrast, for concepts with a family resemblance structure, because there aren’t any attributes that
are common to all cases, the conceptual theme should specify the set of attributes/characteristics that
are shared by subsets of cases of the concept and the central attributes shared by the most cases of the
concept. In addition, regardless of which conceptual structure is used to define the concept, Jaccard
and Jaccoby (2010) have noted that it is also helpful to instantiate the concept by ‘‘specifying con-
crete instances of abstract concepts in order to help clarify their meaning’’ (p. 76). For concepts with
a family resemblance structure, these examples should be the most prototypical cases. However,
although providing examples is valuable, researchers are cautioned to avoid defining a concept
solely in terms of examples because as noted earlier, there is no way to know whether the exemplars
provide a complete listing of the concept’s domain and/or whether new exemplars should be
excluded from the concept’s domain without a clear statement of the concept’s underlying theme
that ties the exemplars together (cf. MacKenzie, 2003).
When the concept is multidimensional, care should be taken to specify the intension of each sub-
dimension or facet in the manner previously described. Specifically, researchers need to describe the
property, entity, and conceptual theme of each individual subdimension. This is important because,
as noted by Messick (1995), a clear definition of each subdimension is necessary to avoid contam-
ination or deficiency in its measurement. In addition, care should also be taken to clarify how the
subdimensions are theorized to be ontologically related to the higher-order construct in order to
properly specify the measurement model (Goertz, 2006; MacKenzie et al., 2011). As noted by
MacKenzie et al. (2011), this means specifying whether the subdimensions are conceptualized as
manifestations or defining characteristics of the higher-order concept.
In addition, the conceptual definition also needs to specify the extent to which the concept is sta-
ble over time and generalizable across situations and cases. For example, some concepts like person-
ality traits are expected to be relatively stable over time and across situations, but these traits are not
necessarily expected to be stable across individual cases (i.e., people). In contrast, some concepts
like mood states are expected to be dynamic over time, across situations, and across individual cases
(i.e., people). Too often, authors fail to specify these aspects of the conceptual definition, which is
problematic because they have implications for the manner in which the concept should be measured
(Baumgartner & Steenkamp, 2006) and tested (Lazarus, 2000; Lazarus & Folkman, 1984). Indeed,
Table 8. An Example of Identifying the Key Attributes of a Construct by Contrasting It With Its Opposite
(from Organ, 1988).
Organizational Citizenship Behavior (OCB)a Task Performanceb
OCB represents individual behavior that is Task performance is individual behavior that is a
discretionary, required part one’s job.
not directly or explicitly recognized by the formal Task performance is generally the basis for
reward system . . . organizational rewards.
By discretionary, we mean that the behavior is not an Task performance is: (a) generally considered an
enforceable requirement of the role or job enforceable requirement of a person’s role in the
description, that is, the clearly specifiable terms oforganization, (b) often included in one’s job
the person’s employment contract with the description, and (c) is often clearly specified in the
organization; person’s employment contract with the
organization.
The behavior is rather a matter of personal choice, Task performance is generally not considered a matter
such that its omission is not general understood as of personal choice. Employees are expected to
punishable. perform their tasks and may be subject to
punishment when they do not do so.
a
Definition taken from Organ (1988, p. 4). bTask definition provided by current authors.
this issue of stability is a primary focus of the extensive literature on state versus trait models (e.g.,
Chen, Gully, Whiteman, & Kilcullen, 2000; George, 1991; Judge, Scott, & Ilies, 2006).
Next, it is also important to differentiate the focal concept from other, related concepts. Indeed,
the very act of defining/labeling something (saying what it is) requires distinguishing it from other
things (saying what it is not). The benefit of doing this is that it (a) helps to distinguish the attributes
that define the focal concept from the attributes that define other, related concepts; (b) diminishes
the possibility of concept proliferation; and (c) identifies the concepts that could be used in empirical
tests of the measures of the focal concept’s discriminant validity. One technique that might be useful
to distinguish the focal concept from others is to take each defining attribute of the focal concept and
think about whether it is also characteristic of the other related concepts and if it’s not, then explain
why not in the definition. This is consistent with Churchill’s (1979) recommendation that when spe-
cifying the domain of a concept, ‘‘The researcher must be exacting in delineating what is included in
the definition and what is excluded’’ (p. 67).
An example of the use of this technique is reflected in Organ’s (1988) definition of organizational
citizenship behaviors (OCB). As indicated in Table 8, for the purposes of his definition, Organ
(1988, p. 4) contrasted the defining properties of OCB with those of task performance. Thus,
whereas task performance is generally considered individual behavior that is a required part of one’s
job, OCB was defined by Organ as ‘‘individual behavior that is discretionary.’’ Similarly, whereas
task performance is generally considered an enforceable requirement of a person’s role in the orga-
nization that is often included in a person’s job description and is specified in the person’s employ-
ment contract, OCB is behavior that ‘‘is not an enforceable requirement of the role or job
description, that is, the clearly specifiable terms of the person’s employment contract with the orga-
nization.’’ Finally, whereas task performance is generally not considered a matter of personal choice
and employees are expected to perform their tasks and may be subject to punishment when they do
not do so, Organ considered OCB to be ‘‘a matter of personal choice, such that its omission is not
generally understood as punishable.’’ Morrison (2011) exhibited a similar thought process when she
described the differences between voice behavior and other related constructs (e.g., issue selling,
whistle blowing, upward communication, and prosocial organizational behavior). Although these
authors may not have explicitly acknowledged using this technique, we believe that it is likely that
they implicitly made the contrasts that we discussed previously and that contrasting a focal concept
with others can serve a worthwhile purpose when trying to identify the key attributes of a concept.
In addition, it is important to specify some of the antecedents and consequences of a concept in
order to help to clarify its meaning. Indeed, as noted by Bagozzi (1984),
Whereas a definition specifies what a concept is and perhaps what it is capable of becoming
and doing, its antecedents supply information as to where it has been (that is, its history and
development) and/or how it is formed or influenced . . . . In a parallel fashion, the meaning of
[a concept] is also determined through its relations to consequences . . . . That is, the implica-
tions of [a concept] supply information as to where a phenomenon is going, what it can lead to,
and/or what influence it has. (p. 21)
However, it is important to emphasize that the concept should not be defined solely in terms of its
antecedents or consequences. The reason for this is that once the antecedents or consequences are
eliminated from the definition, there is nothing left. There must be a positive statement about what
the concept is. For example, to define a concept as something that ‘‘arises from experiences on the
job and leads to promotion’’ doesn’t really say anything about what the concept actually is. Is it job
skills, job-related knowledge, familiarity with the supervisor, understanding organizational politics,
or something entirely different? Without a clear statement of the conceptual theme to go along with
statements about the antecedents and consequences, the definition is unclear.
Finally, when developing a preliminary definition of a concept, researchers should take care to
avoid tautological statements (MacKenzie, 2003; Suddaby, 2010). In this context, tautological refers
to a definition that simply restates in different words the very thing that is to be defined (e.g., a ‘‘fre-
quent purchaser’’ of a brand is a person who ‘‘regularly buys’’ the brand). The obvious problem with
a tautological definition is that it explains the meaning of the concept by referring to the concept
itself.
Stage 4: Refining the Conceptual Definition of the Concept

Although it is possible that the concept definition developed in the third stage is adequate, our own
experience suggests that it may need some refinement before it is finalized. As noted by K. Klein and
Zedeck (2004), the reason for this is that
A construct that seems clear and meaningful to the author who has been thinking about it for
months or years may seem vague and confusing to the reader first exposed to it. The author’s
duty is to be as clear as possible. (p. 932)
Thus, we recommend that researchers developing conceptual definitions take one additional step
before they complete the process—that is to try to refine their preliminary definition.
Perhaps the most obvious thing to do is to ask subject-matter experts, colleagues, or peers to
examine the definition and solicit questions about what you are trying to define. However, before
taking that action, Jaccard and Jacoby (2010) have some suggestions about how to reduce the ambi-
guity and/or jargon that may exist in newly developed concepts. Their first suggestion is to ask the
question, ‘‘What do you mean by that?’’ for as long as there is any ambiguity regarding the words or
terms being used in the definition. The major point is to continue to ask the question about more
specific aspects of the definition until no more ambiguity exists. Another suggestion they make is
to play the role of a journalist who has been asked to write an article for the public regarding the
definition of the concept. This may help reduce the use of jargon in the definition of the concept.
Finally, Jaccard and Jacoby note that the researcher may be able to reduce jargon by placing himself
or herself in a situation where he or she has to explain the concept to someone just learning English.
However, whichever of these strategies is followed, the main point of the exercise is to ensure that
the final version of the conceptual definition is clear, concise, understandable to a broad audience,
and not subject to multiple interpretations.
Tying It Together With Examples From the Literature

To make this discussion more concrete, in this section of the paper we will attempt to provide illus-
trations from the literature of the application of the principles of a good definition that we have iden-
tified. In so doing, we will focus on definitions that have the necessary and sufficient concept
structure because it is considered the standard approach for defining and conceptualizing phenom-
ena and it is clearer than the family resemblance concept structure (Goertz, 2006). Definitions for
each of the concepts discussed in this section of the paper are provided in the Appendix, along with
bracketed letters identifying the portions of the authors’ definitions that relate to various attributes
that are important to creating a clear conceptual definition.
Our first example is Diefendorff, Hall, Lord, and Strean’s (2000, pp. 250-251) definition of the
Action-Orientation (A-O) concept. Consistent with prior research (Kuhl, 1994a, 1994b; Kuhl &
Beckmann, 1994; Kuhl & Goschke, 1994), these authors conceive of A-O as an individual difference
influencing ‘‘goal-striving’’ (i.e., the process of implementing a goal), and as such, it is intended to
capture ‘‘characteristic differences in the enactment and maintenance of goals and the ability to pro-
tect activated goals from competing action tendencies.’’ More specifically, Diefendorff et al. (2000)
propose that
Individuals with a strong action orientation are able to devote their cognitive resources to the
task at hand, thus enabling them to expediently move from a present goal state to some desired
future goal state. These individuals flexibly allocate their attention for the purpose of task exe-
cution and goal attainment. Persons who are more action oriented are characterized by
enhanced performance efficiency (Kuhl, 1994b) and the ability to complete tasks after minor
failures or setbacks. (p. 250)
The authors further clarify the meaning of the A-O concept by contrasting it with its opposite pole
(state-orientation):
Alternatively, individuals with more of a state orientation tend to have persistent, ruminative
thoughts about alternative goals or affective states, which reduces the cognitive resources
available for goal-striving. This reduction of available resources impairs state-oriented indi-
viduals’ ability to initiate activities and to follow tasks through to completion, especially when
the activities are difficult, nonroutine, or both. (p. 250)
Thus, consistent with prior research, Diefendorff et al. define A-O as a ‘‘person’s [entity] character-
istic approach to the process of goal implementation [property].’’ Beyond this, at its most basic level,
they postulate that the A-O concept ‘‘is concerned with individual differences in the ability to initiate
and maintain intentions’’ (p. 250), and as such, A-O is presumably conceived of as being relatively
stable over time and across situations.
As indicated in the Appendix, these authors then go on to distinguish (and define) three separate
attributes or subdimensions of the A-O concept (disengagement, initiative, and task focus) and sum-
marize the distinctions between people who are action-oriented versus state-oriented on these
dimensions by noting that, ‘‘Thus, the three action-orientation dimensions relate to different facets
of the goal-striving process. Action-oriented individuals flexibly disengage from irrelevant concerns
. . . , effectively initiate required actions . . . , and stay focused until tasks are completed.’’ This sug-
gests that disengagement, initiative, and task focus are the necessary and jointly sufficient attributes
of A-O.
As recommended earlier, because A-O is conceptualized as a multidimensional concept, Diefen-
dorff et al. (2000) further clarify the intension of A-O by defining each of these three subdimensions
and their opposite poles. For example, they define the disengagement subdimension as follows:
This dimension, with opposing poles of preoccupation versus disengagement, indicates the
degree to which individuals explicitly process information related to some past, present, or
future state. The action-oriented pole of this dimension (disengagement) refers to the ability
to detach from thoughts about alternative goals or undesirable events that may interfere with
progress on the task at hand. In contrast, the state-oriented pole is associated with impaired
effectiveness due to the perseveration of thoughts related to some unpleasant experience (real
or simulated), often involving failure (Kuhl, 1994a, p. 251).
In addition, they note that as a ‘‘goal striving’’ behavior, A-O differs from goal setting behavior, dis-
positional affect, and cognitive ability. For example, Diefendorff et al. (2000) argue that
Goal setting variables are more discrete in nature and are not tied directly to task implemen-
tation, but rather they reflect cognitions about the likelihood of achieving a goal or a goal’s
value. In contrast, goal-striving approaches to motivation are more closely related to issues
of the self-regulation of thoughts and actions. (p. 250)
Diefendorff et al. (2000) also specify an extensive set of nomological relationships for A-O that
includes both correlates (e.g., self-focus and self-regulatory variables) and consequences (e.g., job
involvement, job satisfaction, work centrality, job performance, and OCB). For example, Diefen-
dorff et al. describe how A-O ought to relate to various work-related outcomes as follows:
Because a key aspect of most jobs is the initiation and maintenance of productive behavior, it
is likely that individual differences in action-state orientation are related to job performance.
Holding other relevant individual characteristics such as ability, knowledge, and experience
equal, employees who are more action-oriented should complete more assigned tasks, have
less difficulty meeting deadlines, and because of their more effective allocation of cognitive
and emotional resources to the task, produce higher quality outputs. This relationship should
hold at least as strongly for discretionary tasks in the workplace, such as organizational citi-
zenship behaviors (OCBs). Because OCBs are more voluntary, there are fewer structural
mechanisms such as production goals or company procedures and policies built into the work-
place to guard against self-regulatory failure. Thus performance of OCBs is expected to be
more dependent on self-regulatory mechanisms internal to the individual employee. Thus,
in general, we expect a moderate, positive relationship between action-state orientation and
supervisor ratings of job performance, indicating that more action-oriented employees per-
form better. This relationship should hold regardless of whether the performance ratings are
for required job behaviors or for more voluntary behaviors such as OCBs. To the extent that
higher performance leads to more tangible and intangible benefits, such as monetary rewards,
social recognition, and personal competence, it is additionally expected that action-state orien-
tation will also be positively related to job attitudes such as job satisfaction, work centrality,
and job involvement. (p. 251)
Although we believe Diefendorff et al. have generally done a good job of defining the action-
orientation concept, one aspect that could use some improvement relates to their labeling of the sub-
dimensions. Despite the fact that we referred to the three subdimensions of action-orientation as dis-
engagement, initiative, and task focus, in their paper the authors refer to these subdimensions by
their state-oriented poles (preoccupation, hesitation, and volatility). We believe this has the potential
to create some confusion because the focus of the paper is on action-orientation, but the subdimen-
sion labels focus on the state-oriented poles of these subdimension. Indeed, we found this somewhat
confusing when interpreting the signs of the correlations between A-O and other concepts. However,
in their defense, the authors undoubtedly did this to be consistent with the work of Kuhl and his
colleagues.
Another fairly good example of a conceptual definition is provided by MacKenzie and Lutz
(1989) in their discussion of attitude toward an advertisement (Aad) in the marketing literature.
According to these authors,
Aad is defined as a predisposition to respond in a favorable or unfavorable manner to a par-

ticular advertising stimulus during a particular exposure occasion (see Lutz, 1985). This con-
ceptual definition of Aad is consistent with Fishbein and Ajzen’s (1975) definition of attitude
in the sense that it views Aad as comprising solely an evaluative or affective response to the
commercial stimulus and does not refer to cognitive or behavioral responses. (p. 49)
Thus, Aad is an evaluative or affective response to an advertisement (property) experienced by an

individual person (entity). Note that according to this definition, there is really only one essential
attribute of Aad (i.e., the evaluative/affective response to an ad). These authors further clarify that
their definition ‘‘does not distinguish between affect and evaluation, nor does it embrace the notion
of a two-component Aad construct.’’ Indeed, they distinguish Aad from a person’s cognitive or
behavioral responses to an advertisement by arguing that the former is an antecedent and the latter
is a consequence of Aad. However, one sentence of their definition makes this somewhat ambigu-
ous: ‘‘The distinctions between cognitive, evaluative, and affective reactions to commercial stimuli
are treated as antecedents to the general attitudinal response denoted as Aad.’’ To eliminate this
ambiguity, this sentence probably should have said that ‘‘the cognitive reactions to commercial sti-
muli are treated as antecedents to the general attitudinal response denoted as Aad.’’ Finally, they
comment on the temporal and situational stability of Aad by noting that it
pertains to a particular exposure to a particular ad and not to consumers’ attitudes toward

advertising in general or even their attitudes toward the ad stimulus of interest at another point
in time (say, after repeated exposure to the ad). Hence, Aad is construed as a situationally
bound construct, an attitudinal reaction to the ad generated at the time of exposure.
One remaining limitation of MacKenzie and Lutz’s (1989) definition is that some researchers may
question the use of the word predisposition to define Aad because they reserve this term for rela-
tively stable, trait-like tendencies. However, it is important to note that MacKenzie and Lutz’s use
of the word is consistent with the how attitudes have generally been defined in the field of psychol-
ogy. Indeed, these authors’ definition of Aad was explicitly developed from Fishbein and Ajzen’s
(1975) widely accepted definition of an attitude ‘‘as a learned predisposition to respond in a con-
sistently favorable or unfavorable manner with respect to a given object’’ (p. 6.). Therefore,
although their use of the word predisposition could suggest to some that Aad is a stable trait, they
clarify that this meaning is not intended (because it is pertains to a particular exposure to a particular
ad) and their use of the word predisposition in this context is consistent with attitude definitions in
the broader field of psychology.
A more macro-oriented example of a relatively good concept definition has been provided by
Zaheer, McEvily, and Perrone (1998). They define interorganizational trust as ‘‘the expectation that
an actor (1) can be relied on to fulfill obligations . . . , (2) will behave in a predictable manner, and
(3) will act and negotiate fairly when the possibility for opportunism is present’’ (p. 143). In addi-
tion, they propose that it is ‘‘a collectively-held trust orientation toward a partner firm’’ (Zaheer
et al., 1998, p. 143). Thus, interorganizational trust is an expectation about a partner firm (property)
that is held by members of an organization (entity). More specifically, the necessary and jointly suf-
ficient conditions for interorganizational trust are that one firm must believe that its partner will act
reliably, predictably, and will negotiate fairly. Zaheer et al. further distinguish that interorganiza-
tional trust is relational rather than dispositional in nature because
dispositional trust is an individual trait reflecting expectancies about the trustworthiness of

others in general . . . [whereas] relational forms of trust pertain specifically to the counterpart
in the dyad. That distinction is important because high relational trust does not imply naively
trusting all exchange partners. Rather, relational trust is likely to be based on experience and
interaction with a particular exchange partner. (p. 143)
Thus, Zaheer et al.’s definition of interorganizational trust clearly describes the concept’s fundamen-
tal nature, the entity to which it applies, and its essential attributes and attempts to distinguish it from
dispositional trust. However, one potential concern with this definition is that, although Zaheer et al.
do distinguish their concept of ‘‘relational trust’’ from ‘‘dispositional trust,’’ they do not distinguish
their concept of trust from other ‘‘relational’’ concepts like dependence, credibility, responsibility,
and so on. For example, they might have argued that ‘‘relational trust’’ is a component of ‘‘credibil-
ity’’ in the sense that both ‘‘trust’’ and ‘‘knowledge’’ are necessary for ‘‘credibility’’ (cf. Hovland,
Janis, & Kelley, 1953). Similarly, it is not clear how trust relates to distrust: Are they opposite ends
of a continuum, or are they distinct constructs (cf., Lewicki, McAllister, & Bies, 1998). Another
potential concern is that the stability of their concept of interorganizational trust is somewhat ambig-
uous because although it is obvious that it can change (i.e., it explicitly allows for betrayal), it is
unclear how malleable it is over time and across situations. Finally, although Zaheer et al. define
trust as being a shared or ‘‘collectively-held’’ expectation, when testing their hypotheses, they oper-
ationalized it from the perspective of only one person in the organization (i.e., a key informant)
rather than as a ‘‘collectively-held’’ judgment of multiple people in the organization.
A final example is Robinson and Bennett’s (1995) definition of the employee deviance concept.
According to these authors, employee deviance is ‘‘voluntary behavior that violates significant orga-
nizational norms and in so doing threatens the well-being of an organization, its members, or both’’
(p. 556). Thus, deviance is behavior (property) engaged in by an employee (entity). The necessary
attributes of this behavior are that it must be (a) voluntary, (b) violate significant organizational
norms, and (c) threaten the well-being of an organization. They clarify the meaning of each of these
attributes using reasonably clear, concise, and unambiguous terms. For example, they explain that
the organizational norms violated by this employee behavior are ‘‘those prescribed by formal and
informal organizational policies, rules, and procedures’’ because ‘‘deviance must be defined in terms
of the standards of a specified social group rather than in reference to a system of absolute moral
standards ([H. B.] Kaplan, 1975)’’ (Robinson & Bennett, 1995, p. 556). Beyond this, they distinguish
deviant behavior from ethical behavior when they say that
the former focuses on behavior that violates organizational norms, whereas the latter focuses
on behavior that is right or wrong when judged in terms of justice, law, or other societal guide-
lines determining the morality of behavior (Lewis, 1985). Thus, although a particular behavior
can be both deviant and unethical, the two qualities are not inevitably linked. (p. 556)
Moreover, they clarify this distinction with an example:
For example, dumping toxic waste in a river is not deviant if it conforms with the policies of
one’s organization. However, most people would probably agree that this act is unethical.
Conversely, reporting this dumping to authorities may be an ethical act, but it would also
be a deviant act in this particular example if it violated organizational norms. (pp. 556-557)
Finally, these authors inductively derive a typology of deviant behaviors consisting of production
deviance, property deviance, political deviance, and personal aggression. Thus, Robinson and Ben-
nett’s definition of employee deviance satisfies many of the criteria for a good conceptual definition
that we have described.
However, two criticisms of their definition of employee deviance might be offered. First, the
definition doesn’t explicitly address the issue of whether deviance is stable over time or across
situations (although it might be implicitly assumed to change over time because it is voluntary).
Second, it defines deviance partially in terms of its consequences through the requirement that
deviant behavior must be ‘‘usually or directly harmful to most organizations.’’ This requirement
suggests that (a) the impact of a behavior must be known a priori in order to determine whether
it qualifies as deviant behavior and (b) the same behavior might be deviant (i.e., harmful to an
organization) in the short run but not in the long run or deviant for one organization but not to
another.
General Summary and Conclusions

We believe our research makes several contributions to the literature. First, it attempts to mitigate
the ongoing problem in the organizational literature of poor conceptual clarity (Locke, 2012; Molloy
et al., 2011; Suddaby, 2010). Second, it integrates the diverse literature on the concept development
process and introduces researchers in the organizational sciences to the discussions of this issue in
other disciplines (e.g., Gerring, 2012; Goertz, 2006; Jaccard & Jacoby, 2010; Loken & Ward, 1990;
Sartori, 1970, 1984). Third, this research clarifies the differences between the necessary and suffi-
cient and family resemblance concept structures and highlights the implications of these differences
for defining concepts (Goertz, 2006; Rosch & Mervis, 1975; Wittgenstein, 1953). Finally, and most
importantly, we recommend a step-by-step procedure to develop better concept definitions and pro-
vide concrete examples from the literature that researchers can refer to when developing their con-
ceptual definitions.
Our discussion of these steps raises a series of questions that authors, reviewers, and editors
would be wise to consider when evaluating conceptual definitions in the literature.
1. Are the techniques used to collect a representative set of definitions (e.g., survey the liter-
ature, interview subject matter experts, compare the construct with its opposite pole, etc.)
described in the paper? (If so, is it clear how the researchers used the techniques to develop
the concept’s definition?)
2. Does the definition describe the type of property the concept represents? (Does the def-
inition specify the nature of the phenomenon (e.g., intrinsic characteristics, thoughts,
feelings, perceptions, actions, or performance metrics) to which the focal concept
refers?)
3. Does the definition describe the entity to which the property applies? (Does the definition
specify the object or event [e.g., person, task, process, relationship, dyad, group, team, orga-
nization, culture, etc.] to which the property applies?)
4. Does the definition describe the theme of the concept and identify its necessary and jointly
sufficient or central attributes (depending upon the concept structure)? (For a concept
having a necessary and sufficient concept structure, does the definition specify: [a] the
concept’s essential attributes/characteristics and [b] the concept’s unique attributes/
characteristics? For a concept having a family resemblance concept structure, does the def-
inition: [a] specify the set attributes that are shared by subsets of cases of the concept,
[b] identify the central attribute[s] shared by the most cases of the concept, and [c] identify
the most prototypical cases [i.e., those that possess the greatest number of shared
attributes?])
5. Is the concept defined solely in terms of examples? (Although it is helpful to provide con-
crete examples of the focal concept through the process of instantiation, the examples
should not serve as the primary mechanisms for defining the concept.)
6. Does the definition specify the dimensionality of the concept? (If so, are the properties, enti-
ties, and conceptual themes of the subdimensions adequately described, and is it clear
whether the subdimensions are conceptualized as manifestations or defining characteristics
of the higher-order concept?)
7. Does the definition specify the stability of the concept? (That is, does the definition specify
whether the concept is considered to be fairly stable or fairly dynamic over time and across
situations?)
8. Does the definition explain how the focal concept differs from related concepts? (Does the
definition identify which attributes of the focal concept are not possessed by [or shared
with] related concepts and include a discussion of this in the paper?)
9. Is the concept defined solely by reference to its consequences or antecedents? (Although it
is important to specify parts of the nomological network of the concept in order to help clar-
ify the theoretical relationship between the focal concept and other concepts, the conceptual
definition should not be based exclusively on a description of the antecedents and conse-
quences of a concept.)
10. Does the definition of the concept use ambiguous, vague, or ill-defined terms? (Is the con-
cept’s definition clear and concise and relatively devoid of technical jargon? Can a layper-
son, not expert in the domain, understand the definition?)
Although much more could obviously be said about how to properly define a concept, we
believe that focusing attention on these questions will help to improve the quality of conceptual
definitions and thereby improve research in the organizational, behavioral, and social sciences
in several ways. First, having a sound conceptual definition should make it easier to distinguish
the focal concept from similar concepts in the field. Second, it should diminish concept ‘‘prolif-
eration,’’ which obscures the pattern of findings in the literature, results in the development of
multiple or conflicting measures of the concept, and impedes theoretical progress. Third, a clearer
conceptual definition should increase the likelihood of a match between the concept and its mea-
sures or manipulations. Fourth, it should be easier to specify and test a nomological network for
the focal concept because having a clearer idea of what the concept means makes it easier to iden-
tify related concepts and to specify whether they are antecedents, consequences, or correlates of it.
Finally, having a good conceptual definition should also decrease the likelihood that measures or
manipulations of the concept will be deficient or contaminated. We hope that this manuscript sti-
mulates thought and changes in the way researchers specify concepts in the organizational, beha-
vioral, and social sciences.
Appendix. Examples of Good Concept Definitions in the Literature
Concept and Definition

Example 1: Action Versus State Orientation (Diefendorff, Hall, Lord & Strean, 2000,
pp. 250-251)
Action-state orientation was construed by Kuhl (1994b; Kuhl & Beckmann, 1994) as an individual difference
variable [S] influencing volitional processes, or what Lewin, Dembo, Festinger, and Sears (1944) referred to
as ‘‘goal-striving.’’ [P&E] Goal-striving refers to the process of implementing a goal and reflects a continuous,
dynamic interplay of events and cognitions whereby information is monitored over time, and actions are
taken based on changing circumstances. [P&E]
Thus, in the context of a broader model of self-regulation, action-state orientation is concerned with
characteristic differences in the enactment and maintenance of goals and the ability to protect activated
goals from competing action tendencies through information-processing mechanisms such as allocation of
attention, inhibition of extraneous cognitions, and emotion control mechanisms (Kuhl, 1994b, Kuhl &
Goschke, 1994). [P&E]
Although individuals likely fall on a continuum of action-state orientation, in further describing the
characteristic it is helpful to contrast those individuals who are more action oriented versus those who are
more state oriented. Individuals with a strong action orientation are able to devote their cognitive resources
to the task at hand, thus enabling them to expediently move from a present goal state to some desired
future goal state. These individuals flexibly allocate their attention for the purpose of task execution and goal
attainment. Persons who are more action oriented are characterized by enhanced performance efficiency
(Kuhl, 1994b) and the ability to complete tasks after minor failures or setbacks. Alternatively, individuals
with more of a state orientation tend to have persistent, ruminative thoughts about alternative goals or
affective states, which reduces the cognitive resources available for goal-striving. This reduction of available
resources impairs state-oriented individuals’ ability to initiate activities and to follow tasks through to
completion, especially when the activities are difficult, nonroutine, or both. (Brunstein & Olbrich, 1985;
Goschke & Kuhl, 1993; Kuhl, 1981; Kuhl, 1994b). [D] In addition to the general notion of action-state
orientation, Kuhl (e.g., Kuhl & Beckmann, 1994) further suggests that three separate aspects or dimensions
can be distinguished: preoccupation, hesitation, and volatility.
Preoccupation dimension. This dimension, with opposing poles of preoccupation versus disengagement,
indicates the degree to which individuals explicitly process information related to some past, present, or
future state. The action-oriented pole of this dimension (disengagement) refers to the ability to detach from
thoughts about alternative goals or undesirable events that may interfere with progress on the task at hand.
In contrast, the state-oriented pole is associated with impaired effectiveness due to the perseveration of
thoughts related to some unpleasant experience (real or simulated), often involving failure (Kuhl, 1994a).
[P&E, DM]
Hesitation dimension. This dimension, with opposing poles of hesitation versus initiative, refers to the degree
to which individuals have difficulty initiating intended goal-directed activities. Action-oriented individuals on
this dimension (initiative pole) are able to easily initiate work on tasks. [P&E, DM] Although both the
hesitation and the preoccupation dimensions are similar in their focus on the ability to move forward with
tasks and to escape from state-oriented processing when necessary, they have different bases. Specifically,
the preoccupation dimension is concerned with whether distracting thoughts interfere with initiating action,
whereas the hesitation dimension emphasizes the behavioral capacity to initiate action. [D]
Volatility dimension. This dimension, with opposing poles of volatility versus persistence, is concerned with
the ability to stay in the action-oriented mode when necessary. It refers to the degree to which individuals
become distracted when working on an interesting or necessary task. More action-oriented individuals
(continued)
Appendix (continued)
(persistence pole) are able to effectively maintain focus on an intention until the task is complete, whereas
more state-oriented individuals are easily pulled off-task, impairing their overall performance. [P&E, DM]
State-oriented processing associated with the volatility dimension may be due to an over activity of the
action initiation system (and hence a tendency to inappropriately initiate new tasks), rather than the
underactivity of this system seen in the state-oriented poles of the preoccupation and hesitation dimensions
(Kuhl, 1994a). [D]
Thus, the three action-orientation dimensions relate to different facets of the goal-striving process. Action-
oriented individuals flexibly disengage from irrelevant concerns (preoccupation), effectively initiate required
actions (hesitation), and stay focused until tasks are completed (volatility). [P&E, DM]
To date, no published study has examined the relationship between action-state orientation and work-
related outcomes. Because a key aspect of most jobs is the initiation and maintenance of productive
behavior, it is likely that individual differences in action-state orientation are related to job performance.
Holding other relevant individual characteristics such as ability, knowledge, and experience equal,
employees who are more action-oriented should complete more assigned tasks, have less difficulty meeting
deadlines, and because of their more effective allocation of cognitive and emotional resources to the task,
produce higher quality outputs. This relationship should hold at least as strongly for discretionary tasks in
the workplace, such as organizational citizenship behaviors (OCBs). Because OCBs are more voluntary,
there are fewer structural mechanisms such as production goals or company procedures and policies built
into the workplace to guard against self-regulatory failure. Thus performance of OCBs is expected to be
more dependent on self-regulatory mechanisms internal to the individual employee. [N]
Thus, in general, we expect a moderate, positive relationship between action-state orientation and
supervisor ratings of job performance, indicating that more action-oriented employees perform better. This
relationship should hold regardless of whether the performance ratings are for required job behaviors or
for more voluntary behaviors such as OCBs. To the extent that higher performance leads to more tangible
and intangible, benefits, such as monetary rewards, social recognition, and personal competence, it is
additionally expected that action-state orientation will also be positively related to job attitudes such as job
satisfaction, work centrality, and job involvement. [N]
(Source: Diefendorff, J. M., Hall, R. J., Lord, R. G., & Strean, M. L. [2000]. Action-state orientation: Construct
validity of a revised measure and its relationship to work-related variables. Journal of Applied Psychology, 85,
250-263. Copyright 2000 by the American Psychological Association. Reprinted with permission.)
Example 2: Attitude Toward an Advertisement (MacKenzie & Lutz, 1989, p. 49)
Throughout our discussion, Aad is defined as a predisposition to respond in a favorable or unfavorable

manner [P&E] to a particular advertising stimulus during a particular exposure occasion (see Lutz 1985). [S]
This conceptual definition of Aad is consistent with Fishbein and Ajzen’s (1975) definition of attitude in the
sense that it views Aad as comprising solely an evaluative or affective response to the commercial stimulus
[P&E] and does not refer to cognitive or behavioral responses. [D] The definition adopted here does not
distinguish between affect and evaluation, nor does it embrace the notion of a two-component Aad
construct. [D] Instead, the distinctions between cognitive, evaluative, and affective reactions to commercial
stimuli are treated as antecedents to the general attitudinal response denoted as Aad. [D, N] Finally, it is
important to note that the definition of Aad pertains to a particular exposure to a particular ad and not to
consumers’ attitudes toward advertising in general [D] or even their attitudes toward the ad stimulus of
interest at another point in time (say, after repeated exposure to the ad). [S] Hence, Aad is construed as a
situationally bound construct, an attitudinal reaction to the ad generated at the time of exposure.
(continued)
(Source: MacKenzie, S. B., & Lutz, R. J. (1989). An empirical examination of the structural antecedents of
attitude toward the ad in an advertising pretesting context. Journal of Marketing, 53, 48-65. Copyright 1989
by the American Marketing Association. Reprinted with permission.)
Example 3: Interorganizational Trust (Zaheer, McEvily, & Perrone, 1998, p. 143)
The literature on interorganizational relations provides two general definitions of trust: confidence or
predictability in one’s expectations about another’s behavior, and confidence in another’s goodwill (Ring
and Van de Ven 1992). We use and extend both conceptualizations and define trust as the expectation that
an actor (1) can be relied on to fulfill obligations (Anderson and Weitz 1989), (2) will behave in a predictable
manner, and (3) will act and negotiate fairly when the possibility for opportunism is present (Anderson and
Narus 1990, Bromiley and Cummings 1995). Our conceptualization of trust as an expectation rather than a
conviction reflects an uncertain anticipation of the referent’s future behavior. [P&E] Our definition explicitly
allows for the possibility of betrayal, which we view as an inherent feature of trust. This probabilistic
element of trust is analogous to making a ‘‘leap of faith’’ by placing confidence in a referent without knowing
with absolute certainty that the referent’s future actions will not produce unpleasant surprises. [S] Our
definition therefore characterizes interorganizational trust as being based on three components—reliability,
predictability, and fairness—that capture some of the complexity of trust (Lewis and Weigert 1985b,
Rempel and Holmes 1986, Rempel et al. 1985). [DM]
Our conceptualization of trust also explicitly recognizes the distinction between relational and dispositional
trust. Whereas dispositional trust is an individual trait reflecting expectancies about the trustworthiness of
others in general (Rotter 1971), relational forms of trust pertain specifically to the counterpart in the dyad.
That distinction is important because high relational trust does not imply naively trusting all exchange
partners. Rather, relational trust is likely to be based on experience and interaction with a particular
exchange partner (Ring and Van de Ven 1992). [D]
The second issue involved in developing a working definition of the trust construct concerns the level of
analysis involved. The Oxford English Dictionary defines trust as ‘‘confidence in or reliance on some quality
or attribute of a person or thing’’ (emphasis added), highlighting the fact that the referent of trust may vary. In
other words, it is conceptually consistent to view trust as being placed in another individual or in a group of
individuals such as the partner organization. However, the same cannot be said for the origin of trust. In our
view, trust has its basis in individuals, although individuals in an organization may share an orientation toward
another organization. From this perspective, interorganizational trust describes the extent to which
organizational members have a collectively-held trust orientation toward the partner firm, which is quite
different from saying that organizations trust each other. In contrast, we view interpersonal trust as also
made up of the three elements—reliability, predictability, and fairness—but with an individual as both the
referent and origin of trust. [D]
(Source: Zaheer, A., McEvily, B., & Perrone, V. (1998). Does trust matter? Exploring the effects of
interorganizational and interpersonal trust on performance. Organization Science, 9, 141-159. Copyright
1998 by the Institute of Management Sciences. Reprinted with permission.)
Example 4: Deviant Workplace Behavior (Robinson & Bennett, 1995, pp. 556-557)
Employee deviance is defined here as voluntary behavior that violates significant organizational norms and in
so doing threatens the well-being of an organization, its members, or both. Employee deviance is voluntary
in that employees either lack the motivation to conform to normative expectations of the social context or
become motivated to violate those expectations ([H. B. Kaplan, 1975). Organizational norms, or those
prescribed by formal and informal organizational policies, rules, and procedures, are specified here because
deviance must be defined in terms of the standards of a specified social group rather than in reference to a
(continued)
system of absolute moral standards ([H. B.] Kaplan, 1975). We focused on the violation of norms espoused
by the dominant administrative coalitions of organizations rather than the norms of work groups or
subcultures. [P&E]
The study of workplace deviance is distinct from the study of ethics in that the former focuses on behavior
that violates organizational norms, whereas the latter focuses on behavior that is right or wrong when
judged in terms of justice, law, or other societal guidelines determining the morality of behavior (Lewis,
1985). Thus, although a particular behavior can be both deviant and unethical, the two qualities are not
inevitably linked. For example, dumping toxic waste in a river is not deviant if it conforms with the policies of
one’s organization. However, most people would probably agree that this act is unethical. Conversely,
reporting this dumping to authorities may be an ethical act, but it would also be a deviant act in this
particular example if it violated organizational norms. [D]
Deviant behavior also has the potential to harm an organization, its members, or both. The term deviant is
usually reserved for acts that violate significant norms (Cohen, 1966) and result in an ‘‘unacceptable
violation . . . believed to threaten society’s well-being’’ (Best & Luckenbill, 1982: 4). Consistent with this
focus, our definition of workplace deviance focuses on violations of norms that threaten the well-being of an
organization. [P&E] Hence, employee deviance excludes minor infractions of social norms, such as wearing a
suit of the wrong style to the office, that are not usually or directly harmful to most organizations. [D]
(Source: Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant workplace behaviors—A
multidimensional-scaling study. Academy of Management Journal, 38, 555-572. Copyright 1995 by the
Academy of Management. Reprinted with permission.)
Note: Letters in brackets [] identify portions of the authors’ definitions that relate to the following attributes of a conceptual
definition: P&E ¼ identifies the property the concept represents and the entity to which it applies; EA ¼ specifies the essential
attributes and conceptual theme of the concept; DM ¼ specifies the dimensionality of the concept; S ¼ specifies the stability
of the concept over time and across situations; N ¼ specifies the nomological net; D ¼ differentiates the focal concept from
other concepts.
Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or
publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
References
Adcock, R. (2005). What is a concept? Political concepts: A working paper series of the committee on con-
cepts and methods. Retrieved from http://www.concepts-methods.org/Files/WorkingPaper/PC%201%20
Adcock.pdf
Adcock, R., & Collier, D. (2001). Measurement validity: A shared standard for qualitative and quantitative
research. American Political Science Review, 95, 529-546.
Aquino, K., & Thau, S. (2009). Workplace victimization: Aggression from the target’s perspective. Annual
Review of Psychology, 60, 717-741.
Bacharach, S. B. (1989). Organizational theories: Some criteria for evaluation. Academy of Management
Review, 14, 496-515.
Bagozzi, R. P. (1984). A prospectus for theory construction in marketing. Journal of Marketing, 48, 11-29.
Bagozzi, R. P. (2007). On the meaning of formative measurement and how it differs from reflective measure-
ment: Comment on Howell, Breivik, and Wilcox (2007). Psychological Methods, 12, 229-237.
Barker, W. O., Lane, J. R., Holbrook, D. P., Vadrevu, N. R., & Padalino, L. T. (2005). Preventative mainte-
nance: A proactive customer service. Bell Labs Technical Journal, 9, 187-200.
Baumgartner, H., & Steenkamp, J.-E.E.M. (2006). An extended paradigm for measurement analysis of market-
ing constructs applicable to panel data. Journal of Marketing Research, 43, 431-442.
Berry, L. L., & Leighton, J. A. (2004). Restoring customer confidence. Marketing Health Services, 24, 14-19.
Binning, J. F., & Barrett, G. V. (1989). Validity of personnel decisions: A conceptual analysis of the inferential
and evidential bases. Journal of Applied Psychology, 74, 478-494.
Blalock, H. M., Jr. (1968). The measurement problem: A gap between the languages of theory and research. In H.
M. Blalock & A. B. Blalock (Eds.), Methodology in social research (pp. 5-27). New York, NY: McGraw-Hill.
Bledow, R., & Frese, M. (2009). A situational judgment test of personal initiative and its relationship to per-
formance. Personnel Psychology, 62, 229-258.
Bollen, K. A., & Bauldry, S. (2011). Three Cs in measurement models: Causal indicators, composite indicators,
and covariates. Psychological Methods, 16, 265-284.
Bollen, K. A., & Lennox, R. (1991). Conventional wisdom on measurement—A structural equation perspec-
tive. Psychological Bulletin, 110, 305-314.
Bolton, R. N., Lemon, K. N., & Verhoef, P. C. (2008). Expanding business-to-business customer relationships:
Modeling the customer’s upgrade decision. Journal of Marketing, 72, 46-64.
Challagalla, G., Venkatesh, R., & Kohli, A. K. (2009). Proactive postsales service: When and why does it pay
off? Journal of Marketing, 73, 70-87.
Chen, G., Gully, S. M., Whiteman, J.-A., & Kilcullen, R. N. (2000). Examination of relationships among trait-
like individual differences, state-like individual differences, and learning performance. Journal of Applied
Psychology, 85, 835-847.
Churchill, G. A., Jr. (1979). A paradigm for developing better measures of marketing constructs. Journal of
Marketing Research, 16, 64-73.
Cohen, M., & Nagel, E. (1934). An introduction to logic and scientific method. New York, NY: Harcourt, Brace.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design & analysis issues for field settings.
Boston, MA: Houghton Mifflin.
Crie, D. (2003). Consumers’ complaint behavior: Taxonomy, typology and determinants: Towards a unified
ontology. Journal of Database Marketing and Customer Strategy Management, 11, 60-79.
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52,
281-302.
Delery, J. E., & Doty, D. H. (1996). Modes of theorizing in strategic human resource management: Tests of
universalistic, contingency, and configurational performance predictors. Academy of Management
Journal, 39, 802-835.
Diamantopoulos, A., & Papadopoulos, N. (2010). Assessing the cross-national invariance of formative
measures: Guidelines for international business researchers. Journal of International Business Studies,
41, 360-370.
Diamantopoulos, A., Reifler, P., & Roth, K. P. (2008). Advancing formative measurement models. Journal of
Business Research, 61, 1203-1218.
Diefendorff, J. M., Hall, R. J., Lord, R. G., & Strean, M. L. (2000). Action-state orientation: Construct validity
of a revised measure and its relationship to work-related variables. Journal of Applied Psychology, 85,
250-263.
DiRenzo, G. J. (1966). General introduction. In G. J. DiRenzo (Ed.), Concepts, theory, and explanation in the
behavioral sciences (pp. ix-xii). New York, NY: Random House.
Dubin, R. (1969). Theory building. New York, NY: Free Press.
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review, 14,
532-550.
Eisenhardt, K. M., & Graebner, M. E. (2007). Theory building from cases: Opportunities and challenges.
Academy of Management Journal, 50, 25-32.
Fishbein, M., & Ajzen, L. (1975). Belief, attitude, intention, and behavior: An introduction to theory and
research. Reading, MA: Addison-Wesley.
Frese, M., & Fay, D. (2001). Personal initiative: An active performance concept for work in the 21st century. In
B. M. Staw & R. M. Sutton (Eds.), Research in organizational behavior (Vol. 23, pp. 133-187). Amsterdam,
Netherlands: Elsevier Science.
Gable, S. L., Reis, H. T., & Elliot, A. J. (2003). Evidence for bivariate systems: An empirical test of appetition
and aversion across domains. Journal of Research in Personality, 37, 349-372.
George, J. M. (1991). State or trait: Effects of positive mood on prosocial behaviors at work. Journal of Applied
Gerring, J. (1999a). Social science methodology: A criterial framework. Cambridge, UK: Cambridge
University Press.
Gerring, J. (1999b). What makes a concept good?: An integrated framework for understanding concept forma-
tion in the social sciences. Polity, 31, 357-393.
Gerring, J. (2012). Social science methodology: A unified framework (2nd ed.). Cambridge, UK: Cambridge
University Press.
Ghiselli, E. E., Campbell, J. P., & Zedeck, S. (1981). Measurement theory for the behavioral sciences. San
Francisco, CA: W.H. Freeman.
Gilbert, C. G. (2005). Unbundling the structure of inertia: Resource versus routine rigidity. Academy of
Management Journal, 48, 741-763.
Goertz, G. (2006). Social science concepts: A user’s guide. Princeton, NJ: Princeton University Press.
Goertz, G., & Mahoney, J. (2012). A tale of two cultures: Qualitative and quantitative research in the social
sciences. Princeton, NJ: Princeton University Press.
Gorden, W. I. (1988). Range of employee voice. Employee Responsibilities and Rights Journal, 1, 283-299.
Grapentine, T. (2006). Unconventional wisdom. Marketing Research, 18, 27-31.
Green, D. P., Goldman, S. L., & Salovey, P. (1993). Measurement error masks bipolarity in affect ratings.
Journal of Personality and Social Psychology, 64, 1029-1042.
Hicks, A., Misra, J., & Ng, T. N. (1995). The programmatic emergence of the social-security state. American
Sociological Review, 60, 329-349.
Hirschman, A. O. (1970). Exit, voice, and loyalty: Responses to decline in firms, organizations, and states.
Cambridge, MA: Harvard University Press.
Holmes, T. H., & Rahe, R. H. (1967). The social readjustment rating scale. Journal of Psychosomatic Research,
11, 213-218.
Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and persuasion: Psychological studies of
opinion change. New Haven, CT: Yale University Press.
Hulleman, C. S., Schrager, S. M., Bodmann, S. M., & Harackiewicz, J. M. (2010). A meta-analytic review of
achievement goal measures: Different labels for the same constructs or different constructs with similar
labels? Psychological Bulletin, 136, 422-449.
Huselid, M. A. (1995). The impact of human-resource management practices on turnover, productivity, and
corporate financial performance. Academy of Management Journal, 38, 635-672.
Ilies, R., Johnson, M. D., Judge, T. A., & Keeney, J. (2011). A within-individual study of interpersonal conflict
as a work stressor: Dispositional and situational moderators. Journal of Organizational Behavior, 32, 44-64.
Ilies, R., Keeney, J., & Scott, B. A. (2011). Work-family interpersonal capitalization: Sharing positive work
events at home. Organizational Behavior and Human Decision Processes, 114, 115-126.
Jaccard, J., & Jacoby, J. (2010). Theory construction and model-building skills: A practical guide for social
scientists. New York, NY: Guilford Press.
Johnson, J. A. (1999). Persons in situations: Distinguishing new wine from old wine in new bottles. European
Journal of Personality, 13, 443-453.
Johnson, R. E., Rosen, C. C., & Levy, P. E. (2008). Getting to the core of core self-evaluation: A review and
recommendations. Journal of Organizational Behavior, 29, 391-413.
Judge, T. A., & Kammeyer-Mueller, J. D. (2012). General and specific measures in organizational behavior
research: Considerations, examples, and recommendations for researchers. Journal of Organizational
Behavior, 33, 161-174.
Judge, T. A., Locke, E. A., & Durham, C. C. (1997). The dispositional causes of job satisfaction: A core eva-
luations approach. Research in Organizational Behavior, 19, 151-188.
Judge, T. A., Scott, B. A., & Ilies, R. (2006). Hostility, job attitudes, and workplace deviance: Test of a multi-
level model. Journal of Applied Psychology, 91, 126-138.
Kaplan, A. (1964). The conduct of inquiry: Methodology for behavioral science. San Francisco, CA: Chandler.
Kaplan, H. B. (1975). Self-attitudes and deviant behavior. Pacific Palisades, CA: Goodyear.
Kehoe, R., & Wright, P. (2013). The impact of high-performance human resource practices on employees’ atti-
tudes and behaviors. Journal of Management, 39, 366-391.
Kinicki, A. J. (1985). Personal consequences of plant closings: A model and preliminary test. Human Relations,
38, 197-212.
Klein, H. J., Molloy, J. C., & Cooper, J. T. (2009). Conceptual foundations: Construct definitions and theoretical
representations of workplace commitments. In H. J. Klein, T. E. Becker, & J. P. Meyer (Eds.), Commitment in
organizations: Accumulated wisdom and new directions (pp. 3-36). New York, NY: Routledge/Taylor.
Klein, J. P., Goertz, G., & Diehl, P. F. (2008). The Peace Scale: Conceptualizing and operationalizing non-
rivalry and peace. Conflict Management and Peace Science, 25, 67-80.
Klein, K., & Zedeck, S. (2004). Theory in applied psychology: Lessons (re)learned. Journal of Applied
Komatsu, L. K. (1992). Recent views of conceptual structure. Psychological Bulletin, 1992, 500-526.
Kristjansson, K. (2012). Positive psychology and positive education: Old wine in new bottles? Educational
Psychologist, 47, 86-105.
Kuhl, J. (1994a). Action versus state orientation: Psychometric properties of the action control scale (ACS-90).
In J. Kuhl & J. Beckmann (Eds.), Volition and personality: Action versus state orientation (pp. 47-59).
Seattle, WA: Hogrefe & Huber.
Kuhl, J. (1994b). A theory of action and state orientations. In J. Kuhl & J. Beckmann (Eds.), Volition and
personality: Action versus state orientation (pp. 9-46). Seattle, WA: Hogrefe & Huber.
Kuhl, J., & Beckmann, J. (Eds.). (1994). Volition and personality: Action versus state orientation. Seattle, WA:
Hogrefe & Hube.
Kuhl, J., & Goschke, T. (1994). A theory of action control: Mental subsystems, modes of control, and volitional
conflict-resolution strategies. In J. Kuhl & J. Beckmann (Eds.), Volition and personality: Action versus state
orientation (pp. 93-123). Seattle, WA: Hogrefe & Huber.
Law, K. S., & Wong, C. S. (1999). Multidimensional constructs in structural equation analysis: An illustration
using the job perception and job satisfaction constructs. Journal of Management, 25, 143-160.
Law, K. S., Wong, C. S., & Mobley, W. H. (1998). Toward a taxonomy of multidimensional constructs.
Academy of Management Review, 23, 741-755.
Lauer, R. H., & Thomas, R. (1976). A comparative analysis of the psychological consequences of change.
Human Relations, 29, 239-248.
Lazarus, R. S. (2000). Toward better research on stress and coping. American Psychologist, 55, 665-673.
Lazarus, R. S., & Folkman, S. (1984). Coping and adaptation. In W. D. Gentry (Ed.), The handbook of beha-
vioral medicine (pp. 282-325). New York, NY: Guilford.
Le, H., Schmidt, F. L., Harter, J. K., & Lauver, K. J. (2010). The problem of empirical redundancy of constructs
in organizational research: An empirical investigation. Organizational Behavior and Human Decision
Processes, 112, 112-125.
Lewicki, R. J., McAllister, D. J., & Bies, R. J. (1998). Trust and distrust: New relationships and realities.
Academy of Management Review, 23, 438-458.
Lewis, P. V. (1985). Defining ‘‘business ethics’’: Like nailing jello to a wall. Journal of Business Ethics, 4,
377-383.
Liang, J., Farh, C. I. C., & Farh, J-L. (2012). Psychological antecedents of promotive and prohibitive voice: A
two-wave examination. Academy of Management Journal, 55, 71-92.
Liu, S., Wang, M., Zhan, Y., & Shi, J. (2009). Daily work stress and alcohol use: Testing the cross-level mod-
eration effects of neuroticism and job involvement. Personnel Psychology, 62, 575-597.
Locke, E. A. (2012). Construct validity vs. concept validity. Human Resource Management Review, 22, 146-148.
Locke, E. A. (2003). Good definitions: The epistemological foundation of scientific progress. In J. Greenberg (Ed.),
Organizational behavior, state of the science (pp. 415-444). Mahwah, NJ: Lawrence Erlbaum Associates.
Loken, B., & Ward, J. (1990). Alternative approaches to understanding the determinants of typicality. Journal
of Consumer Research, 17, 111-126.
LoSavio, S. T., Cohen, L. H., Laurenceau, J-P., Dasch, K. B., Parrish, B. P., & Park, C. L. (2011). Reports of
stress-related growth from daily negative events. Journal of Social and Clinical Psychology, 30, 760-785.
Lutz, R. J. (1985). Affective and cognitive antecedents of attitude toward the Ad: A conceptual framework. In
L. F. Alwitt & A. A. Mitchell (Eds.), Psychological processes and advertising effects: Theory, research and
application (pp. 45-63). Hillsdale, NJ: Lawrence Erlbaum Associates.
MacDuffie, J. P. (1995). Human resource bundles and manufacturing performance: Organizational logic and
flexible production systems in the world auto industry. Industrial and Labor Relations Review, 48, 197-221.
MacKenzie, S. B. (2003). The dangers of poor construct conceptualization. Journal of the Academy of
Marketing Science, 31, 323-326.
MacKenzie, S. B., & Lutz, R. J. (1989). An empirical examination of the structural antecedents of attitude
toward the ad in an advertising pretesting context. Journal of Marketing, 53, 48-65.
MacKenzie, S. B., Podsakoff, P. M., & Podsakoff, N. P. (2011). Construct measurement and validation proce-
dures in MIS and behavioral research: Integrating new and existing techniques. MIS Quarterly, 35, 293-334.
Martinko, M. J., & Gardner, W. L. (1985). Beyond structured observation—Methodological issues and new
directions. Academy of Management Review, 10, 676-695.
Maxham, J. G., & Netemeyer, R. G. (2002). A longitudinal study of complaining customers’ evaluations of
multiple service failures and recovery efforts. Journal of Marketing, 66, 57-71.
Maynes, T. D., & Podsakoff, P. M. (2014). Speaking more broadly: An examination of the nature, antecedents, and
consequences of an expanded set of employee voice behaviors. Journal of Applied Psychology, 99, 87-112.
Merton, R. K. (1958). Social theory and social structure. New York, NY: Free Press.
Merton, R. K., Fiske, M., & Kendall, P. L. (1990). The focused interview: A manual of problems and procedures
(2nd ed.). New York, NY: Free Press.
Messick, S. (1995). Validity of psychological assessment: Validation of inferences from person’s responses and
performances as scientific inquiry into score meaning. American Psychologist, 50, 741-749.
Mill, J. S. (1882). A system of logic, ratiocinative and inductive, being a connected view of the principles of
evidence, and the methods of scientific investigation. New York, NY: Harper & Brothers.
Mintzberg, H. (1975). The manager’s job: Folklore or fact? Harvard Business Review, 53, 49-61.
Molloy, J. C., Chadwick, C., Ployhart, R. E., & Golden, S. J. (2011). Making intangibles ‘‘tangible:’’ A multi-
disciplinary construct validation process. Journal of Management, 37, 1496-1518.
Molloy, J. C., & Ployhart, R. E. (2012). Construct clarity: Multidisciplinary considerations and an illustration
using human capital. Human Resource Management Review, 22, 152-156.
Morrison, E. W. (2011). Employee voice behavior: Integration and directions for future research. Academy of
Management Annals, 5, 373-412.
Morrow, P. C. (1983). Concept redundancy in organizational research: The case of work commitment. Academy
of Management Review, 8, 486-500.
Netemeyer, R. G., Bearden, W. O., & Sharma, S. (2003). Scaling procedures: Issues and applications.
Thousand Oaks, CA: Sage.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York, NY: McGraw Hill.
Nye, C. D., Brummel, B. J., & Drasgow, F. (2014). Understanding sexual harassment using aggregate construct
models. Journal of Applied Psychology, 99, 1204-1221.
Organ, D.W. (1988). Organizational citizenship behavior: The good soldier syndrome. Lexington, MA:
Lexington Books.
Osigweh, C. A. B. (1989). Concept fallibility in organizational science. Academy of Management Review, 14,
579-594.
Osterman, P. (1994). How common is workplace transformation and who adopts it? Industrial and Labor
Relations Review, 47, 173-188.
Podsakoff, N. P., Podsakoff, P. M., MacKenzie, S. B., & Klinger, R. L. (2013). Are we measuring what we say
we’re measuring? Using video techniques to supplement traditional construct validation procedures.
Journal of Applied Psychology, 98, 99-113.
Popper, K. (2002). The logic of scientific discovery. New York, NY: Routledge.
Riggs, F. W. (1975). The definition of concepts. In G. Sartori, F.W. Riggs, & H. Teune (Eds.), Tower of Babel:
On the definition and analysis of concepts in the social sciences (pp. 39-76). Pittsburgh, PA: University of
Pittsburgh.
Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant workplace behaviors—A multidimensional-
scaling study. Academy of Management Journal, 38, 555-572.
Rodale, J. I. (1978). The synonym finder. Emmaus, PA: Rodale Press.
Rosch, E., & Mervis, C. B. (1975). Family resemblances: Studies in the internal structure of categories.
Cognitive Psychology, 7, 573-605.
Russell, J. A., & Carroll, J. M. (1999). On the bipolarity of positive and negative affect. Psychological Bulletin,
125, 3-30.
Sartori, G. (1970). Concept misinformation in comparative politics. The American Political Science Review, 64,
1033-1053.
Sartori, G. (1984). Guidelines for concept analysis. In G. Sartori (Ed.), Social science concepts: A systematic
analysis (pp. 15-85). Beverly Hills, CA: Sage Publications.
Sartori, G., Riggs, T. W., & Teune, H. (1975a). Introduction: Toward the mastery of obstacles to conceptual
clarity. In G. Sartori, F. W. Riggs, & H. Teune (Eds.), Tower of Babel: On the definition and analysis of
concepts in the social sciences (pp. 7-37). Pittsburgh, PA: University of Pittsburgh.
Sartori, G., Riggs, T. W., & Teune, H. (1975b). Tower of Babel: On the definition and analysis of concepts in
the social sciences. Pittsburgh, PA: University of Pittsburgh.
Schwab, D. P. (1980). Construct validity in organizational behavior. In L. L. Cummings & B. M. Staw (Eds.),
Research in organizational behavior (Vol. 2, pp. 3-43). Greenwich, CT: JAI Press.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for gen-
eralized causal inference. Boston, MA: Houghton Mifflin.
Shoemaker, P. J., Tankard, J. W., & Lasorsa, D. L. (2004). How to build social science theories. Thousand
Oaks, CA: Sage.
Short, J. C., Broberg, J. C., Cogliser, C. C., & Brigham, K. H. (2010). Construct validation using computer-
aided text analysis (CATA). Organizational Research Methods, 13, 320-347.
Singh, J. (1991). Redundancy in constructs: Problem, assessment, and an illustrative example. Journal of
Business Research, 22, 225-280.
Sober, E. (1981). The principle of parsimony. British Journal for the Philosophy of Science, 32, 145-156.
Sonenshein, S., DeCelles, K. A., & Dutton, J. E. (2014). It’s not easy being green: The role of
self-evaluations in explaining support of environmental issues. Academy of Management Journal,
57, 7-37.
Spell, C. S. (2001). Management fashions: Where do they come from, and are they old wine in new bottles?
Journal of Management Inquiry, 10, 358-373.
Suddaby, R. (2010). Construct clarity in theories of management and organization. Academy of Management
Review, 35, 346-357.
Tax, S. S., Colgate, M., & Bowen, D. E. (2006). How to prevent your customers from failing. MIT Sloan
Management Review, 47, 30-38.
Tay, L., & Drasgow, F. (2012). Theoretical, statistical, and substantive issues in the assessment of construct
dimensionality: Accounting for the item response process. Organizational Research Methods, 15, 363-384.
Tepper, B. J., & Henle, C. A. (2011). A case for recognizing distinctions among constructs that capture inter-
personal mistreatment in work organizations. Journal of Organizational Behavior, 32, 487-498.
Tesser, A., & Krauss, H. (1976). On validating a relationship between constructs. Educational and
Psychological Measurement, 36, 111-122.
Tharenou, P., & Harker, P. (1982). Organizational correlates of employee self-esteem. Journal of Applied
Tuli, K. R., Kohli, A. K., & Bharadwaj, S. G. (2007). Rethinking customer solutions: From product bundles to
relational processes. Journal of Marketing, 71, 1-17.
Van Dyne, L., Ang, S., & Botero, I. C. (2003). Conceptualizing employee silence and employee voice as multi-
dimensional constructs. Journal of Management Studies, 40, 1359-1392.
Van Dyne, L., & LePine, J. A. (1998). Helping and voice extra-role behaviors: Evidence of construct and pre-
dictive validity. Academy of Management Journal, 41, 108-119.
Vicino, F. L., & Bass, B. M. (1978). Lifespace variables and managerial success. Journal of Applied
Viswanathan, M. (2005). Measurement error and research design. Thousand Oaks, CA: Sage.
Weber, M. (1949). The methodology of the social sciences. New York, NY: Free Press.
Weiss, H. M., Ilgen, D. R., & Sharbaugh, M. E. (1982). Effects of life and job stress on information search beha-
viors of organizational members. Journal of Applied Psychology, 67, 60-66.
Welch, C., Rumyantseva, M., & Hewerdine, L. J. (2015). Using case research to reconstruct concepts:
A methodology and illustration. Organizational Research Methods, 19(1), 111-130.
Williams, L. J., Vandenberg, R. J, & Edwards, J. R. (2009). Structural equation modeling in management
research: A guide for improved analysis. The Academy of Management Annals, 3, 543-604.
Wittgenstein, L. (1953). Philosophical investigations (Trans., G. E. M. Anscombe). Oxford, UK; Basil Blackwell.
Yin, R. K. (2014). Case study research: Design and methods. Thousand Oaks, CA: Sage.
Youndt, M. A., Snell, S. A., Dena, J. W., & Lepak, D. P. (1996). Human resource management, manufacturing
strategy, and firm performance. Academy of Management Journal, 39, 836-866.
Zaheer, A., McEvily, B., & Perrone, V. (1998). Does trust matter? Exploring the effects of interorganizational
and interpersonal trust on performance. Organization Science, 9, 141-159.
Zaltman, G., Pinson, C. R. A., & Angelmar, R. (1973). Metatheory and consumer research. New York, NY:
Holt, Rinehart, & Winston.
Zohar, D., Tzischinski, O., & Epstein, R. (2003). Effects of energy availability on immediate and delayed emo-
tional reactions to work events. Journal of Applied Psychology, 88, 1082-1093.
Author Biographies
Philip M. Podsakoff is the Brian R. Gamache Professor in the Warrington College of Business at the Univer-
sity of Florida. He received his doctoral degree at Indiana University, and his research interests include the ante-
cedents and consequences of organizational citizenship behaviors, leadership effectiveness, and organizational
research methods.
Scott B. MacKenzie is the Neal Gilliatt Chair and Professor of Marketing at the Kelley School of Business at
Indiana University. He received his PhD from the University of California at Los Angeles, and his research
interests include organizational citizenship behavior, leadership, and research methodology.
Nathan P. Podsakoff is an Associate Professor in the Department of Management and Organizations at the
University of Arizona. He earned his PhD from the University of Florida. His research interests include
employee citizenship behaviors, the effects of work-related stressors, and issues related to the methods of beha-
vioral research.

Podsakoff Et Al 2016 Recommendations For Creating Better Concept Definitions in The Organizational Behavioral and

Uploaded by

Copyright:

Available Formats

Podsakoff Et Al 2016 Recommendations For Creating Better Concept Definitions in The Organizational Behavioral and

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Podsakoff Et Al 2016 Recommendations For Creating Better Concept Definitions in The Organizational Behavioral and

Uploaded by

Copyright:

Available Formats

Article

Organizational Research Methods

Philip M. Podsakoff1, Scott B. MacKenzie2,

Concepts—scientific language—comprise the indispensable tools of science, and the irredu-

Necessary and Sufficient Concepts Family Resemblance Concepts

Logical rule AND OR

The Functions of Theoretical Concepts in Scientific Endeavors

Problems With the Lack of Conceptual Clarity

What Makes Defining Concepts So Difficult?

Figure 1. Summary of stages for developing good conceptual definitions.

Recommendations for Developing Good Conceptual Definitions

Stage 1: Identify Potential Attributes of the Concept and/or Collect a Representative

Table 2. Summary of Attributes of Employee Voice Conceptualizations.

Author(s) Conceptualization of Voice Behavior Key Attributes

Hirschman (1970) Voice behavior

Author(s) Conceptualization of Voice Behavior Key Attributes

Behaviors include expressing ideas that shift attention Motivated by fear

Proactive Prevention Proactive Education Proactive Feedback Seeking

Defining characteristics Proactively . . . Proactively . . . Proactively . . .

Illustrative examples (from B2B B2B B2B

Proactive Prevention Proactive Education Proactive Feedback Seeking

Newspaper Evidence Examples: Financial and Operational Constraints

Table 6. Identifying Necessary and Sufficient Attributes of the Concept of a ‘‘Submarine’’.

Anonymous Whistle Helping

Stage 3: Develop a Preliminary Definition of the Concept

Organizational Citizenship Behavior (OCB)a Task Performanceb

Stage 4: Refining the Conceptual Definition of the Concept

Tying It Together With Examples From the Literature

Aad is defined as a predisposition to respond in a favorable or unfavorable manner to a par-

Thus, Aad is an evaluative or affective response to an advertisement (property) experienced by an

pertains to a particular exposure to a particular ad and not to consumers’ attitudes toward

dispositional trust is an individual trait reflecting expectancies about the trustworthiness of

Moreover, they clarify this distinction with an example:

General Summary and Conclusions

Appendix. Examples of Good Concept Definitions in the Literature

Concept and Definition

Example 2: Attitude Toward an Advertisement (MacKenzie & Lutz, 1989, p. 49)

Throughout our discussion, Aad is defined as a predisposition to respond in a favorable or unfavorable

Example 3: Interorganizational Trust (Zaheer, McEvily, & Perrone, 1998, p. 143)

Declaration of Conflicting Interests

You might also like