Intern. J. of Research in Marketing 29 (2012) 310–321
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
Identity-based consumer behavior
Americus Reed II
a, 1
, Mark R. Forehand b, Stefano Puntoni c, Luk Warlop d,⁎
a
The Wharton School, University of Pennsylvania, Philadelphia, PA 19104, USA
Foster School of Business, University of Washington, Seattle, WA 98195, USA
c
Rotterdam School of Management, Erasmus University, PO Box 1738, 3000 DR, Rotterdam, The Netherlands
d
Faculty of Economics and Business, KU Leuven, Naamsestraat 69, B-3000 Leuven, Belgium and BI Norwegian Business School, 0484 Oslo, Norway
b
a r t i c l e
i n f o
Article history:
First received in 18, May 2012 and was under
review for 2 months
Available online 18 September 2012
Keywords:
Identity
Consumer behavior
a b s t r a c t
Although the influence of identity on consumer behavior has been documented in many streams of literature,
the absence of a consistent definition of identity and of generally accepted principles regarding the drivers of
identity-based behavior complicates comparisons across these literatures. To resolve that problem, we propose a simple but inclusive definition of identity. Identity can be defined as any category label with which
a consumer self-associates that is amenable to a clear picture of what a person in that category looks like,
thinks, feels and does. Building from this definition, we propose the following five basic principles that can
help researchers model the process of identity formation and expression: (1) Identity Salience: identity processing increases when the identity is an active component of the self; (2) Identity Association: the non-conscious
association of stimuli with a positive and salient identity improves a person's response to the stimuli; (3) Identity
Relevance: the deliberative evaluation of identity-linked stimuli depends on how diagnostic the identity is in the
relevant domain; (4) Identity Verification: individuals monitor their own behaviors to manage and reinforce
their identities; and (5) Identity Conflict: identity-linked behaviors help consumers manage the relative prominence of multiple identities. To illustrate the potential usefulness of these principles for guiding identity research,
we discuss new avenues for identity research and explain how these principles could help guide investigations
into these areas.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
It is a fundamental human drive to understand who one is, what
one believes and what one does. Therefore, pointing out that consumers like products, brands and consumption behaviors that are
linked to category labels with which they self-associate is rather
uncontroversial. For example, if consumers view themselves as “athletes”, they are likely to behave in ways that are consistent with what
it means to “be” an athlete. This general drive produces a wide range of
“identity driven effects”, including increased attention to identity-related
stimuli (these consumers are more likely to notice and evaluate athletic products), a preference for identity-linked brands (a preference
for athlete-focused Gatorade over brands like Vitaminwater that
have no obvious link to athletes), more positive reactions to advertisements featuring spokespeople who possess the desired identity (pro
athletes are preferred to award-winning actors), the selection of media
catering to the identity (ESPN over CNN), the adoption of behaviors
linked to an identity (wearing equipment such as a distance-running
watch to signal their interest in running) and biased attention toward
⁎ Corresponding author.
E-mail addresses: amreed@wharton.upenn.edu (A. Reed), forehand@u.washington.edu
(M.R. Forehand), spuntoni@rsm.nl (S. Puntoni), luk.warlop@kuleuven.be (L. Warlop).
1
Tel.: +1 215 898 0651; fax: +1 215 898 2534.
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.08.002
identity-consistent memories (increased ease of recalling past athletic
triumphs). These types of identity-driven behaviors have been observed
across numerous identities, and an increasing interest in these effects
has emerged in the academic marketing literature over the last two
decades (see Fig. 1).
2. The three goals of this paper
Although six decades of research on constructs such as the
self-concept, identity, and attitude functions has provided clear evidence that identity often drives behavior, the common processes underlying the influence of identity on behavior are often obscured by
differences in the terminology that is used in these different streams
of literature. In a recent review, Oyserman (2009) took a first step toward unifying these disparate streams of research by arguing that
identities can be cued outside of conscious awareness, that identities
are sensitive to situational factors, and that identities drive many decisions. Similarly, our first objective in this article is to present an inclusive definition of identity in which identity is defined as any
category label to which a consumer self-associates that is amenable to
a clear picture of what the person in the category looks like, thinks,
feels and does. We hope that this definition will subsume various
discipline-based approaches to identity-based behavior that originated
in social psychology (Oyserman, 2007; Tajfel & Turner, 1986),
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
311
Fig. 1. Number of articles published in mainstream consumer research journals (Identity and the self as a percentage of total articles in that particular year).
Consumer Culture Theory (Arsel & Thompson, 2011), and sociology
(Callero, 2003; Howard, 2000); as well as the approaches that arose
from more specific investigations of identity in areas like impression
management (Barreto & Ellemers, 2000) and implicit social cognition
(Greenwald, Pickrell, & Farnham, 2002).
A second objective of this article is to identify a series of important
“identity principles” that connect the various streams of literature
and to thereby provide a more refined structure for the important
processes and mechanisms that have emerged from this literature up
until the present time. These principles are the following: (1) Identity
Salience: identity processing increases when the identity is an active
component of the self; (2) Identity Association: the non-conscious association of stimuli with a positive and salient identity improves a
person's response to the stimuli; (3) Identity Relevance: the deliberative evaluation of identity-linked stimuli depends on how diagnostic
the identity is in the relevant domain; (4) Identity Verification: individuals monitor their own behavior to manage and reinforce identities;
and (5) Identity Conflict: identity-linked behaviors help consumers
manage the relative prominence of multiple identities. It is beyond
the scope of this project to review all of the research that supports
these principles, but a summary of the most notable research support
for these principles is provided in Fig. 2.
The basic identity principles reviewed in this article are the foundations upon which researchers can build to further examine the
theoretical underpinnings of identity-based consumption. The final
objective of this paper is to extend these principles into avenues of future research on identity that hold great promise (see Kirmani, 2009)
and that may, in particular, be critical areas of inquiry for research
in consumer behavior. From a substantive point of view, the basic
principles can also serve as points of departure for future research
to achieve a better understanding of how an identity perspective
can address important managerial and public-policy problems. The
goal in these sections of the article will be to illustrate the usefulness
of these principles for guiding and illuminating future identity research
in marketing and consumer behavior. We selected current trends
that we deemed unprecedented in human history (at least in scale
and pace) and that also have wide-ranging implications for identitybased consumption as they relate to the five aforementioned identity
principles.
Every issue that is discussed to illustrate the way the identity principles can be applied relates either to globalization or to technological
progress, especially in computer-mediated communication. Globalization refers to the increasing interconnection of economic, social
and technological processes across regions and countries. Although
rapid globalization did occur in some earlier historical eras (e.g., the
late colonial period), the current scale and pace of globalization are
unprecedented. Globalization has wide-ranging consequences for both
psychological processes (Arnett, 2002) and for consumer responses to
market offerings (Alden, Steenkamp, & Batra, 1999). Importantly, globalization often requires consumers to make difficult trade-offs and to
hold potentially conflicting beliefs (e.g., Strizhakova, Coulter, & Price,
2012; van Ittersum & Wong, 2010). This opens up many exciting new
areas of research. One of the major engines of globalization, in present
times as in the past, is technological progress. In particular, improvements in computing and communication technology are radically
changing people's lives by introducing new ways of working and communicating and by leading to a reassessment of established behaviors.
Furthermore, improvements in information technology, transport, and
other forms of technology make it possible for more and more people
to interact and to join an integrated world economy. By creating a
“global village” (McLuhan, 1964), computer-mediated communication
channels (social networks, email, and any Internet-based communication platform) are changing the way people around the world relate
to each other. These trends have important ramifications for identity
processes (McKenna & Bargh, 2000). For example, the possibility for a
person to develop a “digital self” (Schau & Gilly, 2003) that can differ
in important respects from the person's “offline” persona opens many
exciting new research questions. These trends are already impacting
312
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
(1) Identity Salience
McGuire, McGuire, Child, & Fujioka, 1978; Grier & Brumbaugh, 1999; Hong, Morris, Chiu, & Benet-Martínez, 2000; Chatman & von Hippel, 2001; Forehand & Deshpandé, 2001;
Grier & Deshpandé, 2001; Forehand, Deshpandé, & Reed, 2002; Reed, 2004; DeMarree, Wheeler, & Petty, 2005; Escalas & Bettman, 2005; Wheeler, Petty, & Bizer, 2005; Wheeler,
DeMarree, & Petty, 2007; Forehand, Perkins, & Reed, 2011; Puntoni, Sweldens, & Tavassoli, 2011; Antioco, Vanhamme, Hardy, & Bernardin, 2012–this issue.
(2) Identity Association
Bem 1981; Schlenker, 1985; Markus & Nurius, 1987; Schmitt, Leclerc, & Dube-Rioux, 1988; Cadinu & Rothbart, 1996; Hugenberg & Bodenhausen, 2004; Jones, Pelham,
Carvallo, & Mirenberg, 2004; Brendl, Chattopadhyay, Pelham, & Carvallo, 2005; Lowery, Unzueta, Knowles, & Goff, 2006; Gawronski, Bodenhausen, & Becker, 2007;
Mercurio & Forehand, 2011; Perkins & Forehand, 2012.
(3) Identity Relevance
Smith, Bruner, & White, 1956; Levy, 1959; Cialdini, Borden, Thorne, Walker, Freeman, & Sloan, 1976; Belk, Mayer, & Bahn, 1982; Solomon, 1983; Wilder & Shapiro, 1984;
Shavitt, 1990; Deshpandé & Stayman, 1994; Wooten, 1995; Shavitt & Nelson, 2000; Kallgren, Reno, & Cialdini 2000; Sirgy, Grewal, & Mangleburg, 2000; Muñiz & O'Guinn
2001; Escalas & Bettman 2003; Gregan-Paxton & Moreau 2003; Ashmore, Deaux, & McLaughlin-Volpe, 2004; Visser & Mirabile, 2004; Ahuvia, 2005; Chaplin & Roedder
John, 2005; Cohen & Reed, 2006; Lowery, Unzueta, Knowles, & Goff, 2006; White & Dahl, 2007; Epp & Price, 2008; Shang, Reed, & Croson, 2008; Aaker and Akutsu,
2009; Zhang & Khare, 2009; Finnel, Reed, Volpp, & Armstrong, 2012; Stokburger-Sauer, Ratneshwar, & Sen, 2012–this issue; Winterich, Zhang, & Mittal, 2012–this issue.
(4) Identity Verification
Rokeach, 1973; Hirschman, 1981; Solomon, 1983; Swann, 1983; Deshpandé, Hoyer, & Donthu,1986; Belk 1988; Williams & Qualls, 1989; Baran, Mok, Land, & Kang, 1989;
Kleine, Kleine & Kernan, 1993; Kleine, Kleine, & Allen, 1995; Aaker 1997; Swann, 1997; Duck, Hogg, & Terry 1999; Laverie, Kleine & Kleine, 2002; Reed 2002; Berger & Heath
2007; Swann, Chang-Schneider, & McClarty, 2007; Rucker & Galinsky, 2008; Gao, Wheeler, & Shiv, 2009; McFerran, Dahl, Fitzsimons, & Morales, 2009; Avery, 2012–this
issue; Claus, Geyskens, Millet, & Dewitte, 2012–this issue; Klesse, Goukens, Geyskens, & de Ruyter, 2012–this issue; Mochon, Norton, & Ariely, 2012–this issue.
(5) Identity Conflict
Higgins, 1986; Markus & Ruvolo, 1989; Benet-Martínez, Leu, Lee, & Morris, 2002; Greenwald, Pickrell, & Farnham, 2002; Roccas & Brewer, 2002; Norman & Aron, 2003;
Benet-Martínez & Haritatos, 2005; Cheng, Lee, & Benet-Martínez, 2006; Amiot, de la Sablonnière, Terry, & Smith, 2007; Lane & Scott, 2007; Luna, Ringberg, & Peracchio,
2008; Mok & Morris 2009; Chen & Bond, 2010; LeBoeuf, Shafir, & Bayuk, 2010; Mok & Morris 2010; McConnell, 2011; Patrick & Hagtvedt, 2012–this issue.
Fig. 2. Research foundation for identity principles.
identity-based consumption patterns in myriad ways, both large and
small, and the pace of change is still increasing. For each of the basic
principles discussed earlier, we review one particular application and
the open theoretical questions that are associated with it. Given the
breadth of topics, the applications are obviously not intended to exhaustively cover the range of implications for future research. Instead,
they serve as illustrative highlights.
3. A parsimonious definition of identity
Our definition of identity starts with the idea that consumers can
potentially identify with a nearly limitless array of different category
labels. Accordingly, we define an identity as any category label to which
a consumer self-associates either by choice or endowment. The category
label invokes a mental representation (i.e., a clear picture) of what
that “kind” of person looks like, thinks, feels, and does (Oyserman,
2009). Some of these identities are relatively stable and “objective”
(e.g., mother, daughter, friend, African-American, etc.), while others
may be more transitory, fluid, and “subjective” (e.g., Republican, athlete,
lawyer, Mac-user, etc.). Although consumers can potentially selfidentify with (or in opposition to) every possible category label, not
all category labels will be central to the consumer's self-definition
(Kihlstrom, 1992). The important point here is that a category label becomes an identity once the consumer has begun to incorporate it into his
or her sense of who he or she is and has initiated the process to become
that kind of person. For example, when individuals transition from voting for a candidate from a particular political party to viewing themselves as members of that political party, they shift from being
unaffiliated individuals who happened to behave in a certain way to
being individuals who embrace a full-fledged political identification.
Once that political identification is formed, the individuals will begin
looking for internal and external feedback to reinforce their identity
(e.g., self-perception processes as well as opportunities to signal their
new identification to others). It is at this point that the identity principles we will propose across various literatures become important.
3.1. Definition properties: group agnostic, bounded and operationalizable
In our view, this definition has three desirable properties. First, it
relaxes the assumption that identities must be indexed to a specific
group of individuals. As a result, there is no need to create different
terms such as “social” identity versus “self” identity versus “personal”
identity because each of these is effectively the same concept applied
to a slightly different domain. The key distinction becomes not
whether an identity has its origin in a social or personal sphere, but
rather how individuals process feedback about all their identities in
their “reflected appraisal” of how well they are enacting those identities (Laverie, Kleine, & Kleine, 2002). For example, individuals often
appraise their standing across multiple identities simultaneously
(e.g., a working parent is likely to concurrently appraise her standing
as a mother, a colleague, a supervisor, and a spouse even though
all these identities exist at different levels of abstraction). This definition also embraces research suggesting that the “self-concept” is
multidimensional and made up of numerous identities. This approach
helps to reduce the surface conflict between the different “types” of
identities that have been studied across various literatures. Although
some slight definition-based differences exist between concepts like
“role identity,” “social identity,” and “self-identity,” the underlying
associative basis of each is largely the same.
Building on this point in our proposed definition, identities can
be “anchored” in different types of referents: objective membership
groups (e.g., gender or family, as discussed in Epp & Price, 2008), culturally determined membership groups (e.g., ethnicity and religion,
as discussed in Dong & Tian, 2009), abstracted role ideals (e.g., mother,
friend, philanthropist), groups premised on association with a known
individual (e.g., a graduate advisor), with an individual who is not
known personally (e.g., Tiger Woods), or with dimensions of self that
are indexed by an imagined other (Deaux, Reid, Mizrahi, & Ethier,
1995). Although there are many types of identities (as described
above), it should be noted that a single identity term can bridge various
classifications. For example, the term “mother” can refer simultaneously to a person's status as a member of an objective group or
as an abstracted role ideal, and it is premised on an association with a
known individual (i.e., the focal person's child or children). To provide
a more comprehensive sense of the types of identities that this definition encompasses, Fig. 3 provides a typology with explicit examples and illustrative features.
Second, the definition we adopt here is broad enough to cover
the various terms and definitions that have been used in different research domains, but it is also concrete enough to differentiate between
what is and what is not an identity (see Cohen, 1989). For example, Gao,
Wheeler, and Shiv (2009) demonstrated that consumers pursue products related to a self-conception if that particular “self-view” has been
threatened. Based on our definition, such a self-view can certainly rise
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
to the level of an identity if a consumer's behaviors related to that particular self-conception become part of how the individual labels himself
or herself, a process that is likely to occur if such behavior is embraced
by a referent social group, a specific individual, or an abstracted ideal.
However, identity is not equivalent to a self-view because identification
involves more than the mere existence of some positive association between a concept and the self. Meaningful identification requires the
adoption or an endowment of a category label that can represent a
rich and clear picture of the type of person to whom the category
label applies, even if such identification is never expressed publicly or
consciously perceived. As a result, self-views can be identities, but that
is not automatically the case.
Third, the definition can be easily represented, as shown in Fig. 4.
Once an identity becomes central to the consumer's self-conception
(I), many secondary associations may also gain prominence in the
individual's self-conception (Oyserman, 2009). For example, individuals
may integrate attitudinal and behavioral norms (Cohen & Reed, 2006;
Oyserman, 2009), emotion profiles (Verrochi-Coleman & Williams,
2012), and a variety of other identity-linked concepts in memory
(Mercurio & Forehand, 2011). These elements structurally define the
normative beliefs, attitudes, emotions and behaviors that define
what that type of person is likely to think, feel, and do. They are
also the building blocks that allow the focal individual to generate
a firm mental representation of that identity and to assess his or her
progress toward enacting it. We depict these various elements as a
bundle of associations (A). Each unique identity has (X) number of
these associations.
3.2. Definition implications: time and the relationship between associations
across identities
There are three important implications of the definition we have
proposed. First, the specific associations for different identities may
change over time as cultural and social factors dictate. In this sense,
cultural and socialization factors serve as meta-constraints on the
types of category labels (and their respective content) that people
can learn about as their self-concepts form, change and evolve over
the course of their lives. For example, gender identity elicits very
different sets of attitudinal and behavioral norms across cultures
(e.g., Luna, Ringberg, & Peracchio, 2008), and the notion of being a
“father” changes over time as a person's relationship with his children
evolves from the time of their childhood to their adulthood. The second important implication is that one or more unique identities may
have different degrees of “overlapping” associations (Deaux et al.,
1995). For example, being “aggressive” may be associated with a
person's professional identity as a Wall Street trader; but it may
also be associated with the person's identity as a “weekend warrior
athlete”. In this sense, the degree of overlap may create correlations
between identities and may play a role in maintaining a sense of
“consistency” within the self (a notion we will explore in more
depth below). The flip side of this is the third implication, namely
the possibility that some of an individual's identity-linked associations may interfere with other associations that are linked to another
identity the person wishes to possess. Because both identities may be
desirable to a person, the possibility that the associations of one identity may interfere with the associations of another identity may produce identity conflict. This is an important principle that we will
discuss in greater depth below.
4. Unifying the literature: Five identity principles
4.1. Identity salience
4.1.1. The identity salience principle
Identity salience exists when an identity is readily accessible to
a consumer and, similar to activation, exists on a continuum from
313
low accessibility to high accessibility. The latter condition is often
referred to as “chronic accessibility” (Forehand, Deshpandé, & Reed,
2002). Although identity salience is not a strictly necessary condition
to observe identity effects (Laverie et al., 2002), the probability of observing identity effects increases as identity salience rises (Puntoni,
Sweldens, & Tavassoli, 2011; Reed, 2004) or with general shifts in
the active self-concept (Wheeler, DeMarree, & Petty, 2007).
Identity salience principle: factors that increase the salience of a
particular identity within a person's self-concept will increase
the probability that the identity will have a subsequent influence
on the person's attitudes and behavior.
Moving beyond this basic principle, it is important to identify the
precursors to identity salience (Reed & Forehand, 2012). One factor
that clearly influences momentary identity salience is the chronic association between the self and the identity domain, which is often referred to as the “strength of identification” (Stayman & Deshpandé,
1989). This kind of association produces a stable and enduring sensitivity to identity-related information (Reed, 2004).
A second factor that may influence identity salience is the composition of an individual's environment. Distinctiveness theory (McGuire,
1984; McGuire, McGuire, Child, & Fujioka, 1978) argues that the salience of personal traits and identities (e.g., gender, race) depends on
the numerical distinctiveness or relative rarity of the traits. The salience
to individuals of their specific memberships depends on how “distinctive” or unusual the trait that is the basis for the specific membership
appears in the immediate environment. For example, Cota and Dion
(1986) placed female participants in groups that varied in their gender
composition and observed an increase in the salience of participants'
gender identity when females were a minority. Similarly, the literature
on self-categorization (Oakes, 1987) argues that a personal characteristic is more likely to be the basis for self-definition when it distinguishes
those who have that characteristic from others who do not.
Third, research has found that identity salience increases in response
to a wide variety of stimulus cues including reference group symbols
(Smith & Mackie, 1995), symbols relating to out-groups (Forehand
et al., 2002), out-group members themselves (Marques, Yzerbyt, &
Rijsman, 1988), and even visual images and words (Hong, Morris,
Chiu, & Benet-Martínez, 2000).
4.1.2. Illustrating the salience principle: English and global consumer
culture
The identity-salience principle asserts that increases in the salience
of an identity within a person's self-concept increase the likelihood of
subsequent identity-driven attitudes and behavior. Because language
is intimately related to culture, language cues are one means of increasing the temporary salience of a cultural identity and of thereby
increasing the influence of that culture on cognitions and behavior.
Among biculturals, exposure to words in a particular language
activates the mental frames of the culture with which that language
is associated. A growing body of work demonstrates that bicultural
individuals think differently when they speak different languages
(Chen & Bond, 2010; Luna et al., 2008; Ogunnaike, Dunham, &
Banaji, 2010; Pavlenko, 2005).
The potential influence of language on identity salience is highlighted by the emergence of English as today's lingua franca. An increasing
number of messages in English (brand names, slogans, product package
information, etc.) reach consumers who are not native English speakers.
For example, in many countries a very large percentage of advertising
messages feature at least some words in English (e.g., Gerritsen,
Korzilius, van Meurs, & Gijsbers, 2000; Lee, 2006). More research is
needed to explore the influence of English being used in marketing
messages in countries where English is not the native language of most
consumers (e.g., De Langhe, Puntoni, Fernandes, & van Osselaer, 2011;
Puntoni, De Langhe, & van Osselaer, 2009).
314
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
Type of
Identity
Abstracted Referent(s)
Fictitious
Characters
Description
Imaginary
social
constructions
created by
marketers,
culture, and the
popular media
Relevant
Extant
Literature(s)
Comic book,
cartoon, story,
film characters
(e.g., Robin
Hood,
Superman,
Neo from the
Matrix,
Gap Kids)
All Members Not
Known
(Large)
Large Social
Collective(s)
Actual
individuals with
whom the person
has no personal
contact, but
wishes to
emulate in
certain ways
Small actual
membership
reference groups
typified by
interaction
among all group
members
Large
membership
groups
characterized by
limited
interaction, may
be personified by
a particular
exemplar
Extremely large,
amorphous and
abstract groups of
individuals, not
likely to be
personified by a
particular
exemplar
Impression
management
theory
Reference Group Selftheory
Categorization
Theory
Spokespersons
(e.g., Tiger
Woods)
musicians and
artists, other pop
cultural icons
(e.g.,
The Pope,
Britney Spears
Peer Group(s)
Immediate
Family
Graduate
Student Cohort
Neighborhood /
Community
Associations
“Trench coat
Mafia”
Known
Not Known
Social
construction of
action patterns
of consistent
behaviors
defined within
membership in a
particular selfcategorization
Actual
individuals with
which the
person has
personal
contact with
and wishes to
maintain a selfdefining
relationship
Identification
based influence
Ph.D. Advisor
Big Brother
Mentor
High School
Counselor
Other
individual role
models
Familial Roles
(e.g., Mother,
Father, son,
daughter)
Occupational
Roles, relational
roles (e.g.,
husband,
girlfriend, etc.)
Group Referent(s)
All Members
Known
(Small)
Roles
Role Identity
theory
Example(s)
Individual Referent(s)
Gender Identity
(male, female)
Athlete Identity
Ethnic Identity,
Political
Identities
Social Identity
Theory
National
Identities (e.g.,
European,
American),
Self as human
being, etc.)
Fig. 3. Taxonomic categorization of identities.
Another important consequence of globalization is the creation
of a global consumer culture and a large group of consumers for
whom a cosmopolitan identity constitutes an important part of their
self-concept (Alden et al., 1999; Grinstein & Wathieu, 2012–this
issue). Whereas much literature in this area focuses on individual differences and conceptualizes global and local consumer culture as the
opposite ends of a single bipolar continuum, identification with a
local culture and identification with a global culture are best understood as separate processes that are at least partially independent
(Steenkamp & de Jong, 2012). Many individuals are likely to endorse
both a local and a global culture, at least to some degree, and to have
both mental frames available to them (Arnett, 2002). For example,
Zhang and Khare (2009) used a priming procedure to increase the
Self-Concept
I1
A1 A2 A3
I2
I3
I4…................IN
A4....AX
Fig. 4. Consumer self concepts comprise adopted/endowed social category labels
(identities) linked to clusters of associations.
accessibility of either a local consumer identity or a global consumer
identity, and they were able in that way to produce differences in
the subjects' expressed preferences for local and global brands.
Linking these areas of research, we propose that the English language may often serve as a cue for a person's cosmopolitan identity.
For example, Alden et al. (1999) argue that because “English has
come to signal modernism and internationalism to many consumers”
an important way for brands to communicate their position in a global
consumer culture is to “use English words, written and/or spoken,
in [their] communications” (p. 77). In other words, in the same way as
Chinese has been shown to cause Chinese-English bilinguals to “think
Chinese” (Chen & Bond, 2010), it seems plausible that English may
sometimes cause Dutch consumers, for example, to “think global”.
If this reasoning is correct, what are the likely consequences for this
new breed of “bicultural” consumers of being confronted with messages
in English? The features of a cosmopolitan identity may vary among
people of different generations in different countries. In some contexts,
a cosmopolitan identity is likely to be associated with tolerant and ecumenical values. In these cases, we predict that individuals may display
less discrimination and bias in an English-language environment. In
some other contexts, a cosmopolitan identity may be strongly associated with the US culture and values. In these cases, exposure to English
may trigger, for example, a more individualistic self-construal.
Another interesting area for future research is the way language
activates consumer culture among individuals depending on whether
global and local consumer cultures are in a harmonious relationship
or a conflicting relationship. In many countries, global consumer culture is often portrayed as being in conflict with traditional (e.g., religious) values. In such cases, it is possible that English may trigger
more complex identity effects for some consumers. In particular, for
consumers who experience global consumer culture and local consumer cultures as conflicting, it is possible that English leads to behaviors that are consistent with local culture rather than with the global
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
consumer culture, in contrast to what one might predict strictly on the
basis of the salience principle (cf., Benet-Martinez, Leu, Lee, & Morris,
2002). This discussion provides an example of how the identitysalience principle is constrained by other identity-based mechanisms.
We will return to the issue of identity conflict later in the paper.
4.2. Identity association
4.2.1. The identity association principle
The first way that salient identities may lead to identity-driven effects is through a simple associative transfer. For example, recent research on implicit partisanship and implicit egotism has shown that
individuals automatically react positively to in-group individuals and
to stimuli associated with the in-group because of the transfer of positive affect from the self-concept to these newly associated stimuli
(Greenwald et al., 2002; Pinter & Greenwald, 2004; Pelham, Carvallo,
& Jones, 2005). Other research has shown that the strength of these
effects increases in accordance with the positivity of the individual's
self-esteem (Gawronski, Bodenhausen, & Becker, 2007; Perkins &
Forehand, 2012). For example, if an arbitrary number (the stimulus) becomes associated with an individual's name (a representation of the self)
a person's response to that number and to people linked to that number
improves (Jones, Pelham, Carvallo, & Mirenberg, 2004). If such automatic
effects are possible with abstract numbers, the association of identity dimensions with the self should produce even more pronounced positive
effects.
Identity association principle: when stimuli become associated
with a positively regarded identity, those stimuli will receive more
positive evaluations and can acquire other identity-related content independent of any explicit processing of the association.
Mercurio and Forehand (2011) recently documented such identitydriven associational learning and transfer in research on the effects
of identity cuing on recall of advertising messages. Building from traditional associative network models (Anderson & Bower, 1973), they
found that identity activation at the time of encoding and retrieving information influenced the association of the new content to the identity
and thereby influenced the participants' subsequent responses. Given
the power of identity to shape the encoding and retrieval of information, it is logical that identities can influence association transfer in
the same way that a person's overall self-concept does. Taken together,
these results suggest that associating new stimuli with an identity can
automatically encourage the transfer of affect and other associations
from the identity to the stimulus.
4.2.2. Illustrating the association principle: ‘not me’ identities
The identity association principle states that objects and concepts
contextually associated to an identity gradually acquire corresponding meanings and evaluations. Brand names, for instance, that are
systematically processed in the context of a particular identity will
gradually be perceived as connoting the identity and its evaluative
meaning. The associative learning process will be largely nondeliberate,
but will be greatly facilitated if the associated concepts are part of the
common discourse of a group that embodies the identity. Social media
are particularly suited to developing identity-associated discourse.
Facebook specifically offers a convenient toolbox for its users to coalesce into ‘like’ groups. These groups make it possible for their members to affirm and enact an identity (Hollenbeck & Kaikati, 2012–this
issue), but the identity-related discourse in these groups also affords an
efficient selection of concepts (objects, people, places, brands, slang
words, …) that come to connote the identity and will receive the associated affective meaning. The continuously reinforced array of
meaningful associations will in turn strengthen the identity itself.
Interestingly, Facebook also offers opportunities to gather in
oppositional communities against people, practices, or particular
315
companies (e.g., the ‘I hate Starbucks’ group). By creating a structured
environment for discourse and associative learning, social media platforms might create ways to shape oppositional or ‘not me’ identities
(Hogg, Banister, & Stephenson, 2009), just like they do ‘me’ identities.
Whereas desired selves allow many opportunities for people to join
collectives defined around a common goal or common means to achieve
a goal (e.g., products or brands), undesired selves by themselves might
not offer the impetus for group identification or for the formation of a
stable identity. People who dislike, avoid, or even hate something do
not necessarily have any goals in common that could allow them to
easily find each other or to find a common language to communicate
with each other.
We propose, however, that oppositional communities on social
networks like Facebook have created unprecedented opportunities
for ‘not me’ identities to take shape in the same way as affirmative
identities do. On Facebook, users with a shared dislike can easily
find each other, engage in frequent conversation, and thereby develop a joint vocabulary. We suggest that the online activity of members
of ‘not me’ groups in social media will lead to the emergence of a
common vocabulary. The availability of a common language will in
turn facilitate a common understanding of how visceral, emotional
and cognitive reactions to the ‘not me’ entity (e.g., Starbucks, and its
customers) are interrelated. Making sense of one's personal oppositional experience as part of a group of like-minded individuals would then
facilitate the adoption of a ‘not me’ or oppositional identity.
It would be useful to investigate the role of the identity-association
principle in the relative stability of ‘not me’ identities as shaped by activity on social media platforms, as well as the marketing implications
of these identities. Intuitively, it might seem that the most stable
identity-shaping ‘not-me’ communities bring the most damage to the
opposed entity. However, we suggest that the opposite is quite plausible. If the discourse of poorly integrated groups spreads outside the
oppositional community, or if it is picked up by mainstream media, it
can bring reputational damage to the attacked ‘not-me’ entity (e.g.,
the opposed public figure or company), but it might not allow the
owner of the ‘not me’ entity to react or counteract. From a marketing
perspective, the attacked entity may actually be better off with social
media opposition that is well organized and has a coherent discourse.
Such well organized groups and their identifying members will be
well differentiated from other groups (the ‘not-not-me groups’). Marketing may then react either by taking action to satisfy the members
of the oppositional (not me) groups or by increasing their attractiveness
towards the ‘not-not-me’ groups in the population.
4.3. Identity relevance
4.3.1. The identity-relevance principle
At a conscious level, stimuli often become linked to identities
when they symbolize the consumer's own personality traits (Aaker,
1997), reflect a desirable self-image, or embody the “type” of person
that the consumer aspires to think, feel and be like (Belk, Mayer, &
Bahn, 1982). In these situations, the consumer's identity is the motivational impetus that drives him or her to form, hold, and express
identity-oriented beliefs and behaviors that connect or separate him
or her from real or imagined others (Escalas & Bettman, 2005).
These identity-based attitudes and behaviors not only help consumers classify themselves but also may come to embody the target
reference group itself (Escalas & Bettman, 2003; White & Dahl,
2007). For example, a strong association of the self with an “American” identity not only encourages a positive attitude to patriotic US
symbols, but can also create a prototype of what a “true American” is.
In that regard, consumers who possess an “American” identity can use
what they believe it means to be “American” to facilitate product
choices that will reinforce the identity (e.g., purchasing a domestic
automobile or avoiding French brie). These identity-based attitudes
are quite resistant to counter-persuasion (Bolton & Reed, 2004), and
316
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
this resilience is at least partially attributable to their shared acceptance
within a network of similar others (Visser & Mirabile, 2004).
The influence of deliberatively processed identity information depends directly on the applicability of the identity to the domain of
evaluation (e.g., an “athlete” identity would be relevant to evaluating
athletic shoes, but is unlikely to be relevant to evaluating kitchen appliances) and the degree to which identity-related information allows
discrimination between options (e.g., an “athlete” identity might help
discriminate between a pair of Nike shoes and a pair of Keds shoes,
but may not help discriminate between a pair of Nike shoes and a
pair of Adidas shoes). These criteria are broadly influenced by five
forms of relevance: object relevance, symbolic relevance, goal relevance, action relevance and evaluation relevance.
Object relevance exists when the object being evaluated is part of
the symbolic constellation of products that define an identity (Kleine,
Kleine, & Kernan, 1993; Reed, 2004). As a case in point, a consumer
who perceives herself as a working mother may be more favorable
to an automobile that emphasizes safety and practicality. These evaluations are particularly common with brands that come to symbolize particular user groups or “fit” with a particular identity (Stokburger-Sauer,
Ratneshwar, & Sen, 2012–this issue).
Symbolic relevance exists when the expression of a belief or the
possession of an object communicates or reinforces one's identity in
the eyes of others (Belk, 1988; Shavitt & Nelson, 2000). Observers readily make impression judgments about others based on their knowledge
of other people's purchase decisions. Therefore, products provide a
“social stock of knowledge that people use in typifying those they
meet” (Shavitt & Nelson, 2000, p. 40). Moreover, this general concept
of symbolic congruence has been used to explain consumer attraction
to products, brands and retail environments (Malhotra, 1988; Sirgy,
Grewal, & Mangleburg, 2000).
Goal relevance exists when a potential belief or behavior is related
to an issue or outcome that is important to the individual's accessible
identity. These beliefs or behaviors could include the expression of an
attitude, specific group-related behaviors, or simply affiliation with a
product or brand. For example, a consumer who benefits from affirmative action would encounter greater goal relevance during an affirmative action debate than would an individual who is unaffected by
such policies (Lowery, Unzueta, Knowles, & Goff, 2006). Applied to
consumption contexts, goal relevance clearly influences relative preference for US car brands in areas where US automakers manufacture.
Ownership of a US automobile in these areas signals external support
for the industry and can reinforce the consumer's identification with it.
Such effects are particularly strong when the identity is self-important
(Gollwitzer, Sheeran, Michalski, & Seifert, 2009).
Stimuli may possess action relevance for an identity if the stimuli
allow an individual to perform some action related to a particular
identity. For example, a “baseball player” may require a bat, glove
and cleats to perform within that identity (Kleine et al., 1993). Similar
to “behavioral involvement” (Ashmore, Deaux, & McLaughlin-Volpe,
2004), action-relevant objects and behaviors allow the consumer to
perform behavioral functions associated with a particular identity. A
corollary of this idea is that the more identity-related possessions a
consumer has, the more empowered a consumer should feel about
his ability to perform in that identity (Ahuvia, 2005) and the more
confident the consumer will be that he holds appropriate opinions
(Jones & Gerard, 1967).
Finally, evaluation relevance refers to the extent to which the
evaluative content of the identity has sufficient clarity and specificity
to inform the consumer's evaluation of the object or brand and to
guide a behavioral response. As a case in point, when an “urban
youth” evaluates shoe brands, he or she may find several brands
that have co-opted young, urban imagery in their advertising and
are thus not differentiable on this identity dimension. In this situation, the absence of a clear identity-related norm provides her with
an inadequate basis for choice (Kallgren, Reno, & Cialdini, 2000),
and thus her identity therefore fails to discriminate between the
available options.
Identity relevance principle: when identity information is deliberatively processed, its influence will be greatest on stimuli that possess object relevance, symbolic relevance, goal relevance, action
relevance or evaluation relevance to the identity.
4.3.2. Illustrating the relevance principle: anthropomorphism and social
robotics
The identity-relevance principle argues that an activated identity
will influence judgment and behavior to the extent that the identity
is relevant to the domain at hand. This influence is not just associative and automatic, but deliberately motivated by the identity. Relevance is typically understood in terms of an instrumentality
relationship between the identity holder and the identity-relevant
object. Using sophisticated kitchen appliances, for instance, may
serve as a means to develop and strengthen the identity of an amateur
chef, and the ‘amateur chef’ identity will drive the evaluation of the appliances' tool-value. This in turn allows marketers to present products
as tools to enact an identity.
It is interesting to consider the consequences of the rapid technological developments that bring artificial intelligence into consumers' lives. Robotic appliances for household tasks like lawn
mowing or vacuum cleaning have already found their way into the
market. Further developments will inevitably lead to the appearance
of products with an ever-increasing level of autonomy and a capacity
for learning and independent decision making. More generally, rapid
technological development in many product categories is transforming
the effectiveness and functional boundaries of consumer products. For
example, anglers today can buy a wide selection of hi-tech products—
from sonar fish-finders to computers that estimate optimal bait—that
make recreational fishing vastly more effective. What are the likely consequences of this increasing autonomy and effectiveness of products?
On the one hand, products will increase their identity relevance by
becoming ever more useful tools. On the other hand, the increasing autonomy and effectiveness of products are likely in some cases to undermine key aspects of what it means to perform a certain identity. For
example, going back to the case of high-tech fishing tools, “if you are
going to use GPS to take you to a location, sonar to identify the fish
and a lure which reflects light that humans can't even see, you may as
well just go to McDonald's and order a fish sandwich” (The
Economist, 2012). In these cases, the products' effectiveness may ironically cause them to loose identity-relevance. Many recreational fishermen would be unwilling to adopt certain high-tech devices such as
sonar fish finders.
Another interesting issue relating to artificial intelligence and autonomous machines arises from the fact that in many cases such machines adopt anthropomorphic features. Robot scientists speculate
that android robots, which in appearance and behavior should be
nearly indistinguishable from humans, will be best suited for assuming
roles that require individual social relationships such as nursing, teaching, or intimate companionship (Duffy, 2003). A successful social robot
will invoke a relational context in which the boundary between its
nature as an instrument and its social identity as a person may become
entirely blurred. Although the widespread availability of android robots
lies in a relatively distant future, many of the more autonomous and intelligent machines available today already have anthropomorphic features. Examples include GPS navigator systems that speak with a
human voice and website shopping aids like Anna, IKEA's “online assistant”. Extending the discussion about identity relevance, the anthropomorphism of such advanced technologies invites questions about
social comparison. Because our current pre-robotic machines do not
have a personal identity, we do not feel threatened if they happen
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
to be stronger or faster than we are, or if they have better memories
and sensory capacities. Anthropomorphic intelligent machines,
however, may be more likely to trigger disadvantageous social comparisons. These machines are not constrained by our physical and
mental limitations, so their anthropomorphic features may highlight
our inadequacies. Thus, anthropomorphic intelligent machines may
have the paradoxical consequence of making consumers feel less confident and empowered. This may already be happening to anyone who,
upon taking the wrong turn against the advice of the navigation system,
seems to detect a hint of condescension in the GPS speaker's voice.
4.4. Identity verification
4.4.1. The identity-verification principle
Once an identity has become salient, consumers will actively monitor the extent to which they have stayed true to the identity. In this
way, a sought-after identity operates very similarly to an “ideal” self
(Higgins, 1986). Higgins argued that as the perceived distance between a consumer's actual and ideal selves increases, the consumer's
motivation to exert effort to reach the ideal also increases. To the extent that a specific identity becomes a cornerstone of a person's ideal
self, aspiring to that identity can become a core driver of behavior. For
example, many consumers aspire to be thin and can imagine what
it would be like to achieve a “thin and healthy” identity. This desire
can provide the motivation to diet, to exercise, or to select products
and services linked to a “thin” identity (McFerran, Dahl, Fitzsimons,
& Morales, 2009). One prominent example is the tendency of consumers to frequent retail outlets whose clothing sizes are shifted
downward from the industry average. Although the consumer's physique is presumably identical at every retail establishment, the consumer is drawn to clothing that claims to be a smaller size (Hoegg,
2012). By wearing this clothing, the consumer is able to move his
self-conception one step closer to his aspirational “thin” identity.
A great deal of advertising and mass-media communication also appeals to these aspirational self-conceptions by presenting models that
symbolize what the consumer wishes to be (Klesse, Goukens, Geyskens,
& de Ruyter, 2012–this issue; Richins, 1991).
The development of an idealized set of identifications can also
increase a consumer's desire to accurately convey these identifications
to the self and to others. This identity pursuit is driven by a selfverification process in which individuals strive to be seen by others
in the way they see themselves (Swann, 1983). The desire to verify
one's identification becomes particularly pronounced when one's identification is threatened (Avery, 2012–this issue; Swann, ChangSchneider, & McClarty, 2007). In fact, recent research has shown that
casting doubt on how consumers view themselves in relation to various
attribute dimensions can lead to choices that reinforce the challenged
attributes (Gao et al., 2009; Rucker & Galinsky, 2008).
Identity-verification principle: feedback from the external environment will be introspectively processed to determine progress
toward the ideal representation of an identity.
However, there are motivational limits to the efficacy of a discrepancy between the actual representation of a consumer's identity and
the ideal. Research has shown that when individuals do not feel
they have the power to achieve an identity ideal, their motivation to
persist in their pursuit of the ideal declines (Chan, Karbowski, Monty,
& Perlmuter, 1986; Norman & Aron, 2003). This suggests that brands
should be careful not to highlight imagery that is aspirational but
unattainable (Klesse et al., 2012–this issue).
The assumption that a person's self-concept contains multiple identities raises a challenge because it introduces complexity in how people
“manage” these different (moving) parts of who they are. Validation
of an identity begs the question of what exactly is being validated and
317
on what criteria. We assume here, based on typical models of
self-regulation (Carver & Scheier, 1998), that consumers constantly
engage in a reinforcement monitoring process in which they retrospectively inspect associations within their identity to make sure that they
are behaving in a consistent manner when enacting that identity.
4.4.2. Illustrating the verification principle: exploring identities
The identity-verification principle asserts that individuals seek validation on how they are enacting a particular identity and that a person's
progress toward the ideal representation of that identity is evaluated relative to the environment. One way in which technology dramatically influences this verification process is by offering individuals an opportunity
to present themselves to others in ways that differ from what is possible
or feasible for them in normal, real-life interactions. Online, a person can
be literally anything. The Internet offers a variety of forums for identity
expression, including avatars in online gaming and virtual realities, personal web pages and blogs, personal profile pages on social networks,
and endless chat rooms and related channels of communication. A rapidly growing body of research investigates self-presentation in online
environments such as personal webpages (Schau & Gilly, 2003), dating
sites (Toma, Hancock, & Ellison, 2008), and social networks (Hollenbeck
& Kaikati, 2012–this issue). Less attention has been paid to the intrapersonal consequences of identity expression in computer-mediated communication. In particular, computer-mediated communication may
bring about important changes on the identity verification process.
Online environments offer people new channels of communications
(McKenna & Bargh, 2000). In particular, computer-mediated communication makes it possible for individuals to belong to groups that are not accessible in real life, even if such belonging is only virtual and for a
limited time. For example, a runner who can no longer practice the
sport because of an injury can nevertheless interact with other
runners online and belong to a community by posting messages and
being active on websites. This social contact can help individuals
reaffirm, and hence verify, their identity. Connecting this discussion
to globalization, computer-mediated communication offers people
in far-away places an easily accessible way to verify and express
their cultural identity. We hypothesize that online identity verification may therefore have important implications for the acculturation
processes of migrants and ethnic minorities. More generally, the
real-life behavioral consequences of this online identity verification
process are also an important area for future research. For example,
acting like a younger person online may have assimilative consequences for how a person feels in his or her real life (i.e., acting younger online could lead people to feel younger in general—e.g., dressing
differently) but perhaps also contrastive consequences (i.e., acting
younger online may in fact underline the discrepancy between
one's hopes and reality), leading to faster adoption of the undesired
identity.
Perhaps most interestingly, computer-mediated communication can
offer individuals a way to experiment with alternative identities in a
safer environment. Identities are often held and developed through a
trial-and-error process (Ibarra, 1999; Markus & Wurf, 1987). The anonymity of computer-mediated communication offers unprecedented opportunities for people to express ideas and enact behaviors with little
concerns about accountability and the ramifications of their behavior
for their real-life social relationships (Kernis & Goldman, 2003). The
Internet removes barriers and lowers the cost of the trial-and-error
processes that are involved in identity verification and identity construction, especially for identities that present individuals with difficult
trade-offs between inclusion and exclusion. Therefore, online platforms have important consequences for people's ability and motivation to identify with identities that mainstream society considers
deviant or undesirable, such as extreme political and alternative sexual
identities (McKenna & Bargh, 1998). The greater likelihood of adopting
a stigmatized identity that participation in online forums can lead to has
318
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
implications for consumption (e.g., magazine subscriptions) and, more
generally, for marketers (e.g., minority targeting and media planning).
This discussion is also relevant for consumption-based communities
that place consumers outside of the mainstream, such as the “persecuted”
Apple Newton community (Muñiz & Schau, 2005).
4.5. Identity conflict
4.5.1. The identity conflict principle
Following from the previous discussion, the final principle that is
addressed in this article relates to the fact that any given identity is not
possessed in isolation; instead, each identity is one of many held identities that must be integrated into a person's overall self-conception.
Research on the interplay of multiple identities generally suggests
that individuals seek to maintain harmony between their various identities (Amiot, de la Sablonnière, Terry, & Smith, 2007; Roccas & Brewer,
2002). Consumption can often lead to identity conflict, or provide
ways to resolve it. This is most likely best exemplified by the stream of
interpretive marketing research that shows how the process of developing and enacting a person's identity is “marked by points of conflict”
(Arnould & Thompson, 2005, p. 871). For example, a company's attempt
to gender-bend a masculine brand can constitute a threat to males
who identify with the brand and can trigger a range of coping responses
(Avery, 2012–this issue).
Research has suggested that individuals typically adopt one of two
specific identity structures to achieve harmony among their held
identities. Harmony is greatest when an individual's various identities
suggest norms for behavior that are consistent with one another
(Amiot et al., 2007; Benet-Martinez et al., 2002; Roccas & Brewer,
2002). However, when the norms of the two identities conflict with
one another individuals may resolve the conflict by using a variety of
self-regulatory processes.
Identity conflict principle: individuals are motivated to reduce
conflict across multiple identities and can do so by managing the
relative salience of their various conflicting identities.
One self-regulatory process that can help reduce conflict is to
maintain balanced salience between the potentially conflicting identities. Evidence for such a balancing process is apparent in research
on bi-cultural individuals. Research has found that priming one component cultural identity in bi-cultural individuals prompts behaviors
consistent with the primed cultural identity and avoidance of behaviors consistent with the second cultural identity, but this pattern applies only when the two cultural identities are integrated. When
cultural inputs from the two cultural identities conflict with one another, cultural primes may, instead, prompt avoidance of the primed cultural identity and pursuit of the second cultural identity (Benet-Martinez et
al., 2002; Mok & Morris, 2009, 2010; Zou, Morris, & Benet-Martinez,
2008). For example, after being primed with an Asian cultural cue,
Asian-American consumers whose Asian and American cultural identities are not well integrated pursue more American behaviors including
an increased propensity for internal attributions (Benet-Martinez et
al., 2002; Zou et al., 2008), uniqueness-seeking, and extroversion (Mok
& Morris, 2009). This response pattern has been attributed to a desire
to keep the component identities in balance and to avoid overt conflict
and the dissonance that would result from that conflict.
4.5.2. Illustrating the conflict principle: aging and lifespan of identities
The identity-conflict principle argues that individuals seek to reduce conflict across multiple identities and that conflicts that are
difficult to resolve will often invoke self-regulatory processes to restore balance. In this paragraph, we consider the potential for identity
conflict that is inherent in the current trends that have accompanied
the aging populations. The adoption of healthier lifestyles, together
with technological progress, has resulted in an increase in the life
expectancy enjoyed by consumers in many countries. People live longer, but, equally important, they also live better. Technology makes a fundamental contribution to the continuing increase in “quality-adjusted life
years.” Examples include healthcare procedures (e.g., knee-replacement
surgery), monitoring devices (e.g., emergency alert devices), transportation systems (e.g., personal vehicles), and diagnostic tools (e.g., devices
for measuring blood pressure). People today can expect to continue
enjoying a high quality of life until an age when few people in earlier
generations would have been alive.
Identification processes are affected by the passing of time, either
because a particular identity is directly linked to a life-stage (e.g., grandparent identity, youth identity) or because aging affects the possibility,
or even the appropriateness, of enacting a particular identity or of
engaging in activities that are associated with the identity (e.g., sportrelated or work-related identities). By affecting the way people transition away from identities that are central to a particular maturity
phase in a person's lifespan, the emergence of a healthier population
of elderly people has implications for consumer identification with
identities that are associated with the later stages of life. For example,
an athletic baby boomer can expect to continue enjoying mountain hiking trips into her seventies. As a result, she will be able to maintain her
identity as an athletic and outdoorsy person until late in her life. The
identity conflict principle argues that people are motivated to monitor
and reduce the perceived conflict between identities. The longer life
span and improved quality of life enjoyed by older people today
means that it is becoming more likely that individuals approaching
old age will perceive an inconsistency between the identities they
are used to (and that they can continue to endorse) and the identities
that society expects people of their age to endorse. This may lead people
to seek ways to combine or amalgamate identities to reduce the perceived conflict, similar to the bicultural identity integration observed
among ethnic minorities (Benet-Martinez et al., 2002). For example,
people may try to “reconstruct” their idea of what it means to age by
contesting the association between wisdom and family support on
one hand and, for example, reduced physical activity on the other. However, given the prevalence of negative stereotypes about aging and
elderly people (North & Fiske, 2012) identity conflict may often simply
result in a generally lower level of identification with identities that are
assumed to be typical for later life stages. For example, anecdotal evidence suggests that some youthful-looking baby boomers recoil
at being called “grandma”.
Although the elderly stereotype tends to feature many undesirable
associations, it is also associated with positive concepts such as
wisdom, family orientation, and support (Hummert, 1990; North &
Fiske, 2012). A focus on nurturing goals and on building meaningful
emotional connections with others, especially with younger relatives,
is often a key feature of identities associated with late-life stages (e.g.,
a grandparent identity). In many countries, for example, older
individuals make an important contribution to family life by helping
with child care. If people perceive an expansive (vs. finite) time horizon
until later in life, they will also tend to adopt goals typical of elderly people at a much later stage in their lives (Carstensen, 2006). This may have
important repercussions for intergenerational relations because identity conflict may motivate aging individuals to identify later, or to a lesser
degree, with identities associated with late life-stages, and this may
make them less inclined to offer support to younger family members.
This, in turn, might make it less likely that younger relatives will be
willing to provide care and support later, when such care is needed.
Therefore, whereas the late adoption of identities associated with
late-life stages is likely to lead to greater life satisfaction for individuals in their “young old age”, it may lead to a reduction in life satisfaction later in life. Although these considerations are speculative,
intergenerational relations represent an increasingly important area of
research because of the dire state of many countries' public finances
and the increasingly skewed age distributions. It is also an area that
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
has received virtually no attention in psychological research (for a recent discussion, see North & Fiske, 2012. An identity perspective could
provide a fruitful approach to investigate intergenerational relations
in a consumer context.
5. Concluding remarks
Our sense of who we are has a large influence on our thoughts,
feelings and behavior. While a variety of labels can be associated
with the self, chronically or in specific situations, a self-label becomes
an identity as soon as it becomes sufficiently central to a person's
self-concept that he or she starts striving to “be” that type of person.
This perspective on identity (“what identity is for”) unites the conceptualizations of identity that can be found in many different streams of
research. Whatever the source of the adopted label that constitutes
one's identity at any moment or in any context, the downstream consequences of identification can be summarized in five principles that
characterize identity-related behavior.
Once a category label is adopted or endowed by an individual as
an identity, factors that increase the salience of this identity will increase the likelihood that identity-based consumer behavior will be
observed (the salience principle). The adoption of a self-label allows
for the—even nondeliberative—transfer of meaning and affect to objects and concepts that are experienced in association with the
self-label (the association principle). Moreover, once an identity is
adopted, the surrounding environment and the people and objects
in it are evaluated for their relevance with respect to the identity,
and a person will think, feel and behave consistently with the identity whenever it is deemed relevant in that situation (the relevance
principle). People are motivated to behave consistently with their
identities, which become the subject of goal striving and will drive
corrective action or thought whenever the identity is at stake (the
verification principle). Finally, because people may hold multiple identities, while each of the identities is not always consistent with all the
others in its implications, identities may conflict. This in turn will motivate cognitive activity and behavior that aim to resolve such conflict
(the conflict principle) either by active attempts to create a harmonized
personal identity or by compartmentalizing identities into separable
partitions of one's life experience.
We proposed that these five principles can serve as useful points of
departure for examining identity-related aspects of important open research questions in marketing and consumer research. To illustrate the
potential of this approach, we deliberately selected future research opportunities in two domains, namely the implications of technology development and of globalization, that are transforming consumer
behavior at the micro (individual) level as well as the macro (societal)
level and for which an identity perspective holds much unrealized potential. Thus, we sought to illuminate how thinking along the lines of
any of the five identity principles discussed in this article can inspire
new and relevant research questions for anyone who tries to understand how consumers and marketers will react to the developments
in these areas. Our selection was meant to be illustrative, rather than
limiting. We suggest that this parsimonious view of identity, with its
focus on trying to understand how people implicate their identities in
their responses to their outside worlds, will allow a better understanding of emerging trends in the marketplace, both from a consumer perspective and from a marketing perspective.
References
Aaker, J. (1997). Dimensions of brand personality. Journal of Marketing Research, 34,
347–357.
Aaker, J., & Akutsu, T. (2009). Why do people give? The role of identity in giving. Journal
of Consumer Psychology, 19, 267–270.
Ahuvia, A. C. (2005). Beyond the extended self: Loved objects and consumers' identity
narratives. Journal of Consumer Research, 32, 171–184.
319
Alden, D. L., Steenkamp, J. -B. E. M., & Batra, R. (1999). Brand positioning through advertising in Asia, North America, and Europe: The role of global consumer culture.
Journal of Marketing, 63, 75–87.
Amiot, C. E., de la Sablonnière, R., Terry, D. J., & Smith, J. R. (2007). Integration of social
identities in the self: Toward a cognitive-developmental model. Personality and
Social Psychology Review, 11, 364–388.
Anderson, J. R., & Bower, G. H. (1973). Human Associative Memory. Washington, D.C.:
Winston & Sons.
Antioco, M., Vanhamme, J., Hardy, A., & Bernardin, L. (2012). On the importance of social integration for minority targeting effectiveness. International Journal of
Research in Marketing, 29, 380–389 (this issue).
Arnett, J. J. (2002). The psychology of globalization. American Psychologist, 57, 774–783.
Arnould, E. J., & Thompson, C. J. (2005). Consumer Culture Theory (CCT): Twenty years
of research. Journal of Consumer Research, 31, 868–882.
Arsel, Z., & Thompson, C. (2011). Demythologizing consumption practices: How consumers protect their field-dependent capital from devaluing marketplace myths.
Journal of Consumer Research, 38, 791–806.
Ashmore, R. D., Deaux, K., & McLaughlin-Volpe, T. (2004). An organizing framework for
collective identity: Articulation and significance of multidimensionality. Psychological
Bulletin, 130, 80–114.
Avery, J. (2012). Defending the markers of masculinity: consumer resistance to brand
gender-bending. International Journal of Research in Marketing, 29, 322–336 (this issue).
Baran, S. J., Mok, J. J., Land, M., & Kang, T. Y. (1989). You are what you buy: Mass mediated judgments of people's worth. Journal of Communication, 39(2), 46–54.
Barreto, M., & Ellemers, N. (2000). You can't always do what you want: Social identity
and self-presentational determinants of the choice to work for a low status group.
Personality and Social Psychology Bulletin, 26, 891–906.
Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research, 15,
139–168.
Belk, R. W., Mayer, R., & Bahn, K. (1982). Developmental recognition of consumption
symbolism. Journal of Consumer Research, 9, 4–17.
Bem, S. L. (1981). Gender schema theory: A cognitive account of sex typing. Psychological
Review, 88, 354–364.
Benet-Martinez, V., Leu, J., Lee, F., & Morris, M. W. (2002). Negotiating biculturalism:
Cultural frame switching in biculturals with oppositional versus compatible cultural
identities. Journal of Cross-Cultural Psychology, 33, 492–516.
Benet-Martinez, V., & Haritatos, J. (2005). Bicultural Identity Integration (Bii): Components and psychosocial antecedents. Journal of Personality, 73, 1015–1049.
Berger, J., & Heath, C. (2007). Where consumers diverge from others: Identity signaling
and product domains. Journal of Consumer Research, 34, 121–134.
Bolton, L. E., & Reed, A., II (2004). Sticky priors and identification based judgments.
Journal of Marketing Research, 41, 397–410.
Brendl, M. C., Chattopadhyay, A., Pelham, B. W., & Carvallo, M. (2005). Name letter
branding: Valence transfers when product specific needs are active. Journal of
Consumer Research, 32, 405–415.
Cadinu, M. R., & Rothbart, M. (1996). Self-anchoring and differentiation processes in
the minimal group paradigm. Journal of Personality and Social Psychology, 70,
661–677.
Callero, P. L. (2003). The political self: Identity resources for radical democracy. In P. J.
Burke, T. J. Owens, R. S. Serpe, & P. A. Thoits (Eds.), Advances in identity theory and
research (pp. 57–70). : Kluwer/Plenum Press.
Carstensen, L. L. (2006). The influence of a sense of time on human development.
Science, 312, 1913–1915.
Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. New York: Cambridge
University Press.
Chan, F., Karbowski, J., Monty, R. A., & Perlmuter, L. C. (1986). Performance as a source
of perceived control. Motivation and Emotion, 10, 59–70.
Chaplin, L. N., & Roedder John, D. (2005). The development of self-brand connections in
children and adolescents. Journal of Consumer Research, 32, 119–129.
Chatman, C. M., & von Hippel, W. (2001). Attributional mediation of in-group bias.
Journal of Experimental Social Psychology, 37, 267–272.
Chen, S. X., & Bond, M. H. (2010). Two languages, two personalities? Examining language effects on the expression of personality in bilingual context. Personality
and Social Psychology Bulletin, 36, 1514–1528.
Cheng, C. -Y., Lee, F., & Benet-Martinez, V. (2006). Assimilation and contrast effects
in cultural frame switching: Bicultural identity integration and valence of cultural
cues. Journal of Cross-cultural Psychology, 37, 742–760.
Cialdini, R. B., Borden, R. J., Thorne, A., Walker, M. R., Freeman, S., & Sloan, L. R. (1976).
Basking in reflected glory: Three (football) field studies. Journal of Personality and
Social Psychology, 34, 366–375.
Claus, B., Geyskens, K., Millet, K., & Dewitte, S. (2012). The referral backfire effect: the
identity threatening nature of referral failure. International Journal of Research in
Marketing, 29, 370–379 (this issue).
Cohen, J. B. (1989). An over-extended self? Journal of Consumer Research, 16,
125–128.
Cohen, J. B., & Reed, A., II (2006). A multiple pathway anchoring and adjustment
(MPAA) model of attitude generation and recruitment. Journal of Consumer Research, 33, 1–15.
Cota, A. A., & Dion, K. L. (1986). Salience of gender and sex composition of ad hoc
groups: An experimental test of distinctiveness theory. Journal of Personality and
Social Psychology, 50, 770–776.
De Langhe, B., Puntoni, S., Fernandes, S., & van Osselaer, S. M. J. (2011). The anchor contraction effect in international marketing research. Journal of Marketing Research,
48, 366–380.
Deaux, K., Reid, A., Mizrahi, K., & Ethier, K. (1995). Parameters of social identity. Journal
of Personality and Social Psychology, 68, 280–291.
320
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
DeMarree, K. G., Wheeler, S. C., & Petty, R. E. (2005). Priming a new identity:
Self-monitoring moderates the effects of nonself primes on self-judgments and behavior. Journal of Personality and Social Psychology, 89, 657–671.
Deshpandé, R., Hoyer, W. D., & Donthu, N. (1986). The intensity of ethnic affiliation:
A study of the sociology of Hispanic consumption. Journal of Consumer Research,
13, 214–220.
Deshpandé, R., & Stayman, D. M. (1994). A tale of two cities: distinctiveness theory and
advertising effectiveness. Journal of Marketing Research, 31, 57–64.
Dong, L., & Tian, K. (2009). The use of western brands in asserting Chinese national
identity. Journal of Consumer Research, 36, 504–523.
Duck, J. M., Hogg, M. A., & Terry, D. J. (1999). Social identity and perceptions of media
persuasion: Are we always less influenced than others? Journal of Applied Social
Psychology, 29, 1879–1899.
Duffy, B. D. (2003). Anthropomorphism and the social robot. Robotics and Autonomous
Systems, 42, 177–190.
Epp, A. M., & Price, L. L. (2008). Family identity: A framework of identity interplay in
consumption practices. Journal of Consumer Research, 35, 50–70.
Escalas, J. E., & Bettman, J. R. (2005). Self-construal, reference groups, and brand meaning. Journal of Consumer Research, 32, 378–389.
Escalas, J. E., & Bettman, J. R. (2003). You are what they eat: The influence of reference
groups on consumers' connections to brands. Journal of Consumer Psychology, 13,
339–348.
Finnel, S., Reed, A., II, Volpp, K., & Armstrong, K. (2012). The joint influence of past,
present and future selves on hypertension medication compliance. Working paper.
Forehand, M. R., & Deshpandé, R. (2001). What we see makes us who we are: Priming
ethnic self-awareness and advertising response. Journal of Marketing Research, 38,
336–348.
Forehand, M. R., Perkins, A., & Reed, A., II (2011). When are automatic social comparisons not automatic? The effect of cognitive systems on user imagery-based
self-concept activation. Journal of Consumer Psychology, 21, 88–100.
Forehand, M. R., Deshpandé, R., & Reed, A., II (2002). Identity salience and the influence
of differential activation of the social self-schema on advertising response. Journal
of Applied Psychology, 87, 1086–1099.
Gao, L., Wheeler, C. S., & Shiv, B. (2009). The ‘shaken self’: Product choices as a means of
restoring self-view confidence. Journal of Consumer Research, 36, 29–38.
Gawronski, B., Bodenhausen, G. V., & Becker, A. P. (2007). I like it, because I like myself:
Associative self-anchoring and post-decisional change of implicit evaluations. Journal of Experimental Social Psychology, 43, 221–232.
Gerritsen, M., Korzilius, H., van Meurs, F., & Gijsbers, I. (2000). English in Dutch commercials:
Not understood and not appreciated. Journal of Advertising Research, 40, 17–31.
Gollwitzer, P., Sheeran, P., Michalski, V., & Seifert, A. (2009). When intentions go public: Does social reality widen the intention-behavior gap? Psychological Science, 20,
612–618.
Greenwald, A. G., Pickrell, J. E., & Farnham, S. D. (2002). Implicit partisanship: Taking
sides for no reason. Journal of Personality and Social Psychology, 83, 367–379.
Gregan-Paxton, J., & Moreau, P. (2003). How do consumers transfer existing knowledge?
A comparison of analogy and categorization effects. Journal of Consumer Psychology,
13, 422–430.
Grier, S. A., & Brumbaugh, A. M. (1999). Noticing cultural differences: Ad meanings
created by target and non-target markets. Journal of Advertising, 28, 79–93.
Grier, S. A., & Deshpandé, R. (2001). Social dimensions of consumer distinctiveness:
The influence of group social status and identity on advertising persuasion. Journal
of Marketing Research, 38(2), 216–224.
Grinstein, A., & Wathieu, L. (2012). Happily (mal)adjusted: cosmopolitan identity and
expatriate adjustment. International Journal of Research in Marketing, 29, 337–345
(this issue).
Higgins, E. T. (1986). Self-discrepancy: A theory relating self and affect. Psychological
Review, 94, 319–340.
Hirschman, E. C. (1981). American Jewish ethnicity: Its relationship to some selected
aspects of consumer behavior. Journal of Marketing, 45, 102–110.
Hoegg, J., Maura, L., Scott, M. L., Andrea, C., Morales, A. C., & Darren, W. (2012). The effect of vanity-sizing on product evaluations. Working paper.
Hogg, M. K., Banister, E. N., & Stephenson, C. A. (2009). Mapping symbolic (anti-) consumption. Journal of Business Research, 62, 148–159.
Hollenbeck, C., & Kaikati, A. M. (2012). Consumers' use of brands to reflect their actual
and ideal selves on Facebook. International Journal of Research in Marketing, 29,
395–405 (this issue).
Hong, Y., Morris, M. W., Chiu, C., & Benet-Martínez, V. (2000). Multicultural minds: A
dynamic constructivist approach to culture and cognition. American Psychologist,
55, 709–720.
Howard, J. A. (2000). Social psychology of identities. Annual Review of Sociology, 26,
367–393.
Hugenberg, K., & Bodenhausen, G. V. (2004). Category membership moderates the
inhibition of social identities. Journal of Experimental Social Psychology, 40(2),
233–238.
Hummert, M. L. (1990). Multiple stereotypes of elderly and young adults: A comparison of structure and evaluations. Psychology and Aging, 5, 182–193.
Ibarra, H. (1999). Provisional selves: Experimenting with image and identity in professional adaptation. Administrative Science Quarterly, 44, 764–791.
Jones, E. E., & Gerard, H. B. (1967). Foundations of social psychology. New York: Wiley.
Jones, J., Pelham, B., Carvallo, M., & Mirenberg, M. (2004). How do I love thee? Let me
count the Js: Implicit egotism and interpersonal attraction. Journal of Personality
and Social Psychology, 85, 665–683.
Kallgren, C. A., Reno, R. R., & Cialdini, R. B. (2000). A focus theory of normative conduct:
When norms do and do not affect behavior. Personality and Social Psychology Bulletin,
26, 1002–1012.
Kernis, M. H., & Goldman, B. M. (2003). Stability and variability in the self-concept and
self-esteem. In M. R. Leary, & J. P. Tangey (Eds.), Handbook of self and identity
(pp. 106–127). New York, NY: Guilford Press.
Kihlstrom, J. F. (1992). Dissociation and dissociations: A commentary on consciousness
and cognition. Consciousness and Cognition, 1, 47–53.
Kirmani, A. (2009). The self and the brand. Journal of Consumer Psychology, 19, 271–275.
Kleine, R. E., Kleine, S. S., & Kernan, J. B. (1993). Mundane consumption and the self:
A social identity perspective. Journal of Consumer Psychology, 2, 209–235.
Kleine, S. S., Kleine, R. E., & Allen, C. T. (1995). How is a possession ‘Me’ or ‘Not Me’?
Characterizing types and an antecedents of material possession attachment. Journal
of Consumer Research, 22, 327–343.
Klesse, A., Goukens, C., Geyskens, K., & de Ruyter, K. (2012). Repeated exposure to the
thin ideal and its implications for the self: two weight loss program studies.
International Journal of Research in Marketing, 29, 355–362 (this issue).
Lane, V. R., & Scott, S. G. (2007). The neural network model of organizational identification. Organizational Behavior and Human Decision Processes, 104, 175–192.
Laverie, D. A., Kleine, R. E., & Kleine, S. S. (2002). Re-examination and extension of
Kleine, Kleine, and Kernan's social identity model of mundane consumption: the
mediating role of the appraisal process. Journal of Consumer Research, 29, 659–699.
LeBoeuf, R. A., Shafir, E., & Bayuk, J. B. (2010). The conflicting choices of alternating
selves,". Organizational Behavior and Human Decision Processes, 111, 48–61.
Lee, J. S. (2006). Linguistic constructions of modernity: English mixing in Korean television commercials. Language in Society, 35, 59–91.
Levy, S. J. (1959). Symbols for sale. Harvard Business Review, 37(4), 117–124.
Lowery, B. S., Unzueta, M. M., Knowles, E. D., & Goff, P. A. (2006). Concern for the
in-group and opposition to affirmative action. Journal of Personality and Social Psychology, 90, 961–974.
Luna, D., Ringberg, T., & Peracchio, L. (2008). One individual, two identities: Frame
switching among biculturals. Journal of Consumer Research, 35, 279–293.
Malhotra, N. K. (1988). Self-concept and product choice: An integrated perspective.
Journal of Economic Psychology, 9, 1–28.
Markus, H., & Nurius, P. (1987). Possible selves: The interface between motivation and
the self-concept. In K. Yardley, & T. Honess (Eds.), Self and identity: Psychosocial
perspectives (pp. 157–172). New York, NY: John Wiley and Sons.
Markus, H., & Ruvolo, A. (1989). Possible selves: Personalized representations of goals.
In L. A. Pervin (Ed.), Goal concepts in personality and social psychology. Hillsdale,
N.J.: Lawrence Erlbaum.
Markus, H., & Wurf, E. (1987). The dynamic self-concept: A social psychological perspective. Annual Review of Psychology, 38, 299–337.
Marques, J. M., Yzerbyt, V. Y., & Rijsman, J. B. (1988). Context effects of intergroup
discrimination: In-group bias as a function of experimenter's provenance. British
Journal of Social Psychology, 27, 301–318.
McConnell, A. R. (2011). The multiple self-aspects framework: Self-concept representation and its implications,". Personality and Social Psychology Review, 15, 3–27.
McFerran, B., Dahl, D. W., Fitzsimons, G. J., & Morales, A. C. (2009). I'll have what she's
having: Effects of social influence and body type on the food choices of others.
Journal of Consumer Research, 36, 915–929.
McGuire, W. J. (1984). Search for the self: Going beyond self-esteem and the reactive
self. In R. A. Zucker, J. Aronoff, & A. I. Rabin (Eds.), Personality and the prediction
of behavior (pp. 73–120). New York: Academic Press.
McGuire, W. J., McGuire, C. V., Child, P., & Fujioka, T. (1978). Salience of ethnicity in the
spontaneous self-concept as a function of one's ethnic distinctiveness in the social
environment. Journal of Personality and Social Psychology, 36, 511–520.
McKenna, Y. A., & Bargh, J. A. (1998). Coming out in the age of the Internet: Identity
“demarginalization” through virtual group participation. Journal of Personality
and Social Psychology, 75, 681–694.
McKenna, Y. A., & Bargh, J. A. (2000). Plan 9 from cyberspace: The implications of
the Internet for personality and social psychology. Personality and Social Psychology
Review, 4, 57–75.
McLuhan, M. (1964). Understanding media. New York: McGraw-Hill.
Mercurio, K., & Forehand, M. (2011). An interpretive frame model of identity dependent learning: The moderating role of content–identity association. Journal of
Consumer Research, 38, 555–577.
Mochon, D., Norton, M. I., & Ariely, D. (2012). Bolstering and restoring feelings of competence via the IKEA effect. International Journal of Research in Marketing, 29,
363–369 (this issue).
Mok, A., & Morris, M. W. (2009). Cultural chameleons and iconoclasts: Assimilation and
reactance to cultural cues in biculturals' expressed personalities as a function of
identity conflict. Journal of Experimental Social Psychology, 45, 884–889.
Mok, A., & Morris, M. W. (2010). Asian-Americans' creative styles in Asian and American situations: Assimilative and contrastive responses as a function of bicultural
identity integration. Management and Organization Review, 6, 371–390.
Muñiz, A. M., Jr., & O'Guinn, T. C. (2001). Brand Community. Journal of Consumer Research,
27, 412–431.
Muñiz, A. M., & Schau, H. (2005). Religiosity in the abandoned Apple Newton brand
community. Journal of Consumer Research, 31, 737–747.
Norman, C. C., & Aron, A. (2003). Aspects of possible self that predict motivation to
achieve or avoid it. Journal of Experimental Social Psychology, 39, 500–507.
North, M. S., & Fiske, S. T. (2012). An inconvenienced youth? Ageism and its potential
intergenerational roots. Psychological Bulletin, 138, 982–997.
Oakes, P. (1987). The salience of social categories. In J. C. Turner (Ed.), Rediscovering the
group: A self-categorization theory (pp. 117–141). New York: Basil Blackwell.
Ogunnaike, O., Dunham, Y., & Banaji, M. R. (2010). The language of implicit preferences.
Journal of Experimental Social Psychology, 46, 999–1003.
Oyserman, D. (2009). Identity-based motivation: Implications for action-readiness, procedural readiness, and consumer behavior. Journal of Consumer Psychology, 19, 250–260.
A. Reed II et al. / Intern. J. of Research in Marketing 29 (2012) 310–321
Oyserman, D. (2007). Social identity and self-regulation. In A. W. Kruglanski, & E. T.
Higgins (Eds.), Social psychology: Handbook of basic principles, Vol. 2, New York:
Guilford Press.
Patrick, V., & Hagtvedt, K. (2012). How to say “No”: Conviction and identity attributions
in persuasive refusal. International Journal of Research in Marketing, 29, 390–394
(this issue).
Pavlenko, A. (2005). Emotions and multilingualism. Cambridge: Cambridge University Press.
Pelham, B. W., Carvallo, M., & Jones, J. T. (2005). Implicit egotism: Current directions.
Psychological Science, 14, 106–110.
Perkins, A., & Forehand, M. (2012). Implicit self-referencing: the effect of non-volitional
self-association on brand and product attitude. Journal of Consumer Research, 39,
142–156.
Pinter, B., & Greenwald, A. G. (2004). Understanding implicit partisanship: Enigmatic
(but genuine) group identification and attraction. Group Processes and Interpersonal
Relations, 7, 283–296.
Puntoni, S., De Langhe, B., & van Osselaer, S. M. J. (2009). Bilingualism and the emotional intensity of advertising language. Journal of Consumer Research, 35, 1012–1025.
Puntoni, S., Sweldens, S., & Tavassoli, N. (2011). Gender identity salience and perceived
vulnerability to breast cancer. Journal of Marketing Research, 48, 413–424.
Reed, A., II (2002). Social identity as a useful perspective for self-concept based consumer research. Psychology and Marketing, 19, 1–32.
Reed, A., II (2004). Activating the self-importance of consumer selves: Exploring identity salience effects on judgments. Journal of Consumer Research, 31, 286–295.
Reed, A., II, & Forehand, M. R. (2012). Identity salience and self-regulation. Working Paper.
Richins, M. L. (1991). Social comparison and the idealized images of advertising. Journal of Consumer Research, 18, 71–83.
Roccas, S., & Brewer, M. (2002). Social identity complexity. Personality and Social Psychology Review, 6, 88–106.
Rokeach, M. (1973). The nature of human values. New York: Free Press.
Rucker, D. D., & Galinsky, A. D. (2008). Desire to acquire: Powerlessness and compensatory consumption. Journal of Consumer Research, 35, 257–267.
Schau, H. J., & Gilly, M. C. (2003). We are what we post? Self-presentation in personal
web space. Journal of Consumer Research, 30, 385–404.
Schlenker, B. R. (1985). The self and social life. New York: Mc Graw-Hill.
Schmitt, B. H., Leclerc, F., & Dube-Rioux, L. (1988). Sex typing and consumer behavior:
A test of gender schema theory. Journal of Consumer Research, 15, 122–128.
Shang, J., Reed, A., II, & Croson, R. (2008). Identity congruency effects on donations.
Journal of Marketing Research, 45, 351–361.
Shavitt, S. (1990). The role of attitude objects in attitude functions. Journal of Experimental Social Psychology, 26, 124–148.
Shavitt, S., & Nelson, M. R. (2000). The social-identity function in person perception:
Communicated meanings of product preferences. In G. Maio, & J. M. Olson (Eds.),
Why we evaluate: Functions of attitudes (pp. 37–57). Mahwah, NJ: Lawrence
Erlbaum Associates, Inc.
Sirgy, J. M., Grewal, D., & Mangleburg, T. (2000). Retail environment, self-congruity,
and retail patronage: An integrative model and a research agenda. Journal of Business Research, 49, 127–138.
Smith, W. B., Bruner, S. J., & White, R. W. (1956). Opinions and personality. New York:
John Wiley & Sons.
Smith, E. R., & Mackie, D. M. (1995). Social Psychology. New York: Worth Publishers.
Solomon, M. R. (1983). The role of products as social stimuli: A symbolic interactionism
perspective. Journal of Consumer Research, 10, 319–329.
Stayman, D. M., & Deshpandé, R. (1989). Situational ethnicity and consumer behavior.
Journal of Consumer Research, 16, 361–371.
321
Steenkamp, J. -B. E. M., & de Jong, M. G. (2012). A global investigation into the constellation of consumer attitudes toward global and local products. Journal of Marketing,
74, 18–40.
Stokburger-Sauer, N., Ratneshwar, S., & Sen, S. (2012). Drivers of consumer-brand
identification. International Journal of Research in Marketing, 29, 406–418 (this
issue).
Strizhakova, Y., Coulter, R. A., & Price, L. L. (2012). The young adult cohort in emerging
markets: Assessing their global cultural identity in a global marketplace. International Journal of Research in Marketing, 29, 43–54.
Swann, W. B., Jr. (1983). Self-verification: Bringing social reality into harmony with the
self. In J. Suls, & A. G. Greenwald (Eds.), Social psychological perspectives on the self,
Vol. 2. (pp. 33–66)Hillsdale, NJ: Erlbaum.
Swann, W. B., Jr. (1997). The trouble with change: Self-verification and allegiance to
the self. Psychological Science, 8, 177–180.
Swann, W. B., Jr., Chang-Schneider, C., & McClarty, K. (2007). Do our self-views matter?
Self-concept and self-esteem in everyday life. American Psychologist, 62, 84–94.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of inter-group behavior. In
S. Worchel, & W. G. Austin (Eds.), Psychology of intergroup relations. Chigago, IL:
Nelson-Hall.
The Economist (2012). The one that didn't get away: Fancy fishing gadgets that work.
http://www.economist.com/node/21557306 (last accessed 5 September 2012)
Toma, C., Hancock, J., & Ellison, N. (2008). Separating fact from fiction: An examination
of deceptive self-presentation in online dating profiles. Personality and Social
Psychology Bulletin, 34, 1023–1036.
van Ittersum, K., & Wong, N. (2010). The Lexus and the olive tree? Trading off between
global convergence and local divergence. International Journal of Research in Marketing, 27, 107–118.
Verrochi-Coleman, N. M., & Williams, P. (2012). Feeling like myself: Emotion regulation and social identity. Working paper.
Visser, P., & Mirabile, R. R. (2004). Attitudes in the social context: The impact of social
network composition on individual-level attitude strength. Journal of Personality
and Social Psychology, 87, 779–795.
Wheeler, S. C., DeMarree, K. G., & Petty, R. E. (2007). Understanding the role of the
self in prime-to-behavior effects: The active-self account. Personality and Social
Psychology Review, 11, 234–261.
Wheeler, S. C., Petty, R. E., & Bizer, G. Y. (2005). Self-schema matching and attitude
change: Situational and dispositional determinants of message elaboration. Journal
of Consumer Research, 31, 787–797.
White, K., & Dahl, D. W. (2007). Are all out-groups created equal? Consumer identity
and dissociative influence. Journal of Consumer Research, 34, 525–536.
Wilder, D. A., & Shapiro, P. N. (1984). Role of out-group cues in determining social identity. Journal of Personality and Social Psychology, 47, 342–348.
Williams, J. D., & Qualls, W. J. (1989). Middle-class black consumers and intensity of
ethnic identification. Psychology and Marketing, 6, 263–286.
Winterich, K. P., Zhang, Y., & Mittal, V. (2012). How political identity and charity positioning increase donations: Insights from Moral Foundations Theory. International
Journal of Research in Marketing, 29, 346–354 (this issue).
Wooten, D. B. (1995). One-of-a-Kind in a Full House: Some consequences of ethnic and
gender distinctiveness. Journal of Consumer Psychology, 4, 205–224.
Zhang, Y., & Khare, A. (2009). The impact of accessible identities on the evaluation of
global versus local products. Journal of Consumer Research, 524–537.
Zou, X., Morris, M. W., & Benet-Martínez, V. (2008). Identity motives and cultural priming:
cultural (dis)identification in assimilative and contrastive responses. Journal of Experimental Social Psychology, 44, 1151–1159.
Intern. J. of Research in Marketing 29 (2012) 322–336
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
Defending the markers of masculinity: Consumer resistance to brand gender-bending
Jill Avery ⁎
Simmons School of Management, 300 The Fenway, M-336, Boston, MA 02115, USA
a r t i c l e
i n f o
Article history:
First received in 17, June 2011 and was under
review for 4 months
Available online 23 September 2012
Keywords:
Brand management
Brand communities
Gender
Masculinity
Self/brand connection
Identity
Consumer behavior
a b s t r a c t
I study the Porsche Cayenne SUV launch to ethnographically analyze how men consuming a gendered brand
respond to perceived brand gender contamination. The consumers' communal gender work in a Porsche
brand community is analyzed to uncover brand gender contamination's effects on the identity projects of
consumers, the brand as an identity marker, and the prevailing gender order in the group. Through the promulgation of gender stereotypes, Porsche owners stratify themselves along gender lines and create an
ingroup that is sharply defined by masculinity and an outgroup that is defined by femininity. The construction of social barriers limits access to Porsche's meanings to those who achieve masculine ideals and causes
the SUV owners to resort to hyper-masculine behaviors to combat exclusion. The consumers' gender work
reverses the firm's efforts to gender-bend the brand, reinstates Porsche as a masculine marker, and reifies
particular definitions of masculinity in the community.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
“It's a boy!” claims Volkswagen in the 2012 relaunch of its Beetle.
The Beetle's more aggressive design was devised to bring men into
the female-skewed customer base. Harley-Davidson once claimed
that it made “big toys for big boys”; today, the company is trying to
attract women. Marketers are gender-bending their brands, taking
products that had been targeted to one sex and targeting them to
the other. In the postmodern era, both men and women have engaged
in gender-bending consumption, co-opting the consumption practices and products of the opposite sex to play with definitions of gender and support new ideologies. Consumer researchers argue that we
are in a post-gender period in which the stark lines that have historically divided men's and women's consumption are blurring (Firat,
1994; Patterson & Elliott, 2002). Have we finally reached a time
when gender does not matter in consumption? Can brands transcend
their gendered roots and become neither masculine nor feminine, but
an androgynous mixture of both?
Throughout history, our consumption has been gendered and consumers have relied on gendered products and brands as props to perform
their gender identities. In 1994, Fischer and Gainer (1994: 101) hypothesized that gendered consumption was ubiquitous and enduring:
If we find…that most consumption domains are gendered, and not
particularly susceptible to revision in this regard, we will begin to
have a better appreciation of the ways that consumption practices
⁎ Tel.: +1 617 521 3853.
E-mail address: jill.avery@simmons.edu.
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.04.005
are shaped by and support other mutually reinforcing social practices that contribute to the current gender order.
Gender still takes center stage in many brand narratives, and
brands often find it difficult to attract the opposite sex to brands associated with one sex. Pepsi and Coke launched new brands to attract
men to diet soda after their efforts to sell Diet Pepsi and Diet Coke
to men were unsuccessful. PepsiMax, Coke Zero, and Dr. Pepper Ten
work hard to distance themselves from feminine diet sodas by
claiming that they are the “diet cola for men” and “it's not for
women.” When entering the more feminine body wash category, Gillette infused its advertising with masculine images of power tools and
footballs and encouraged men to “wash like a man, feel like a man,”
while Dove for Men featured men in traditional roles, acting as a protector and lifting weights, reassuring men that their masculinity
would be preserved if they used the product. Despite consumers'
gender-bending consumption, gender appears to remain an important organizing construct in branding.
In this article, I analyze consumers' lived experience and collective
response to the gender-bending of their brands by exploring consumption in situ to understand the role that gender plays in today's
postmodern world. I explore two research questions: How do consumers respond to the appropriation of their brands by the opposite
sex and how does their response affect the brands' potency as a gendered identity marker? I first illuminate how consumers use brands
to enact gender and why the gender contamination of brands occurs.
Then, I describe my design for the study of consumers' response to
the gender-bending of Porsche via the launch of an SUV that attracted
women to the masculine brand. I outline the SUV's identity threat
and analyze the existing owners' collective impression management
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
practices. I show that the consumers who rely on a brand's gendered
identity respond to gender-bending by fighting for their brand rather
than discarding it. In fighting, the consumers strengthen the brand's
gendered meanings, paradoxically increasing its potency as an identity marker as the other sex encroaches. The consumers' gender work
reverses the firm's efforts to gender-bend the brand, reinstates the
brand as a masculine marker, and reifies the particular definitions
of masculinity in the brand community. I conclude with a discussion
of the theoretical and managerial implications of gender-bending
branding.
2. Theoretical foundations
2.1. Enacting gender through consumption
Our most salient and central identity in the multitude of identities
that define us is our sense of ourselves as being male or female. We
rely heavily on gender to define ourselves and to classify and understand others (Deaux & LaFrance, 1998). Gender, unlike sex, is not biologically determined; instead, it is a socially accomplished, culturally
constituted ongoing construction project. We are not granted a gender at birth; we perform our gender through situated, symbolic social
interaction (West & Zimmerman, 1987). We construct gender in social interactions by tailoring our actions to conform (or not) to the
normative conceptions of masculinity and femininity that exist in
our culture (Gherardi, 1995) by choosing from a cultural repertoire
of gendered behaviors (Wetherell & Edley, 1999). These practices, in
turn, create a social gender display that reinforces (or resists) the prevailing conceptions of masculinity and femininity (Butler, 1990;
Lorber, 1994). Although many different forms of masculinity and
femininity exist concurrently in a particular culture (Carrigan,
Connell, & Lee, 1985), one form is held as the taken-for-granted hegemonic standard. All people might not adhere to the hegemonic definitions, but these definitions inform people's actions and how others
interpret them (Spence & Helmreich, 1978).
Gender display is often accomplished through the use of props
(Lorber, 1994). We create, enhance, and accomplish our gender identities through consumption and, thus, our possessions function as
symbolic gender identity markers. Penaloza (1994) suggests that
there are separate masculine and feminine consumer cultures that
define what is appropriate (and inappropriate) for each gender to
purchase and consume, while others support that possessions,
brands, and consumption behaviors and practices are gendered (cf.
Fischer & Arnould, 1990; Sherry, Kozinets, Duhachek, et al., 2004;
Wallendorf & Arnould, 1991). Gendered brands contain either masculine or feminine identity meanings that are socially shared among the
members of a culture. We adorn our gender displays with these
brands as tangible markers; gendered brands help materialize gender, enlivening who we are as men or women. Brands populate the
universe of choices that we make to express our gender, and men
and women generally prefer and choose brands, possessions, and activities that reflect their gender identity (as reviewed in Palan, 2001;
Stern, 1988).
2.2. The constraining forces of gendered consumption
The gendering of brands constrains what we buy, given the strong
polarization of gender in many cultures, where what is feminine is
understood to be the antithesis of what is masculine (Bem, 1993).
When we do gender “appropriately” through consumption, our practices sustain and reinforce the prevailing roles (Fischer & Gainer,
1994; Gherardi, 1995) and people understand who we are or would
like to be. When we engage in gender-bending practices by coopting the products of the opposite sex, we are often called to account
and our gender identity is questioned (Kramer, 2005). Through our
performative acts, we reinforce the regulative gender discourses
323
(Butler, 1990) the cultural narratives that outline proper behavior
(Borgerson & Schroeder, 2004).
The political nature of gender has a differential effect on men and
women. Androcentrism reigns in most cultures, where masculinity is
more highly valued and is seen as normal, while femininity is deviant
from and less than masculinity (Kramer, 2005). Hegemonic masculinity in many cultures has been defined as misogynistic and patriarchal
(Connell, 1993) and includes “the dread of and the flight from
women,” (Donaldson, 1993: 645). Women and women's things signify not only femininity but also a lack of masculinity. The most important rule of manhood is to not be like a woman; real men “must never,
never resemble women, or display strongly stereotyped feminine
characteristics” (Brannon, 1976:14). Most men manage their masculinity through consumption to ward off fears that others will see
them as effeminate or gay (Carrigan et al., 1985; Kimmel, 1996). For
men striving to achieve masculinity, success largely depends upon renouncing the feminine (Conway-Long, 1994; Cornwall & Lindisfarne,
1994), and this constraint circumscribes male consumers' choices to
those that fall within the regulative boundaries (Borgerson &
Schroeder, 2004; Rinallo, 2007).
Research suggests that using feminine brands carries a greater stigma for men than using masculine brands does for women. While both
men and women prefer brands that express their own gender identity,
women are more likely to purchase masculine brands than men are
likely to purchase feminine brands (Fry, 1971; Vitz & Johnson, 1965;
Worth, Smith, & Mackie, 1992). Penaloza (1994: 366, 374) attributes
this preference to differential power dynamics at work:
Because most of those with money and power are men, the crossing of women into the male domain by wearing clothes associated
with the masculine is viewed as rational and is naturalized,
whereas for men, to cross into the feminine domain by wearing
clothing associated with the feminine is to willingly pursue its
stigma and downward mobility, which is viewed as irrational
and it goes against individual male privilege and the male dominated culture.
2.3. Gender-bending consumption and gender contamination
First and second wave feminists appropriated men's consumption
symbols to fight for gender equality by tossing aside the vestiges of
femininity (Hollows, 2000; McCracken, 1988) and adopting short
haircuts, cigarette smoking, and masculine fashion styles. Even the
most masculine of enclaves, the Harley-Davidson brand community
(Martin, Schouten, & McAlexander, 2006) and The Citadel, the allmale military college (Addelston & Stirratt, 1996), were infiltrated
by women, many of whom used the oppositional tension generated
by their consumption to create alternative femininities.
The emergence of the metrosexual discourse in the 1990s appeared to
usher in the deconstruction of gender and a move toward the androgynous consumption that was promised by postmodern theory in which
bricoleur consumers circumvent constraining categories by composing
a multifaceted and creative self through their consumption (Holt &
Thompson, 2004; Thompson & Haytko, 1997). The metrosexual was a
media-proliferated masculine ideology (Simpson, 1994) that emerged
as an alternative to traditional masculinity and gave license to men to
pursue consumption activities that were never before deemed acceptable. Young, urban, heterosexual men began buying products that had
been associated with women or homosexual men (Bordo, 1999; Crane,
1999). Through their countercultural consumption, metrosexuals
redefined the boundaries of masculine consumption by disarticulating
practices that were previously associated with women and homosexuals and resignifying them as appropriate for heterosexual men.
However, given the political and power disparity that still exists
between men and women, women's gender-bending consumption
324
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
might be more dangerous to men than men's gender-bending consumption is to women. The gender contamination of brands might
occur as men searching for masculine distinction work to avoid
brands that are used by women and/or abandon previously masculine
brands that have been infiltrated by women (McCracken, 1988;
Tuncay & Otnes, 2008). Anthropologists have documented genderrelated totems and talismans from multiple cultures and historical
periods that are associated with men, which women are forbidden
to see, touch, or use (Herdt, 1981). Addelston and Stirratt (1996) document a contemporary masculinity marker, the Citadel graduation
ring, that is contaminated by the touch of a woman.
The symbolic interactionists (Mead, 1934) theorized that the
meaning of a symbolic gesture, such as the use of a brand, is a social
product revealed in an audience's response to it. Meaning is dynamic
and socially constituted. The fact that a brand's identity meanings are
common knowledge is a necessary condition for the use of the brand
as an identity marker; consumers not only have to interpret the
brand's identity meanings for themselves, but they have to know
that other people in their relevant social audiences will interpret
them in the same way.
Gender contamination threatens this shared interpretation of a
brand's meaning, and people might find it more difficult to create
their identity through consumption as the practices, products, and
brands that are traditionally associated with masculinity or femininity are appropriated by the opposite sex. Gender-bending branding
might generate reflected appraisals that are inconsistent with the
message that the brand's customers are trying to send. People feel
“out of face” (Goffman, 1959) when their reflected appraisals do not
reflect who they want to be. Feeling out of face is psychologically uncomfortable and motivates identity practices to alleviate the incongruence (Burke, 1991). Distress increases when the identity in question is
a central part of the person's overall identity and when the person is
highly committed to the identity, as is often the case with gender
(Burke, 1991; Swann & Ely, 1984). Saving face (Goffman, 1959) involves
practices designed to influence the audience so that their reflected appraisals move back into congruence with the desired identity.
Under identity threat, people tend to immerse themselves in social environments in which their desired identities are supported,
and they isolate themselves from the environments in which they
are not (McCall & Simmons, 1966). Research on the Star Trek
(Kozinets, 2001), Apple Newton (Muniz & Schau, 2005), and Hummer
(Luedicke, Thompson, & Giesler, 2009) brand communities shows
that consumers look for the collective support of their brand brethren
to defend against stigmatizing brand meanings. In today's world, saving face is likely to be both an individual and a collective practice undertaken in a brand's communities.
3. Method
Through a multimethod design, I analyzed existing consumers' phenomenological response to the gender-bending of their brand. This design allowed me to analyze the consumers' construction of their
identity using brands as the brands moved through two moments of
cultural production, following methods used by cultural historians
(Denning, 1987; Illouz, 1997; Radway, 1984) and brought to consumer
research by Holt and Thompson (2004). First, I observed the consumers'
construction of identity through brands as it intersected with mass culture discourses, which serve as a repository of masculine meanings that
are available for use. Second, I observed the consumers' construction of
identity as it was enacted in day-to-day consumption practices.
criteria. First, the Porsche brand serves as a strong example of a gendered brand, with a positioning described by prior researchers as “extreme masculinity” (Stuteville, 1971). Cars are an important part of
many men's extended selves, and Belk (2004: 273) has claimed that
“extreme identification with automobiles is a predominantly male
fascination… There are few possessions as important to many adult
American men as their cars.” Some men experience their cars as a sexual extension of themselves (Belk, 2004; Lane & Sternberg, 1985).
Sports cars, in particular, can serve as phallic symbols (Thompson &
Holt, 2004), which increase the probability that identity concerns will
be triggered if the gender meanings associated with a sports car brand
are altered. Finally, the press surrounding the launch of the Porsche
Cayenne suggested that the brand's gender identity meanings were
changing, “Porsche goes soccer mom… Has Porsche lost its soul?”
claimed Forbes, while The New York Times found, “There may be no vision more heretical to a testosterone-poisoned 911 owner than that of
a suburban mother loading groceries into the back of her Porsche after
dropping her children off at soccer practice.”
3.2. Cultural discourse analysis
For some brands, especially those that have achieved the status of a
cultural icon, gender identity meanings are reflected in and/or emerge
from the mass mediated image that the brand enjoys in the culture
(Holt, 2004). To understand brand meaning as conveyed by mass
media depictions of the brand, one must understand the cultural
stories in which the brand performs a supporting role by interpreting
the symbolism of the brand's products as presented in cultural texts
(Hirschman, 1988; Holbrook & Grayson, 1986). To understand
Porsche's gender identity meanings, I conducted a cultural analysis of
the Porsche brand in popular culture. I analyzed major American
movies and television programs over a forty-year time period (see
Table 1) that featured Porsche vehicles and extracted the identity characteristics shared by the characters driving them. I began my analysis
using lists constructed by the brand community members and
supplemented them with information from an online movie database.
3.3. Brand identity meaning elicitation experiment
To better understand Porsche's existing gender identity meanings
as well as the new gender identity meanings brought by the SUV, I
conducted an experiment. First, I elicited Porsche brand and SUV category gender identity meanings from consumers using an unaided
elicitation task: “describe the type of person who drives a Porsche
(an SUV).” The elicited meanings formed the basis of a study administered to MBA students at a large U.S. business school. The study
consisted of a randomly assigned, between-subjects design that manipulated which portion of the Porsche product line was shown.
Thirty-nine participants (sports car group) were given the Porsche
brand logo and the photographs and names of the sports cars in the
Porsche product line. Thirty-four participants (SUV group) were
given the Porsche brand logo and the photograph and name of the
Porsche Cayenne SUV. Gender and car/SUV ownership were balanced
across the conditions, and no significant differences across groups
were observed. Participants were asked to “describe the type of person
who purchases and drives these Porsche vehicles” by completing a set of
23 semantic differential items consisting of bipolar scales anchored by
contrasting meanings on each end of a 7-point scale (e.g., male/female).
Additionally, participants used a 7-point Likert scale to measure their
agreement with each of five additional identity meanings for which direct opposites did not exist (e.g., soccer mom).
3.1. Natural experiment of observation
3.4. Brand community netnography
Porsche shocked the world when it announced the launch of the
Porsche Cayenne SUV, the first non-sports car in its seventy-year history. This event was chosen for this study based on the following
Last, an historical archive of unfolding consumer-to-consumer conversations about the launch in an online Porsche brand community
325
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
Table 1
Cultural analysis of Porsche identity meanings.
Movie/television show
Porsche
(Actor or role)
Identity characteristics
Harper (1966)
The Quiller Memorandum (1966)
Downhill Racer (1969)
Le Mans (1971)
The Return of the Dragon (1974)
Grand Theft Auto (1977)
Annie Hall (1977)
FM (1978)
Good Guys Wear Black (1978)
Private Benjamin (1980)
Middle Age Crazy (1980)
American Gigolo (1980)
Dogs of War (1980)
Arthur (1981)
Modern Problems (1981)
Modern Romance (1981)
Condorman (1981)
The Soldier (1982)
48 Hrs. (1982)
Risky Business (1983)
Scarface (1983)
The Big Chill (1983)
Breathless (1983)
Class (1983)
Flashdance (1983)
Bachelor Party (1984)
Sixteen Candles (1984)
LA Law (1986)
Highlander (1986)
Top Gun (1986)
The Hidden (1987)
Less Than Zero (1987)
Three for the Road (1987)
No Man's Land (1987)
Mystic Pizza (1988)
Bull Durham (1988)
Tequila Sunrise (1988)
Dirty Rotten Scoundrels (1988)
Her Alibi (1989)
Three Fugitives (1989)
Seinfeld (1989–1998)
The Simpsons (1989–pres)
Mr. Destiny (1990)
Opportunity Knocks (1990)
Pacific Heights (1990)
Father of the Bride (1991)
Highlander II (1991)
Bill and Ted's Bogus Journey (1991)
Doc Hollywood (1991)
Patriot Games (1992)
The Player (1992)
Terminal Bliss (1992)
Father Hood (1993)
The Specialist (1994)
Angels in the Outfield (1994)
Beverly Hills Cop 3 (1994)
Honor Thy Father and Mother (1994)
Nine Months (1995)
Thinner (1996)
From Dusk till Dawn (1996)
A Time to Kill (1996)
Kingpin (1996)
Kiss The Girls (1997)
Playing God (1997)
Batman and Robin (1997)
A Life Less Ordinary (1997)
Civil Action (1998)
Dirty Work (1998)
Office Space (1999)
Friends (1999)
The Thirteenth Floor (1999)
Eye of the Beholder (1999)
For the Love of the Game (1999)
Deuce Bigalow, Male Gigolo (1999)
Disney's the Kid (2000)
356
356
911
911
356
911
911
911
930
911
928
911
356
924
911
911
911
930
Speedster
928
928
911
356
911
911
911
944
911
356
Speedster
928
911
911
911
911
911
911
911
356
911
911
911
550
911
911
911
911
911
Speedster
911
911
928
928
968
911
968
911
911
928
964
356
993
911
356
550
928
911
Paul Newman
George Segal
Robert Redford
Steve McQueen
“Bruce Lee”
Cynical, cool P.I.
British spy, loner, unorthodox, dangerous, deadly
Quietly cocky ski racer
Race car driver
Martial artist, fighter, good guy, super-hero
Snooty rich boy
Psycho, strange, suicidal, Yankee
Radio station manager
Ex U.S. army commando
Wealthy, self-absorbed, handsome, adulterer
40, having affair & mid-life crisis
Bad guy pimp
Mercenary, revolutionary hero
Wealthy, Lazy, playboy
Shady, egotistical author
Jealous, self-centered, single, successful film editor
Bad guys
Anti terrorist soldier
Foul-mouthed slick convict
“daddy's car”, smart, sexy, naïve, wealthy, Princeton
Cocaine dealer, unhinged, tough, violent, ruthless, savvy, mob boss
Yuppies, midlife crisis
Slick Las Vegas hustler with a good side
Suave, sophisticated, wealthy, mischievous rebel
Construction company boss
Trust Fund Baby, Military
High school jock, sensitive
Sleazy, womanizing divorce lawyer
Romantic action hero, great powers, strength, immortal
Powerful, smart, pretty, in charge
FBI Agent
Ivy League preppy, wealthy
Steals it
Wealthy, thief, party-guy, icy smooth, risk taker, unhinged
Wealthy, uber-WASP, rebellious law school drop-out, pretty
Star pitcher, sweet but stupid
Drug dealer, evil, handsome
Wealthy, suave con man
Writer of bad novels looking for love
Steals it
Wannabe owner, urban, funny, nerdy
Scary, crazy clown, criminal, boozer, gambler, dirtbag
Mid-life crisis, unhappy, powerless man relives youth
Joy ride
Manipulative, deadly con man
Upper middle class dad in mid-life crisis when daughter leaves nest
Romantic action hero, great powers, strength, immortal
Steal it for a joyride
Hot shot big city doctor
Doctor, mother
Trying to pick-up girl
Affluent, handsome single guy fighting over a girl
Small time criminal
Manic villain, psychotic hired gun with a vendetta
Baseball manager, skeptic, heart of gold, father
Rebel, irreverent scammer
Young, kills parents, buys 911 with insurance money
Commitment phobic single psychiatrist becomes a dad
Doctor
Bank robber on the lam
Energetic, ambitious, razor-sharp law student
Bad guy, goon
Detective and author, father, cool and professional, widower
LA surgeon on drugs joins the mob
Handsome hero, rich, brave
Dentist
Lawyer, desires material success, goes into ruin to help
Snotty rich guy
Suspender wearing yuppie boss
Single actor who pretends it is his to pick up girls
Young dashing scientist
Female serial killer
40 y/o pitcher in mid-life crisis
Debonair gigolo
High powered image consultant
911
996
911
928
993
Boxster
996
Christopher Walkin
“Jeff Dugan”
Chuck Norris
“Henri Alan Tremont”
Bruce Dern
Christopher Walken
Dudley Moore
Dabney Coleman
Albert Brooks
Ken Wahl
Eddie Murphy
Tom Cruise
Al Pacino
Richard Gere
Rob Lowe
“Nick”
Tom Hank's rival
“Jake Ryan”
“Arnie Becker”
Connor Macleod
Kelly McGillis
“Good alien”
Andrew McCarthy
Charlie Sheen
Charlie Sheen
Charles Gordon Windsor
Tim Robbins
“Mac”
Tom Selleck
Martin Short
“Jerry”
“Krusty the Clown”
James Belushi
Dana Carvey
Michael Keaton
Steve Martin
Connor MacLeod
“Bill and Ted”
Michael J. Fox
Anne Archer
Luke Perry
Patrick Swayze
James Woods
Danny Glover
Eddie Murphy
Erik Menendez
Hugh Grant
George Clooney
Sandra Bullock
“Stan”
Morgan Freeman
David Duchovny
“Batman”
John Travolta
“Bill Lumberg”
“Joey”
Craig Bierko
Ashley Judd
Kevin Costner
“Antoine”
Bruce Willis
(continued on next page)
326
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
Table 1 (continued)
Movie/television show
Porsche
American Psycho (2000)
Hollow Man (2000)
Mission Impossible 2 (2000)
Space Cowboys (2000)
Along Came a Spider (2001)
Blow (2001)
Friends (2001)
Hannibal (2001)
James Dean (2001)
Legally Blonde (2001)
Life as a House (2001)
Smallville (2001)
Die Another Day (2002)
Enough (2002)
Anything Else (2003)
Intolerable Cruelty (2003)
Old School (2003)
The Simple Life (2003)
Spy Game (2001)
Barbershop 2: Back in Business (2004)
Cellular (2004)
Cellular (2004)
In Good Company (2004)
Be Cool (2005)
Bewitched (2005)
Just Friends (2005)
The Transporter 2 (2005)
Big Momma's House 2 (2006)
The Sopranos (2006)
The Sopranos (2006)
Cars (2006)
911
996
996
356
911
911
911
Boxster
550
Boxster
911
996
993
996
356
996
911
Boxster
912
Boxster
Cayenne
911
911
996
911
Cayenne
911
Cayenne
Boxster
996
(Actor or role)
Kevin Bacon
Tom Cruise
Clint Eastwood
Morgan Freeman
Johnny Depp
“Jack Geller”
Ray Liotta
James Dean
Reese Witherspoon
Lex Luthor
“Zao”
Woody Allen
George Clooney
Craig Killborn
Paris Hilton
Robert Redford
“competitor”
Kim Basinger
Topher Grace “Carter”
Harvey Keitel
Will Ferrell
Ryan Reynolds
Martin Lawrence
“Carmella”
“Rosalie”
“Sally Carrera”
was used to reconstruct a longitudinal netnography designed to capture
the existing consumers' response to the SUV. The four-year longitudinal
design allowed for the capture of consumers' responses in real time,
rather than retrospective reflection, and permitted the observation of
both the immediate and the longer term responses. The tenets of
netnographic community selection (Kozinets, 2002) were used to identify a specific Porsche online community for study. Rennlist, an international community of Porsche enthusiasts, was chosen because it is the
largest automotive special interest website on the internet, enjoying
65,000 subscribers at the time of this study. Rennlist features a large
number of members who post to the consumer-to-consumer web forums, and it consists of a diverse group of posters including current
and prospective Porsche owners from 29 countries. At the time of this
study, over 36,000 participants were talking with each other on the
Rennlist web forums, while their conversations were being read by an
estimated 200,000 additional people.
Car brands often elicit strong, active brand community participation among their consumers (Algesheimer, Dholakia, & Herrmann,
2005). Rennlist is a brand community that exhibits the three markers
of brand community (Muniz & O'Guinn, 2001). First, the Rennlist
community exhibits a consciousness of kind. Strong ties exist
among the members who experience a group identity; many have
participated in the community for many years and communicate
with each other daily. Second, the Rennlist community has numerous
rituals and traditions that include greeting rituals, norms that govern
online conversations and racing etiquette, status monikers, traditional ride events, clubs consisting of members who share a particular interest, and tangible markers such as decals and t-shirts. Finally, the
members of the Rennlist community have a shared sense of moral responsibility toward their fellow community members and toward the
community itself. Community members log on daily to ask for purchase or technical advice but also to discuss how to navigate the social challenges of owning a Porsche.
I unobtrusively observed the conversations on the Rennlist web forums as a nonparticipating observer. Fifty-five months of historical
data spanning from May 2001 through November 2005 and consisting
Identity characteristics
Yuppies, Urban, Wealthy
Brilliant, cocky scientist, insensitive egoist, lack of ethics, narcissistic misogynist
Heroic, brave, smart, risk taker spy
Hot shot test pilot, astronaut
Retired detective/psychologist
Cocaine smuggler, poor roots, wealthy, party-animal, addict
Dorky middle aged father, lawyer, inappropriate comments, Jewish
FBI Agent
Sexy, dangerous, rebel
Beautiful ditzy blonde sorority girl law student
Rich teenager
Trust fund, evil criminal mastermind, angry, vengeful, playboy
Bad guy
Wealthy wife beater
Paranoid, odd professor
Wealthy, has it all, LA divorce attorney in midlife crisis
Boyfriend who fools around with the waitress
Bubble headed blonde heiress
Retiring CIA agent
Barbershop owner
High school biology teacher
Obnoxious, smarmy lawyer
Young, business school prodigy, dumped by his wife
Sleazy record label executive
Self-absorbed actor
Rich, shy, good looking music executive, smooth talking lady killer
“bad guy”
Wife pressures To give up 911 for SUV, cross dressing FBI agent
Mob boss' wife, soccer mom, suburban, Italian
Mobster's wife, newly single, Italian, suburban
Porsche as anthropomorphized woman, lawyer, rebel's love interest
of 181,000 conversations (threads) were manually combed through
and read during a five-month netnographic immersion in the archives.
Field notes were written throughout to capture any emerging insights
based on the identification of recurrent behaviors across people. Emerging grounded theory was tested by returning to the data to scrutinize
constructs and their relationships, to unearth counterexamples, and to
affirm, challenge, elaborate, and refine the conceptual model (Glaser &
Strauss, 1967). The data were reread, and conceptual constructs and relationships were iteratively adjusted until sufficient interpretative convergence was achieved. Following the conclusion of the analysis, I
purchased a Porsche 911, participated in dealer and driver education
events with other owners, and began conversing on Rennlist to add participant–observer ethnographic data to the analysis, following the participant observation method pursued by Muniz and Schau (2005).
Finally, I conducted five three-hour member check interviews with
Porsche owners to confirm and refine my understanding of the online
conversations.
4. Results
4.1. Porsche as a marker of masculinity
My cultural analysis of the brand yielded shared identity meanings that are evident across all of the characters driving Porsche vehicles in American movies and television programs. Porsche drivers are
overwhelmingly male: 91% of the characters portrayed were men.
Porsche helps men express their masculinity by tapping into two
strains of masculinity that are present in contemporary culture and
summarized by Holt and Thompson (2004): the “Breadwinner” and
the “Rebel.” Contemporary men often use their consumption practices to tack between these two myths to shape their masculinity
into a new gender ideology that exhibits a combination of the two,
the man-of-action hero. Breadwinner masculinity is characterized by a
man who “pursues organization success by following rules, [and who is]
responsible, respectable, rational, reserved, [a] community pillar, safe,”
while rebel masculinity is characterized by a man who “pursues
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
autonomy by defying institutional rules, [and who is] creative, confident,
self-directed, potent, youthful, dangerous” (Holt & Thompson, 2004:
437). The Porsche brand encompasses many of the expectations associated with masculinity that underpin the breadwinner and the rebel myths,
such as “No Sissy Stuff” (anti-feminine), “The Big Wheel” (wealth and
power), “The Sturdy Oak” (toughness and dependability), and “Give
'Em Hell” (aggression and rebellion) suggested by Brannon (1976), the
“Playboy” or “Macho Man” suggested by Lindsey (1997) and Osgerby
(2001), and the “Peter Pan” (childlike and irresponsible) summarized
by Register (2001).
Wealthy, professional white collar men drive Porsche sports cars.
These men are doctors, lawyers, scientists, and businessmen, living
successful breadwinner lives. Their Porsche sports cars are external
markers of their success and wealth. A darker side to Porsche is also
evident that helps the brand to express rebel masculinity. Villains,
such as cocaine smugglers, hustlers, and criminals drive Porsche
sports cars, giving the brand a risky, dangerous edge. These Porsche
men are cocky, irreverent, rebellious, risk-seeking loners who buck
the traditional establishment. Tough “good guys” drive Porsche sports
cars, including detectives, military men, CIA operatives, and spies.
Porsche men are sexy and sexual, both in appealing and not so appealing ways, as adulterers or womanizers. These men are often egotistical,
narcissistic, or spoiled brats and coping with midlife crises during which
they engage in irresponsible behavior. Movies illustrate the symbolic
value of a Porsche as an antidote for aging and the resolution of
mid-life crises. They also use it as a symbol for the loss of manhood associated with marriage and family in which characters feel pressured to
give up their Porsche sports car for a more sensible family car.
I compared the findings of the cultural analysis with my
netnographic analysis of the brand community to see if the lived experiences of the consumers were consistent with the mass mediated
brand meanings surrounding masculinity and to understand how the
identity meanings of Porsche contained in the mass culture discourse
are interpreted by men in everyday consumption practice, following
Holt and Thompson (2004). The brand community frequently engages with mass culture discourses in their communal discussions
of their brand, creating a discursive field fueled by both the mass mediated meaning and their own consumption situated personalized
meanings. In conversations, the movies that contain Porsche cars
are dissected, and media-incited stereotypes are identified and
then grounded by personal stories.
The brand community revels in the masculine exclusivity of
Porsche, and masculine ideologies are often at the crux of the consumers' stories about their ownership experience. The members frequently remind each other that 85% of Porsche owners are men. To
the members of the brand community, buying a Porsche is often
seen as an assertion of one's manhood, whether it is the first purchase
of a used Porsche following an adolescence spent in a room filled with
Porsche memorabilia or whether it is stubbornly holding onto a
cherished Porsche in the face of spousal disapproval. Members of
the brand community struggle against giving up their Porsche cars
when they become fathers, and they mourn the cars that they have
sold to fulfill family obligations. The stereotype of a Porsche as an antidote to a man's midlife crisis is openly acknowledged, and the brand
community abashedly accepts it as accurate, with many owners citing
a midlife crisis as the impetus for their purchase. Porsche cars are
powerful antidotes to the anxieties about aging and the impending
decrease in virility that characterize many midlife crises (Aldwin &
Levenson, 2001) because they allow their owners to regress back to
an earlier age in which they enjoyed few responsibilities and entanglements and when they were in their sexual prime.
Porsche cars are viewed by many in the community as tools to attract women. Throughout their conversations, sexual meanings are
introduced and debated. Some members wear these meanings proudly, while others are somewhat embarrassed by them. The comments
below capture owners' perceptions of the transformative sexual
327
power of their cars. Women's responses to their cars bolster their
masculinity and allow them to engage in fantasy play.
Earlier this week I took whifey for a spin in the [Porsche] 930 and
whilst sitting at a traffic light she suddenly exploded…“Them
bitches!” she said, pointing at the two hotties in the Honda next
to us checking out the Porsche and making eyes at me… Hell, I
didn't even notice them before. Now, I'm not a man who screws
around. Quite happily married for 17 years now. But I must admit,
getting the old ego stroked at intersections reminds me of high
school and kinda makes me feel like a real stud…Oh shit, is this
what they call a “midlife crisis”? (Pierre Martins)
There's super hot women everywhere, ignoring us. Keep in mind, I
don't exactly fit in with the more well dressed clientelle at this establishment as I am mostly very happy in a nice t-shirt and a pair
of jeans. Anyway, tons of hotties, no love (and in some cases a bit
of attitude and contempt) from any of them, until we decided to
leave. We walk outside and head across the street to my [Porsche]
996. When it is apparent as to which car we are going to, I hear,
‘Mr. bald guy, where are you going, come back’ from a group of
about 10 absolutely drop dead beautiful, tightly clothed, amazingly sexy women. These very same woman only moments before
were not paying us any attention. (mastermind)
Mastermind's story demonstrates the transformation of an emasculated man who is failing to keep up with the men who surround
him due to his attire and his balding pate. His Porsche becomes a
magical transport, bringing him back into the game and enabling
him to capture the attention of women.
A Porsche's technical performance, including its power, its speed,
and its handling, is a source of pride for Porsche owners. The brand
community revels in the fact that driving a Porsche requires a higher
level of skill and that women (and lesser men) have difficulty in driving them. This attitude is evidenced in the following post that argues
that only “real men” can handle the car. Driving a Porsche involves
the physical mastery of the machine, which requires its drivers to
be strong, powerful, and mechanically skilled, traits strongly associated with the cultural definitions of masculinity (Mishkind, Rodin,
Silberstein, & Moore-Striegel, 1987).
Man, that's basically what Porsche is all about! Porsches & especially the early ones were made for REAL drivers, not every ‘average Joe’ out there. There are plenty of cars that are easy & boring to
drive and requires almost no driving skills at all. So one who can't
handle Porsche, can get a Honda or something. (Flying Finn)
Hence, prior to the launch of the SUV, in the media and in the situated use in the lives of consumers, the Porsche brand serves as a
strong marker of masculinity for its owners.
4.2. The gender contamination of the Porsche Cayenne SUV
Merely 9% of the movies and television programs that I analyzed in
the cultural analysis featured female Porsche drivers. Two of the
women drove the Porsche Cayenne SUV. Carmella in The Sopranos is
a middle aged, suburban housewife married to a mob boss, while
Kim Basinger plays a middle aged teacher who is kidnapped and
fights to save her family in Cellular. These two roles suggest that the
SUV is bringing new identity meanings to the brand, which include
middle aged, suburban motherhood, indicating that the Porsche
brand might be becoming gender contaminated.
To confirm that the Porsche Cayenne SUV brought new gender identity meanings to the Porsche brand, I used a one-way between-subjects
multivariate analysis of variance (MANOVA) to analyze the data from
the brand identity meaning elicitation experiment, using all of the 28
328
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
items as dependent variables. The descriptive statistics for each of the 28
items by condition are contained in Table 2. A multivariate test of the full
model showed that these 28 dependent variables were significantly affected by the experimental condition (Wilks' lambda=.002, F (56,
160)=2.194, pb 0.001). While Porsche sports cars and the Porsche Cayenne SUV share some identity meanings, the tests of between-subject effects showed that they differ on eleven, including seven that are closely
tied to sex or gender identity (Male/Female, Masculine/Feminine, Soccer
Mom, Playboy) or that are correlated with gender identity based on commonly accepted gender stereotypes, (Dangerous/Cautious, Reckless/Safe,
Car Lover), (Bem, 1974; Edelbrock & Sugawara, 1978; Spence &
Helmreich, 1978). Porsche sports cars owners are perceived to be more
male, more masculine, more like a playboy, more dangerous, more reckless and more of a car lover than the Porsche Cayenne SUV owners, who
are perceived to be more female, more feminine, more like a soccer mom,
more cautious, and more safe. The absence of differences for the
remaining items indicates that some of Porsche's identity meanings, particularly those related to economic status, self-esteem, and social display
behaviors, remain intact with the SUV launch.
Consistent with the findings suggested by the cultural analysis and
the experiment, my analysis of the netnographic data from the Porsche
brand community indicates that the launch of the Porsche Cayenne contaminated Porsche as a marker of masculinity and threatened the masculine gender display of the existing Porsche owners. Across the myriad
of conversations in my dataset, I uncovered a dominant theme that
discussed the alternative gender identity meanings brought by the
Porsche Cayenne SUV, their threat to the masculine meanings previously associated with Porsche, and their stigmatizing impact on the members of the brand community. Many of the existing owners lamented
that the Porsche Cayenne SUV not only feminized the Porsche brand
but also introduced alternative, subordinate versions of masculinity to
it. The brand community's analysis expresses alarm over the femininity
of the brand extension.
Table 2
Porsche identity meanings (sports cars vs. SUV).
Soccer mom
Playboy
Car lover
Yuppie
Status oriented
Male/female
Young/old
Single/married
Urban/suburban
Poor/rich
Has kids/has no kids
Wants to fit in/wants to stand out
Hip/mainstream
Dangerous/cautious
Exciting/boring
Fun loving/serious
Discriminating tastes/follows the
crowd
Cool/uncool
Stylish/unstylish
Arrogant/humble
Showy/modest
Snobby/down to earth
Insecure/secure
Practical/impractical
Inferiority/superiority complex
Follower/leader
Masculine/feminine
Reckless/safe
a
Sports car
meana
Sports car
SD
SUV
meana
SUV
SD
Sig.b
6.41
2.86
2.51
2.51
1.81
2.24
4.35
3.35
3.43
6.49
5.05
5.51
3.76
3.30
2.97
3.05
3.68
.798
1.378
1.387
1.070
.995
.830
.949
1.207
1.642
.559
1.201
1.346
1.722
1.024
1.142
1.290
1.811
3.88
4.97
3.75
2.24
1.69
3.71
4.56
5.09
4.88
6.51
3.18
5.34
3.29
4.32
3.50
3.97
3.41
1.822
1.507
1.785
1.208
.739
1.404
1.106
1.379
1.533
.633
1.642
1.223
1.169
1.364
.862
1.141
1.258
***
***
**
3.16
2.70
1.78
1.76
2.38
3.00
5.41
3.84
4.22
2.68
3.59
1.259
1.222
.821
.683
.953
1.291
1.066
1.500
1.357
.944
.985
3.29
2.74
2.24
1.94
2.53
3.24
4.96
4.50
4.50
3.85
4.50
1.142
.931
.955
.776
.961
1.046
1.170
1.022
.929
1.395
1.080
**
**
**
**
**
**
†
I am actually excited about seeing them [the Porsche Cayenne SUV]
driving around on the roads, but I'm not excited at all about seeing
who's driving them. Hopefully I will see a competent-looking driver,
but I know I'll see a soccer mom like I see in every [Hummer] H2
around here. It just gives me a heeb. Like this morning, I was behind
a BEAUTIFUL brand new [Porsche] 911 turbo for like 10 minutes! I
flipped my shit and couldn't take my eyes off it. Then I passed by
at a light when it turned, only to find a 50 year old-ish looking woman with poofy fake looking hair driving it… I had to hold back the
tears. Anyone else feel this way when they see out-of-place Porsche
drivers? Like people you KNOW right when you look at em — that
they don't give a shit about having the privilege to drive a precision
engineered German sports car!!! (TarHeel 944S)
You guys forget that you have the wrong hormones flowing through
your veins to understand why Porsche is putting out this SUV. My
wife would snatch one up in a heartbeat. Someone mentioned the
worst scenario would be soccer moms driving these. Well, get ready
to meet your nightmare. It'll be sitting at a stop light near you. (Bart)
I heard a rumor that Porsche was devoting their entire racing budget
to the Cayenne Challenge, an event that pits homemakers against
each other in a challenging obstacle course. …events include: 1000
metre grocery dash. Drivers launch from a standing start and drive
furiously to a simulated shopping center where 4 bags of groceries
are loaded before the driver hurtles their 2+ ton vehicle back to
the start/finish line… Mutitasking behind the wheel… Drivers negotiate a difficult city street simulation while talking on a cell phone,
cradling a Grande Non-Fat, No-Whip Mocha in their lap, and applying mascara or lipstick. (H20NOO)
Porsche's marketing efforts confirm the existing owners' suspicions
that the Porsche Cayenne SUV is targeted toward women. Many of the
early photographs released by PAG depict women drivers, local Porsche
dealerships advertise special “soccer mom promotions,” and PAG sponsors women's professional tennis. These marketing efforts further the
feminization of the brand and causes consternation among the Porsche
men.
They say that they are aiming this at 42 to 47 yr old males with
porsches…but, all the photos ive seen have girls driving it and that
putz shwab [President of Porsche North America, Fred Schwab]
says that its great for the alpha female/wife. (RobertG)
It is not merely the introduction of women to the brand that
alarms the existing owners; it is that Porsche has introduced its first
non-sports car with the introduction of the SUV. SUV's are perceived
by the brand community as “practical vehicles” that are popular with
mainstream consumers, while Porsche sports cars are valued for the
fact that they are fun, frivolous, and rare. The introduction of the
SUV weakens the association between the brand and its masculine
sports car heritage. Porsche owners begin to fear that the launch of
the Porsche Cayenne SUV opens the door for future non-sports car
models from Porsche that will push it into even more feminine categories, such as minivans or station wagons, which open the brand to a
broader audience, thereby diminishing its mystique.
If this trend keeps up — Porsche will be building station wagons to
compete with the Subaru Outback and then Minivans to compete
with the Chrysler. (DoubleNutz)
**
**
For first five single items, 7-point Likert scale, 1 anchors “agree” and 7 anchors “disagree”. For all remaining items, 7-point scale, 1 anchors first trait of pair and 7 anchors
second trait of pair.
b
† = p b .10, * = p b .05, ** = p b .01, and *** = p b .001.
Yeah, I can't wait for the day when Porsche offer a car for each
group of buyer. A van for the soccer mom, a Lexus like porsche sedan for those who prefer comfort over performance, a porsche
hatchback for the younger buyers who can't pay a lot. Yeah,
everybody's driving a porsche and they are all happy and some
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
of them will one day drive a [Porsche] 911…but wait, what does a
911 mean to them anyway? We are all driving porsches, and my
porsche van is better than your 9 what? So much for the porsche
mystique. (Cloud964)
Finally, while some SUV brands, such as Hummer and Jeep, function as masculine markers and convey their masculinity with offroad adventure mythology and an aggressive physical design, others
are associated with more feminine imagery. The brand community
largely finds the Porsche SUV to be less masculine than the Hummer,
Jeep, and Range Rover SUVs, and labels it as more reminiscent of a
station wagon or a minivan, both of which are vehicles that are associated with women. In the following posts, the brand community members connect competitive SUVs with masculine themes (Lamborghini =
sex, Hummer = military tanks, Suburban = cowboy) but are unable to
do the same for the Porsche SUV.
A truck/SUV should look like a truck–large in dimensions and boxy,
with a strong upright grille–lots of right angles. Rugged… I identify
with SUV's like the Suburban (we used to call them Cowboy
Cadillacs)… Also, the Cayenne looks very feminine to my eyes,
which doesn't jive with my sense of a traditional, rugged SUV. (Anir)
Mr. Lamborghini said to his engineers… I want you to build a car
that when I say to my wife ‘I'm going out to buy cigarettes’, is so good
looking that first good looking lady will come with me to the hotel,
and is so fast when I'll come back, my wife says: ‘that was fast honey!’… Little different than what boys at [Porsche] told their designers…we want to have a car, that…every soccer mom likes, but
25 year old hotties couldn't care less and has so much cargo space
that we can fit 30 grocery bags easily. (Flying Finn)
Even though I personally do not like the Hummer, I believe it is selling
well because of its tough, masculine image. It appeals to those who
want the biggest, baddest, Schwartznegger-ish truck on the planet.
OTOH, the Cayenne is actually a bit feminine in its lines. It looks more
like a minivan or tall station wagon than a traditional truck. (Anir)
Many of the brand community members fret that the Cayenne is
feminine, citing features of the vehicle that suggest that the SUV
was designed for women: its automatic tiptronic transmission, its
cupholders, and its feminine design. The automatic transmission obliterates the physical challenge of mastering Porsche's difficult handling,
while the cupholders add unnecessary weight to the vehicle, heretical
in a race car. Both of these features decrease the product's masculinity
quotient by distancing the SUV from Porsche's sports cars, as evidenced
in the following brand community conversation. Note how the SUV is
associated with women, metrosexuals, and men “without balls,” distancing the SUV drivers from masculinity narratives. Porsche's executives are feminized and accused of eating and dressing in unmanly ways:
On the subject of when did carmakers lose their balls… When the
average male became a metrosexual and lost his. It is true… Men
are no longer men though. They want powered cup holders, and
soft touch interiors so they don't damage their manicured nails…
Then to top it off, German car makers start acting like Asian car
makers. They now wear suits fit for Milan while eating sushi for
lunch. And this is supposed to be the Porsche I looked forward to
when I was a kid? I want the company I worshiped when young
but was too young to drive. The whole situation makes me sick.
AHHH. That's it, I am going to go take my uncivilized, no-sissyneed-aply, non-muffled [Porsche] 911 out for a wild ride and try
to forget the whole thing… Little did I know then that most cars
today would be built for sissy wimps with no taste, driving skill
and automotive historical knowledge by companies lacking the required cojones to stick to their original ideals. It is sad. In my opinion, cupholder doesn't belong in a Porsche (or any sports car for
329
that matter), nor does tiptronic [automatic transmission] etc.…
(Ed Bighi)
However, all that sort of things seem now almost fine, when I'm
comparing them to a Cayenne. You're absolutely right on that
one… The day they started rolling Porsches off the line with
Tiptronic and cupholders was a sad day indeed… It pains me to see
what was one a no nonsense pure race car now plagued with useless
extra weight and soft suspensions to make that trip to the grocery
store more comfy. (Flying Finn)
And a 6 speed [manual transmission]? You know, soccer moms
don't want to shift — it may interfere with their quality cell phone
time :rolleyes: (mpm '95 C4)
The introduction of the Porsche Cayenne SUV threatens the masculine identity of the brand's users. The owners voice their concerns that
the buyers of the SUV are incongruent with their desired masculine
identity. The excerpts from the brand community conversations below
are indicative of the negative emotional response that this incongruence
elicits and are indicative of the out-of-face emotional disturbance that
the identity threat brings. The owners experience the SUV as a personal
affront, a “slap in the face” or an “ugly sister” that reflects badly on them:
And it is impossible for me to dispassionately talk about this slapin-the-face soccer-mom-tank-barge from the accountants who now
run Porsche. To find that Porsche is now selling SUV's, and soon
pick-ups, toothbrushes, and maybe perfume, is atrocious. They have
destroyed the brand and killed its mystique. My days of reading the
race results from Le Mans to figure out how many of the top positions
were Porsches is long dead — and the accountants and marketers at
Porsche are dancing on the grave, while they try to figure out how
to make more money from the brand name. I'm sick… (mike_la_jolla)
The Cayenne remains the ‘ugly sister’, a necessary blemish on, an
up to now, untarnished marque. (Robbo 66)
While research has suggested that physical damage to a man's car
can represent castration (Belk, 2004), the Porsche brand community
shows that symbolic damage to a man's car brand can be just as painful.
4.3. Defending masculinity markers through collective identity work
In the sections below, I describe the collective face-saving identity
work that the brand community engages in as a defensive justifying
practice (Schau, Muniz, & Arnould, 2009) to alleviate the gender identity crisis brought on by the SUV. This identity work involves attaching
derogatory gender stereotypes to SUV owners that psychologically
and socially establish an ingroup/outgroup. This behavior is consistent
with social psychological research on stereotyping, which shows that
group threat leads to the stereotyping of others and a prejudicial affect
toward them. Stereotyping places others into categories, and this categorization leads to the construction of boundaries among groups and
the creation of ingroup/outgroup social structures (Tajfel & Turner,
1979) in which people surround themselves with others who resemble
them and distance themselves from those who do not as a way of declaring who one is (and who one is not).
4.3.1. Collective gender stereotyping
In their online conversations, the brand community members stereotype Porsche Cayenne SUV owners, claiming that the product is purchased by people who are very different from the existing owners,
specifically “soccer moms” who are looking for a trendy SUV to avoid
driving a minivan to the grocery store, shopping mall, and school playgrounds, “poseurs” who are interested in obtaining the image of a
Porsche but who are not true car enthusiasts and cannot properly handle
330
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
the cars, and “yuppies” who are merely seeking a status symbol. The
gender significance of each of these three stereotypes is discussed below.
As observed by Martin, Schouten, and McAlexander (2006) in
their study of female Harley-Davidson owners, women who purchase
symbols of masculinity engage with and co-opt masculinity discourses through their consumption of these products. Through the
gender stereotyping that links the Porsche Cayenne SUV to a derisive
soccer mom stereotype, the existing Porsche owners block this transfer of masculinity to the women who buy the SUV by supplanting
positive masculine associations with negative feminine associations,
thereby making the product less attractive to women for gender experimentation. At the same time, the linkage of the SUV to a feminine
ideology further diminishes its attractiveness to men because it strips
the product of its masculinity, thereby limiting the product's appeal
and its lifespan in the marketplace. The following post is an example
of the usage of the “soccer mom” stereotype.
… My GOD say it ain't so! I keep telling myself that if I wish and hope
and pray hard enough… that it won't happen. That this SUV thing is
all just a bad dream.... And you were there, Auntie Lame, and Uncle
Killporsche… That perhaps this is all just an elaborate hoax and
coverup for what what is really going to be the greatest Porsche
since the [Porsche] 956. That maybe Porsche Marketing will point
at everyone and laugh saying, C'mon… You *really* believed all that
CRAP about SUVs?! This is Porsche for cry'n out loud. We're proud
Germans! We don't make jokes! We make race bread and race
WINNING sports cars! NOT MINI Vans! I can just see it know… Some
rich Soccer Mom driving the damn thing to pick up her kids with a
bumper sticker that reads, Proud parent of honor student and a Baby
on Board sticker next to the Barney stick on sunshade with a baby
seat in back… NO! WAIT! The ULTIMATE killjoy! Porsche adds a fold
down babyseat standard in each one! …Oh the humanity! …I wonder how many cup holders it will have??? Just shoot me now! (Doc)
The “soccer mom” stereotype gains cultural resonance from its
connections to the cultural discourses surrounding the identity
meanings attached to SUV owners. The New York Times columnist
Bradsher (2002) connects SUVs to an emerging social group defined
as “soccer moms” in the cultural vernacular, calling them the vehicle
of choice for women who are trying to escape the stigma of driving a
minivan. This description, ironically, transfers the negative associations that the women were trying to avoid in the minivan category
to the SUV category. “Soccer mom” is a pejorative term for third
wave or post-feminist, upper class motherhood and homemaking. A
soccer mom is an affluent homemaker, often very highly educated,
who has left behind a successful professional career to voluntarily
stay home and take care of her children. The soccer mom stereotype
hearkens back to the housewife and Stepford Wife stereotypes
berated by second wave feminists such as Betty Friedan (1963),
who depicted the home as a prison in which women were subjugated
to servitude to meet their family's needs and who depicted the housewife as a passive, dehumanized robot. Updated in contemporary television shows such as Desperate Housewives, today's soccer moms are
portrayed as boring, bored, frustrated, and depressed, with no identity of their own and with few responsibilities or achievements outside
of those of their family.
The soccer mom ideology emerged in American culture during third
stage feminism, a movement characterized by Faludi (1992) as a backlash against the definitions of success put forth by the second stage feminists. “Opting out” of the public sphere and an identity built around
professional success, soccer moms returned to the home and embraced
voluntary domesticity and the traditional roles of wife and mother, encouraged by alternative versions of femininity offered by media celebrities such as Martha Stewart and Nigella Lawson who celebrated
women's roles in the private sphere (Brunsdon, 2006). Given that
these soccer moms had grown up during three decades of feminist
critique against the traditional definitions of femininity and the subjugation of women in the home, opting out was often accompanied
by strong feelings of guilt and regret about giving up their hard earned
professional identities, of accepting less than their mothers, and of surrendering the ground won by second wave feminists (Hollows, 2006).
The attachment of the soccer mom stereotype to the Porsche
Cayenne SUV associates the vehicle not only with femininity but
with a particular feminine ideology that is not attractive to men or
women. The women who purchase the SUV must contend with
being labeled a “soccer mom,” a stereotype that prevents them from
gaining access to the masculine meanings that are traditionally associated with Porsche. Through the use of the “soccer mom” stereotype,
the existing Porsche owners relegate the Porsche Cayenne SUV
owners to the bottom of a gender power structure. By linking the
product to negative conceptions of domesticity and motherhood
that hearken back to an era when women had significantly less
power and were relegated to the private, rather than the public
sphere, the existing Porsche owners increase the power distance between the Porsche sports cars and the SUV and, hence, between
themselves and the Porsche Cayenne SUV owners.
The “poseurs” stereotype has existed in the Porsche brand community since its inception. Throughout history, following the launch
of controversial Porsche vehicles, the brand community has engaged
in debates in which the members argue over whether the owners
that the new product brings to the brand are “real” Porsche men or
mere poseurs. Brand community members depict the “poseurs” as
deviant from the traditional definitions of masculinity and emasculate them by referring to their lack of understanding of the technical
workings of the car, their inability and/or unwillingness to push the
car to its limits, or their lack of appreciation for the brand's illustrious
racing heritage. It has been shown that automobiles can be regarded
as sacred objects by their devotees (Belk, Wallendorf, & Sherry,
1989), with rituals devoted to caring for them, to showing them off
to others, and to driving them in a manner deemed appropriate by
the brand's evangelists (Belk, 2004). As Leigh, Peters, and Shelton
found in their study of MG owners (2006), being accepted as an authentic member of a brand community requires much more than
just buying the brand; it requires driving (and respecting) the brand
in the appropriate way. Porsche owners adhere to similar rituals
and demean those who do not follow them. SUV owners are considered inauthentic: they do not drive in the appropriate manner, driving too slowly or carefully; they do not care for their vehicle in a
respectful manner, letting it get dirty or parking it too close to other
cars; and they do not respect the heritage of the brand and its racing
roots. The post below is an example of the “poseurs” stereotype in the
brand community conversations.
That's why they're making Cayenne… There are plenty of people
(with money) who only want to drive Porsche because there is a
badge on the hood. These people would be driving happily Honda
Civic's w/ the Porsche badge & hefty price. They don't care/know
what good handling is, why ignition key is on left, what's so great
about going over 160 and hearing that wonderful flat six ‘singing’ behind you etc. Only thing they care is that name. They're cruising
around 20 mph and that's it for them. Irony is that they want to drive
Porsche because of that name, mystic etc. that has come from Porsche
being true sportscar company, winning LeMans etc. Now, making
this change (to a wrong direction), that mystic, name that they want
to show off to their buddies, might not be like that in future. It soon
might be just another car company. To many people, I'm sure, it's just
an transportation to go from A to B & to show it off, but for me, it's not
just a car, and that's why this bothers me. (Flying Finn)
The poseurs' stereotype indicates that one cannot merely buy masculinity through the purchase of a Porsche; rather, to “grab the phallus”
(Thompson & Holt, 2004), one must embody the masculine ideals of the
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
brand. The decision of whether one is worthy of Porsche's gender identity meanings is adjudicated by the members of the Porsche brand community, who designate which owners belong in the group and which do
not. Although one can aspire to masculinity via the purchase of a
Porsche, it is only through acceptance by the brand community that
one rises above the rank of poseur (Leigh, Peters, & Shelton, 2006).
Through brand community discussions that label the SUV owners as poseurs, the Porsche sports car owners further disassociate themselves
from them.
The “yuppies” stereotype also has subcultural resonance in the
Porsche brand community because it is a stigmatized identity that
has been linked with the brand since the 1980s. During the 1980s,
Porsche became the car of choice for “yuppies”: young, upwardly mobile professionals. It is interesting to notice in the post below how the
“yuppie” is depicted as deviant from the traditional definitions of
masculinity and feminized by referencing the yuppies' preoccupation
with fashion and taste for feminine beverages.
The Cayenne started life a few years ago with 450 BHP from its
4.7 litre V8.... And what do they do with this modern masterpiece
of engineering? They put the bloody thing in a glorified tractor.
90% of which are used to convey their Armani-suited owners to
shopping malls to order cafes \**&^+#$! latte or \*%$+ !!* chardonnay or very occasionally some seriously rough stuff on the driveway
to the+ *^^%!\\ golf club! AAARRGGHH. Words fail me! Oh life!
Deep drives thy sting… This thing makes a joke of us. (ColinB)
While the “yuppie” stereotype contains breadwinner masculinity, as
noted by the upper class references contained in above post (Armani,
country club), it is deficient in rebel masculinity, leaving the men who
purchase the Porsche Cayenne SUV lacking the blend necessary to
achieve hegemonic masculinity through Porsche ownership.
Through stereotyping, the members of the Porsche brand community
thwart the SUV owners' efforts to “grab the phallus.” Both men and
women who purchase the SUV are labeled with stereotypes that distance
them from the existing sports car owners, designating them as “the
other.” These stereotypes are summarized in the following post that
uses humor to paint SUV owners as lacking in masculinity. Specifically,
Cayenne owners are linked to such feminine consumption practices as
wearing fashionable and precious clothing, having a manicure, drinking
designer coffees, embracing feng-shui design, and attending a spa.
101 Projects for Your Cayenne… 1. Adding air to the tires on your
own. Cost: $27 ($5 tip to gas station attendant to instruct use of air
hose; $20 dry cleaning for rustic-looking wool gabardine pants; $2
for latex exam gloves to protect manicure). 2. Getting Coffee While
Someone Else Changes Your Oil. Cost $125 (Visit to doctor's office
to ensure you are OK after having learning that you consumed
non-designer brand of coffee in the waiting room = $120, Bottle of
Pepto Bismol = $5). 3. Modifying Cup Holder to Accommodate Mega
Grande Latte. Cost= $700 (Porsche Upgrade Coffee Holder
Kit = $300, Custom fitted overalls with embroidered Porsche
crest= $200; Neiman-Marcus signature screwdriver set = $100;
Massage Therapist to massage aching limp wrist = $100). 4.
Installing Floor Mats. Cost = $450 (Porsche Factory Mats $200;
Feng-Shui Consultant to assist with color choice and alignment= $250). 6. Upgrading GPS Software: Cost= $600 (Worldwide
Starbuck's locations Software = $120; Spa Locator Software= $200;
Polo Club Registry Software = $100). 7. Installing Lacrosse Stick
Holder: Cost= $675 (Porsche Techquipment Lacrosse Stick Holder
Kit = $500, Neiman Marcus Signature Screwdriver Set (exclude if already purchased for cup holder project) = $100; Color coordinated
calfskin mechanic's gloves = $75). (Rich911E, Tumalo)
Psychologically constructing a barrier between themselves and
the women (and feminized men) who join the brand community
through the purchase of an SUV, the existing Porsche owners resolve
331
the identity dissonance associated with the SUV by viewing themselves as very different from the new Porsche owners who do not
live up to their version of masculinity. This separation allows them
to psychologically renew their commitment to the Porsche brand as
a masculine identity marker without accepting its dilution. Because
the stereotyping is a communal practice that is achieved through
public conversations and is thus not confined to the minds of individual consumers, its inculcation has social effects. The stereotyping
clearly specifies the characteristics and practices that define a real
Porsche man, establishing norms of behavior for the members of the
brand community. Through these acts, the meanings associated
with the Porsche Cayenne SUV that challenge the traditional definitions of masculinity are suppressed and contained.
In summary, the Porsche owners' gender stereotyping of the SUV
owners stratifies the Porsche brand community along gender lines;
the existing Porsche sports car owners populate the ingroup and
enjoy access to Porsche's masculine identity meanings, while the
Porsche Cayenne SUV owners and the feminine identity meanings
that are attached to them populate the outgroup. Symbolically, the
Porsche sports car owners construct a gated community around themselves where the gates that mark the boundary between them and the
SUV owners impede the transfer of masculine identity meanings from
the Porsche sports cars to the Porsche SUV and, more importantly, impede the undesirable transfer of feminine identity meanings from the
Porsche SUV to the Porsche sports cars. Importantly, the ingroup is
sharply defined by and dedicated to hegemonic masculinity, while the
outgroup is defined by femininity and subordinate masculinities.
Hence, Porsche's masculine meanings are protected from dilution
through the incursion of undesirable others by acts of containment;
through stereotyping, Porsche owners psychologically and socially
keep the masculine identity meanings in and the feminine meanings
out. Through their actions, the Porsche owners thwart the desire of
the firm to open the brand to women, reverse the evolution of the
brand's gender identity meanings, and reinstate Porsche as a masculine
identity marker. The angry backlash among the existing customers of
the brand more firmly defines Porsche as a symbol of masculinity and
strengthens the ties between “real men” and the brand.
Despite the fact that brand communities are often conceptualized as
egalitarian, populist, and operating outside of the constraints of the external culture's status system (Cova & Cova, 2002), most brand communities exhibit well-defined and rigid status hierarchies that reflect the
members' different levels of adherence to the ethos of the community
(Schouten & McAlexander, 1995). The members earn status through
their participation within the group, their level of ownership, their idiosyncratic knowledge and experience with the brand or the community, and their allegiances with status-endowed others. Researchers have
conceptualized this status structure as a series of concentric circles, at
the center of which are the “hard-core” consumers (Fox, 1987). It is
these hard-core consumers who play the major role in creating and arbitrating brand meaning (Fox, 1987; Holt, 2004; Schouten & McAlexander,
1995). By collectively responding to the identity threat of the Cayenne,
the existing Porsche owners leverage the status system of their brand
community and create new status distinctions that place Cayenne owners
at the bottom of the hierarchy. Social stratification limits the contamination of the Porsche brand by reinforcing gender distinctions and boundaries and circumscribing the Porsche Cayenne SUV to contain its
feminizing influence.
The emergence of a social hierarchy strengthens the ties among the
privileged members (sports car owners) and perpetuates their own interpretations of the brand's meaning, thereby increasing their commitment to the community and facilitating the identity projects that they
are pursuing through use of the brand. Concurrently, the ingroup/
outgroup social structures increase the social distance among the member subgroups, as the less privileged groups (Cayenne owners) are ostracized, precluded from participation in rituals and practices, and
denied access to the brand meanings that they need for their own
332
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
identity work. Hence, the less privileged members are not able to derive
social benefits from their brand ownership, losing the communal and
identity benefits of being part of a brand community.
What makes the Porsche activity different from other settings in
which car club participants exclude replicas or reproduction cars from
participating in their activities (Belk, 2004) is that the Porsche SUV is
an authentic Porsche. As Leigh, Peters, and Shelton remind us, “When
brand communities and subcultures are operative in a brand's ‘meaning’ ecology, managers no longer solely own the brand, its associations,
or the right to position it for strictly commercial purposes.” (2006: 492).
Now, it is the Porsche brand community that wields control.
4.3.2. Trying to fit in through hyper-masculinity
Some men in the brand community purchase the Porsche Cayenne
SUV, and the negative stereotyping and status stratification practices
of their fellow owners force them to do identity work of their own to
save face within the group with regards to their purchase. Male Porsche
Cayenne SUV owners engage in hyper-masculine behaviors to compensate for their purchase and to gain entrée into the ingroup, similar to the
practices observed in men who do not conform to hegemonic definitions of gender in the occupational, athletic, or personal spheres, such
as male clerical workers, gay athletes, and men with physical disabilities, who negotiate social acceptance through overemphasizing their
masculinity in other spheres (Anderson, 2002; Gershick & Miller,
1995; Henson & Rogers, 2001). Male Porsche Cayenne owners try to
combat the anti-masculine stereotypes, positioning themselves as fellow enthusiasts and citing other Porsche sports cars that they own to
claim legitimacy. They tell stories about how they take their SUVs
off-road, use them to tow their race cars or their boats, or use them to
access adventure sports such as skiing, hiking, hunting, and fishing,
thereby tapping into other markers of masculinity. These owners refer
to their SUVs as sports cars, speaking about them using sports car
terminology.
I drive around all winter, up and down canyons, go skiing, into remote lakes on dirt roads and get paged to sprint back to the hospital all the time…Would I love cruising around in a cayenne
knowing that if it dumps while I'm up at snowbird that I'll safely
and surely make it back home or to the hospital no matter what
and if it doesn't that I'll be able to push down those awesome
twisty roads in little cottonwood? Definitely… Just so you don't
think I'm trolling or have no experience with porsche… I have
[Porsche] 996tt, used to have 84 [Porsche] turbo with upgraded
turbo etc., had 81 [Porsche] euro 911 sc before that and have been
reading everything I could get my hands on about them since I
was 12. I love the car and think they've brought something new
to the SUV/multipurpose car category with the cayenne turbo
with its offroad abilities, handling and performance… I think the
cayenne will get me most of the places this takes me (except
one remote lake and MOAB trails) and get me home with more
handling and zip.
The male Porsche Cayenne SUV owners brag about the technical
performance of the vehicles and tell stories about beating sports cars
at track events. With aggressive posturing, many of these men put
down the other Porsche models. This behavior is an example of masculine one-upmanship, where men fighting to achieve status and recognition within a social group attempt to demonstrate their masculinity by
conquering or dominating the weaker members of the group (Iacuone,
2005). Sensing an opportunity for one-upmanship, given the Porsche
brand community's longstanding ostracism of the Porsche Boxster and
944/951 owners, the Porsche Cayenne SUV owners pit themselves
(and their vehicles) against these weaker owners in an aggressive
show of force to curry favor with the Porsche 911 owners who reign
at the top of the status hierarchy.
The base Cayenne is still a 350 hp v8, and it definitely faster than a
[Porsche] 944 is…And you'd be rather surprised at the Cayenne on
a real track! There was an S model out at Sebring and the thing
was blowing by [Porsche] Boxsters and [Porsche] 930s, and yes,
[Porsche] 951s… Like I said, go drive one. It'll surprise the hell
outta you. (dave120)…on a real track, smart stuff, it will eat your
[Porsche] 951 alive.
Welcome to the 21st century, where the technology now has
surpassed your and my 1980s technology… You continue to emit
garbage, and until you go experience one of these ‘school busses’ for
yourself you can't say what it is or is not capable of. Go, drive, believe.
Till then shut your hole about oh the cayenne is a mistake this, and oh
if I ran porsche things would be so much better… You seem sore that
its faster than a [Porsche] boxster, I'm sure yours will edge it out, but
still, faster than many sports cars. How many companies offer off
road performance and blistering track times from the same vehicle?
One. Porsche. The cayenne is here to stay, so why whine. (ERAU-944)
Despite this show of hypermasculinity, the Porsche sports car
owners deny the Porsche Cayenne SUV owners entrée into the
ingroup and pull their previously maligned Boxster brethren closer
in a show of sports car family support.
No way, no how, could the Cayenne S beat the [Porsche] Boxster S
on any smooth road or track. Absurd. It won't even begin to handle
with it, won't begin to accelerate with it, has a lower top speed,
and 60% more weight on a far higher center of gravity. I have driven the Cayenne S, it has neither the handling nor power of an Audi
S6 Avant, and I can tell you the S6 won't catch a Boxster S anywhere either. (NMoore)
The technical performance of the Cayenne is largely irrelevant to
the existing owners, who focus on the illegitimate owners who are
trying to steal the sports car owners' masculinity marker. In this
post, the Cayenne is portrayed as a “strap-on for soccer moms and effeminate stockbrokers,” illustrating its illegitimacy as a false phallus.
Porsche screwed the pooch on the Cayenne. It's great that it handles
well… it's great that it goes fast… blah friggin blah. 99% of these
abortions will be owned and driven by people who have no business
with a fast anything, the only thing that will save them is that it can
actually handle. The real issue is that Porsche was a company with a
tremendous heritage of building one-off race cars for the street. It's
entire existence is almost contrary to the last 40 years of Porsche
history. How can you say, Porsche, there is no substitute about a vehicle that is so obviously nothing special? I can tow more with a
Chevy, I can carry more in a [Ford] Expedition. Who cares how fast
I can go, if I am going camping with a boat on the back? Only a
fucktard would haul ass with his family in the car. I can get more luxury from Lexus in the LX470, I can get more bling in a [MercedesBenz] ML55. I can go further offroad in a Jeep, or even an International Scout. So what's left? An expensive strap-on for soccer-moms and
effeminate stock-brokers. (Water944t)
In summary, the face-saving practices of the Cayenne owners
highlight that these mistreated owners try to validate their ownership and their place in the group by closely adhering to the group's
masculine identity standards. Facing exclusion, the men who purchase the SUV and who are relegated to the outgroup resort to a display of hyper-masculine behaviors to try to gain entry into the
ingroup. These actions further reinforce the standards for masculine
behavior in the brand community. The act of defining the ingroup
causes the groups of consumers who previously saw themselves as
heterogeneous (Porsche 911 and Boxster owners) to come together
and celebrate their similarities. Adding gender diversity to the
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
community counter intuitively makes the group more homogeneous
and devoted to the traditional notions of masculinity. The managers'
attempts to gender-bend the Porsche brand paradoxically spur the
brand community to solidify and defend the brand's status as a masculine identity marker.
5. Discussion
5.1. Theoretical contributions
First, this research shows that gender continues to exert a significant
force on people in their choice and usage of brands. Despite the claims
that we are living in a postmodern era of gender experimentation, the
men in this study continue to value and protect the masculine identity
meanings of their brands. When sufficient numbers of women appropriate a brand that is being used as a male identity marker, it disrupts
the men's identity performances. Rather than making the brand “gender neutral” as suggested by Fischer and Gainer (1994) or “opening
up the possibility for diverse gestures and characteristics” as suggested
by Butler (1997), this incursion initiates collective self-presentation
practices that solidify the brand's male-ness. Female gender contamination, initially recorded in ancient civilizations, is as potent today as it
was in the past.
While instances of people using gender-bending consumption to
experiment with their identities abound in the popular and academic
literature (cf. Martin et al., 2006), this research shows that when
male-gendered brands target the opposite sex, their male consumers
can and do fight back. The fight over Porsche could reflect the greater
battle that men are fighting to maintain their dominant position in
the broader social hierarchy. When the boundaries between groups
of unequal status and power become more permeable and when status
hierarchies become unstable, the members of high status groups become motivated to maintain the status distinctions, increase ingroup
identification, and increase outgroup discrimination (Sachdev &
Bourhis, 1991). In contemporary society, these conditions are in
place; the boundaries between men and women are becoming
more permeable in the workplace, in the home, and in consumption,
and the prevailing patriarchal power structure is under attack from
women, gay men, and alternative definitions of manhood. Kimmel
(1996) claims that rather than becoming more androgynous in response to the movements that are blurring gender boundaries,
men, in an angry response to a world in which the attainment of hegemonic masculinity remains out of reach, are regressing toward behaviors that reinforce the traditional definitions of masculinity.
This behavior makes gendered consumption more important today
than it was in times when gender roles were less blurry. Masculine
and feminine identity markers are more valuable to consumers
when gender roles are permeable, making gendered consumption a
powerful force in the postmodern era.
Second, these findings illustrate how consumers in brand communities engage in gender production. The consumers' communal response
to gender-bending produces and reproduces the gender distinctions
and hierarchies. The gender-bending of the Porsche brand, rather than
blurring gender lines and contributing to a less hierarchical gender
order, strengthens the gender distinction by reifying and reproducing
hegemonic masculinity and solidifying its position at the top of the social hierarchy in the brand community. The reinstatement of the
Porsche brand as a masculine identity marker shows that identitydriven consumption contributes to the regulative gender discourses
that constrain both men and women in contemporary society.
Gender-bending brands that are used by consumers for identity purposes ignite a discursive process, marked by power, that determines
the gendered meaning of the brand in culture and reinforces the prevailing power structures. This process shows how the normative and
ideological beliefs perpetuated by brand communities through the moralistic identity work highlighted by Luedicke et al. (2009) connect to
333
larger and more stable sociological constructs and structures, such as
gender narratives.
Third, this research illuminates how the existing consumers phenomenologically respond to brand gender-bending practices, and it
identifies a different response than that predicted by the prior experimental research. Prior research has shown that men choose possessions
associated with masculinity and avoid possessions associated with femininity; therefore, researchers have hypothesized that men will abandon possessions once they have been contaminated by women.
However, this research finds that the men who have invested themselves in a brand will fight for its meaning, territorially protecting
their masculine identity markers from becoming associated with
women. Rather than leaving their brand when it has been gender contaminated, the Porsche men engage in defensive practices within their
brand community to actively defend their masculine identity marker
and combat the feminization of their brand. This strategy supports our
understanding of the co-creative process of brand meaning management
among the consumers of a brand and underscores the importance of
brand communities as spaces for creating value for consumers outside
of, and sometimes in conflict with, the value provided by marketers.
The response of Porsche consumers contrasts with the findings
from prior research that has focused on the appropriation of the identity markers of subordinate (rather than dominant) groups. Studies of
subcultures whose symbols are appropriated by the mainstream
(Clarke, 1986) find that members abandon identity markers once
they are appropriated by the mainstream and find new ones with
which to distinguish themselves. Rinallo (2007) shows that homosexual men, however, embrace the appropriation of their fashion
and grooming markers by heterosexual men, viewing it as a way to
gain mainstream acceptance of their alternative version of masculinity. In contrast, this article studies the appropriation of the identity
markers of a dominant group, namely white, heterosexual men. This
dominant group actively resists the appropriation of their identity
marker brands and works to limit access to the brand's masculine
identity meanings to those who embody their definition of masculinity. The distinctions between the findings of these diverse studies reminds us that gender politics still matter; the men who are at the top
of the power structure actively resist when those on the rungs below
them (including women and less masculine men) use their brands.
While subcultures abandon their products when the dominant
groups abscond with them, the dominant groups actively fight to
maintain semiotic control of their markers of identity and actively
work to prevent subordinate groups from appropriating them.
Fourth, the Porsche consumers' battle for their brand contributes
to our understanding of how consumption practices impose social
structure and social distinction within brand communities. Politicized
ingrouping/outgrouping processes have been previously documented
in brand community research (Muniz & O'Guinn, 2001; Muniz &
Schau, 2005) that shows that the community members often view
themselves as an ingroup and those who use the competitive brands
as an outgroup, an inter-brand effect. However, the Porsche results
show that this process can also be an intra-brand effect. Regardless
of the brand moniker that one buys, entry into the ingroup is a communally negotiated process (Leigh et al., 2006). Consumers who buy
the Porsche brand expecting entry into a social collective might find
themselves excluded and ridiculed by those in the brand's communities who are pursuing their own political agendas.
When men refuse to share their masculine identity markers with
women (or with men whom they deem inferior), these men reinforce
their power advantage and buttress hegemonic masculinity as the
prevailing cultural standard in their brand community. However, at
the same time, through the gender stereotyping that elicits and reinforces recognizable pejorative subcategories within the gender spectrum, the existing Porsche owners contribute to the creation of a
complex gender status hierarchy where the more nuanced gender
identities beyond the male/female dichotomy, such as the soccer
334
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
mom, the poseur, and the yuppie, are defined and stratified. By
defending their masculine identity markers, the men defend their position at the top of the gender order and suppress the alternative definitions of gender that are emerging for both men and women.
In conclusion, this research shows that the social practices that
constitute gender continue to shape the meaning of brands and concomitantly shape the use and value of brands as gender identity
markers by consumers. It also illustrates how the practices of gendered brand consumption contribute to the gendering of the social hierarchy and the regulative discourses that constrain both men and
women in brand communities. This research helps us to better understand the continued presence and importance of gendered consumption in the postmodern world.
5.2. Future research
It is important to note that this study focuses on the common
identity needs of a largely homogeneous population: the majority of
the Porsche brand community is Caucasian, heterosexual, upper
class, highly educated, white collar professionals. Their collective response to the launch of the SUV represents how men who are well
endowed with the markers of masculinity respond to brand gender
contamination. Men pursuing alternative forms of masculinity or
those who are further from the hegemonic ideal are not represented,
and research suggests that their responses could differ. The men who
do not fit the typical profile associated with traditional masculinity
can use gender-bending consumption to explore alternative forms
of masculinity. Homosexual men can engage in feminine consumption practices to distance themselves from hegemonic definitions
and to create their own definition of gay masculinity (Kates, 2002).
African-American men can pursue a black masculinity populated
with consumption items that would typically be classified as feminine
such as jewelry (Weems, 2005). Women's responses to brand gender
contamination are also not captured in this study. Women, as a subordinated group, might be more fluid when playing in the spaces between the dichotomous poles of masculine and feminine (Borgeson &
Rehn, 2004) and might therefore perceive less of a threat from brand
gender contamination.
The moderating role of other identities that intersect with gender is
also left unexplored. Interesting intersections of gender with race and
social class were evident in the data and suggest that stereotyping is
used to define not only gender hierarchies but also racial and class
ones. The Porsche brand community attaches pejorative racial and social class stereotypes to the SUV owners, such as the hip-hop rapper
and the Southern redneck, to distance these users from the brand's
more traditional masculine meanings. Further research that explores
identity intersectionality and the simultaneity of gender, race, class,
and sexual identities, and intersectionality's interaction with consumption behavior would be useful to further develop the theory because a
brand's identity meanings are multifaceted and represent a desired
identity defined by more than one identity category.
5.3. Managerial implications
Stagnant sales in many mature product categories are causing
managers to look for new ways to increase business. A gendered
brand necessarily ignores half of a potential audience, making it a
ripe target for expansion. Gender-bending a gendered brand allows
the marketers to tap into the large consumer segment occupied by
the opposite sex. However, this decision is where the trouble often
begins. When a gendered brand's identity meanings are diluted by
its appropriation by the opposite sex, its existing consumers have
two options: to discard the brand because it no longer delivers their
desired gender identity (a strategy suggested by previous research)
or to stay and fight for the gendered meanings that they use to
present their identities, as presented here in this article. Either option
thwarts the manager's objective of increasing the brand's appeal to
both sexes.
This study offers insight into the challenges of gender-bending
brands. Before gender-bending, the managers need to assess how
much of their brand equity derives from the brand's gendered identity meanings. For brands such as Porsche that have gendered identity
meanings that are actively utilized by their consumers, genderbending could be a high risk strategy. In this case, the managers
should consider alternative branding strategies, such as the creation
of a new brand to target the opposite sex; for example, Procter and
Gamble offers its Old Spice brand for men and its Secret brand for
women, hiding the corporate affiliation between the two brands
from consumers to avoid gender contamination. If launching a new
brand is too costly, marketers can launch gendered derivatives of
their existing brands; for example, Gillette's women's products are
branded “Gillette for Women,” while Dove's men's products are
branded “Dove for Men” to distinguish the brands and to reassure
consumers that they are not using the same products as the opposite
sex. For brands that are not used by consumers to communicate their
gender identities, gender-bending is a more amenable strategy because the brand equity is not reliant on the gendered identity meanings of the brand. Thus, determining how central the gendered
meanings are to the existing consumers is a key first step in deciding
whether to gender-bend a brand.
Second, managers need to determine how contaminating usage by
the opposite sex will be. When women ride Harley-Davidson's, they
engage in gender tourism and empower their own femininities
through a hyper-masculine display (Martin et al., 2006). The brand
is useful to these women because of its masculine identity meanings,
and their masculine creative play does not threaten the gendered
meaning of the brand but rather reinforces it, as they are not trying
to feminize the brand through their usage. However, the Porsche
case illustrates the dangers when brand managers or the women
they target position or use a masculine identity marker in more feminine ways. When a brand's manager actively courts women with
feminized product designs, features, or brand narratives, the inherent
gendered meanings of the brand are more likely to be diluted and
male consumers might fight to protect them. As long as women respect
the masculinity of the brand and act appropriately with it in their usage
practices, Martin's research suggests that men will allow a certain degree
of female patronization of their brand. However, when an incursion of
women into the brand fundamentally changes the perceived masculinity
of the brand in the eyes of the social audience, the men in brand communities mobilize to protect the identity meanings that they hold dear.
Hence, women can use masculine brands as long as they respect their
masculinity; the trouble begins when women (or the brand's managers)
try to feminize the masculine identity markers.
Third, managers need to be ready for a backlash when they try to
gender-bend masculine brands. This study shows that masculine brands
become paradoxically more masculine following gender-bending brand
management programs. Managers still need to heed the asymmetry inherent in gender relations and its differential impact on consumer behavior (Bristor & Fischer, 1993). Despite media-fueled images of genderbending that suggest that the gender boundaries are breaking down,
this article shows that contemporary men remain protective of their
brands and will actively reject efforts to feminize them. Male consumers
are likely to have negative responses to these gender-bending practices
and can use tactics such as negative stereotyping to preclude women
from accessing the brand's identity meanings. These actions reinforce
the masculinity of the brand and cause consumers to display hypermasculine behaviors to gain entry into the brand's community of
owners. As such, the managers of masculine identity marker brands
should be wary of gender-bending branding practices. Rather than
stretching the brand's meaning to make it more flexible, these practices
might paradoxically make the brand more male than ever before, precluding future brand expansion opportunities.
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
Acknowledgments
This paper evolved from the author's dissertation at the Harvard
Business School. The author extends her deepest appreciation to the
members of her dissertation committee, John Deighton, John Gourville,
Susan Fournier, and Douglas Holt, for their inspiration and guidance, to
her Simmons colleagues, Stacy Blake-Beard and Deborah Kolb for their
encouragement and ideas, and to the reviewers and participants of
the 2008 ACR Gender, Marketing, and Consumer Behavior for their
helpful advice.
References
Addelston, J., & Stirratt, M. (1996). The last bastion of masculinity: Gender politics at
the citadel. In C. Cheng (Ed.), Masculinities in organizations (pp. 54–76). Thousand
Oaks, CA: Sage.
Aldwin, C. M., & Levenson, M. R. (2001). Stress, coping, and health at midlife: A developmental perspective. In M. E. Lachman (Ed.), Handbook of midlife development. : John Wiley.
Algesheimer, R., Dholakia, U. M., & Herrmann, A. (2005). The social influence of brand
community: Evidence from European car clubs. Journal of Marketing, 69, 19–34.
Anderson, E. (2002). Openly gay athletes: Contesting hegemonic masculinity in a homophobic environment. Gender and Society, 16, 860–877.
Belk, R. W. (2004). Men and their machines. Advances in Consumer Research, 31,
273–278.
Belk, R. W., Wallendorf, M., & Sherry, J. F., Jr. (1989). The sacred and the profane in consumer behavior: Theodicy on the odyssey. Journal of Consumer Research, 16, 1–38.
Bem, S. L. (1974). The measurement of psychological androgny. Journal of Consulting
and Clinical Psychology, 42, 155–162.
Bem, S. L. (1993). The Lenses of Gender: Transforming the Debate on Sexual Inequality.
New Haven, CT: Yale University Press.
Bordo, S. (1999). The male body: A new look at Men in public and private spaces. New
York: Farrar, Straus and Giroux.
Borgerson, J., & Schroeder, J. E. (2004). Identity and Iteration: Marketing and the Constitution
of Consuming Subjects. Paper presented at the ACR Conference on Gender, Marketing,
and Consumer Behaviour.
Borgeson, J., & Rehn, A. (2004). General economy and productive dualisms. Gender,
Work and Organization, 11(4), 455–474.
Bradsher, K. (2002). High and mighty: The dangerous rise of the SUV. New York: Public
Affairs.
Brannon, R. (1976). The male Sex role: Our culture's blueprint of manhood and what
it's done for us lately. In D. David, & R. Brannon (Eds.), The forty-nine percent majority: The male sex role. New York: Random House.
Bristor, J. M., & Fischer, E. (1993). Feminist thought: Implications for consumer research. Journal of Consumer Research, 19, 518–536.
Brunsdon, C. (2006). The feminist in the kitchen: Martha, Martha and Nigella. In J.
Hollows, & R. Moseley (Eds.), Feminism in popular culture. New York: Berg.
Burke, P. J. (1991). Identity processes and social stress. American Sociological Review, 56,
836–849.
Butler, J. (1990). Gender trouble: Feminism and the subversion of identity. New York:
Routledge.
Butler, J. (1997). The future of gender, seminar at the pembroke center. December 12,
1997. Providence, Rhode Island: Brown University.
Carrigan, T., Connell, B., & Lee, J. (1985). Toward a new sociology of masculinity. Theory
and Society, 5, 551–604.
Clarke, J. (1986). Style. In S. Hall, & T. Jefferson (Eds.), Resistance through rituals: Youth
subcultures in post-war britain. London: Hutchinson.
Connell, R. W. (1993). The big picture: Masculinities in recent world history. Theory
and Society, 22, 597–623.
Conway-Long, D. (1994). Ethnographies and masculinities. In H. Brod, & M. Kaufman
(Eds.), Theorizing masculinities: Research on Men and masculinities. London: Sage.
Cornwall, A., & Lindisfarne, N. (1994). Dislocating masculinity. : London Routledge.
Cova, B., & Cova, V. (2002). Tribal marketing: The tribalization of society and its impact
on the conduct of marketing. European Journal of Marketing, 36(5/6), 595–620.
Crane, D. (1999). Diffusion models and fashion: A reassessment. Annals of the American
Academy of Psychology and Social Science, 566, 13–24.
Deaux, K., & LaFrance, M. (1998). Gender. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.),
(4th Edition). The Handbook of Social Psychology, Vol. 1. (pp. 788–827)Boston, MA:
McGraw-Hill.
Denning, M. (1987). Mechanic accents: Dime novels and working class culture. New York:
Verso.
Donaldson, M. (1993). What is hegemonic masculinity? Theory and Society, 22,
643–657.
Edelbrock, C., & Sugawara, I. (1978). Acquisition of sex-typed preferences. Developmental
Psychology, 14, 614–623.
Faludi, S. (1992). Backlash: The undeclared war against women. London: Vintage.
Firat, A. F. (1994). Gender and consumption: Transcending the feminine? In J. A. Costa
(Ed.), Gender and consumer behavior. Thousand Oaks, CA: Sage.
Fischer, E., & Arnould, S. J. (1990). More than a labor of love: Gender roles and christmas
gift shopping. Journal of Consumer Research, 17(3), 333–345.
Fischer, E., & Gainer, B. (1994). Masculinity and the consumption of organized sports.
In J. A. Costa (Ed.), Gender and consumer behavior. Thousand Oaks, CA: Sage.
335
Fox, K. J. (1987). Real punks and pretenders: The social organization of a counterculture.
Journal of Contemporary Ethnography, 16, 733–745.
Friedan, B. (1963). The sexual sell. In J. B. Schor, & D. B. Holt (Eds.), The consumer society
reader (pp. 26–46). New York: The New Press.
Fry, J. N. (1971). Personality variables and cigarette brand choice. Journal of Marketing
Research, 8, 298–304.
Gershick, J. T., & Miller, S. A. (1995). Coming to terms: Masculinity and physical disability.
In D. Sabo, & D. F. Gordon (Eds.), Men's health and illness: Gender, power, and the body
(pp. 183–204). London: Sage.
Gherardi, S. (1995). Gender, symbolism and organizational cultures. Thousand Oaks, CA: Sage.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for
qualitative research. New York, NY: Aldine De Gruyter.
Goffman, E. (1959). The presentation of self in everyday life. New York: Anchor Books.
Henson, K. D., & Rogers, J. K. (2001). Why Marcia you've changed! male clerical temporary workers doing masculinity in a feminized occupation. Gender and Society, 15,
218–238.
Herdt, G. H. (1981). Guardians of the flutes: Idioms of masculinity. New York: McGraw-Hill.
Hirschman, E. C. (1988). The ideology of consumption: A structural–syntactical analysis of Dallas and dynasty. Journal of Consumer Research, 15, 344–359.
Holbrook, M. B., & Grayson, M. W. (1986). The semiology of cinematic consumption: Symbolic consumer behavior in Out of Africa. Journal of Consumer Research, 13, 374–381.
Hollows, J. (2000). Feminism, femininity, and popular culture. New York: Manchester
University Press.
Hollows, J. (2006). Can I go home Yet? feminism, post-feminism and domesticity. In J.
Hollows, & R. Moseley (Eds.), Feminism in popular culture. New York: Berg.
Holt, D. B. (2004). How brands become icons: The principles of cultural branding. Boston,
MA: Harvard Business School Press.
Holt, D. B., & Thompson, C. J. (2004). Man-of-action heroes: The pursuit of heroic masculinity in everyday consumption. Journal of Consumer Research, 31(2), 425–440.
Iacuone, D. (2005). “Real Men are Tough Guys”: Hegemonic masculinity and safety in
the construction industry. Journal of Men's Studies, 13(2), 247.
Illouz, E. (1997). Consuming the romantic utopia: Love and the cultural contradictions of
capitalism. Berkeley, CA: University of California Press.
Kates, S. M. (2002). The protean quality of subcultural consumption: An ethnographic
account of gay consumers. Journal of Consumer Research, 29(December), 392–429.
Kimmel, M. S. (1996). Manhood in America. New York: Free Press.
Kozinets, R. V. (2001). Utopian Enterprise: Articulating the Meanings of Star Trek's
Culture of Consumption. Journal of Consumer Research, Vol. 28, (1), (2001, June),
pp. 67–88.
Kozinets, R. V. (2002). The field behind the screen: Using netnography for marketing
research in online communities. Journal of Marketing Research, 39(1), 61–72.
Kramer, L. (2005). The sociology of gender. Los Angeles, CA: Roxbury Publishing
Company.
Lane, R. C., & Sternberg, D. (1985). The use of the automobile as a narcissistic object,
extension of the self, and a personal metaphor. In H. S. Strean (Ed.), Love and
narcissism in psychoanalytic practice (pp. 81–97). New York: Haworth Press.
Leigh, T. W., Peters, C., & Shelton, J. (2006). The consumer quest for authenticity: The
multiplicity of meanings within the MG subculture of consumption. Journal of the
Academy of Marketing Science, 34(4), 481–493.
Lindsey, L. L. (1997). Gender roles. Upper Saddle River, NJ: Prentice-Hall.
Lorber, J. (1994). Paradoxes of gender. New Haven, CT: Yale University Press.
Luedicke, M. K., Thompson, C. J., & Giesler, M. (2009). Consumer identity work as moral
protagonism: How myth and ideology animate a brand-mediated moral conflict.
Journal of Consumer Research, 36(April).
Martin, D. M., Schouten, J. W., & McAlexander, J. H. (2006). Claiming the throttle: Multiple
femininities in a hyper-masculine subculture. Consumption, Markets and Culture, 9(3),
171–205.
McCall, G. J., & Simmons, J. L. (1966). Identities and interactions: An examination of
human associations in everyday life. New York: Free Press.
McCracken, G. (1988). Culture and consumption: New approaches to the symbolic character
of consumer goods and activities. Bloomington, IN: Indiana University Press.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Mishkind, M. E., Rodin, J., Silberstein, L. R., & Moore-Striegel, R. H. (1987). The embodiment of masculinity: Cultural, psychological, and behavioural dimensions.
In M. S. Kimmel (Ed.), Changing men: New directions in research on men and
masculinity. London: Sage Publications.
Muniz, A., & O'Guinn, T. C. (2001). Brand community. Journal of Consumer Research, 27,
412–432.
Muniz, A., & Schau, H. J. (2005). Religiosity in the abandoned apple Newton brand
community. Journal of Consumer Research, 31(4), 737–747.
Osgerby, B. (2001). Playboys in paradise. New York: Oxford International Publishers.
Palan, K. M. (2001). Gender identity in consumer behavior research: A literature review
and research agenda. Academy of Marketing Science Review, 2001(10).
Patterson, M., & Elliott, R. (2002). Negotiating masculinities: Advertising and the
inversion of the male gaze. Consumption, Markets and Culture, 5, 231–246.
Penaloza, L. (1994). Crossing boundaries/drawing lines: A look at the nature of gender
boundaries and their impact on marketing research. International Journal of Research
in Marketing, 11, 359–379.
Radway, J. (1984). Reading the romance. Chapel Hill, NC: University of North Carolina
Press.
Register, W. (2001). Everyday Peter Pans: Work, manhood, and consumption in urban
America, 1900–1930. In R. Horowitz (Ed.), Boys and their toys (pp. 199–228). New
York: Routledge.
Rinallo, D. (2007). Metro/fashion/tribes of men: Negotiating the boundaries of men's
legitimate consumption. In B. Cova, R. V. Kozinets, & A. Shankar (Eds.), Consumer
tribes. Oxford: Elsevier.
336
J. Avery / Intern. J. of Research in Marketing 29 (2012) 322–336
Sachdev, I., & Bourhis, R. Y. (1991). Power and status differentials in minority and majority group relations. European Journal of Social Psychology, 17, 277–293.
Schau, H. J., Muniz, A. M. J., & Arnould, E. J. (2009). How brand community practices
create value. Journal of Marketing, 73, 30–51.
Schouten, J. W., & McAlexander, J. H. (1995). Subcultures of consumption: An ethnography
of the new bikers. Journal of Consumer Research, 22(June), 43–61.
Sherry, J. F., Jr., Kozinets, R. V., Duhachek, A., DeBerry-Spence, B., Nuttavuthisit, K., &
Storm, D. (2004). Gendered behavior in a male preserve: Role playing at ESPN
zone Chicago. Journal of Consumer Psychology, 14(1 & 2), 151–158.
Simpson, M. (1994). Here come the mirror men. November 15, 1994.
Spence, J. T., & Helmreich, R. L. (1978). Masculinity & femininity: Their psychological
dimensions, correlates, and antecedents. Austin, TX: University of Texas Press.
Stern, B. (1988). Sex role self concept measures and marketing: A research note.
Psychology and Marketing, 5(1), 85–99.
Stuteville, J. R. (1971). Sexually polarized products and advertising strategy. Journal of
Retailing, 47(2), 3–13.
Swann, W. B. J., & Ely, R. J. (1984). A battle of wills: Self-verification versus behavioral
confirmation. Journal of Personality and Social Psychology, 46, 1287–1302.
Tajfel, H., & Turner, J. (1979). An integrative theory of intergroup conflicts. In W. G. A. a. S.
Worchel (Ed.), The social psychology of intergroup relations (pp. 33–47). Monterey,
CA: Brooks-Cole.
Thompson, C. J., & Haytko, D. L. (1997). Speaking of fashion: Consumers' Uses of fashion
discourses and the appropriation of countervailing cultural meanings. Journal of
Consumer Research, 24, 15–42.
Thompson, C. J., & Holt, D. B. (2004). How do men grab the phallus? Gender tourism in
everyday consumption. Journal of Consumer Culture, 4(3), 313–338.
Tuncay, L., & Otnes, C. (2008). The use of persuasion management strategies by
identity-vulnerable consumers: The case of urban heterosexual male shoppers.
Journal of Retailing, 84(4), 487–499.
Vitz, P. C., & Johnson, D. (1965). Masculinity of smokers and the masculinity of cigarette
images. Journal of Applied Psychology, 49(3), 155–159.
Wallendorf, M., & Arnould, E. J. (1991). ‘We Gather Together’: The consumption rituals
of thanksgiving day. Journal of Consumer Research, 18(June), 13–31.
Weems, R. E., Jr. (2005). “Bling-bling” and other recent trends in African American consumerism. In C. Conrad, J. Whitehead, P. Mason, & J. Stewart (Eds.), African Americans
in the U.S. economy. : Rowman and Littlefield.
West, C., & Zimmerman, D. H. (1987). Doing gender. Gender and Society, 1, 125–151.
Wetherell, M., & Edley, N. (1999). Negotiating hegemonic masculinity: Imaginary positions and psycho-discursive practices. Feminism and Psychology, 9,
335–356.
Worth, L. T., Smith, J., & Mackie, D. M. (1992). Gender schematicity and preference for
gender-typed products. Psychology and Marketing, 9(1), 17–30.
Intern. J. of Research in Marketing 29 (2012) 337–345
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
Happily (mal)adjusted: Cosmopolitan identity and expatriate adjustment
Amir Grinstein a, b, 1, Luc Wathieu c,⁎
a
b
c
Guilford Glazer Faculty of Business and Management, Ben Gurion University of the Negev, Beer Sheva 84105, Israel
Faculty of Economics and Business Administration, VU University Amsterdam, De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands
Georgetown University, McDonough School of Business, Washington, DC 20057, United States
a r t i c l e
i n f o
Article history:
First received in 15, July 2011 and was under
review for 5 months
Available online 17 September 2012
Keywords:
Consumer identity
Cosmopolitanism
Expatriate adjustment
Horizon effects
a b s t r a c t
As an increasing portion of the world's population identifies itself as cosmopolitan, we examine whether cosmopolitan identity involves openness and adaptability to new environments or instead favors maintaining a
global lifestyle that persists across environments. Based on a field study of expatriates, we find that the
expected duration of sojourn is a crucial moderator of cosmopolitan behavior. In short-duration sojourns,
cosmopolitans adjust more to new environments than non-cosmopolitans. In long-duration sojourns,
non-cosmopolitans adjust more to the host country while cosmopolitans tend to retreat into a global lifestyle.
We find that these adjustment choices are correlated with well-being, contrary to the claims in existing literature on expatriates that adjustment should be the preferred behavior regardless of consumer identity.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
“Where are you from?” is one of the most commonly asked identity
questions in everyday life, a question that is generally useful to predict
relevant traits like taste and attitude. Research and practice have shown
that much can be inferred from someone's current address
(Steenburgh, Ainslie, & Engebretson, 2003) and history of acculturation
(e.g., Birman & Trickett, 2001; Kang, 2006; Penaloza, 1994). However, in
an increasingly globalized society, many individuals claim a cosmopolitan identity that is more relevant to them than the idiosyncrasies of
their personal journey across locations. These individuals favor consumption and lifestyle habits that transcend local conditions. Their
values, interests, and behaviors are cross-cultural, and they view themselves as ‘citizens of the world’. While there is a body of research on various sociological aspects of the cosmopolitan identity (see, for example,
Gouldner, 1957; Hannerz, 1990), its implications for consumer behavior
remain largely uncharted (exceptions include Cannon & Yaprak, 2002;
Riefler & Diamantopoulos, 2009; Thompson & Tambyah, 1999).
In this paper, we study the adjustment behavior of expatriate consumers in connection with their cosmopolitan identity. A basic tenet of
previous literature is that successful expatriates need to adjust their lifestyle to the host country. Failure to do so has been associated with mental health issues, identity confusion, dissatisfaction, poor performance,
and thus high turnover for the sponsoring organization (e.g., Birdseye
& Hill, 1995; Black, Mendenhall, & Oddou, 1991; Black & Stephens,
1989; Bolino, 2007; Stahl, Miller, & Tung, 2002; Takeuchi, Shay, & Li,
⁎ Corresponding author. Tel.: +1 202 687 2783.
E-mail addresses: gramir@bgu.ac.il (A. Grinstein), lw324@georgetown.edu
(L. Wathieu).
1
Tel.: +972 8 6461221.
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.03.003
2008). However, although intuition suggests that cosmopolitan expatriates should be adept at relocation, it is not clear what adjustment strategy we might actually expect from them. ‘Citizens of the world’ might be
more open and adaptable than their non-cosmopolitan counterparts,
but they are also likely to seek global consumption patterns that they
carry with them wherever they reside. Interestingly, the prominent
global cities of the world (e.g., London, New York, Hong Kong, Paris, Singapore, Tokyo, Sydney, Milan, Shanghai, etc., see Sassen, 1991) usually
combine strong local flavors and characteristics with exceptional access
to global products and services. Many banks, academic institutions,
housing developments, retailers, hotels and restaurants, among others,
target cosmopolitan consumers, and additional insights into their preferences should prove useful.
Our analysis reveals that the cosmopolitan identity is associated
with a specific pattern of expatriate adjustment depending on the
expected duration of sojourn. We find that cosmopolitans engaged in
a short sojourn adjust more to local patterns of consumption than
non-cosmopolitans, whereas cosmopolitans engaged in a long sojourn
adjust less to these local patterns than their non-cosmopolitan counterparts. We predict (and show) that these adjustment strategies are driven by differences in the determinants of well-being. Non-cosmopolitans
derive well-being from the status quo, adjusting only if duration makes
the effort worthwhile, whereas cosmopolitans derive well-being from
the pursuit of a global lifestyle that can nevertheless be enriched by
short-term experimentation with local choices.
Our study contributes to several theoretical and practical perspectives. First, we identify two relevant dimensions (i.e., degree of cosmopolitanism and expected duration of sojourn) that should be
taken into account when addressing expatriate consumer segments.
Second, we challenge the nearly universally accepted idea in the organizational behavior and management literatures that expatriates
338
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
are made better off by adjusting to their host country (e.g., Black et al.,
1991; Stahl et al., 2002; Takeuchi et al., 2008; Van Vianen, De Pater,
Kristfof-Brown, & Johnson, 2004). Finally, we offer a better understanding of the behaviors that evolve from a cosmopolitan identity.
The study of expatriate adjustment as a function of the expected duration of sojourn leads to a better understanding of preference persistence or status quo biases in the presence of horizon effects (Shu &
Gneezy, 2010; Wathieu, 1997). Our findings show that individuals
hold identity beliefs that significantly recast intertemporal decisions
into larger orientations.
In the following sections, we develop our hypotheses, outline our
empirical approach and results, and draw conclusions that highlight
limitations and avenues for future research.
2. Expatriates, cosmopolitan identity, and adjustment behavior
Expatriates are people who take a work or study assignment in a foreign location, independently or sent by an organization to which they
belong—often a multinational corporation (Harzing, 2001; Inkson,
Arthur, Pringle, & Barry, 1997). Expatriation is a dramatically increasing
phenomenon, and expatriates have a significant impact on the global
economy, especially through the execution of international business
strategies and the transfer of knowledge across countries (Richardson
& McKenna, 2006; Stahl et al., 2002; Suutari & Brewster, 2000). Expatriates are distinct from immigrants in that their sojourn has a finite time
duration that is likely to influence adjustment to the host country. Our
analysis draws a distinction between two types of expatriates: the
(baseline case) non-cosmopolitan expatriate and the cosmopolitan expatriate. We propose that the costs associated with adjustment and
non-adjustment to the host country, and the evolution of these costs
over time, are different for each type of expatriate.
Starting with the baseline case, we propose that non-cosmopolitan
expatriates will find adjustment costly at first, but increasingly less costly as adaptation begins to occur. Consumers plunged into a new environment ordinarily prefer to persist with previous choices, even when
these are dysfunctional or suboptimal with respect to the new environment. Preference persistence has been attributed to loss aversion
(Samuelson & Zeckhauser, 1988) or to aversion toward uncertainties
and ambiguities characterizing new environments (Muthukrishnan,
1995; Muthukrishnan, Wathieu, & Xu, 2009). Adjusting to the host
country thus involves mental effort, in addition to more concrete search
costs (Samuelson & Zeckhauser, 1988). Both of these costs will diminish
over time, which is captured by the downward-sloping marginal cost
curve in Fig. 1.
In contrast, the costs associated with no adjustment should remain relatively incompressible over time. Such costs might include
shopping at specialty stores featuring products imported from the
home country, transporting and keeping stocks from the home country, or postponing activities (e.g., a visit to the dentist) until the next
trip home. As suggested in the organizational behavior and management literatures, maladjustment can also cause mental health issues,
identity confusion, dissatisfaction, poor performance, turnover, and
overall poor well-being (Birdseye & Hill, 1995; Black & Stephens,
1989; De Cieri, Dowling, & Taylor, 1991; Jun, Gentry, & Hyun, 2001;
Takeuchi, Wang, Marinova, & Yao, 2009).
We propose that the expected sojourn duration will be a key determining factor in the decision to adjust, as captured in Fig. 1: adjustment
for non-cosmopolitan expatriates would not be worthwhile in the context of a short sojourn, but it becomes more likely when the expected
sojourn is sufficiently long and allows adaptation to take place.
Expatriates with a cosmopolitan identity are much less likely to be
driven by preference persistence and gradual adaptation. Indeed, we
expect cosmopolitans to be driven by curiosity and their adventurous
nature, open to the panoply of transcultural diversity (Holt, 1997).
They consume cultural differences in a reflective, intellectualizing
manner and can afford to experiment with local experiences while
they (by definition) are reluctant to take root and permanently adjust
to any local condition (Hannerz, 1990; Thompson & Tambyah, 1999).
Cosmopolitans have been described as variety-seekers, early adopters
of innovations, open-minded, extensive travelers and positive
thinkers (Holt, 1997; Riefler & Diamantopoulos, 2009; Yoon, 1998).
Therefore, expatriation for a cosmopolitan individual is less likely to
coincide with perceptions of ambiguity and loss of subjective competence. Instead, we propose that a short-term adjustment will be experienced as a beneficial process of discovery and learning rather than
as a cost. However, in the course of an extended sojourn, cosmopolitans will progressively come to view adjustment as a loss of their
identity, a costly confinement away from their preferred international
consumption patterns and global lifestyle standards and expectations. This idea is captured in Fig. 2 by a hockey stick-shaped curve
that represents the total cost of adjustment as a function of time
spent in the host country. The graphical analysis then suggests that
Fig. 1. Stylized representation of the baseline effect of expected duration of sojourn on the decision to adjust.
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
339
Fig. 2. Stylized representation of the effect of expected duration of sojourn on the decision to adjust in the cosmopolitan identity case.
a cosmopolitan embarking on a long sojourn will avoid adjustment and
maintain a global lifestyle, while a short sojourn (in sharp contrast with
the baseline case) will involve an embrace of local choices. An alternative
reading of Fig. 2, assuming that expatriates make incremental decisions
(instead of forming an initial adjust/not adjust policy), would predict
that more cosmopolitan individuals adjust for a short period of time
before switching back to their global habits and values, which would
still be consistent with an overall pattern of adjustment in short-term
sojourns contrasting with a reluctance to adjust in long-term sojourns.
The discussion above can be summarized in the form of a testable
hypothesis:
H1. In the presence of a shorter (longer) sojourn, expatriates who
identify themselves as more cosmopolitan adjust more (less) to the
host country, compared with less-cosmopolitan expatriates.
Hypothesis H1 contradicts previous research that almost unanimously represented expatriate adjustment as experientially desirable
(e.g., Bhaskar-Shrinivas, Harrison, Shaffer, & Luk, 2005; Black &
Gregersen, 1991; Mezias & Scandura, 2005). If we are correct, the
well-being outcome of a sojourn to a new country is affected by an
alignment among cosmopolitan identities, expected duration of sojourn, and adjustment to the new environment. This alignment exists
in two scenarios: more-cosmopolitan expatriates will feel more comfortable adjusting when they expect a short-term sojourn and
less-cosmopolitan expatriates will feel more comfortable adjusting
when they expect a long-term sojourn. In contrast, two other scenarios constitute a discrepancy among the same factors that are expected
to have a negative impact on well-being: more-cosmopolitan expatriates will feel less comfortable adjusting when they expect a
long-term sojourn and less-cosmopolitan expatriates will feel less
comfortable adjusting when they expect a short-term sojourn.
This characterization of the preferences associated with the cosmopolitan identity is captured in the following testable hypothesis:
H2. The impact on well-being of a direct relationship between individuals' expected duration of sojourn and their adjustment to the
host country is positive for less-cosmopolitan expatriates and negative for more-cosmopolitan expatriates.
3. A study of expatriates in the United States
3.1. Sample and data collection
We collected data on self-initiated and business expatriates.
Self-initiated expatriates are people who have sought an overseas position independently such as academic researchers, students, entrepreneurs, and to some extent, company employees (Inkson et al., 1997;
Suutari & Brewster, 2000). Business expatriates are employees and
managers sent to a foreign country by their organization with the intent
of controlling company operations or providing technical and administrative services (Harzing, 2001; Jun et al., 2001). While prior research on
expatriates has typically focused on one of the two groups (e.g., Black &
Stephens, 1989; Peltokorpi, 2008), our more holistic sampling approach
has several advantages. First, it ensures an inclusive view of the studied
phenomenon and enhances the generalizability of the study. Second, it
enhances heterogeneity with respect to the key factors studied (i.e., adjustment, cosmopolitan identity, expected duration of sojourn). This
study is interested in consumer adjustment, and consumption is a behavior shared by all types of expatriates (whether students, visiting
professors or managers in multinational corporations). Nevertheless,
we conducted a series of independent-samples t-tests to compare the
two groups of expatriates on all the variables in the study, making
sure that the groups are not different.
We used three sources to generate our sample (N = 260). We first
identified two large foreign communities residing in the area where
the authors lived during the time of the research (near Boston, Massachusetts): an Israeli and a Belgian community. Through the help
of the local Israeli and Belgian community associations, lists of Israelis
and Belgians residing in the area were retrieved, and 750 invitations
were sent via email asking self-initiated and business expatriates to
participate in our survey. The lists included different types of expatriates (e.g., those sent by their organization, students, entrepreneurs,
visiting scholars) and to some extent immigrants. The latter, however, selected themselves out in view of the email invitation and the
questions asked in the survey. Ninety-seven expatriates from these
two groups responded to the survey (a 13% response rate). Approximately 70% of these respondents defined themselves as selfinitiated expatriates, while 30% identified as business expatriates.
340
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
Focusing on this single key dimension of the construct is likely to enhance
the reliability of the measure. As we later describe in the Measurement
model validation section, we made sure that this conceptualization is indeed empirically manifested as a single dimension. The first item asked
respondents to read a text featuring a characterization of a cosmopolitan
(Cannon & Yaprak, 2002: 37) and rate their agreement with the text on a
seven-point Likert scale (anchored by 1 = “strongly disagree,” 7 =
“strongly agree”). The text emphasizes cross-cultural openness and diversity in the context of consumption, social interactions and experiences
with foreign countries and products. It reads as follows: “Many of my
friends long to go home. I have no home. Or rather, I feel like my home is
the world. I like to experience the best of what the world has to offer. I have
friends all over the world, and I enjoy the texture and variety the world has
to offer. I have my favorite foods, but they don't represent any particular culture. I simply like them because they are good. The same is true of my friends.
I savor them when we can get together, even if that is only once every 10 or
20 years. We may communicate by email, or call, if the occasion arises, but I
don't pine away when they are not around me. Life is too short for that. I live
in Singapore, but I could live just as well in New York, Kiev, or Rio de Janeiro. I
like to visit smaller, less cosmopolitan cities because they are interesting. But I
wouldn't want to live there. They don't offer the variety or the level of excellence I have come to enjoy. Most of all, I like to travel. If I want culture, I go to
London; if I want art, I go to Paris; if I want intellect, I go to Moscow. It's all out
there. You just have to find out where.” The other three items were adopted
from Cleveland and Laroche's (2007) global consumer culture scale and
from Jain and Etgar's (1977) cosmopolitanism scale. These items are in
line with the basic premise that “cosmopolitans are individuals with
broad experience in many countries” (Haas, 2006: 367) and that they
have interest in travels and stays abroad (Riefler & Diamantopoulos,
2009). These items were also measured using seven-point Likert agreement scales. The reliability of this measure, 0.67, is similar to that of the
cosmopolitanism scale with the highest reliability that is currently available in the literature (Riefler & Diamantopoulos, 2009).
For expected duration of sojourn, respondents were asked to assess,
in years, the initial expected duration for their current assignment. To
make it easier for respondents to assess the expected duration, we
used a six-point scale with time categories. The reasons for using
this categorization rather than a continuous measure were twofold.
First, past academic and managerial reports tend to categorize expatriates' assignment duration (e.g., Bhaskar-Shrinivas et al., 2005;
Bolino, 2007; GMAC Global Relocation Services, 2008). Second, following interviews with expatriates, we realized that it is sometimes
difficult for them to accurately forecast their expected assignment duration. Rather, they felt much more comfortable when considering
The third source of our sample involved a group of internationals registered at the behavioral lab of the business school with which the authors were affiliated during the time of performing this research. The
behavioral lab's list included both U.S. and non-U.S. citizens, in addition to students and some non-students (typically people who work
at or near the university). Here, a list of 1200 non-U.S. born subjects
(both students and non-students) received via email an invitation
to participate in our study. One hundred and sixty-three agreed to
participate (a 14% response rate). The above response rates are comparable to other studies on expatriates (Birdseye & Hill, 1995;
Gregersen, Hite, & Black, 1996). As in the first sub-sample, approximately 70% of respondents in this group defined themselves as
self-initiated expatriates, and 30% identified as business expatriates.
In the analyses reported later, we controlled for differences between
the two sources of our sample (using a dummy for each of the
sub-samples: the Israeli sub-sample, the Belgian sub-sample and the
sub-sample based on the list from the behavioral lab), and between
self-initiated and business expatriates, showing that our results are
robust across the groups. Furthermore, we made sure that
non-response bias is not an issue. Because the Belgian sub-sample
was relatively small (N = 11), we focused for this purpose on the
Israelis (N = 81) and those sampled by the behavioral lab (N = 168).
Specifically, we compared early and late respondents (Armstrong &
Overton, 1977) on three key variables for which data were available
(i.e., age, time already on assignment, and nature of expatriation—
self-initiated or business) and found that there were no significant
differences between them in all cases. These findings minimize concerns for a non-response bias.
3.2. Measures
Whenever possible, we adopted established survey measures. For an
overview of these measures, the items and loadings on the measures,
and their reliabilities, see Appendix A. The correlation matrix appears
in Table 1.
The cosmopolitan identity construct is still in need of a well-accepted
scale (Cannon & Yaprak, 2002; Haas, 2006, Riefler & Diamantopoulos,
2009). In this study, we adopted a composite measure of a cosmopolitan
identity based on four items derived from a number of sources. The decision to integrate items from different sources is based on the evidence
that prior measurement efforts of cosmopolitanism exhibited very poor
psychometric properties (Riefler & Diamantopoulos, 2009). An effort
was made to identify items that correspond with a key conceptualization
of cosmopolitanism—the notion of cross-cultural openness and diversity.
Table 1
Correlation matrix and descriptive statistics of the study's key constructs.
Constructs
1
2
3
4
5
6
7
8
Cosmopolitan identity
Expected duration of sojourn
Adjustment to the host country
Well-being
Family status
Time already in host country
Type of expatriate: Student
Type of expatriate: multinational corporation
manager/employee
9 Source of sample: Israeli sample
10 Source of sample: Behavioral lab sample
11 Gender
12 Long-term orientation
Mean
SD
†
p b 0.10.
⁎ p b 0.05.
⁎⁎ p b 0.01.
1
2
3
4
5
6
7
8
0.03
0.11†
1
0.14⁎
−0.03
−0.12⁎
0.24⁎⁎
−0.23⁎⁎
0.17⁎⁎
1
0.15⁎
0.08
0.13⁎
−0.05
0.10
1
−0.06
0.05
−0.13⁎
0.09
1
0.04
0.14⁎
−0.09
1
−0.11†
0.03
1
−0.49⁎⁎
1
−0.35⁎⁎
0.28⁎⁎
0.04
−0.18†
4.87
1.39
0.02
−0.07
−0.06
0.27⁎⁎
3.73
1.24
−0.01
−0.04
−0.01
−0.30⁎⁎
5.02
1.31
−0.34⁎⁎
0.32⁎⁎
0.03
0.04
–
–
−0.11
−0.09†
0.10
0.10
43.28
54.84
1
0.03
0.26⁎⁎
0.18⁎⁎
0.21⁎⁎
0.13⁎
−0.20⁎⁎
0.13⁎
0.12⁎
0.07
5.33
1.06
†
−0.28⁎⁎
0.34⁎⁎
−0.02
0.06
–
–
−0.03
0.05
−0.11†
−0.15
–
–
9
10
11
12
1
−0.89⁎⁎
0.02
n.a.
–
–
1
0.03
n.a.
–
–
1
−0.12
–
–
1
57.20
34.42
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
time categories (e.g., between 3 and 5 years, up to one year). Specifically, our scale included the following categories: “less than 1 year,”
“approximately 1 year,” “approximately 2 years,” “approximately
3–5 years,” “approximately 5–10 years” and “10 years or more.”
In addition, we conducted the following robustness check for this
measure. We followed the expatriate literature, which observes that
assignments lasting about one year or less are typically considered
short-term (Bhaskar-Shrinivas et al., 2005; Bolino, 2007). We therefore conducted all the analyses using a dummy code for long versus
short expected sojourn, yielding similar results (additionally, we
conducted all the analyses while using the more conservative approach of 2 years as the cut-off point and obtained similar results).
The measure of adjustment to the host country was adopted from
Black (1988). It included 5 items (e.g., “adjustment to housing conditions”) and was measured using a seven-point Likert scale (anchored
by 1 = “no adjustment,” 7 = “total adjustment”).
Well-being was measured using the ‘satisfaction with life’ scale,
which was adopted from Diener, Emmons, Larsen, and Griffin
(1985). This three-item measure is designed to assess cognitive aspects of well-being (e.g., “the conditions of my life are excellent”). It
was measured using a seven-point Likert agreement scale (anchored
by 1 = “strongly disagree,” 7 = “strongly agree”).
We also collected data on a number of control variables that are
often used in the expatriate literature or that are especially appropriate for this study. We controlled for time already in the host country,
in months, because some of the respondents were already in the
midst of their sojourn, whereas others had just started theirs. We
controlled for family status (dummy for accompanied by relatives
(coded as ‘0’) versus not accompanied by relatives (coded as ‘1’))
based on Van Vianen et al. (2004). We controlled for gender (male
‘0’ and female ‘1’) following Peltokorpi (2008) and Takeuchi et al.
(2009). We monitored for type of expatriate (a dummy for students
(typically self-initiated expatriates), a dummy for business expatriates (typically managers and employees of multinational corporations), and a dummy for “others”) following Richardson and
McKenna (2006), and Jokinen, Brewster, and Suutari (2008). To control for differences among the respondents' cultural backgrounds in
the specific case at hand, we controlled for variability in subjects'
long-term orientation (Hofstede, 1980). This construct can be especially relevant when considering expatriates' time expectations and
time-related decisions. Indeed, a long-term orientation emphasizes
a focus on the future and adaptation of one's own traditions to new
circumstances. People with a long-term orientation believe that the
most important events in life will occur in the future, and they are
more adaptable (Van Evergdingen & Waarts, 2003). Specifically, we
added a variable that captures Hofstede's measure. Since Hofstede
collected data on this variable only in a limited number of countries,
we conducted the analysis twice, with the original Hofstede data
and after replacing the missing values using the linear trend at
point method (Sheskin, 2004). Both analyses yielded similar results.
Finally, because we generated our sample from different sources, we
controlled for this difference using two of the following three
dummy variables—the Israeli sample, the Belgian sample and the
sample based on the university behavioral lab.
3.3. Measurement model validation
The confirmatory factor analysis (i.e., the measurement model
that includes all our scales) is extremely satisfactory: χ 2(51) =
109.998, p b 0.05; CFI = 0.938; IFI = 0.940; RMSEA = 0.067. Using the
estimated measurement model, we assessed the convergent validity
of the constructs. All the item-to-construct loadings were found to
be significant—the lowest t-value was 5.49, demonstrating adequate
convergent validity. We also confirmed the discriminant validity of
the constructs by running a series of models involving a comparison
between pairs of constructs. Specifically, models with construct
341
correlations constrained to 1.00 (suggesting that two constructs can
actually be represented by a single construct) were compared to
unconstrained models. In all cases, this led to a significant increase
in Chi-square and LM-tests, revealing that these constraints should
be removed. Thus, all constructs exhibited discriminant validity.
A specific implication of the above validation process is the confirmation that our cosmopolitan identity measure, the only measure we
are using which is still not established in the literature, involves a single key dimension, which captures the cross-cultural openness and
diversity of cosmopolitans.
3.4. Analysis
We used Structural Equation Modeling (SEM), AMOS 17.0, to test
our hypotheses. All models are satisfactory in terms of their fit indices
(CFI, IFI and RMSEA; see Table 2). To test H1 we ran Model 1, which
involved expatriates' adjustment to the host country as the dependent variable and the two-way interaction between expatriates' cosmopolitan identity and their expected duration of sojourn as the
key explanatory variable. To test H2 we ran Model 2, which involved
expatriates' well-being as the dependent variable and the three-way
interaction among expatriates' cosmopolitan identity, their expected
duration of sojourn, and their adjustment to the host country as the
key explanatory variable. Recall that our theory predicts different effects on well-being of the pair-wise combination of adjustment and
expected duration of sojourn, depending on whether the expatriate
possesses a cosmopolitan identity.
Following convention (Aiken & West, 1991), we mean-centered
the variables constituting the interaction terms and then tested for
multicollinearity, making sure that variance of inflation factors (VIF)
and tolerance are below harmful levels.
3.5. Common method bias
Several steps were taken in the design of the questionnaire to
prevent common method bias (Rindfleisch, Malter, Ganesan, &
Moorman, 2008). First, each of the dependent and independent variables were on different pages of the electronic questionnaire with different instructions. Second, each of these variables employed distinct
endpoints (for example, the well-being measure included the
end-points “strongly agree” versus “strongly disagree,” while the adjustment measure included the end-points “total adjustment” versus
“no adjustment”).
To check for common method bias, we first conducted a Harman's
one-factor test (i.e., factor analysis without rotation). The results indicated that no single general factor existed. Rather, three distinct factors representing our multiple-item scales emerged, and none of the
factors accounted for the majority of the variance among the measures (the most dominant factor accounted for only 23.6% of the variance) (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Second, we
performed the same test on pairs of multiple item scales and found
the test to always produce two distinct factors (Kyriakopoulos &
Moorman, 2004). Third, following Podsakoff et al.'s (2003) and
Netemeyer, Boles, McKee, and McMurrian's (1997) guidelines, we incorporated a ‘same-source’ factor (i.e., a single-common-methodfactor) into the indicators of all our theoretical constructs in each of
the two structural models we ran. In all of the cases, the hypothesized
relationships were not influenced by the same-source factor and
retained their influence and direction. Moreover, in the vast majority
of cases, the indicator loadings to the theoretical constructs remained
significant (only one indicator in one of the models became insignificant). Finally, only a minority of indicator loadings on the samesource factor were significant. We thus conclude that there is no
ground for concern regarding a common method bias in our analysis.
Furthermore, the presence of significant interactions among the
model constructs as described below also supports a conclusion that
342
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
Table 2
Results of SEM models.
Control variables
Family status
Time already at the host country
Type of expatriate: student
Type of expatriate:
multinational corporation
manager/employee
Source of sample: the Israeli
sample
Source of sample: behavioral
lab sample
Gender
Long-term orientation
Main effects
Cosmopolitan identity
Expected duration of sojourn
Adjustment to host country
Interaction effects
Cosmopolitan
identity × expected
duration of sojourn
Cosmopolitan
identity × adjustment to
host country
Expected duration of
sojourn × adjustment to
host country
Cosmopolitan
identity × expected
duration of
sojourn × adjustment to
host country
Model fit indices
χ2(df)
p-Value
CFI
IFI
RMSEA
Adjustment to the
host country (Model 1)
Well-being
(Model 2)
−0.036
(0.151)
0.049
(0.001)
−0.096
(0.185)
0.010
(0.187)
−0.055
(0.178)
−0.026
(0.002)
−0.155⁎
(0.216)
−0.047
(0.221)
−0.132
(0.380)
−0.109
(0.354)
0.090
(0.142)
0.051
(0.003)
−0.005
(0.454)
−0.058
(0.417)
−0.055
(0.170)
−0.148⁎⁎
(0.004)
0.218⁎⁎
(0.109)
0.126⁎
(0.060)
-
0.272⁎⁎
(0.141)
0.015
(0.075)
0.072
(0.085)
−0.174⁎⁎
(0.045)
−0.122
(0.053)
–
−0.074
(0.054)
–
0.136⁎⁎
(0.059)
–
−0.154⁎⁎
(0.037)
240.55(107)
0.000
0.909
0.914
0.069
401.25(193)
0.000
0.892
0.900
0.065
We report standardized coefficients (standard errors in parentheses).
We highlight in bold the hypothesized effects.
⁎⁎ p b 0.05.
⁎ p b 0.1.
the respondents reliably captured effects of the phenomena without
substantial bias (Blalock, 1971).
4. Results
Our results are reported in Table 2. Accordingly, H1 is supported as
the tendency of expatriates to adjust to the host country is shown to depend on the combination of our key constructs of interest: expatriates'
cosmopolitan identity and their expected duration of sojourn. Specifically, we find, as the results of Model 1 illustrate, that the more (less) cosmopolitan the expatriate, the more she/he will tend to adjust to the local
environment when expecting a short-term (long-term) sojourn in the
host country (β= −0.174, p =0.009). A breakdown of the results,
displayed in Fig. 3, shows how cosmopolitan expatriates were more likely to adjust than non-cosmopolitan expatriates when expecting a
short-term sojourn (Mcosmopolitans =5.88, Mnon-cosmopolitans =4.86,
t105 =5.304, pb .0001). In contrast, non-cosmopolitan expatriates were
more likely to adjust than cosmopolitan expatriates when expecting a
long-term sojourn (Mnon-cosmopolitans =5.52, Mcosmopolitans =5.20, t151 =
2.725, p=.007).
Our results also support H2, as the three-way interaction is significant and negative (β= −0.154, p = 0.030). Accordingly, as Model 2
suggests, expatriates with varying cosmopolitanism levels generally report higher levels of well-being in the course of adjusting to local conditions only when there is an alignment between cosmopolitan identity
and the expected duration of the sojourn. A breakdown of the results,
displayed in Figs. 4a and 4b, sheds more light on the results of Model
2. As predicted, among more-cosmopolitan expatriates, general lifestyle
adjustment increases well-being only during expected short-term sojourns (Mcosmopolitans/high adjustment = 5.04, Mcosmopolitans/low adjustment =
4.71, t54 = 2.251, p = .028). The opposite is true for expected
long-term sojourns where adjustment among more-cosmopolitan
expatriates decreases well-being (Mcosmopolitans/high adjustment =
4.49, Mcosmopolitans/low adjustment = 5.92, t65 = 3.981, p b .0001).
For less-cosmopolitan expatriates, general life-style adjustment
increases well-being only during expected long-term sojourns
(Mnon-cosmopolitans/high adjustment =5.58, Mnon-cosmopolitans/low adjustment =
4.51, t82 =5.592, pb .0001). During expected short-term sojourns, higher
or lower levels of adjustment among less-cosmopolitan expatriates lead
to similar degrees of well-being (Mnon-cosmopolitans/high adjustment =5.22,
Mnon-cosmopolitans/low adjustment =4.94, t49 =1.131, p=.261).
5. Discussion
Our study of expatriates supports the suggestion that a cosmopolitan expatriate expecting a longer duration of sojourn and a
less-cosmopolitan expatriate expecting a shorter duration of sojourn
would be less inclined to adjust to their host country, where the reluctance to adjust in both cases is typically associated with greater
well-being. This result conflicts with existing assumptions and findings in acculturation and expatriation literatures, which consider adjustment as unambiguously good. Nevertheless, the result matches
our theoretical model, which interprets cosmopolitan identity as involving a preference for lifestyles and habits that transcend local conditions. In the context of a de-localization, the more-cosmopolitan
expatriates will be curious at first and then eventually become disinclined toward local habits. In contrast, less-cosmopolitan expatriates
will start with a period of costly adjustment before they enjoy their
new place. Our finding is in fact consistent with some recent studies
that show expatriate adjustment to be a more ambiguous issue than
prior studies had suggested (e.g., Bolino, 2007; Takeuchi et al., 2008).
Our study constitutes an unusual attempt to analyze preference
adaptation and persistence in a real-life intertemporal context.
People's adaptation when they move to a different context has been
shown in experimental research to be stronger when feelings toward
the earlier context are more ambiguous (Muthukrishnan, 1995). Our
research adds the new realistic insight that preference persistence
is also influenced by the anticipated length of sojourn in the new
context: a longer sojourn will reduce (improve) the degree of adaptation of more- (less-) cosmopolitan expatriates. This finding might
also be reminiscent of the psychological theory of temporal construals
(e.g., Liberman, Trope, & Wakslak, 2007), which suggests that an
abstract identity concept such as cosmopolitanism will be more salient for individuals who contemplate long-term horizons (Wakslak,
Nussbaum, Liberman, & Trope, 2008).
From a managerial point of view, our findings pose an interesting
dilemma. On the one hand, cosmopolitans form a prime target for expatriate assignments; they have a global perspective, and because their
motivation for expatriation is presumably more personal/experiential
(Stahl et al., 2002; Tung, 1998) it can compensate for the perceived career risk implied by overseas assignments (Stahl et al., 2002). On the
other hand, our findings suggest that cosmopolitans might require
and enjoy longer assignments only at the expense of predictably
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
343
7
5.88
Adjustment
5.52
High cosmopolitan identity
5.20
4.86
Low cosmopolitan identity
1
Short expected sojourn Long expected sojourn
Fig. 3. Two-way interaction—cosmopolitan identity times expected duration of sojourn on adjustment.
lower degrees of adjustment to local conditions. By the same logic, organizations and the expatriate herself/himself should carefully consider
short-term assignments when the expatriate is less cosmopolitan as
she/he might enjoy shorter assignments only at the expense of a
lower degree of adjustment. Furthermore, in the context of the trend toward shorter international assignments, which is part of a movement to
reduce expenses in response to economic conditions (GMAC Global
Relocation Services, 2008), organizations may consider paying special
attention to level of cosmopolitanism among their potential expatriates.
Based on our findings, cosmopolitans are likely to be more suitable candidates for short-term expatriations.
While our results may suggest important implications for multinational corporations and expatriates in terms of expatriate selection,
assignment duration, and expectations on performance in international assignments, they may also have implications for choosing
the location of expatriate assignments. Non-cosmopolitans on short
assignment and cosmopolitans on long assignments might both be
preferably directed to cosmopolitan cities. Less-cosmopolitan cities
might be more suitable for cosmopolitans on short assignment or
non-cosmopolitans on longer terms.
The study may also have implications for brand managers who
target the often-affluent population of expatriates in different geographical contexts (Andreasen, 1966).
6. Limitations and future research directions
This field-based study has allowed us to discuss important variables in real contexts where they matter most. However, one drawback of such studies is that causality is not being addressed. To
address this issue, longitudinal research designs may be more relevant. Additionally, our results might be subject to self-selection
and endogeneity biases, e.g., if cosmopolitans who do not need adjustment naturally value (or are preferably selected for) assignments of long duration. There might be specific heterogeneity
a
7
5.92
Well-being
5.04
4.71
4.49
Higher adjustment
Lower adjustment
1
Short expected sojourn
b
Long expected sojourn
7
5.58
Well-being
5.22
4.94
4.51
Higher adjustment
Lower adjustment
1
Short expected sojourn
Long expected sojourn
Fig. 4. a: Three-way interaction—cosmopolitan identity times expected duration of sojourn times adjustment on well-being: the high cosmopolitan identity case. b: Three-way
interaction—cosmopolitan identity times expected duration of sojourn times adjustment on well-being: the low cosmopolitan identity case.
344
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
effects associated with self-initiated vis-à-vis business expatriates,
although our findings are just as robust when controlling for their
effects. Future work may consider a systematic comparison between
the two groups of expatriates, as we are not aware of such studies at
present.
In addition, our dependent variables are subjective in nature and
may benefit from some objective measurement to validate our subjects' responses (see for example Takeuchi et al., 2008, 2009). Nevertheless, efforts were made to ensure that common method bias was
not a concern, starting with the design of the questionnaire and
finishing with a series of tests for common method bias.
Lack of consensus on the conceptualization and measurement of
the cosmopolitanism construct and poor performance of prior measures led us to develop our own cosmopolitan identity measure, focusing on the underlying meaning of the construct and one of its
key dimensions—the notion of cross-cultural openness and diversity.
While it is based on items taken from different sources, our measure
exhibits reliability that is similar to that of the cosmopolitanism scale
with the highest reliability available in the literature. Nevertheless,
there is room for further research on the development and validation
of a cosmopolitanism measure. A relevant point of departure is the
work by Riefler and Diamantopoulos (2009), and that by Riefler,
Diamantopoulos, and Siguaw (2012), which offer guidance for future
studies involving a cosmopolitanism scale.
The unique nature of the dominant groups among the sampled expatriates—Israelis and Belgian—may be a source of concern. Specifically, Israelis and Belgians may have characteristics in common that
differentiate them from citizens of many other (often larger and
presumably less cosmopolitan) countries. However, in the context
of our sample, note that: (a) Israelis and Belgians represent only
part of the sample (81 and 11, respectively, in a total sample size of
260) which is in fact composed of a variety of expatriates from
many countries, (b) we control for these sub-samples' main effects
and observe that our results hold above and beyond these controls.
Similarly, the location of our study—the Boston area, with its unique
characteristics—may limit the generalizability of our study to other
locations. Yet the Boston area may have actually provided us with
some advantages: Boston is a large-enough city—it is neither too
small nor too large, it is neither clearly cosmopolitan nor clearly
non-cosmopolitan, and it is home to a variety of expatriate types (academic, high-tech, etc.).
Other relevant and potentially interesting venues for future research might be as follows.
First, cultural differences and background conditions involving
uncertainty and ambiguity might moderate the effects found here.
There might be effects from the proximity or distance between the
home and the host country. As a further example, the study of the
impact of expected sojourn duration on expatriates' behavior and
performance may benefit from exploring the role of relevant cultural constructs based on Hofstede's (1980) cultural framework. We
studied subjects' long-term orientation, but other cultural dimensions such as uncertainty avoidance may also play a crucial part.
Similarly, economic and educational background conditions can
play a role in explaining expatriates' behavior and may be given
more attention.
Second, the study of social networks is gaining interests and relevance in marketing and management (e.g., Stam & Elfring, 2008;
Van den Bulte & Wuyts, 2007). Studying the social networks of
expatriates may be helpful in understanding their impact on expatriates' decisions and behaviors. The specific effects of such networks, and the more general study of expatriates' communities,
on well-being and assignment performance would itself be a matter
of interest.
Third, a better understanding of the phenomenon studied may be
gained by studying relevant mediators. Such potential mediators may
include a host of expatriates' perceptions and capabilities, such as
expatriates' expectations regarding the outcomes associated with
long- and short-term sojourns, and their level of “comfort” with
their level of self-perceived adjustment or their learning capabilities.
In sum, it appears that studies on expatriates (not only crosssectional but also longitudinal studies) can be fertile grounds for insights that are not only useful for contemporary multinational corporations but also elucidate variables of great general theoretical
interest such as psychological and physical distance, or the general
phenomenon of preference and habit adaptation.
Acknowledgment
The authors would like to thank Liat Sadot for her invaluable help
in the data collection process. They would also like to acknowledge
helpful comments from Alan Andreasen, Ann Kronrod and seminar
participants in HEC Montréal, Tel-Aviv University and Ben-Gurion
University.
Appendix A. The measures
Measures and their items
Cosmopolitan identity
My experience is similar to the one
described in the attached text (“being
a cosmopolitan”).
I have extensively visited countries other
than my home country.
I have resided in countries other than my
home country.
In the future, there is great likelihood that
I will live in countries other than my
home country.
Expected duration of sojourn
(1) less than 1 year (2) approximately
1 year (3) approximately 2 years (4)
approximately 3–5 years (5)
approximately 5–10 years (6)
10 years or more
Adjustment to the host country
Living conditions in general
Housing conditions
Food
Shopping
Entertainment
Sources
λ = 0.58
Cannon and
Yaprak (2002)
Cleveland and
λ = 0.43
Laroche
(2007)
Jain and Etgar λ = 0.56
(1977)
λ = 0.64
0.67
–
–
–
Black (1988)
λ = 0.87
λ = 0.88
λ = 0.50
λ = 0.46
λ = 0.61
0.83
λ = 0.92
λ = 0.77
λ = 0.73
0.83
–
–
–
–
–
–
–
–
–
Peltokorpi
(2008)
Takeuchi et al.
(2009)
Hofstede
–
(1980)
–
Well-being (satisfaction with life)
In most ways, my life is close to my ideal. Diener et al.
The conditions of my life are excellent.
(1985)
If I could live my life over, I would
change almost nothing.
Control variables
Family situation (accompanied/not
accompanied by relatives)
Time already in the host country
Type of expatriate (student/business
expatriate/other)
Source of sample (Israeli, Belgian or
university behavioral lab)
Gender
Long-term orientation
Loadings Cronbach's
alpha
Van Vianen et
al. (2004)
–
Jokinen et al.
(2008)
Richardson
and McKenna
(2006)
–
–
A. Grinstein, L. Wathieu / Intern. J. of Research in Marketing 29 (2012) 337–345
References
Aiken, L., & West, S. (1991). Multiple regression: Testing and interpreting interactions.
Newbury Park, CA: Sage Publications.
Andreasen, A. R. (1966). Geographic mobility and market segmentation. Journal of
Marketing Research, 3, 341–348.
Armstrong, S., & Overton, T. (1977). Estimating nonresponse bias in mail surveys. Journal
of Marketing Research, 14, 396–400.
Bhaskar-Shrinivas, P., Harrison, D., Shaffer, M. A., & Luk, D. M. (2005). Input-based and
time-based models of international adjustment: Meta-analytic evidence and theoretical extensions. Academy of Management Journal, 48(2), 257–281.
Birdseye, M. G., & Hill, J. S. (1995). Individual, organizational/work and environmental
influences on expatriate turnover tendencies: An empirical study. Journal of International Business Studies, 26, 787–813.
Birman, D., & Trickett, E. J. (2001). Cultural transitions in first-generation immigrants:
Acculturation of Soviet Jewish refugee adolescents and parents. Journal of
Cross-Cultural Psychology, 32(4), 456–477.
Black, J. S. (1988). Workrole transitions: A study of American expatriate managers in
Japan. Journal of International Business Studies, 19, 277–294.
Black, J. S., & Gregersen, H. B. (1991). Antecedents to cross-cultural adjustment for
expatriates in Pacific Rim assignments. Human Relations, 44, 497–515.
Black, J. S., Mendenhall, M. E., & Oddou, G. (1991). Toward a comprehensive model of
international adjustment: An integration of multiple theoretical perspectives.
Academy of Management Review, 16, 291–317.
Black, J. S., & Stephens, G. K. (1989). The influence of the spouse on American expatriate
adjustment and intent to stay in Pacific Rim overseas assignments. Journal of Management, 15(4), 277–294.
Blalock, H. M. (1971). Causal models involving unmeasured variables in stimulus–
response situations. In H. M. BlalockJr. (Ed.), Causal models in the social sciences
(pp. 335–347). New York: Aldine-Atherton.
Bolino, M. C. (2007). Expatriate assignment and intra-organizational career success:
Implications for individuals and organizations. Journal of International Business
Studies, 38, 819–835.
Cannon, H. M., & Yaprak, A. (2002). Will the real-world citizen please stand up! The
many faces of cosmopolitan consumer behavior. Journal of International Marketing,
10(4), 30–52.
Cleveland, M., & Laroche, M. (2007). Acculturation to the global consumer culture:
Scale development and research paradigm. Journal of Business Research, 60,
249–259.
De Cieri, H., Dowling, P. J., & Taylor, K. F. (1991). The psychological impact of expatriate
relocation on partners. International Journal of Human Resource Management, 2(3),
377–414.
Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life
scale. Journal of Personality Assessment, 49(1), 71–75.
GMAC Global Relocation Services (2008). Global Relocation Trends 2008 Survey Report. Oak
Brook, IL: GMAC Global Relocation Services, National Foreign Trade Council, and SHRM
Global Forum Available at: http://www.gmacglobalrelocation.com/insight_support/
global_relocation.asp.
Gouldner, A. W. (1957). Cosmopolitans and locals: Toward an analysis of latent social
roles. Administrative Science Quarterly, 2, 281–306.
Gregersen, H. B., Hite, H. B., & Black, J. S. (1996). Expatriate performance appraisal in
U.S. multinational firms. Journal of International Business Studies, 27, 711–738.
Haas, M. (2006). Acquiring and applying knowledge in transnational teams: The roles
of cosmopolitans and locals. Organization Science, 17(3), 313–332.
Hannerz, U. (1990). Cosmopolitans and locals in a world culture. Theory Culture and
Society, 7, 237–251.
Harzing, A. -W. (2001). Of bears, bumble-bees, and spiders: The role of expatriates in
controlling foreign subsidiaries. Journal of World Business, 36, 366–379.
Hofstede, G. (1980). Culture's consequences. Beverly Hills, CA: Sage.
Holt, D. B. (1997). Poststructuralist lifestyle analysis: Conceptualizing the social patterning of consumption in postmodernity. Journal of Consumer Research, 23(4),
326–350.
Inkson, K., Arthur, M. B., Pringle, J., & Barry, S. (1997). Expatriate assignment versus
overseas experience: Contrasting models of international human resource development. Journal of World Business, 32, 351–368.
Jain, A. K., & Etgar, M. (1977). Overlap in retail outlet and product innovator characteristics. Journal of Retailing, 53(2), 63–75.
Jokinen, T., Brewster, C., & Suutari, V. (2008). Career capital during international work
experience: Contrasting self-initiated expatriate experience and assigned expatriation. International Journal of Human Resource Management, 19(6), 979–998.
Jun, S., Gentry, J. W., & Hyun, Y. J. (2001). Cultural adaptation of business expatriates in
the host marketplace. Journal of International Business Studies, 32, 369–377.
Kang, S. -M. (2006). Measurement of acculturation, scale formats, and language competence. Journal of Cross-Cultural Psychology, 37(6), 669–693.
345
Kyriakopoulos, K., & Moorman, C. (2004). Tradeoffs in marketing exploitation and exploration strategies: The overlooked role of market orientation. International Journal of Research in Marketing, 21, 219–240.
Liberman, N., Trope, Y., & Wakslak, C. (2007). Construal level theory and consumer behavior. Journal of Consumer Psychology, 17(2), 113–117.
Mezias, J. M., & Scandura, T. A. (2005). A needs-driven approach to expatriate adjustment and career development: A multiple mentoring perspective. Journal of International Business Studies, 36, 519–538.
Muthukrishnan, A. V. (1995). Decision ambiguity and incumbent brand advantage.
Journal of Consumer Research, 22(1), 98–109.
Muthukrishnan, A. V., Wathieu, L., & Xu, A. J. (2009). Ambiguity aversion and the preference for established brands. Management Science, 55(12), 1933–1941.
Netemeyer, R., Boles, J., McKee, D., & McMurrian, R. (1997). An investigation into the
antecedents of organizational citizenship behaviors in a personal selling context.
Journal of Marketing, 61(3), 85–98.
Peltokorpi, V. (2008). Cross-cultural adjustment of expatriates in Japan. International
Journal of Human Resource Management, 19(9), 1588–1606.
Penaloza, L. (1994). Border crossing: A critical ethnographic exploration of the consumer
acculturation of Mexican immigrants. Journal of Consumer Research, 21, 32–54.
Podsakoff, P., MacKenzie, S., Lee, J. Y., & Podsakoff, N. (2003). Common method biases
in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903.
Richardson, J., & McKenna, S. (2006). Exploring relationships with home and host countries: A study of self-directed expatriates. Cross Cultural Management, 13(1), 6–22.
Riefler, P., & Diamantopoulos, A. (2009). Consumer cosmopolitanism: Review and replication of the CYMYC scale. Journal of Business Research, 62, 407–419.
Riefler, P., Diamantopoulos, A., & Siguaw, J. (2012). Cosmopolitan consumers as a target
group for segmentation. Journal of International Business Studies, 43, 1–21.
Rindfleisch, A., Malter, A. J., Ganesan, S., & Moorman, C. (2008). Cross-sectional versus
longitudinal survey research: Concepts, findings, and guidelines. Journal of Marketing Research, 45, 261–279.
Samuelson, W., & Zeckhauser, R. (1988). Status quo bias in decision making. Journal of
Risk and Uncertainty, 1, 7–59.
Sassen, S. (1991). The global city: New York, London, Tokyo. Princeton: Princeton University Press.
Sheskin, D. J. (2004). Handbook of parametric and nonparametric statistical procedures.
Boca Raton: Chapman and Hall.
Shu, S., & Gneezy, A. (2010). Procrastination of enjoyable experiences. Journal of
Marketing Research, 47(5), 933–944.
Stahl, G. K., Miller, E. L., & Tung, R. L. (2002). Toward the boundaryless career: A closer
look at the expatriate career concept and the perceived implications of an international assignment. Journal of World Business, 37, 216–227.
Stam, W., & Elfring, T. (2008). Entrepreneurial orientation and new venture performance:
The moderating role of intra- and extraindustry social capital. Academy of Management Journal, 51(1), 97–111.
Steenburgh, T. J., Ainslie, A., & Engebretson, P. H. (2003). Massively categorical variables:
Revealing the information in zip codes. Marketing Science, 22(1), 40–57.
Suutari, V., & Brewster, C. (2000). Making their own way: International expatriates
through self-initiated foreign assignments. Journal of World Business, 35, 417–436.
Takeuchi, R., Shay, J. P., & Li, J. (2008). When does decisions autonomy increase expatriate managers' adjustment? An empirical test. Academy of Management Journal,
51(1), 45–60.
Takeuchi, R., Wang, M., Marinova, S. V., & Yao, X. (2009). Role of domain-specific facets
of perceived organizational support during expatriation and implications for performance. Organization Science, 20(3), 621–634.
Thompson, C. J., & Tambyah, S. K. (1999). Trying to be cosmopolitan. Journal of Consumer Research, 26(3), 214–241.
Tung, R. L. (1998). American expatriates abroad: From neophytes to cosmopolitans.
Journal of World Business, 33, 125–144.
Van den Bulte, C., & Wuyts, S. (2007). Social networks and marketing. Cambridge, MA:
Marketing Science Institute.
Van Evergdingen, Y., & Waarts, E. (2003). The effect of national culture on the adoption
of innovations. Marketing Letters, 14(3), 217–232.
Van Vianen, A. E. M., De Pater, I. E., Kristfof-Brown, A. L., & Johnson, E. C. (2004). Fitting
in: Surface- and deep-level cultural differences and expatriates' adjustment. Academy of Management Journal, 47(5), 697–709.
Wakslak, C., Nussbaum, S., Liberman, N., & Trope, Y. (2008). Representations of the self
in the near and distant future. Journal of Personality and Social Psychology, 95,
757–773.
Wathieu, L. (1997). Habits and the anomalies in intertemporal choice. Management
Science, 43(11), 1552–1563.
Yoon, S. J. (1998). Dimensionality of the cosmopolitanism construct: Cross-method
validation of the Emic and Etic interpretations. Asia Pacific Advancement in Consume Research, 3, 81–88.
Intern. J. of Research in Marketing 29 (2012) 346–354
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
How political identity and charity positioning increase donations: Insights from
Moral Foundations Theory
Karen Page Winterich a,⁎, Yinlong Zhang b, Vikas Mittal c
a
b
c
Smeal College of Business, Pennsylvania State University, 449 Business Building, University Park, PA 16802, United States
University of Texas at San Antonio, One UTSA Circle, San Antonio, TX 78249, United States
J. Hugh Liedtke Professor of Marketing and Management, Jones Graduate School of Business, Rice University, 6100 Main Street, Houston, TX 77005, United States
a r t i c l e
i n f o
Article history:
First received in 30, June 2011 and was under
review for 5½ months
Available online 15 September 2012
Keywords:
Political identity
Liberals
Conservatives
Donation behavior
Moral identity
Moral Foundations Theory
a b s t r a c t
Marketers can strategically target potential donors based on their political identity. Drawing upon the moral
foundations underlying political identity, we examine conditions in which a charity's positioning increases donations based on its alignment with a political identity. In doing so, we demonstrate the moderating role of moral
identity internalization, which is a robust predictor of donation behavior. In three studies, our results reveal that
when the moral foundations of a charity, as evidenced by the charity's management processes (Study 1) or mission (Studies 2 and 3), are aligned with the donor's political identity, donations increase. This effect is enhanced
among those with higher moral identity internalization. These results offer theoretical contributions, as well as
practical implications, for organizations soliciting donations, providing insight into the extent to which charity
positioning based on moral foundations can differentially influence donations based on political identity.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
With Americans 1 donating over $200 billion annually (Giving
USA, 2010), nonprofit organizations must appeal to consumers to donate to their specific charity (Cronin, 2006). Several studies show that
a consumer's identity affects his or her donation behavior (Shang,
Reed, & Croson, 2008; Winterich & Barone, 2011). In particular, a
consumer's moral identity interacts with other identities (e.g., gender; see Winterich, Mittal, & Ross, 2009) and influences his or her giving (Aquino & Reed, 2002) as well as aspects of the donation process itself
(e.g., donating time vs. money; see Reed, Aquino, & Levy, 2007). Clearly,
consumer donations are labile and susceptible to external influences, particularly those related to charity characteristics and consumer identity.
We examine how consumers' political identity – relative conservatism or liberalism – may influence their donations (Brooks, 2006;
Smiley, 2004). More specifically, we examine whether alignment between political identity and a charity's moral foundations or the psychological basis of morality (Haidt & Joseph, 2004) increases donations.
Take a universally important cause: helping children. Would donations
to such a cause be affected by the donor's political identity? If so, do the
moral foundations of this political identity offer a systematic explanation? In addition, does the effect of the alignment between a charity
⁎ Corresponding author. Tel.: +1 814 863 4483.
E-mail addresses: kpw2@psu.edu (K.P. Winterich), yinlong.zhang@utsa.edu
(Y. Zhang), vmittal@rice.edu (V. Mittal).
1
Charitable contributions totaled $303.75 billion in 2009 with $227.41 billion from
individuals, $14.10 billion from corporations, and the remaining $62.24 billion from
bequests and foundations (Giving USA, 2010).
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.05.002
and a political identity on donations persist unconditionally, or is it
moderated by moral identity internalization, which represents the
private, subjective experience of one's moral character and its hidden
cognitions (Aquino, Freeman, Reed, Lim, & Felps, 2009)? That is, is
the enhancing effect of moral identity internalization on donations dependent on the charity's moral alignment with a consumer's political
identity? Answers to these questions are not obvious, though they are
important, given the growing significance of political identity in shaping
the lives of consumers across the world (Abramowitz, 2011; Ford, 2006).
We answer these questions in three studies. Study 1 manipulates the
charity management process (i.e., private or government management)
for a cause that helps to feed hungry children. Studies 2 and 3 manipulate
a charity's moral foundations communicated in its stated mission. These
studies demonstrate that a charity's positioning and political identity interact to influence donation intentions and behavior. Moral identity internalization moderates this effect. Similar to the role of fit between a
brand and a cause (Zdravkovic, Magnusson, & Stanley, 2010), a charity's
alignment with the donor's political identity may garner increased donations. Consequently, the alignment mechanism can be rooted in moral
foundations and may be used by charities to enhance donations.
2. Political identity, moral foundations, and moral
identity internalization
2.1. Political identity
Political identity represents a person's beliefs about the underlying
goals and ideals of a political system (Grove, Remy, & Zeigler, 1974).
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
In Western countries, especially within the United States, there are two
dominant political identities: liberal and conservative (Brooks, 2006;
Reyna, Henry, Korfmacher, & Tucker, 2005; Skitka & Tetlock, 1993).
However, the liberal–conservative continuum for political views is not
limited to the United States (Graham, Haidt, & Nosek, 2009). The present study focuses on donations by American consumers, for whom a liberal identity corresponds to the Democratic party, and the conservative
identity corresponds to the Republican party. Next, we propose that political identities may influence donation behavior, depending on the
moral emphasis of the charity.
2.2. Moral Foundations Theory
Moral Foundations Theory (MFT hereafter) systematically explains
the origins, development, and cultural variations in morality (Haidt &
Joseph, 2004). Drawing upon traditional values research (Rokeach,
1973; Schwartz, 1992), MFT focuses on the psychological basis of morality. Moral foundations are not values or virtues; rather, they are the
psychological systems that guide cognitions: the “first draft” of the
moral mind.
MFT identifies five specific moral foundations — harm/care, fairness/
reciprocity, in-group/loyalty, authority/respect, and purity/sanctity
(Haidt & Graham, 2007; Haidt & Joseph, 2004). These foundations collapse into two super-ordinate foundations labeled the individualizing
and binding foundations. As is explained below, the individualizing
foundation alludes to the protection and fair treatment of individuals.
In contrast, the binding foundation is based on loyalty, tradition, and
purity.
• Harm/care is associated with concern about caring, nurturing, and
protecting vulnerable individuals from harm. Fairness/reciprocity
considers the notions of rights, justice, and what people owe to each
other. These two foundations suggest a general regard for the protection and fair treatment of individuals. As such, they collectively have
been called the “individualizing” foundations.
• In-group/loyalty concerns patriotism and self-sacrifice for one's group.
Authority/respect focuses on the importance of social order and respect for leadership and traditions. Purity/sanctity refers to a more
spiritual mindset that resists the carnal nature of humans. Together,
these three foundations focus not on the protection of individuals
but on group-binding, duty, and the moral systems in which people
are bound in groups and institutions. Therefore, together they have
been called the “binding” foundations.
People's moral foundations can systematically differ by political
identity (Haidt, 2012). Specifically, liberals find issues of harm and
fairness, which represent the basis of individualizing moral foundations,
to be of higher moral relevance to their judgment than conservatives
do. Conversely, conservatives find the binding moral foundations,
which concern issues of authority, group loyalty, and purity, to be of
greater moral relevance than liberals do (Graham et al., 2009; Haidt &
Graham, 2007). Notably, individualizing foundations tend to be of
greater relevance in general. However, this effect is moderated by political identity: the difference between the two foundations is greater for
liberals. Conservatives tend to value these two foundations relatively
equally (Graham et al., 2009). These systematic differences manifest
such that conservatives find binding foundations to be of greater relevance than liberals do, and liberals find individualizing foundations to
be of more relevance than conservatives do.
The binding foundations have been examined previously, though
mostly in the context of immorality. These foundations can result in
unacceptable behavior, such as racism, blind obedience, and stigma
(Graham et al., 2009). Importantly, morality research recognizes that
moral–psychological mechanisms are often double-edged swords (de
Waal, 1996). For example, conservatives tend to have more authoritarian personalities, and studies have found that authoritarianism is
347
associated with religious fundamentalism and prejudice (Altemeyer,
1996). Such findings have led to a perception of “conservatives without
conscience” (Dean, 2006). However, conservatives, who tend to value
individualizing and binding foundations relatively equally, may still engage in moral behavior, such as protecting others and supporting fairness. However, these behaviors may be limited to their group or those
abiding by tradition based on the simultaneous importance of binding
moral foundations.
How do moral foundations offer insight into the effects of political
identity on helping others? Skitka and Tetlock (1993) found that conservatives denied help to those not agreeing to undergo vocational
training. Such denial of aid is consistent with morality based on respect
for authority, a binding foundation held more strongly by conservatives
such that individuals should have been obedient to authority and reformed (Graham et al., 2009). Fowler and Kam (2007) examined the allocation of money in dictator games. When a political identity was
known, Republicans were more likely to give to another Republican
(vs. Democrat). In contrast, Democrats were not more likely to give to
another Democrat (vs. Republican). This study suggests that Republicans have a stronger in-group bias than Democrats, which is consistent
with the in-group loyalty that characterizes conservatives (Graham et
al., 2009). Reyna et al. (2005) found that conservatives were more
opposed to affirmative action for black people than for other groups
(i.e., women). This effect was mediated by group-based stereotypes
based on deservingness and was not mediated by racism or threat to
oneself. This role of group-based stereotypes is also consistent with
the in-group moral foundation held more strongly by conservatives
than by liberals. Notably, this finding also corresponds to past MFT research showing that liberals place greater emphasis on individualizing
moral foundations than on binding foundations (Graham et al., 2009).
This differential emphasis results in favoring affirmative action, regardless of group, because of fairness. In contrast, conservatives may value
both individualizing and binding foundations (Graham et al., 2009).
Nevertheless, their evaluations of fairness may be limited based on
their equal prioritization of binding moral foundations (i.e., only the
in-group; Reyna et al., 2005). Therefore, moral foundations may explain
these findings.
We argue that the extent to which the moral foundations of the
charity are appropriately aligned with those of the consumer's political
identity may predict the level of a donation that a consumer makes to
the charity. Specifically, when a charity is characterized by binding
moral foundations, conservatives may donate more to the charity than
liberals based on the importance of binding foundations to conservatives. Conversely, when a charity is characterized by individualizing
moral foundations, liberals may donate more than conservatives to
the charity, due to the importance of individualizing foundations to liberals. We theorize that individualizing foundations will be more aligned
with liberals than with conservatives for two reasons. First, individualizing foundations are of greater relevance to liberals than conservatives.
Second, individualizing foundations of care and fairness are more inherently present in charities. Therefore, conservatives who value binding
and individualizing foundations equally may need binding foundations
to be of reasonable importance to a charity in addition to the importance of individualizing foundations inherent to charities to donate to it.
2.3. Moderating role of moral identity internalization
Moral identity has been defined as the mental representation of one's
moral character that is held internally (internalization) and projected to
others through moral actions in the world (symbolization; Aquino &
Reed, 2002; Blasi, 1984). Internalization has also been referred to as
moral identity centrality (Aquino, McFerran, & Laven, 2011, Aquino et
al., 2009) because it represents the private experience of moral identity
and is central to one's self-concept. Consistent with this centrality, the internalization dimension is a more consistent predictor of various types of
moral behaviors and behavioral intentions than symbolization (Aquino
348
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
et al., 2009, 2011; Reed & Aquino, 2003; Reed et al., 2007; Reynolds &
Ceranic, 2007). We focus on the internalization dimension, referring to
it as moral identity internalization, as it is more central to the self and
thus more influential on moral behaviors, such as donation behavior.
We theorize that moral identity internalization moderates the effect
of the alignment between political identity and charity moral foundation
on donation behavior such that the effect of alignment on donations is
enhanced when internalization is high. Reed et al. (2007) argue that
“when a consumer's moral identity is highly salient or self-important,
his or her sensitivity to acts that are consistent with moral identity increases” (p. 180). This increased sensitivity to moral acts occurs because
consumers process and prefer to act in ways that are consistent with and
enhance their social identity (Blasi, 1984; Forehand, Deshpandé, & Reed,
2002; Reed, 2004). Along the lines of this argument, consumers may desire to act in ways that help to align the moral foundations underlying
these behaviors with their political identity, and this desire may be moderated by their moral identity internalization. How does this occur?
We propose a two-step process. First, the act must be deemed to be
moral based on the alignment of the act with the moral foundations embedded in the other relevant social identity (i.e., political identity). If the
act is deemed to be moral based on a person's political identity, moral
identity internalization can motivate moral behavior based on the importance to one's self-concept. In other words, moral identity internalization can catalyze charitable acts once they are perceived as moral
based on the alignment between one's political identity and the charity's
moral foundations. Specifically, conservatives are more likely than liberals
to deem causes that are more representative of binding moral foundations to be more moral; in contrast, liberals are more likely than conservatives to deem causes that are more representative of individualizing
foundations to be more moral. However, higher moral identity internalization will increase the motivation to engage in moral action (i.e., donations) when a cause is aligned with one's political identity.
To summarize, our theorizing is as follows. First, donations will increase with the alignment between political identity and charity
moral foundation. Specifically, conservatives will donate more than
liberals to a cause more strongly associated with binding foundations,
and liberals will donate more than conservatives to a cause with individualizing foundations. Second, moral identity internalization moderates the effect of alignment: the alignment effect is enhanced for
those with higher moral identity internalization, thereby increasing
donations. In contrast, donations will not be influenced by political
identity and cause alignment among those with lower moral identity
internalization because their motivation to engage in moral behavior
is weaker. That is, alignment does not motivate donations for those
lacking the perception of the importance of a moral identity.
3. Study 1: alignment of political identity with the charity
management process
Several charitable programs, such as WIC, Planned Parenthood, and
even public parks, receive a large portion of funding through the
government and are conducted as public agencies. Others, such as rehabilitation centers, homeless shelters, and soup kitchens, receive little if
any government support and are primarily funded and managed by
community and religious organizations. These differences may be
particularly important, as consumers' political identities correspond to
the role of government in helping the public. For example, liberals
tend to support government programs more than conservatives, whereas conservatives tend to favor private enterprise more than liberals
(Anderson, Mellor, & Milyo, 2005; Brooks, 2006). This relationship between private enterprise and conservatism may be associated with the
underlying moral foundations of in-group and authority, whereas the
relationship between government programs and liberalism may be associated with moral foundations of harm and fairness. Indeed, liberals
responded more favorably than conservatives to moral judgments regarding harm and fairness with specific items including the following:
“The government must first and foremost protect all people from
harm,” and “When the government makes laws, the number one principle should be ensuring that everyone is treated fairly” (Graham et al.,
2009). Given these beliefs, we expect that a privately managed charity
should be more closely aligned with the moral foundations of conservatives than with those of liberals. In contrast, government-managed charities, which may focus on preventing harm and ensuring fairness, should
be more closely aligned with the moral foundations of liberals than with
those of conservatives. We also propose that moral identity internalization will enhance the effect of political identity and charity alignment on
donations. That is, moral identity internalization will moderate the joint
effect of the charity management process and political identity, resulting
in a three-way interaction.
3.1. Pretest
Participants (n= 56 undergraduates) indicated their agreement
with the following two items: “Most of the time, I feel that liberals are
supportive of government programs for social welfare” (M= 5.26/7,
t(55)= 6.43, p b .0001) and “Conservatives tend to support the endeavors of private organizations for social welfare over those of government organizations” (M= 4.95/7, t(55)= 4.78, p b .0001). The t-tests
indicate that responses were significantly greater than the scale midpoint of 4. Therefore, students in our sample perceive that liberals
support government-run programs and that conservatives support privately run programs. Importantly, responses to these two items were not
correlated with respondents' own political identity (r=.17, p=.22;
r=−.15, p=.27, respectively), indicating that this perception holds regardless of political identity.
We theorized that these differences in the preference for government
versus privately managed organizations between political identities may
be associated with differences in moral foundations. We conducted a separate test (n=41) to test this theory. Participants read one of two charity
descriptions for Save the Children such that both descriptions were for the
same cause but was either managed by private agencies (coded as ‘1’) or
government agencies (coded as ‘-1’), respectively, as described below:
Save the Children creates lasting change for children in need in the
United States and around the world through private [government]
intervention. Save the Children aids children by providing food,
clothing, and medical aid to children in poverty through privately
funded support [government-funded support]. The funds that
Save the Children raises are managed and administered to children
in need through private organizations especially local citizens and
religious organizations [public agencies, especially the U.S. government]. These private [public] agencies will use the funds to help children who are in need of food, clothing, and medical care.
After reading the description, participants indicated the extent to
which they felt that Save the Children was concerned with each of the
following five statements: caring for the vulnerable, protecting the
rights of every individual, protecting group interests, maintaining social
order, and maintaining standards of decency. These items were adapted
from the Moral Foundations Questionnaire developed by Graham et al.
(2011). Responses were on a 9-point Likert scale (1= not at all to 9 =
very much so). The first two statements reflect individualizing moral
foundations (r = .44, p = .01), whereas the last three reflect binding
moral foundations (α = .64).
Participants in the privately managed condition perceived the
charity to have stronger binding moral foundations (M= 6.39) than
those in the government-managed condition (M= 4.85, t(39)= 3.23,
p = .003), as expected. Participants in the government-managed condition perceived the charity as having individualizing moral foundations (M= 6.78) equal to those in the privately managed condition
(M= 6.86, t(39)= −.16, p = .87). Individualizing moral foundations
were not perceived as differing between government and private
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
management. However, this response may have been observed because
the basic needs of food, clothing, and medical care provided by Save the
Children reflect the moral foundations of harm and fairness. Importantly, a difference was found in binding moral foundations between private
management and government management, which is consistent with
the tendency for conservatives to find binding foundations to be more
relevant than liberals (Graham et al., 2009, 2011). Perceptions of charity
moral foundations did not differ by political identity (effect of political
identity on binding foundations: p = .82, individualizing foundations:
p = .86), and the difference in binding moral foundations remained significant when political identity was included as a covariate (p = .004).
Therefore, these results suggest that government versus private management of a charitable organization may differentially align with binding moral foundations.
3.2. Participants and design
Undergraduates (N= 161; 58% female) completed the study for
course credit. Participants were assigned to either a private-managed
charity or a government-managed one with both moral identity and political identity measured as continuous variables. Moral identity was
measured at the beginning of a one-hour research session with other
personality scales. After completing unrelated survey tasks for approximately 20–30 min, participants completed the donation intention
task followed by demographic information, political identity, and the
conservatism scale.
349
3.3.3. Charity management process
Participants read one of the two pretested charity descriptions,
which indicated private (coded as ‘1’) or government (coded as ‘-1’)
management (Jaccard & Turrisi, 2003).
3.3.4. Donation intentions (allocation)
Participants were told the following: “Suppose you have $100 at
your disposal. Please indicate the amount you are willing to donate
to Save the Children” (M = $26.83, SD = $22.00).
3.4. Results
3.4.1. Manipulation check
To verify that participants perceived the charity management process
as intended, they were asked to respond to six statements (1 =strongly
disagree, 7=strongly agree) after completing the dependent measures.
Three items indicated management by private agencies and include
“Private organizations support and fund the Save the Children program.”
Three items indicated management by government agencies and include
“The U.S. government and other public organizations fund the Save the
Children program.” The three private-agency items were reversecoded, and the six items were averaged to create an index (α=.91). Participants in the privately managed condition perceived that the charity
was run by government agencies less than those in the governmentmanaged condition (Mprivate =2.76 vs. Mgovt =4.59, t(159) =10.18,
pb .0001). Therefore, the manipulation of the charity management process was successful.
3.3. Measures
3.3.1. Political identity
We measured political identity with the six-item scale constructed
by Mehrabian (1996). Sample items are as follows: “I am politically
more liberal than conservative,” and “On balance, I lean politically
more to the left than to the right” (1= strongly disagree, 9 =strongly
agree). These six items were averaged to create one conservatism
score (α= .84) which had a mean score of 6.21/9 (SD= 1.72). On this
scale, higher scores represent stronger conservatism. The results
reported below are for the measured conservatism political identity,
but the results were consistent when using self-reported Democratic
(33%) and Republican (67%) party affiliation and are available upon request. As expected, conservatism is positively and strongly correlated
with political identity (r =.73, p b .0001; 0 =Democrat, 1 =Republican).
3.3.2. Moral identity internalization
Participants responded to Aquino and Reed's (2002) Self-Importance
of Moral Identity Scale (1=strongly disagree, 7=strongly agree). The
ten items represent the two five-item subscales measuring internalization and symbolization. Consistent with our theorizing and past research
(Aquino et al., 2011; Reed & Aquino, 2003; Reed et al., 2007), we focus on
internalization. Internalization items include the following: “It would
make me feel good to be a person who has these characteristics,”
“I strongly desire to have these characteristics,” and “Having these characteristics is not really important to me” (reverse-coded). The five internalization items were indexed (α=.75, M=6.49, SD=.51), and this index
was not correlated with political identity (r=.01, p=.82). We also averaged the five symbolization items (α=.79, M=4.88, SD=.96).2
2
We assessed the role of moral identity symbolization by replacing the internalization dimension with the symbolization dimension and conducted the same regression
analysis for donation intentions with all main effects and two-way interactions in addition to the three-way interaction. However, no effects were significant, including the
three-way interaction (p = .75). Moral identity symbolization was not found to have a
significant main effect or interaction effect in subsequent studies. This result is consistent with past research, which did not find consistent significant effects of moral identity
symbolization or find symbolization to alter the effect of internalization (e.g., Aquino et al.,
2011; Reed et al., 2007). Therefore, we do not discuss the symbolization dimension
further.
3.4.2. Donation intentions (Allocation)
We conducted a regression analysis with donation intentions as the
dependent variable and the charity management process (government
vs. private), moral identity internalization, and political identity (both
continuous variables) as the independent variables. The results are
reported in Table 1. First, as expected, the two-way interaction of the
charity management process and political identity was significant
(β = $4.37, t(153) = 2.35, p = .02). However, to understand the moderating role of moral identity internalization, we examined the three-way
interaction of moral identity internalization, the charity management
process, and political identity. This three-way interaction was significant (β = $3.35, t(153) = 1.98, p = .05). To interpret the pattern of this
three-way interaction, we examine the effect of political identity for
each of the charity management processes by low and high levels of
moral identity internalization (i.e., one standard deviation above and
below the mean of moral identity; Aiken & West, 1991). Among those
with high moral identity internalization (+1SD), political identity negatively influenced donation intentions when the charity was managed
by the government (β = −$8.51, t(153)= −2.14, p = .03) such that liberals had higher donation intentions than conservatives. In contrast,
there was a significant positive effect of political identity for donations
to the privately managed charity (β = $6.93, t(153) = 2.03, p =.04)
such that conservatives had higher donation intentions than liberals.
When moral identity internalization was low (−1SD), there was no effect of political identity on the intentions to donate to the governmentmanaged charity (β = −$2.61, t(153) = −.81, p =.42) or the privately
managed charity (β = −$.55, t(153) = −.15, p = .88). See Fig. 1. These
results support our hypothesis.
3.4.3. Religion as a potential alternative explanation
Private agencies, such as religious organizations, were included in
the private management condition. Because religion may be correlated with political identity (Altemeyer, 1996), it may be an alternative
explanation. In our sample, political identity and religion were correlated such that Republicans were more likely to be of Christian faith
(r = .43, p b .0001). When religion was included as a covariate in the
main analysis reported earlier, it did not predict donations (p = .42).
When religion replaced political identity in the regression analysis
350
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
Table 1
Regression coefficients (Studies 1–3).
Predictors
Intercept
Political identity (PI)
Moral identity (MI)
Charity (C)
PI × MI
PI × C
MI × C
PI × MI × C
df
F
R2
Study 1: donation intentions
(allocation)
Study 2: donation intentions
Study 3: donation allocation
behavior
b
t
b
t
b
t
28.47
−1.18
5.81
1.64
.40
4.37
.96
3.35
(7, 162)
2.77⁎⁎
.11
15.90⁎⁎
3.64
−.01
.63
.12
−.00
.40
.20
.51
(7, 92)
3.00⁎⁎
.19
17.76⁎⁎
2.54
.19
.32
−.02
.09
.31
−.08
.42
(7, 190)
2.42⁎⁎
.09
17.62⁎⁎
1.25
2.28⁎⁎
−.10
.54
2.00⁎⁎
−.64
3.16⁎⁎
.92
.23
2.35⁎⁎
.52
1.98⁎⁎
−.04
2.74⁎⁎
.60
−.00
1.90⁎
.86
2.03⁎⁎
−.58
2.62⁎⁎
Note: ⁎⁎p b .05, ⁎p b .10.
predicting donation intentions, no effects were significant, including
the three-way interaction (p = .40). Therefore, religion does not appear to be a viable alternative explanation.
the charity management process. Therefore, we manipulated moral
foundations by the description of the charity's mission, rather than
through the management process, in the next study.
3.5. Discussion
4. Study 2: alignment of political identity with charity's
moral foundations
Donation Intentions (Allocate $100)
50
Donation Intentions (Allocate $100)
A subtle manipulation of the charity management process affected
the donation intentions of conservatives and liberals. Specifically,
when the charity was managed by the government, liberals with high
moral identity internalization had higher donation intentions than
conservatives. In contrast, when the charity was managed privately,
conservatives with high moral identity internalization had higher donation intentions than liberals. When moral identity internalization was
low, donation intentions did not differ, regardless of the alignment of
the charity management process with political identity because these
individuals did not find donation behavior to be important to their
self-concept, specifically, their moral identity. It is likely that perceptions of a charity's moral foundations may differ in ways other than
50
High Moral Identity Internalization
45
40
35
30
25
20
15
10
5
0
4.1. Pretest
Participants (N = 61) read one of two charity descriptions for
Children's Advocates. Although both descriptions emphasized harm
and fairness foundations through a focus on abused, neglected, and
at-risk children, the binding description placed an additional emphasis on purity through spiritual needs and in-group loyalty through
community and leadership. Given these charity distinctions, we anticipate moral foundations to differ in terms of in-group and purity
binding foundations rather than in terms of authority. All participants
read the following passage: “Founded in 1981, Children's Advocates
works to break the cycle of child abuse through prevention, education,
advocacy and funding. Children's Advocates strives to meet the…”
with one of the following passages, depending on condition.
Binding moral foundations
Individualizing moral foundations
Liberal
Conservative
Low Moral Identity Internalization
45
40
Binding moral foundations
35
Individualizing moral foundations
30
[Binding Moral Foundations: coded as ‘1’] “…educational and spiritual needs of abused, neglected, and at-risk children. Children's
Advocates fights for every child's right to be protected from abuse
and neglect and to grow up in a safe, stable, permanent home.
Children's Advocates champions efforts to help young people make
informed and responsible decisions in a safe, community environment through programs on: character and leadership development;
education and career development; and health and life skills.”
[Individualizing Moral Foundations: coded as ‘-1’] “…physical and
emotional needs of abused, neglected, and at-risk children.
Children's Advocates fights to enshrine in the law of the land every child's right to be protected from abuse and neglect. Children's
Advocates' programs and services include: National Child Abuse
Hotline, Good-Touch/Bad Touch, and Children's Advocacy Centers.”
25
20
15
10
5
0
The charity's mission was aligned with either binding (purity, ingroup, and authority) or individualizing (harm and fairness) moral foundations. This study focuses on differences in the binding moral foundations of in-group loyalty and purity. Next, in Study 3, we focus on the
binding moral foundation of authority to increase generalizability.
Liberal
Conservative
Fig. 1. Moderating role of moral identity internalization on the effect of political identity on donation intentions by the charity management process (Study 1).
After reading the description, participants indicated the extent to
which they felt that Children's Advocates is concerned with each of the
following five statements: preventing harm, protecting and ensuring
351
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
A total of 100 adults (Mage =45.40 years, SD=11.94, range=21–71)
from an online panel across the United States participated in the online
study in exchange for entry in a lottery for $50 Amazon.com gift certificates. The study is a one-factor (Charity Moral Foundation: Binding vs. Individualizing) between-subjects design, with both moral identity and
political identity being measured as continuous variables. The online survey was presented as two unrelated short surveys, with the first survey
automatically redirecting to the second survey to be consistent with
the cover story of multiple short surveys. In the first survey, participants
completed the moral identity measure used in Study 1. We used five internalization items with α=.84, M=6.19/7, SD=0.93 along with the
political identity, which was couched within demographic measures.
Political identity was self-reported on a 7-point scale with 1=strongly
liberal to 7=strongly conservative. Responses ranged from 1 to 7 with
an average of 4.01, SD=1.69, indicating that both strong liberals and
strong conservatives were represented in our sample. Political identity
was not significantly correlated with moral identity internalization,
r=−.04, p=.69.
Next, participants were redirected to the second, presumably
unrelated, survey, which contained approximately 15 min of unrelated
questions. Immediately before finishing the second survey, participants
read one of the two descriptions of Children's Advocates presented
in the pretest and responded to a donation request. We measured donation intentions with two items adapted from Reed and Aquino (2003):
“I would like to donate money to this cause” and “I would like to support
this cause.” We selected these two items because they explicitly focus
on consumers' donation intentions. The responses (1= strongly disagree to 7 = strongly agree) to the two items were averaged to create
an index (r = .76, p b .0001, M = 3.82, SD= 2.11).
The two-way interaction of charity moral foundations and political
identity was marginally significant (β =.40, t(92)= 1.90, p = .06), as
expected. However, this two-way interaction was qualified by the
three-way interaction of moral identity internalization, charity moral
foundation, and political identity, which was significant (β = .51,
t(92)= 2.03, p = .04; see Table 1 for regression results). As in Study 1,
we examined the effect of political identity on donation intentions for
each moral foundation at high and low levels of moral identity internalization (+/−1 SD) to understand the pattern of the interaction. Among
those with high moral identity internalization, political identity negatively influenced donation intentions for the charity with an individualizing moral foundation (β = −.92, t(92)= −2.08, p = .04): liberals had
higher donation intentions than conservatives. In contrast, there was a
significant positive effect of political identity for the charity with a binding moral foundation (β =.90, t(92) =2.18, p = .05): conservatives
demonstrated higher donation intentions than did liberals. When
moral identity internalization was low, there was no effect of political
identity on donation intentions to the charity with an individualizing
moral foundation (β = .10, t(92) =.17, p = .86) or the charity with a
binding moral foundations (β= −.12, t(92) = −.29, p = .77); see
Fig. 2. These results are consistent with those of Study 1 in which
moral foundations were manipulated through the charity management
process, rather than through the charity's mission.
Donation intentions increase when the moral foundations of the
charity are aligned with consumers' political identity. Moral identity
internalization enhances this effect. We next examine charity positioning based on differences in the binding moral foundation of authority to provide more generalizable evidence. More importantly,
Study 3 examines donation behavior, as opposed to intentions, to improve the external validity of these results.
7
High Moral Identity Internalization
6
Donation Intentions
4.2. Participants and design
4.3. Results and discussion
5
4
3
2
Binding moral foundations
Individualizing moral foundations
1
7
Liberal
Conservative
Low Moral Identity Internalization
6
Donation Intentions
rights, ensuring respect, protecting group interests, and preventing impurities. The responses were made on a 7-point Likert scale (1 =strongly
disagree to 7= strongly agree). The first two statements concern individualizing moral foundations and were averaged (r =.79, p b .0001),
whereas the last two concern the focal in-group and purity foundations
in the binding charity description (r=.74, p b .0001).
Participants in the binding condition had stronger perceptions
that the charity had binding moral foundations compared with
those in the individualizing condition (M = 4.86 vs. 3.27, t(59) =
3.71, p = .0005), as expected. In contrast, participants in the individualizing condition had stronger perceptions that the charity had individualizing moral foundations compared with those in the binding
condition (M = 5.75 vs. 4.77, t(59) = 2.03, p = .05). We acknowledge
that in the binding condition, individualizing, and binding moral
foundations do not differ (M = 4.77 and 4.86, t(30) = 1.64, p = .11)
as they do in the individualizing condition (M = 5.75 and 3.27,
t(29) = − 8.71, p b .0001). However, this finding is representative of
the inherent emphasis on individualizing moral foundations of harm
and fairness for any non-profit organization seeking to protect abused
children. Moreover, there is a tendency for political distinctions in individualizing moral foundations, specifically harm, to be weakest,
whereas binding foundations, such as purity, tend to be strongest
(Graham et al., 2011). Therefore, consistent with these differences
in moral foundations by political identity, the critical distinction is
the relative emphasis on binding moral foundations between the
two charity descriptions. Given this finding, the descriptions were
perceived as intended. 3
5
4
Binding moral foundations
Individualizing moral foundations
3
2
3
Perceptions of charity moral foundations did not differ by political identity (effect
of political identity for binding: p = .10; individualizing: p = .57), and the perceived
differences in moral foundations remained significant when political identity was included as a covariate (binding: p = .05, individualizing: p = .03).
1
Liberal
Conservative
Fig. 2. Moderating role of moral identity internalization on the effect of political identity on donation intentions by charity moral foundations positioning (Study 2).
352
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
5. Study 3: donation allocation behavior by charity
moral foundations
This study examines donation allocation behavior, rather than donation intention, in the context of an actual charity supporting home
ownership, namely, Rebuilding Together. The descriptions of the
charity differed primarily in terms of the authority moral foundation.
This difference from prior studies should ensure that effects are not
dependent on the specific cause (i.e., supporting children in Studies
1 and 2) or moral foundation. To manipulate the moral foundations,
we emphasized different aspects of the charity according to the information presented on the charity's website.
5.1. Pretest
Participants (N= 47) read one of two descriptions of Rebuilding
Together. One description emphasized the authority dimension
through a focus on individual development and traditions and the
in-group dimension through family and community support. All participants read the following passage: “Rebuilding Together (http://www.
rebuildingtogether.org/) is the nation's leading organization working
to preserve affordable home ownership and revitalize communities. Rebuilding Together believes…” and were randomly assigned to continue
this passage with one of the following statements:
[Binding Moral Foundations: coded as ‘1’] “…in the importance of
a healthy home where families can live together. They provide rehabilitation services to working families who are trying to develop
themselves to follow American traditions and support their
communities.”
[Individualizing Moral Foundations: coded as ‘-1’] “…every person
deserves the protection of a home. They provide free critical repairs to the homes of low-income Americans as well as financial
support to protect low income families who are susceptible to
home loss. This support works to ensure every individual has the
right to a home.”
Participants indicated the extent to which they felt that Rebuilding
Together was concerned with the following factors: caring for the
vulnerable, protecting the rights of every individual, protecting group
interests, supporting tradition, and preventing impurity (1= not at
all to 9 = very much so). The first two statements for individualizing
moral foundations were averaged (r = .43, p = .003), and the third
and fourth statements for binding foundations were averaged (r = .33,
p = .03) because neither description focused on the purity dimension.
Participants in the individualizing condition perceived the charity as
having a stronger individualizing foundation compared with those in
the binding condition (M= 6.98 vs. 5.85, t(45)= 2.19, p = .03). In contrast, those in the binding condition perceived the charity as having a
stronger binding foundation compared with those in the individualizing
condition (M=6.68 vs. 5.09, t(45) = 2.95, p = .005), as expected. 4
5.2. Participants and design
Online panel participants (N= 198, Mage = 42.08 years, SD= 13.18,
range= 18–78) completed the study in exchange for entry in a lottery
for a $50 Amazon.com gift certificate. The study is a one-factor (Charity
Moral Foundation: Binding vs. Individualizing) between-subjects design
with both moral identity and political identity measured as continuous
variables. First, participants read one of the two pretested descriptions
4
Perceptions of charity moral foundations did not differ by political identity (effect
of political identity for binding: p = .35, individualizing: p = .78), and the perceived differences in moral foundations remained significant when political identity was included as a covariate (binding: p = .006, individualizing: p = .03).
of Rebuilding Together. The website for Rebuilding Together was included in the description to indicate that the charity was not fictitious. Next,
the subjects were told they could donate all, some, or none of the $50 gift
card lottery winnings to Rebuilding Together. The donation allocations
were as follows: (a) $50 donation to Rebuilding Together/no gift card,
(b) $40 donation/$10 gift card, (c), $30 donation/$20 gift card, (d) $20
donation/$30 gift card, (e), $10 donation/$40 gift card, and (f) No donation and $50 gift card. Following Freeman, Aquino, and McFerran
(2009), donation allocation was treated as a continuous variable ranging
from 1 to 6, with higher numbers indicating larger donations.
Finally, participants completed a series of personality questionnaires
that included the moral identity scale (five internalization items: α =
.83, M = 6.21/7, SD= 0.83) and the conservatism scale used in Study
1. Political identity, as measured in Study 2, was included with demographic questions, including gender and age (M= 3.45/7, SD = 1.68).
Higher scores indicate a more conservative political identity, which was
not correlated with moral identity internalization (r=−.11, p=.16).
5.3. Results for donation allocation behavior
Following Freeman et al. (2009), we ran a regression analysis (see
Table 1).5 The two-way interaction of charity moral foundation and
political identity was statistically significant (β = .31, t(190) = 2.00,
p = .04). Consistent with Studies 1 and 2, this two-way interaction
was qualified by the three-way interaction of moral identity internalization, charity moral foundation, and political identity (β = .42, t(190) =
2.62, p = .01).
We examined the pattern of this three-way interaction as in Studies 1 and 2. Among those with higher moral identity internalization,
liberals had higher donation allocations than conservatives for the individualizing moral foundations charity (β = − .43, t(190) = − 1.95,
p = .05). In contrast, conservatives had higher donation allocations
than liberals for the binding moral foundations charity (β = .99,
t(190) = 2.99, p = .003). When moral identity internalization was
low, there was no effect of political identity on donation allocations
to the individualizing moral foundation charity (β = .22, t(190) = .58,
p = .56) or the binding moral foundations charity (β = −.01, t(190) =
−.04, p = .97). See Fig. 3. 6 We next discuss the contributions, implications, and limitations of this research.
6. Discussion
We theorize that consumers who self-identify as being on either end
of the political spectrum may be more likely to donate, as long as the
charity is aligned with the moral foundations of these consumers' political identity. This effect of alignment on donations is enhanced by consumers' moral identity internalization. We used different samples,
measures, and charities, demonstrating the robustness of our results.
Two studies (Studies 2 and 3) used a panel of adults. Study 3 examined
actual donation behavior. Political identity is assessed with a conservatism measure (Study 1) and a liberal–conservative continuum (Studies
2 and 3) to ensure that the results are not specific to a particular measure of political identity. The robustness of moral foundations as a positioning strategy is demonstrated using an in-group and purity focus
(Study 2) and an authority and in-group focus (Study 3).
6.1. Contributions to marketing practice and theory
Charities can enhance their donations by positioning their management process or mission to align with the moral foundations underlying
5
The substantive results and conclusions were unchanged when analyzed as an ordered logit.
6
We replicated the results with the conservatism measure used in Study 1: the
three-way interaction was significant, and the pattern remained unchanged.
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
Donation Allocation Behavior
6
5
Binding moral foundations
Individualizing moral foundations
4
3
2
1
6
Donation Allocation Behavior
High Moral Identity Internalization
5
Liberal
Conservative
Low Moral Identity Internalization
Binding moral foundations
Individualizing moral foundations
4
3
2
1
Liberal
Conservative
Fig. 3. Moderating role of moral identity on the effect of political identity on donation
allocation behavior by charity moral foundations positioning (Study 3).
political (and other) donor identities. This implication may be particularly relevant when for-profit organizations partner with non-profit
organizations to increase cause-related marketing (Varadarajan` &
Menon, 1988; Zdravkovic et al., 2010). For instance, Target has experienced customer boycotts because of its donations to a political candidate who opposed same-sex marriage (Hernandez, 2010). Target's
donation to support a political candidate who opposed same-sex marriage likely conflicted with the individualizing moral foundation of fairness. Future research should examine how large organizations, such as
Target, can simultaneously cater to many segments with diverse political identities.
Theoretically, this research contributes to the political identity literature, particularly in the context of donation behavior. Scholars and
authors in the popular press engage in a spirited debate on whether
liberals or conservatives are more generous. Although there are proponents of each ideology (Brooks, 2006; Dean, 2006; MacDonald,
2004; Skitka & Tetlock, 1993), support may be confounded with the
nuances of their claims. For example, when measuring the average
dollar amount donated (Brooks, 2006 7) or the percent of income
given to charity (MacDonald, 2004), households headed by a conservative individual donate more than those headed by a liberal person.
However, when measuring the respondent's support for government
spending on social programs, those with a relatively liberal political
identity are more generous (Anderson et al., 2005; Brooks, 2006). In
7
These results, according to the author, were based on the Social Capital Community
Benchmark Survey in 2000 and were found when controlling for income and other
covariates.
353
the only experimental study in which liberals and conservatives
were randomly assigned to inequality treatments based on unequal
show-up fees, there were no significant differences in giving in the
public goods game based on respondents' political identity (Anderson
et al., 2005). In all three of our studies, we found no main effect of political identity (all p's > .20) on donations. 8 Therefore, our results show
that political identity alone does not make a person more or less generous. Rather, Republicans and Democrats may donate more or less to a
specific charity based on the degree to which the charity's moral foundations are aligned with their own political identity. This issue may be
relevant to examining brand equity, which may be differentially affected by a consumer's identity and moral foundations.
To date, MFT has received limited consideration in the marketing
literature, although binding or individualizing moral foundations
may be influential in other consumer contexts. For example, binding
foundations may be influential in contexts where purity is of great
importance (e.g., cleaning products, as well as food products and services) or where authority is relevant (e.g., banking, insurance, and investment services). In contrast, individualizing foundations may be
influential in contexts focused on fairness (e.g., education) or on protection from harm (e.g., child or elderly care). We believe not only
that the distinctions between binding and individualizing moral foundations are a key conceptual insight of our paper but also that the
model provides actionable guidance to charities. Because these two
sets of moral foundations are distinct constructs (and not merely
the obverse of one another), they should be examined in additional
consumer and marketing contexts.
Moral identity has been examined as an antecedent of donation
behavior (Aquino et al., 2011; Shao, Aquino, & Freeman, 2008 for review), but research on the moderating effect of moral identity on donation behavior is nascent (see Aquino et al., 2011; Winterich et al.,
2009 for exceptions). We contribute to this stream, hoping that future
research will develop a broader model where multiple identities interact with one another to affect key consumer behaviors.
6.2. Limitations and future research considerations
We only measured political identity and moral identity, thereby
limiting causal interpretations of the empirical evidence. Although results based on temporarily activated moral identity and measured
moral identity internalization are highly consistent (Aquino et al.,
2011; Reed et al., 2007; Winterich et al., 2009), research should replicate these results by priming and/or activating political identity.
Such research would also reveal the interplay between chronic and
activated identities. Will the salience of moral identity simply enhance donations among those without a chronically high moral identity when charity moral foundations are aligned with their political
identity? Similarly, it is unclear whether moral identity enhances
the importance of moral foundations that are aligned with political
identity. Alternatively, are moral foundations equally important, regardless of moral identity? If so, moral identity may simply act as a
motivator for moral action, which includes donating to a charity
when moral foundations are aligned with one's political identity. Additionally, symbolization can be examined in future research, as it
may be an influencing factor when donations are recognized or are
publicly visible.
Finally, our focus on samples based in the United States should not
mask the fact that consumers in other countries donate billions of dollars annually as well and may also differ somewhat in terms of their political identity. For instance, in the United Kingdom, over £52 billion
was donated to the 162,400 charities in 2009–2010 (UK charitable
8
The average donations by political identity are as follows: for Study 1, donation allocation intentions (liberal=28.48 vs. conservative=28.40, p=.98); for Study 2, donation
intentions (liberal=3.99 vs. conservative=3.72, p=.53); for Study 3, donation allocation
behavior (liberal=2.53 vs. conservative=2.62, p=.72).
354
K.P. Winterich et al. / Intern. J. of Research in Marketing 29 (2012) 346–354
sector, 2011). The association between political identity assessed by the
liberal-conservative continuum and moral foundations held even when
participants included those from Europe, Canada, Latin America, and
other areas (Graham et al., 2009, 2011). Nonetheless, we believe these
results should be replicated with samples from other countries, particularly those in Europe and Asia.
6.3. Conclusion
We find that aligning the charity's moral foundations with those
underlying the donor's political identity increases donations. This
alignment is enhanced among those with high moral identity internalization. These results should aid marketers in positioning their
charities to align with the target donor's political identity and thus increase donations to their charity.
References
Abramowitz, A. (2011). The disappearing center: Engaged citizens, polarization, and American
democracy. Yale University Press.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions.
Thousand Oaks, CA: Sage.
Altemeyer, B. (1996). The authoritarian specter. Harvard University Press.
Anderson, L. R., Mellor, J. M., & Milyo, J. (2005). Do liberals play nice? The effects of political party and ideology in public goods and trust games. In J. Morgan (Ed.), Advances
in Applied Microeconomics: Experimental and Behavioral Economics (pp. 107–132). San
Diego, CA: Elsevier.
Aquino, K., Freeman, D., Reed, A., II, Lim, V. K. G., & Felps, W. (2009). Testing a social
cognitive model of moral behavior: The interaction of situational factors and moral
identity centrality. Journal of Personality and Social Psychology, 97, 123–141.
Aquino, K., McFerran, B., & Laven, M. (2011). Moral identity and the experience of
moral elevation in response to acts of uncommon goodness. Journal of Personality
and Social Psychology, 100, 703–718.
Aquino, K., & Reed, A. (2002). The self-importance of moral identity. Journal of Personality
and Social Psychology, 83, 1423–1440.
Blasi, A. (1984). Moral identity: Its role in moral functioning. In W. Kurtines, & J. Gewirtz
(Eds.), Morality, moral behavior and moral development (pp. 128–139). New York:
Wiley.
Brooks, A. (2006). Who really cares: The surprising truth about compassionate conservatism
who gives, who doesn't, and why it matters. New York: Basic Books.
Cronin, C. (2006, March 3). Are there too many charities?, BusinessWeek Online. http://
www.businessweek.com/bwdaily/dnflash/mar2006/nf2006033_5498.htm
de Waal, F. B. M. (1996). Good natured: The origins of right and wrong in humans and
other animals. Cambridge, MA: Harvard University.
Dean, J. W. (2006). Conservatives without conscience. Penguin.
Ford, P. (2006, May 3). In Europe, too, a 50/50 political divide. (Cover story). Christian
Science Monitor, 98, 1–11.
Forehand, M., Deshpandé, R., & Reed, A. (2002). Identity salience and the influence of
differential activation of the social self-schema on advertising response. Journal
of Applied Psychology, 87, 1086–1099.
Fowler, J., & Kam, C. (2007). Beyond the self: Social identity, altruism, and political participation. Journal of Politics, 69, 813–827.
Freeman, D., Aquino, K., & McFerran, B. (2009). Overcoming beneficiary race as an impediment to charitable donations: Social dominance orientation, the experience of
moral evaluation, and donation behavior. Personality and Social Psychology Bulletin, 35,
75–84.
Giving USA (2010). Giving USA Foundation: Center on philanthropy at Indiana University.
Graham, J., Haidt, J., & Nosek, B. A. (2009). Liberals and conservatives rely on different sets
of moral foundations. Journal of Personality and Social Psychology, 96, 1029–1046.
Graham, J., Nosek, B. A., Haidt, J., Iyer, R., Koleva, S., & Ditto, P. H. (2011). Mapping the
moral domain. Journal of Personality and Social Psychology, 101(2), 366–385.
Grove, D., Remy, R., & Zeigler, L. H. (1974). The effects of political ideology and educational climates on student dissent. American Politics Research, 2, 259–275.
Haidt, J. (2012). The righteous mind: Why good people are divided by politics and religion.
New York: Pantheon.
Haidt, J., & Graham, J. (2007). When morality opposes justice: Conservatives have
moral intuitions that liberals may not recognize. Social Justice Research, 20, 98–116.
Haidt, J., & Joseph, C. (2004). Intuitive ethics: How innately prepared intuitions generate culturally variable virtues. Daedalus, 133, 55–66.
Hernandez, J. (2010, August 2). Political ads off limits, Goldman promises. The New York
Times. http://www.nytimes.com/2010/08/03/nyregion/03goldman.html?_r=1.
Jaccard, J., & Turrisi, R. (2003). Interaction effects in multiple regression. Thousand Oaks,
CA: Sage.
MacDonald, G. (2004, December 22). Who are the nation's ‘cheapstates’? Try the blue
ones. Christian Science Monitor, 97, 15.
Mehrabian, A. (1996). Relations among political attitudes, personality, and psychopathology assessed with new measures of libertarianism and conservatism. Basic and Applied
Social Psychology, 18, 469–491.
Reed, A., II (2004). Activating the self‐importance of consumer selves: Exploring identity salience effects on judgments. Journal of Consumer Research, 31, 286–295.
Reed, A., II, & Aquino, K. (2003). Moral identity and the expanding circle of moral regard
toward out-groups. Journal of Personality and Social Psychology, 84, 1270–1286.
Reed, A., II, Aquino, K., & Levy, E. (2007). Moral identity and judgments of charitable behaviors. Journal of Marketing, 71, 178–193.
Reyna, C., Henry, P. J., Korfmacher, W., & Tucker, A. (2005). Examining the principles in
principled conservatism: The role of responsibility stereotypes as cues for
deservingness in racial policy decisions. Journal of Personality and Social Psychology,
90, 109–128.
Reynolds, S. J., & Ceranic, T. L. (2007). The effects of moral judgment and moral identity
on moral behavior: An empirical examination of the moral individual. Journal of
Applied Psychology, 92, 1610–1624.
Rokeach, M. (1973). The nature of human values. New York: Free Press.
Schwartz, S. H. (1992). Universals in the content and structure of values. In M. P. Zanna
(Ed.), Advances in experimental social psychology, Vol. 25. (pp. 1–65)New York:
Academic.
Shang, J., Reed, A., & Croson, R. (2008). Identity congruency effects on donations. Journal
of Marketing Research, 45, 351–361.
Shao, R., Aquino, K., & Freeman, D. (2008). Beyond moral reasoning: A review of moral
identity research and its implications for business ethics. Business Ethics Quarterly,
18, 513–540.
Skitka, L., & Tetlock, P. (1993). Providing public assistance: Cognitive and motivational
processes underlying liberal and conservative policy preferences. Journal of Personality
and Social Psychology, 28, 1205–1223.
Smiley, J. (2004). The unteachable ignorance of the Red States. Retrieved November
15, 2007, from http://www.slate.com/id/2109218/
UK charitable sector (2011). The UK charitable sector: A snapshot. Retrieved June 28, 2011,
accessed http://www.philanthropyuk.org/resources/uk-charitable-sector-snapshot
Varadarajan, P. R., & Menon, A. (1988). Cause-related marketing: A coalignment of
marketing strategy and corporate philanthropy. Journal of Marketing, 52, 58–74.
Winterich, K. P., & Barone, M. J. (2011). Warm glow or cold, hard cash? Social identity
effects on consumer choice for donation versus discount promotions. Journal of
Marketing Research, 48, 855–868.
Winterich, K., Mittal, V., & Ross, W. (2009). Donation behavior toward ingroups and
outgroups: The role of gender and moral identity. Journal of Consumer Research,
36, 199–214.
Zdravkovic, S., Magnusson, P., & Stanley, S. M. (2010). Dimensions of fit between a
brand and a social cause and their influence on attitudes. International Journal of
Research in Marketing, 27, 151–160.
Intern. J. of Research in Marketing 29 (2012) 355–362
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
Repeated exposure to the thin ideal and implications for the self: Two weight loss
program studies
Anne-Kathrin Klesse a,⁎, Caroline Goukens b, Kelly Geyskens b, Ko de Ruyter b
a
b
Department of Marketing, Tilburg University, Warandelaan 2,5000 LE Tilburg, The Netherlands
Maastricht University, Department of Marketing and Supply Chain Management, Tongersestraat 53, 6211 LM Maastricht, The Netherlands
a r t i c l e
i n f o
Article history:
First received in 29, June 2011 and was under
review for 6 months
Available online 14 September 2012
Keywords:
Thinness ideal
Exposure to models
Self-concept
Goal attainability
a b s t r a c t
Exposure to thin models results in self-esteem shifts that influence people's motivation to diet. This research
study applies a goal perspective to explain the effect of exposure to thin models on dieters' motivation to lose
weight. Two (one-week) weight loss program studies that included treatment conditions in which participants were repeatedly exposed to either a thin model or to control conditions with either a neutral
dieting-related cue (Study 1) or a normal-sized model (Study 2) were conducted. Female participants who
were exposed to the thin model perceived their dieting goals as less attainable, engaged in more
goal-inconsistent behavior (i.e., consuming unhealthy snacks), and had less success losing weight. The effect
of exposure to thin models on weight loss success is mediated specifically by the perceived attainability of the
thin ideal.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
Each of us has a sense of who we are. The existing literature uses
several different terms to describe this sense of self, self-identity, or
self-concept (e.g., Belk, 1988; Ellemers, Spears, & Doosje, 2002;
Markus & Kunda, 1986). To know one's “self” means that one perceives oneself as possessing (or lacking) certain character traits, skills,
or physical attributes and thus as belonging to certain social groups
(Dittmar, 2008). People frequently derive self-images by comparing
themselves with others (Festinger, 1954), which has direct consequences on how they evaluate themselves, on how satisfied they
are with whom they represent, and on how they ideally would like
to be (Richins, 1991).
Because body image is an important part of identity (Harter,
1999), comparing ourselves with images presented in mass media affects our self-evaluations, usually by establishing an ideal of what we
want to look like. For example, models presented in mass media are
thinner than the average female (Levine & Smolak, 1996). Fashion
models often have body mass indexes (BMI) as low as 16, which is
significantly lower than the lowest point (18.5) in the range of BMI
values that represent a biologically healthy weight (Feldman, 2006).
Comparisons with a (role) model on such salient and relevant dimensions (e.g., thinness) can strongly influence how a person thinks and
feels about the self (Epstude & Mussweiler, 2009). Especially for female dieters, most of whom have internalized a thin ideal to define
⁎ Corresponding author.
E-mail addresses: A.K.Klesse@uvt.nl (A.-K. Klesse),
c.goukens@maastrichtuniversity.nl (C. Goukens), k.geyskens@maastrichtuniversity.nl
(K. Geyskens), k.deruyter@maastrichtuniversity.nl (K. de Ruyter).
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.06.003
their personal appearance value system, body weight and size are
highly accessible domains of identity and the ideal self (Ogden,
1995). Consequently, these women are very susceptible to thin
model cues (Wegner, 1994; Wheeler & Berger, 2007).
The consequences of exposure to thin models have been well
studied, and the literature primarily demonstrates that self-esteem
and body satisfaction decrease among dieters who are exposed to
media images of thin models (see Grabe, Ward, & Hyde, 2008). Yet,
some studies also suggest positive outcomes, particularly that viewing a thin model might have inspirational effects (Collins, 1996;
Joshi, Herman, & Polivy, 2004; Mills, Polivy, Herman, & Tiggemann,
2002). Recent studies explain these contradictory findings by demonstrating that the effects depend on the extremity of the model cue
(Smeesters & Mandel, 2006; Smeesters, Mussweiler, & Mandel,
2010); that is, extreme comparison standards foster a contrast effect
(e.g., Meyers & Biocca, 1992; Richins, 1991) that can induce changes
in self-esteem, motivation to lose weight, and even food intake
(Dittmar & Howard, 2004; Heinberg & Thompson, 1995; Mills et al.,
2002).
However, existing experimental studies on the behavioral consequences of model exposure on eating behavior have been limited to
single-exposure settings in which participants' self-esteem, motivation to diet, or eating behavior is assessed immediately after the exposure. In this study, we argue that dieting and eating behaviors
require a longer period of observation that allows for the concurrent
investigation of the participants' motivation to pursue a weight loss
goal and their corresponding eating behaviors. We propose that just
as exposure to model cues influences eating behavior, one's eating
behavior also serves as a feedback about one's progress in achieving
the relevant goal. If exposure to a thin model causes dieters to
356
A.-K. Klesse et al. / Intern. J. of Research in Marketing 29 (2012) 355–362
disinhibit food intake (Mills et al., 2002), those dieters might be
interpreting their increased consumption as lack of progress in
attaining their goal to lose weight. Such negative feedback regarding
their progress toward a particular goal may then initiate a shift in
their motivation to pursue the goal or to engage in goal-consistent
behavior by dampening their expectations of goal attainability
(Brendl & Higgins, 1995).
To obtain a comprehensive understanding of how model cues influence the motivation to lose weight and the corresponding eating
behavior, we need to adopt a longer perspective. In this article, we investigate the effects of repeated exposures to a thin model on the motivation to diet and on eating behavior over a longer time span, which
enables us to contribute to existing research in several ways. First,
whereas previous studies have ignored the intertwining effects of exposure on eating behavior and on motivation to lose weight, we consider both factors concurrently over a longer time span. Second,
modern consumers cannot realistically be exposed only once to a
thin model cue; they frequently encounter such cues, either intentionally or unintentionally. Dieters who have internalized a thin
ideal should be particularly sensitive to these thin model cues
(Wegner, 1994; Wheeler & Berger, 2007). We therefore investigate
the effects of this constant exposure on participants' eating behaviors
and motivations to lose weight. Third, achieving a goal such as losing
weight does not depend solely on an individual's motivation in any
given situation but rather demands engagement in goal-consistent
behavior over time. Therefore, exploring the effect of exposures to
thin models over time in a setting that mimics typical dieting scenarios seems more adequate.
that self-judgments move in the direction of the comparison standard
(Mussweiler & Strack, 2000). In contrast, comparisons with an extreme
standard (e.g., an extremely thin model) foster standard-inconsistent
self-knowledge and result in contrasts, such that the self is judged as
opposite to the comparison standard (Richins, 1991). If the difference
between the assessor and the model with respect to the standard is significant (e.g., normal-BMI respondents exposed to extremely thin
models), exposure to the model leads to a contrast effect and reduces
self-esteem (Smeesters et al., 2010), which might motivate people to
lose weight (Dittmar & Howard, 2004; Heinberg & Thompson, 1995;
Smeesters et al., 2010) and change their eating behaviors (Anschutz,
Engels, Becker, & van Strien, 2008; Seddon & Berry, 1996; Smeesters
et al., 2010; Strauss, Doyle, & Kreipe, 1994; Warren, Strauss, Taska, &
Sullivan, 2005).
Existing studies have also demonstrated that exposure to thin
models influences subsequent food intake. Nonetheless, whereas
some studies find that people eat less after exposure to thin models
(Smeesters et al., 2010), others find that dieters stop inhibiting their
eating behavior and instead increase their food intake (Mills et al.,
2002). For dieters, exposure to thin models might increase the perceived discrepancy between their actual self and the desired end
state because it fosters contrast effects (Smeesters et al., 2010). The
proposition that the discrepancy between the assessor (e.g., the dieter) and the comparison standard (e.g., a thin model) determines the
consequences of exposure is consistent with studies on goal formation (Brown, 1948; Hull, 1932; Levin, 1938), which demonstrate
that the distance between the current self and a desired end state determines the motivation to pursue a certain goal.
2. Theoretical background
2.2. Goals and attainability
2.1. Social comparison
Theories of goal pursuit emphasize a key driver: the desire to reduce a perceived discrepancy between the actual state and the desired end state (Carver & Scheier, 1998; Kruglanski et al., 2002).
When a desirable end state has been activated and adopted as a
goal, people continuously assess their progress by comparing their actual state (e.g., current weight) with the activated end state (e.g., thin
model's perceived weight). Feedback on perceived progress toward
the goal may influence expectations about goal attainability (Brendl
& Higgins, 1995) and thereby influence the motivation to work toward the goal in two dynamic ways.
On the one hand, proximal goals should increase individual motivation and promote better performance than distal goals. Bandura
and Schunk (1981) find that proximal goals, compared with distal
goals, provide immediate and achievable incentives and are thus
more effective in mobilizing goal-attainment efforts. Similarly,
according to the goal gradient hypothesis (Brown, 1948; Hull, 1932;
Levin, 1938), “goals loom larger” effect (Brendl & Higgins, 1995),
the classic theory of achievement motivation (Atkinson, 1964) and
the theory of reasoned action (Fishbein & Ajzen, 1974), the motivation to attain a certain goal increases as the desired end state approaches. If a goal appears closer and easier to attain, individuals
become more confident (Tubbs, Boehne, & Dahl, 1993) and allocate
more effort to reach it. However, if a goal appears very difficult to attain, continued investments of time and effort seem unlikely to pay
off, and the goal seeker might thus disengage from the effort. Following this reasoning, we posit that the intertwining effects of increased
food intake and exposure to thin models decrease dieters' commitment to their goal.
On the other hand, studies on the unconscious effects of subtle
cues on behavior suggest that primes trigger more enduring effects
when they are perceived as distant from an active self-concept. Sela
and Shiv (2009) demonstrate that situational primes consistent
with self-concepts are more likely to affect behavior in a semantic activation pattern; cues that are discrepant from the self-concept are
likely to affect it in a goal-activation pattern. Thus, a perceived
Appearance, beauty, and the “perfect” body have long been central
influences on women's identities. Women are socialized early to understand that their bodies should be used to attract others (e.g., Thompson,
Heinberg, Altbe, & Tantless-Dunn, 1999), and they learn to see themselves as objects to be evaluated on the basis of their appearance
(Frederickson & Roberts, 1997). This pressure is repeatedly reinforced
by strong sociocultural ideals of female beauty, which have come to
be embodied in ultra-thinness (Wiseman, Gray, Moismann, & Ahrens,
1992). In this context, women develop a natural drive to evaluate
their own attributes and abilities and compare themselves with media
models, who are substantially thinner than the majority of the female
population (Festinger, 1954). These upward comparisons with someone who is superior with regard to the relevant dimension (i.e., a thin
model who is superior on the thinness dimension; Wertheim, Paxton,
Schutz, & Muir, 1997) strongly influence how people think and feel
about themselves (Epstude & Mussweiler, 2009). The effects of upward
comparisons to thin (idealized) models are often conceptualized in
terms of “contrast effects,” which refer to the tendency to evaluate oneself more negatively after viewing highly attractive others (Thornton &
Moore, 1993). Richins (1991) demonstrates that women are less satisfied with their own physical appearance after exposure to thin, attractive models. However, upward comparisons may also have positive
effects (e.g., Collins, 1996; Mussweiler & Strack, 2000). For example,
viewing highly attractive models may have inspirational effects and
may result in positive shifts in self-perception (Collins, 1996). Mills
et al. (2002) show specifically that dieters rate themselves as thinner
after exposures to idealized body images.
Recent studies (Smeesters & Mandel, 2006; Smeesters et al., 2010)
shed further light on these contradictory findings by demonstrating
that the extremity of the model determines the effects of the exposure
on people's beliefs and self-esteem regarding their appearance. Comparisons with a moderate standard (e.g., a moderately thin model) trigger standard-consistent self-knowledge and result in assimilation, such
A.-K. Klesse et al. / Intern. J. of Research in Marketing 29 (2012) 355–362
discrepancy between the active self-concept and the cue signals that
the goal has not been attained, which functions as a motivator
(Dijksterhuis, Chartrand, & Aarts, 2007; Sela & Shiv, 2009). This finding that primes are especially motivating when there is greater discrepancy between the active self-concept and the prime (Sela &
Shiv, 2009) is in line with studies on dynamic self-regulation, which
demonstrate that a focus on what remains to be done to achieve a desired end state is a sign of limited progress and increases the motivation to adhere to the goal (Koo & Fishbach, 2008). Following this
reasoning, the intertwining effects of exposure to thin models and increased food intake should increase the perceived discrepancy between the dieter and the model and thus increase the motivation to
attain the weight loss goal.
These two accounts thus make different predictions about the influences of exposure to thin model cues on a dieter's commitment to
become thinner and on her identity construction. According to studies on modern identity, identity has shifted “from a fixed set of characteristics determined by birth to a reflexive, ongoing, individual
project shaped by appearance and performance” (Zukin & Maguire,
2004, p. 180). That is, a person's sense of self is guided, constantly refined, and updated by what happens and by what he or she has done
and hopes to become. Individual success or failure in achieving the
goal of losing weight should thus have direct consequences for identity formation.
2.3. Current research
We investigate whether being exposed to a very thin model decreases the motivation to lose weight and triggers disengagement
from the goal (prediction I) or increases that motivation and fosters
goal-consistent behavior (prediction II). In this study, we focus on
normal-weight to moderately overweight women who view very
thin model cues, a common situation in everyday life. We specifically
include dieters as participants because they appear most susceptible
to thin model cues as a result of their efforts to become thinner
(Wegner, 1994; Wheeler & Berger, 2007). Because exposure to thin
models might result in more food intake (Mills et al., 2002), which increases the discrepancy between the current self and the desired end
state, it becomes especially relevant to explore how the motivation to
attain a thinner self evolves over time. Unlike existing studies, we do
not rely on a single exposure to a thin model but rather expose participants repeatedly to the thin ideal.
3. Study 1
We organized a weight loss program for students to explore
whether repeated exposure to a thin model, rather than a neutral,
dieting-related cue, influenced the dieters' motivation and their success in becoming thinner. We specifically invited female students
who wanted to lose weight to participate in a one-week program.
They received an eating diary in which they were to note after
every occasion of food consumption (i.e., breakfast, lunch, dinner,
and all snacks) exactly what they ate. Half of the participants received
a diary with a cover picture of a thin model (treatment); the other
half used diaries with a neutral, dieting-related image of a measuring
tape on the cover (control). To avoid any potential bias that might
arise if only the control condition saw a process-related cue, we included the picture of the measuring tape within the diary provided
to the treatment condition participants.
3.1. Participants
Dieting is especially common among young women (Morgan,
2007), so our sample of undergraduate women was highly appropriate. We sent an e-mail to all first-year female students and explained
that we were looking for women who wanted to lose weight and
357
would like to participate in a one-week weight loss program. They
were told that they would receive 50€ for their participation. We
recruited 54 female undergraduate students with an average age
of 19.96 years (SD = 1.70) who wanted to lose an average of
6.85 kg (SD = 4.15) and had an average BMI of 23.25 (SD = 4.00),
which is near the upper bound of the normal and healthy BMI
range (i.e., 18.5–24.9).
3.2. Method
Because losing weight is the top New Year's resolution for many
people (Dorsett, 2010), we considered the beginning of the (calendar) year to be the perfect time to conduct the study. Before the
study began, the participants were asked to complete an online questionnaire that collected some general information and to choose
among four time slots for their first meeting. This choice, unbeknownst to the participant, also assigned her to either the treatment
(model cue) or the control (dieting-related cue) condition. Using the
participants' answers to the online questionnaire, we compared the
participants in the two conditions with respect to how many kilos
they wanted to lose and their motivation to lose weight. None of
the results indicated any significant differences between the two
groups (all Fs b 1).
The weight loss program spanned one week. The available time
slots allowed the participants to participate in meetings at four different times during the day. Each meeting followed a strict procedure:
The participants were informed that many women want to lose
weight but that they unfortunately often use extreme diets, pills, or
medicines that may be effective in the short term but result in the
so-called yo-yo effect in the long run. The researcher explained that
in addition to limiting food intake, it was important for them to monitor their food consumption. Therefore, each participant received an
eating diary and was asked to write down, for one week, everything
she ate and drank. These diaries were created in cooperation with a
renowned health center (Infraligne®) that helps people lose weight.
As we noted, half of the diaries featured a thin model on the cover,
with pictures of the measuring tape on the pages, whereas the other
half showed the neutral, dieting-related picture on the cover and
every note page. An instruction manual explained how the participants should complete their diary and highlighted the importance
of taking the diary with them and noting their consumption right
away, including everything they ate or drank, without exceptions.
At the end of the first session, the participants were weighed and received their eating diaries. They recorded their current weight and
desired weight on the first page of the diary, and they were informed
that they would be weighed again one week later. The researcher
asked them to wear the same clothes for the second meeting to ensure that weight differences could not be attributed to lighter or
heavier clothes. The participants also received a phone number
that they could call any time if they had any questions. Within the
period of the study, we sent them two e-mails and called them
once to confirm that they were writing down everything they ate
and drank.
The second meeting took place exactly one week later. The participants first completed a questionnaire that we designed to investigate
their perceived goal attainability (“I am able to work effectively toward my long-term goal”), motivation to write in their diary (“I was
very motivated to write in my eating diary”), and self-esteem regarding their appearance (e.g., “I am dissatisfied with my weight” from
the Appearance Self-Esteem Scale, Heatherton & Polivy, 1991). The
items were measured on seven-point agree–disagree Likert scales.
Then, the participants assigned a grade to their performance with respect to their goals during the weight loss program (1–10, 10 being
very good). Finally, each participant was weighed again, and this
weight was recorded on the last page of the diary.
358
A.-K. Klesse et al. / Intern. J. of Research in Marketing 29 (2012) 355–362
3.3. Results
Overall, the participants were highly compliant with their diary
entries. Six participants did not seem to take the weight loss program
seriously and wrote only sporadically in their diaries; we excluded
them from the analysis. All of the other participants followed the
rules explained in the manual and indicated that they carefully
noted their consumption.
3.3.1. Perceived attainability
We assessed whether the treatment and control participants differed in their perception of goal attainability. The analysis revealed
a significant difference (F(1,46) = 6.29, p b .05) between the control
participants (Mc = 5.08, SD = 1.50) and the treatment participants
(Mt = 4.08, SD = 1.25). The participants exposed to the thin ideal
were less confident than the participants in the control condition
that they would be able to attain their goal. This finding is in line
with our argument that exposure to thin models causes the goal to
be perceived as more difficult to attain. However, the question
remained as to whether this perception motivated or demotivated
them to work toward their dieting goals.
3.3.2. Goal success
By adopting a broader time span, we were able to investigate the
participants' actual weight loss, which should have signaled their
level of motivation to work toward their dieting goals. We subtracted
the weight measured at the end of the week from the initial weight
measured at the beginning of the study. To express the difference in
weight not in absolute terms but in relation to participants' initial
body weight, we divided the difference by the initial body weight.
The comparison of the control participants (Mc = .87, SD = 1.08)
with the treatment participants (Mt = − .23, SD = 2.11) revealed a
significant difference (F(1, 46) = 5.24, p b .05). Specifically, the participants in the control condition managed to lose weight, whereas
those in the treatment condition were less successful in achieving
their goal and even gained a slight amount of weight. From these results, we were able to infer that the participants exposed to the thin
ideal consumed more food than the participants in the control
condition.
Because snacking (both too much and too often) contributes substantially to weight gain (Science Daily, 2000), we compared the diaries of members of the control and treatment conditions based on the
number of unhealthy snacks (e.g., chocolate, chips, cake, donuts) they
consumed.1 We defined a snack as anything eaten between meals or
after dinner. The participants in the treatment condition ate significantly more snacks (Mt = 10.45, SD = 9.03) than the participants in
the control condition (Mc = 5.91, SD = 4.47; F(1, 46) = 4.89, p b .05).
We acknowledge, however, that this finding does not necessarily indicate that they consumed more calories because the snack count
did not include calorie amounts.
Thus, we conducted further analyses to compare the number of calories in the unhealthy snacks the participants consumed. Three independent judges checked the number of calories in each snack listed in
the diaries and calculated the total number of calories consumed over
the week.2 The participants in the treatment condition (Mt = 1284.75,
SD = 1023.15) consumed significantly (F(1, 46) = 4.92, p b .05) more
calories than those in the control condition (Mc = 754.15, SD =
571.08), which explains why they gained weight. Thus, the weight
loss program that showed them a picture of a very thin model did not
1
We asked three independent judges to count the number of snacks consumed by
each participant. The judges' results were all highly correlated (r = .846, r = .745;
p b .001). Consequently, we added the scores and calculated the average.
2
To determine the number of calories in each snack, the judges used the following
web pages: www.caloriecount.com, www.dietbites.com, and www.livestrong.com.
Their results were highly correlated (r = .856, p b .001), so we used the average for
our analysis.
help them decrease the discrepancy between their actual weight and
their desired weight but instead increased it even more.
The question that arises is whether the consumption of calories was
different from the outset or whether this difference appeared over the
one-week period. Accordingly, we compared the amount of calories
that the participants in the treatment condition consumed in the beginning (days 1 and 2) and at the end (days 7 and 8) of the week to the
amount consumed by the participants in the control condition.3 We
found no significant difference during the initial days of the program;
the participants in the treatment condition consumed snacks containing
122 cal on average (SD= 199.19), and the participants in the control
condition consumed snacks containing approximately 88 cal (SD=
134.19; Fb 1). However, we uncovered a significant difference at the
end of the week: the control condition participants consumed an average of 60 cal (SD= 111.13), whereas the treatment condition participants consumed an average of 174 cal (SD=205.14; F(1, 46) =6.04,
pb .05). Although this finding is surprising, considering that the participants were trying to lose weight, it is in line with our finding that they
perceived the dieting goal as less attainable. Because expectations for
the eventual attainment of a particular goal can determine whether
people persist or disengage (e.g., Carver & Scheier, 1998), and because
the treatment condition caused the goal of losing weight appear
unattainable, the participants considered it reasonable to stop putting
effort into their goal of losing weight.
3.3.3. Self-evaluation
To assess whether the participants exposed to the thin model,
who performed significantly worse than members of the control condition, were aware of their bad performance, we compared the conditions with respect to the grade that participants assigned themselves.
The participants in the treatment (Mt = 7.25, SD = 1.42) and control
(Mc = 7.21, SD = 1.95) conditions assigned almost identical grades
to themselves (F (1, 46) = .007, p = .93). That is, the participants exposed to the thin model did not realize that the model cue negatively
influenced their eating behavior. This is in line with existing studies on
goal activation (Chartrand, Huber, Shiv, & Tanner, 2008), which found
that situational cues influence goal-consistent or goal-inconsistent behavior without the subjects' awareness. This finding is striking: were
participants in the two conditions really pursuing the same goal
throughout the week? We might argue that the participants exposed
to the thin model simply revised their goal downward when they perceived that their initial target was not attainable. The questionnaire
that the participants completed at the end of the study included a question that explicitly asked whether they had changed their goal weight;
only three respondents indicated that they revised their goal (they revised it downward). Thus, the participants in the treatment condition
did not consciously revise their goal and assess their performance
according to a newer, more attainable target.
3.4. Discussion
The results of Study 1 reveal that repeated exposure to a thin
model, rather than a dieting-related cue, triggers perceptions that a
weight loss goal is unattainable. Existing studies similarly demonstrate that people are more confident about attaining goals that require less work to complete (e.g., Tubbs et al., 1993). An extremely
thin model represents a desired end state that is distant from the current self. The large discrepancy between the actual self and the thin
model then emphasizes the significant effort required to achieve the
desired ideal, which causes people to engage in goal-inconsistent behavior (i.e., consuming unhealthy snacks) and makes them less successful in losing weight.
3
For this analysis, we relied on data from one coder; the other two were responsible
only for providing overall numbers.
A.-K. Klesse et al. / Intern. J. of Research in Marketing 29 (2012) 355–362
Accordingly, Study 1 provides evidence in support of our prediction I: being confronted with a thin model prompts dieters to disengage from the goal of losing weight. These findings run counter to
the account provided by Sela and Shiv (2009), who argue that
model cues encourage people to reach their goal by representing a
distant desired end state (prediction II). The significant discrepancy
between the dieter's actual self and the thin model might explain
why our findings support prediction I; because the distant model
cue appears out of reach, the sense of unattainability is likely to trigger loss of motivation in persisting with one's goal. By contrast, prediction II might hold for distant goals that still seem attainable. In
Study 2, we assessed the assumption that it is the perceived
unattainability of the goal that is responsible for the negative effect
of exposure on the treatment participants' success in the program.
4. Study 2
For this study, we specifically controlled for model thinness: the
treatment condition featured a very thin model (different from the
one used in Study 1), whereas the control condition included the
same model, adjusted to appear normal in size. We used Photoshop®
tools to alter some body parts (i.e., arms, belly, legs, and face) of the
thin model so that the model looked normal rather than extremely
thin. Thus, we could test whether it was exposure to a model as
such or only exposure to an unrealistically thin model that triggered
the perception of an unattainable goal and dampened motivation.
The findings from Study 1 also suggest that perceived attainability
determines whether people engage in goal-consistent behavior that
pushes them toward their desired end state. Following this reasoning,
we might have expected that perceived attainability would mediate
the effect of exposure to the thin model on the participants' weight
loss, even though our mediation tests revealed no significant effect.
Our measure of attainability (“I am able to work effectively toward
my long-term goal”) in Study 1 was rather broad and did not explicitly assess the perceived difficulty of losing weight. Therefore, for
Study 2, we measured attainability more precisely (“It is unrealistic
for me to expect to reach my goal of losing weight”) to determine
the potential for mediation.
Our experimental design for Study 2 also helped us address a possible alternative explanation for the Study 1 results: being presented
with a thin model might have evoked an outcome focus, whereas the
dieting-related cue (i.e., measuring tape) might have activated a process focus. Existing studies indicate that a process focus results in different levels of performance than an outcome focus. Gollwitzer and
Brandstätter (1993) show that people complete difficult goals approximately three times more often when they adopted an implementation intention (“I intend to perform goal directed behavior y
when I encounter situation z”) rather than merely goal intentions
(“I intend to achieve x”). Similarly, Pham and Taylor (1999) demonstrate that students study more and earn better grades when they
focus on processes rather than outcomes. Hence, the process focus
stimulated by the measuring tape in the Study 1 control condition
may have driven participants more effectively toward the goal of losing weight than did the outcome focus prompted by the thin model. If
we could replicate our findings when both conditions featured images of the same model that differed only in the thinness, we could
rule out this explanation.
Finally, we explicitly asked the Study 1 participants to monitor
their food consumption behavior. The strict monitoring they experienced might have made them more aware of their consumption behavior. Thus, in Study 2, we did not mention that the diaries had to
be handed in but rather left the participants with the impression
that the diaries would be only for their personal use. We also did
not call or e-mail them during the week; we only indicated that
they could contact us with any questions or problems.
359
4.1. Participants
As with Study 1, we sent an e-mail to female undergraduates and
explained that we wanted to recruit female students who wanted to
lose weight to participate in a one-week weight loss program, for
which they would receive 50€. As a first requirement for participation, they had to complete an online questionnaire, which we used
to prescreen participants with respect to their current weight and
dieting intentions. We recruited 42 female undergraduate students
who expressed a wish to lose weight. The online questionnaire
showed that they are 20.98 years of age on average (SD = 1.41),
that they wanted to lose an average of 4.82 kg (SD = 4.61), and that
they had an average BMI of 22.5 (SD = 3.94), which lies at the
upper boundary of the normal and healthy BMI range.
4.2. Method
We conducted this weight loss program during the last week of
November; the study solicitation communicated that we were
looking for students who wanted to shape up for the holiday season
by losing some weight. Before the actual study began, the participants
completed the same online questionnaire that was used in Study 1 to
provide general information and choose a time slot, which assigned
them randomly to the treatment (thin model cue) or control (normal
model cue) condition. We again used their answers to the online
questionnaire to compare the two groups with regard to how many
kilos they wanted to lose and their motivation; neither analysis
showed any significant differences (all Fs b 1).
The weight loss program lasted one week, and each meeting
followed the same procedure as described for Study 1. However, we
did not stress explicitly that it was important for the participants to
keep a precise record of their food consumption. After the introduction to the study, all of the participants received an eating diary. In
the treatment condition, they received a diary that featured the thin
model, whereas in the control condition, they saw a slightly larger
version of the same model.
A pretest (N = 35) confirmed that the thin model (Mt = 5.8,
SD = 1.38) was perceived as significantly thinner than the normal
model (Mc = 4.8, SD= 1.50; F(1, 56) = 6.81, p b .05). However, the pretest revealed no significant difference between the attractiveness ratings of the thin model (Mt = 5.46, SD =1.45) and those of the normal
model (Mc = 5.70, SD= 1.37; F(1, 56) =.41, p >.05). The pretest participants also did not indicate any differences in how much they desired to
look like either the thin model (Mt = 4.35, SD= 1.62) or the normal
model (Mc = 4.23, SD= 1.50; F(1, 56) = .092, p > .05).
The participants were weighed at the end of the first session, and
their current and desired weight was recorded on the first page of
their diaries. Exactly one week later, the second meeting took place,
during which we measured their perceived goal attainability (“It is
unrealistic for me to expect to reach my goal of losing weight”) explicitly to explore its potential mediating effect. Finally, at the end
of the session, each participant was weighed, and her weight was
recorded on the last page of the diary.
4.3. Results
4.3.1. Perceived attainability
We assessed whether the participants in the two conditions differed in their perception of how attainable the goal was; this analysis
revealed a marginally significant difference (F(1,40) = 63.87, p b .06)
between the control condition participants (Mc = 2.17, SD = .86)
and the treatment condition participants (Mt = 3.04, SD = 1.73). The
participants exposed to the thin model were less confident than the
participants who were confronted with the normal model with regard to their ability to attain their goal.
360
A.-K. Klesse et al. / Intern. J. of Research in Marketing 29 (2012) 355–362
4.3.2. Goal success
In line with Study 1, we examined the participants' success in
achieving their goal to lose weight. We subtracted their weight at the
end of the week from their initial weight, then divided the difference
by their initial body weight. A comparison of the control (Mc = .013,
SD = .009) and treatment (Mt = .003, SD = .013) condition participants
showed a significant difference (F(1, 40) = 6.56, p b .05). The participants in the control condition managed to lose weight, but the participants in the treatment condition were not as successful in losing
weight.
4.3.3. Mediation analysis
As indicated, we intended to test whether perceived attainability
mediated the effect of our manipulation on goal success. Following
Zhao, John, Lynch, and Chen (2010), we applied a bootstrap test to establish mediation. The results revealed a positive and significant
mean indirect effect (a × b = .0022), and the 95% confidence interval
excluded zero (.0051 to .0001).
4.4. Discussion
The results from our second weight loss program study consistently revealed that dieters exposed to a thin model did not succeed
in losing weight, whereas those exposed to a normal-sized model
were able to do so. These results complement and advance the findings from Study 1 in four major ways. First, they demonstrate that it
is not exposure to a model as such that causes dieters to be less successful in attaining their goal but rather being confronted with a very
thin model. Second, the perceived attainability of the dieting goal
drives the effect; being exposed to a thin model decreases the perceived attainability of the desired end state and thereby fosters disengagement from the goal. Measuring perceived attainability more
directly, as in Study 1, helped establish perceived attainability as the
underlying mechanism. Third, the Study 1 results were replicated
even though we did not monitor the participants as closely as we
did in the previous study, which confirms the robustness of the mediating effect of perceived attainability. Even if the participants did not
worry that their diary entries would be checked, the same effects
emerged in terms of goal success. However, because we did not
state explicitly that the diaries were part of the study, we could not
track the effect of exposure to the model cues on actual food consumption, as we did in Study 1. Fourth, using a different model generalized our findings because it showed that it was not the appearance
of the model but rather the perception of her thinness that had a negative effect on the dieters' motivation to lose weight.
5. General discussion
Existing studies have repeatedly demonstrated that exposure to
thin models influences dieters' self-esteem, motivation to diet, and
eating behavior, though only by testing single exposures followed
by immediate assessments of motivational and behavioral consequences. We extended this research stream by showing that it is important to explore the consequences of exposure to thin model cues
over time. In line with existing studies, we have offered two contradictory predictions about the effects of constant exposure to a thin
model on people's commitment to attaining a thinner self. The findings of two separate weight loss program studies provide support
for prediction I: repeated exposure to a thin model fosters disengagement from the goal. Specifically, in Study 1, we showed that dieters
exposed to a thin model ate more unhealthy snacks than dieters exposed to a neutral dieting-related cue. This not only prevented them
from getting closer to their goal of losing weight but even caused
them to gain weight. In Study 2, we complemented and advanced
these findings by showing that it was specifically exposure to a very
thin model that triggered the perception of the goal as unattainable.
The participants exposed to a very thin model perceived their weight
loss goal as less attainable than the participants exposed to a normal
model and therefore did not manage to lose weight. By contrast, their
counterparts who were exposed to a normal model came closer to
attaining their desired weight.
These findings extend previous studies on exposures to thin
models in at least three important ways. First, existing studies rely
predominantly on shifts in self-esteem to explain the consequences
of these exposures on behavioral outcomes. We adopt a goal perspective and thereby highlight that exposure to a very thin model changes
the dieter's belief about the very attainability of a thinner self. Our
findings reveal that the perception that a goal is unattainable
demotivates dieters from investing effort in achieving the goal and
causes them to disengage from the goal. Similarly, existing studies
on goal-related behavior demonstrate that low expectancies of attainability cause people to recognize that their continued investments of time and effort are unlikely to pay off and thus to
eventually give up on that particular goal (Carver & Scheier, 1998).
Although the omnipresence of thin models in contemporary marketing and media fosters desires for a thinner self, it simultaneously
hampers the search for an ideal identity. That is, women adopt the
thin ideal but also recognize it as difficult to attain, so they disengage
from weight loss goals. Self-identity studies also suggest that people
can build a stock of knowledge about themselves and develop multiple self-schemata (Markus, 1977), perhaps by thinking of their recent
experiences of success or failure and the implications of those experiences for future possibilities (Baumeister, 2010). We contribute to
these studies by showing that exposure to thin models alters the importance that people attach to some self-schemata and to the actions
needed to attain this ideal self. In Study 1, for example, the participants in the control condition (Mc = 5.33) reported higher commitment to their dieting goals than the participants in the treatment
condition (Mt = 4.52; F(1, 46) = 4.26, p b .05). The latter engaged in
more goal-inconsistent behavior (i.e., eating unhealthy snacks) after repeated exposures to a thin model, which implies that constant confrontation with an ideal self decreases people's motivation to achieve it.
Second, unlike existing studies, we investigated the effect of repeated exposures to thin model cues over a longer time span. Goals
such as losing weight cannot be achieved immediately, so our tests
of the effect of exposure to thin models on motivation to lose weight
are more realistic. The longer time span enabled us to investigate the
exposures' effects on actual weight fluctuation and to reveal the surprising finding that repeated exposures to thin models while dieting
decreases dieters' success. Our study context is realistic (i.e., normal
to moderately heavy women exposed to very thin model cues), and
our findings imply that the prevalence of extremely thin model images in consumer culture is contributing to rising obesity rates. In
this sense, our results add to the current debate regarding obesity
and its causes (e.g., Flegal, Carroll, Ogden, & Curtin, 2010), with important consequences for companies, policy makers, and legislative
bodies. For example, being exposed to thin images in shows such as
America's Top Model, commercials for diet-related products, and
women's magazines makes it even harder for dieters to reach their
ideal identities. In light of this finding, policy makers might consider
imposing boundaries on how thin models depicted in these settings
should be.
Our study is also realistic because consumers are frequently exposed to thin model cues. For example, a female consumer with a
weight loss goal is likely to confront a thin model on the packaging
of her low-fat breakfast cereal, a multitude of very thin models as
she reads a magazine during her lunch break, and more very thin
models as she watches TV while preparing her dinner. Our study design, in which participants recorded their food consumption in a diary
every time they ate something, thus resembles reality in that participants were repeatedly exposed to the thin model cue both before and
after their food consumption.
A.-K. Klesse et al. / Intern. J. of Research in Marketing 29 (2012) 355–362
Third, we contribute to studies on the influence of exposure to thin
models by highlighting that exposure results in overeating outside a laboratory setting. Contrary to what one might expect, dieters do not compensate for indulging (Dhar & Simonson, 1999; Novemsky & Dhar,
2005) due to their exposure to thin models by eating less in subsequent
consumption situations but rather seem to keep eating, as Study 1
showed. These results may imply a “what the hell effect” (Cochran &
Tesser, 1996), such that increased food intake resulting from exposure
to a thin model cue (Mills et al., 2002) gets interpreted by dieters as
an indication of absolute failure. Then, re-exposure to the thin model
when writing in the diary may make this initial lack of progress more salient and foster defeatist thinking, licensing the dieters to indulge further. This explanation is consistent with studies demonstrating that
overly difficult goals are counterproductive (Soman & Cheema, 2004):
if strivers fail, they suffer demotivation and poorer performance than
people who never considered the goal. In our treatment conditions,
the participants might have interpreted their increased food intake as
a failure to achieve their goal, which reduced their commitment to it.
Being confronted with an extremely thin model does not motivate individuals (prediction II) but rather fosters disengagement from the goal.
As with all studies, the present study needs to be interpreted in the
context of its limitations. First, the attractive model on the diary cover in
Study 1 could have increased the participants' motivation to write in
their diary, such that the treatment condition participants provided
more detailed, explicit descriptions of how much and what they consumed. However, this alternative reasoning cannot explain the difference in weight loss/gain or the fact that we did not find a difference in
the participants' motivation to write in their diary. Second, Mills et al.
(2002) show that restrained eaters experience a self-esteem boost
after being exposed to thin media images, so perhaps participants in
the treatment condition in Study 1 considered themselves more attractive after being exposed to the thin model and no longer experienced a
need to lose weight. This alternative account could explain the decreased commitment to dieting goals; however, the comparison of the
two groups of participants based on their self-esteem (Heatherton &
Polivy, 1991) showed no significant difference. We assume that this deviation from extant studies reflects the longitudinal nature of our study:
exposure to a thin media image might boost dieters' self-esteem immediately, but there seems to be no long-term influence.
Finally, our study suggests several directions for further research.
For example, self-identity literature has shown that recent experiences of success or failure help people update their self-views
(Baumeister, 2010). Failing to inhibit food intake might highlight
the difficulty of reaching a thin ideal and cause people to look for alternative ideals to boost their self-image (Rudman, Dohn, & Fairchild,
2007). Further studies should investigate such compensatory mechanisms and explore whether people might try to compensate for their
lack of motivation to lose weight by, for example, studying harder.
Further research could also explore the consequences of our present findings for marketers and companies that advertise dieting or
low-fat products using thin models. Consider, for example, a dieter
who consumes a low-fat product featuring a thin model on the
cover: does she attribute her lack of success in attaining her weight
loss goal to the product and thus refrain from purchasing it again?
Our study suggests that advertising campaigns or package designs
that feature thin models might be counterproductive and hamper
weight loss success and could thus foster dissatisfaction with the
product.
Acknowledgements
The authors are grateful to the health center Infraligne ® for helping creating the diaries used in the study. They also thank Nina Gros
for her valuable help in preparing the study, and Katja Sillen, Bart
van Kranenburg and Tom Van Laer for judging the diaries.
361
References
Anschutz, D. J., Engels, R. C. M. E., Becker, E. S., & van Strien, E. (2008). The bold and the
beautiful. Influence of body size of televised media models on body dissatisfaction
and actual food intake. Appetite, 51, 530–537.
Atkinson, J. W. (1964). An introduction to motivation. Princeton, NJ: D. Van Nostrand.
Bandura, A., & Schunk, D. H. (1981). Cultivating competence, self-efficacy, and intrinsic interest through proximal self-motivation. Journal of Personality and Social Psychology,
41(3), 586–598.
Baumeister, R. R. (2010). The self. In R. R. Baumeister, & E. J. Finkel (Eds.), Advanced social psychology. The state of the science. (pp. 139–175) Oxford: University Press.
Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research, 1,
26–42.
Brendl, C. M., & Higgins, E. T. (1995). Principles of judging valence: What makes events
positive or negative? Advances in Experimental Social Psychology, 28, 95–160.
Brown, J. S. (1948). Gradients of approach and avoidance responses and their relation
to motivation. Journal of Comparative and Physiological Psychology, 41, 450–465.
Carver, C., & Scheier, M. F. (1998). On the self-regulation of behavior. New York: Cambridge
University Press.
Chartrand, T. L., Huber, J., Shiv, B., & Tanner, J. R. (2008). Nonconscious goals and consumer choice. Journal of Consumer Research, 35, 190–201.
Cochran, W., & Tesser, A. (1996). The “what the hell” effect: Some effects of goal proximity and goal framing on performance. In Leonard L. Martin, & Abraham Tesser
(Eds.), Striving and feeling: Interactions among goals, affect, and self-regulation
(pp. 99–120). Mahwah, NJ: Erlbaum.
Collins, R. L. (1996). For better or worse: The impact of upward social comparison on
self-evaluations. Psychological Bulletin, 119, 51–69.
Dhar, R., & Simonson, I. (1999). Making complementary choices in consumption
episodes: Highlighting versus balancing. Journal of Marketing Research, 36,
29–44.
Dijksterhuis, A., Chartrand, T. L., & Aarts, H. (2007). Automatic behavior. In John A.
Bargh (Ed.), Automatic processes in social thinking and behavior. Philadelphia, PA:
Psychology Press.
Dittmar, H. (2008). Consumer culture, identity and well-being. The search for the ‘good
life’ and the ‘body perfect’. New York: Psychology Press.
Dittmar, H., & Howard, S. (2004). Thin-ideal internalization and social comparison tendency as moderators of media models' impact on women's body-focused anxiety.
Journal of Social and Clinical Psychology., 23, 768–791.
Dorsett, K. (2010). New Year's resolution No. 1: Lose weight. CNN http://edition.
cnn.com/2010/HEALTH/diet.fitness/12/31/lose.weight.new.resolution/index.
html (accessed 27-06-2011).
Ellemers, N., Spears, R., & Doosje, B. (2002). Self and social identity. Annual Review of
Psychology, 53, 161–186.
Epstude, K., & Mussweiler, T. (2009). What you feel is how you compare: How comparisons influence the social induction of affect. Emotion, 9, 1–14.
Feldman, D. (2006). Versace & Prada join fight against size-0 models! Gisele Bündchen
risks…fashion model directory. http://www.fashionmodeldirectory.com/news/
Model+News/versace-prada-join-fight-against-size0-models-gisele-bundchenrisks.htm (accessed 29-06-2011)
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117–140.
Fishbein, M., & Ajzen, I. (1974). Attitudes toward objects as predictors of single and
multiple behavioral criteria. Psychological Review, 81, 59–74.
Flegal, K. M., Carroll, M. D., Ogden, C. L., & Curtin, L. R. (2010). Prevalence and trends in
obesity among US adults, 1999–2008. Journal of the American Medical Association,
303, 235–241.
Frederickson, B. L., & Roberts, T. (1997). Objectification theory: Toward understanding
women's lived experiences and mental health risks. Psychology of Women Quarterly,
21, 173–206.
Gollwitzer, P. M., & Brandstätter, V. (1993). Implementation intentions and effective
goal pursuit. Journal of Personality and Social Psychology, 73, 186–199.
Grabe, S., Ward, L. M., & Hyde, J. S. (2008). The role of the media in body image concerns among women: A meta-analysis of experimental and correlational studies.
Psychological Bulletin, 134, 460–476.
Harter, S. (1999). The construction of the self: A developmental perspective. New York:
Guilford.
Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60, 895–910.
Heinberg, L. J., & Thompson, J. K. (1995). Body image and televised images of thinness
and attractiveness: A controlled laboratory investigation. Journal of Social and Clinical Psychology, 14, 230–242.
Hull, C. L. (1932). The goal-gradient hypothesis and maze learning. Psychological Review, 39, 25–43.
Joshi, R., Herman, C. P., & Polivy, J. (2004). Self-enhancing effects of exposure to
thin-body images. International Journal of Eating Disorders, 35, 333–341.
Koo, M., & Fishbach, A. (2008). Dynamics of self-regulation: How (un)accomplished
goal actions affect motivation. Journal of Personality and Social Psychology, 94,
183–195.
Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun, W. Y., & Sleeth-Keppler,
D. (2002). A theory of goal systems. In Mark P. Zanna (Ed.), Advances in experimental social psychology, 34. (pp. 331–378)San Diego: Academic Press.
Levin, K. (1938). The conceptual representation and the measurement of psychological
forces. Durham, NC: duke University Press.
Levine, M. P., & Smolak, L. (1996). Media as context for the development of disordered
eating. In L. Smolak, M. P. Levine, & R. Striegel-Moore (Eds.), The developmental psychopathology of eating disorders: Implications for research, prevention, and treatment
(pp. 235–257). Hillsdale, England: Lawrence Erlbaum Associates, Inc.
362
A.-K. Klesse et al. / Intern. J. of Research in Marketing 29 (2012) 355–362
Markus, H. R. (1977). Self-schemata and processing information about the self. Journal
of Personality and Social Psychology Bulletin, 10, 203–208.
Markus, H., & Kunda, Z. (1986). Stability and malleability in the self-concept in the perception of others. Journal of Personality and Social Psychology, 51, 858–866.
Meyers, P. N., & Biocca, F. A. (1992). The elastic body image: The effect of television advertising and programming on body image distortions in young women. Journal of
Communication, 42, 108–133.
Mills, J. S., Polivy, J., Herman, C. P., & Tiggemann, M. (2002). Effects of exposure to thin
media images: Evidence of self-enhancement among restrained eaters. Personality
and Social Psychology Bulletin, 28, 1687–1699.
Morgan, C. (2007). 80% of women and 92% of girls are dissatisfied with their bodies.
http://women4hope.wordpress.com/2007/10/18/80-of-women-and-92-of-girls-aredissatisfied-with-their-bodies/ (accessed 27-06-2011)
Mussweiler, Thomas, & Strack, Fritz (2000). The relative self: Informational and judgmental consequences of comparative self-evaluation. Journal of Personality and Social Psychology, 79, 23–38.
Novemsky, N., & Dhar, R. (2005). Goal fulfillment and goal targets in sequential choice.
Journal of Consumer Research, 32, 396–404.
Ogden, J. (1995). Cognitive and motivational consequence of dieting. European Eating
Disorders Review, 24, 228–241.
Pham, L. B., & Taylor, S. E. (1999). From thought to action: Effects of process- versus
outcome-based mental simulations on performance. Personality and Social Psychology Bulletin, 25, 250–260.
Richins, M. L. (1991). Social comparison and the idealized images of advertising. Journal of Consumer Research, 18, 71–83.
Rudman, L. A., Dohn, M. C., & Fairchild, K. (2007). Implicit self-esteem compensation:
Automatic threat defense. Journal of Personality and Social Psychology, 93, 798–813.
Science Daily (2000). http://www.sciencedaily.com/releases/2000/02/000208075502.
htm (accessed 27-06-2011)
Seddon, L., & Berry, N. (1996). Media-induced disinhibition of dietary restraint. British
Journal of Health Psychology, 1, 27–33.
Sela, A., & Shiv, B. (2009). Unraveling priming: When does the same prime activate a
goal versus a trait? Journal of Consumer Research, 36, 418–433.
Smeesters, D. H. R. V., & Mandel, N. (2006). Positive and negative media image effects
on the self. Journal of Consumer Research, 32, 576–582.
Smeesters, D. H. R. V., Mussweiler, T., & Mandel, N. (2010). The effects of thin and heavy
media images on overweight and underweight consumers: Social comparison processes and behavioral implications. Journal of Consumer Research, 36, 930–949.
Soman, D., & Cheema, A. (2004). When goals are counter-productive: The effects of violation of a behavioral goal on subsequent performance. Journal of Consumer Research, 31, 52–62.
Strauss, J., Doyle, A. E., & Kreipe, R. E. (1994). The paradoxical effect of diet commercials
on reinhibition on dietary restraint. Journal of Abnormal Psychology, 103, 441–444.
Thompson, J. K., Heinberg, L. J., Altbe, M., & Tantless-Dunn, S. (1999). Exacting beauty:
Theory, assessment, and treatment of body image disturbance. Washington DC:
American Psychological Association.
Thornton, B., & Moore, S. (1993). Physical attractiveness contrast effect: Implications
for self-esteem and evaluations of the social self. Personality and Social Psychology
Bulletin, 19, 474–480.
Tubbs, M. E., Boehne, D. M., & Dahl, J. G. (1993). Expectancy, valence, and motivational
force functions in goal-setting research: An empirical test. Journal of Applied Psychology, 78, 361–373.
Warren, C. S., Strauss, J., Taska, J. L., & Sullivan, S. J. (2005). Inspiring or dispiriting? The
effect of diet commercials on snack food consumption in high school and
college-aged women. International Journal of Eating Disorders, 27, 266–270.
Wegner, D. M. (1994). Ironic processes of mental control. Psychological Review, 101,
34–52.
Wertheim, E., Paxton, S. J., Schutz, H. K., & Muir, S. L. (1997). Why do adolescent girls
watch their weight? An interview study examining sociocultural pressures to be
thin. Journal of Psychosomatic Research, 42, 345–355.
Wheeler, C. S., & Berger, J. (2007). When the same prime leads to different effects. Journal
of Consumer Research, 34, 357–368.
Wiseman, C., Gray, J., Moismann, J., & Ahrens, A. (1992). Cultural expectations of thinness in women: An update. International Journal of Eating Disorders, 11, 85–89.
Zhao, X., John, G., Lynch, J. R., & Chen, Q. (2010). Reconsidering Baron and Kenny:
Myths and truths about mediation analysis. Journal of Consumer Research, 37,
197–206.
Zukin, Sharon, & Maguire, Jennifer Smith (2004). Consumers and consumption. Annual
Review of Sociology, 30, 173–197.
Intern. J. of Research in Marketing 29 (2012) 363–369
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
Bolstering and restoring feelings of competence via the IKEA effect
Daniel Mochon a,⁎, Michael I. Norton b, Dan Ariely c
a
b
c
Freeman School of Business, Tulane University, 7 McAlister Drive, New Orleans, LA 70118, United States
Harvard Business School, Soldiers Field Road, Boston, MA 02163, United States
Duke University, 1 Towerview Drive, Durham, NC 27708, United States
a r t i c l e
i n f o
Article history:
First received in 3, July 2011 and was under
review for 5 ½ months
Available online 14 September 2012
Keywords:
Effort
Co-creation
Competence
Value
Self-signal
a b s t r a c t
We examine the underlying process behind the IKEA effect, which is defined as consumers' willingness to pay
more for self-created products than for identical products made by others, and explore the factors that influence both consumers' willingness to engage in self-creation and the utility that they derive from such activities. We propose that creating products fulfills consumers' psychological need to signal competence to
themselves and to others, and that feelings of competence associated with self-created products lead to
their increased valuation. We demonstrate that the feelings of competence that arise from assembling products mediate their increased value (Experiment 1), that affirming consumers' sense of self decreases the
value they derive from their creations (Experiment 2), and that threatening consumers' sense of self
increases their propensity to make things themselves (Experiments 3A and 3B).
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
A multitude of companies have emerged that allow consumers to
create and design their own products, such as t-shirts, coffee mugs
and ties. LEGO has a large and profitable online community, where
adult fans of LEGO can generate and submit their own designs. Local
Motors even offers the unique experience of being involved in the assembly of one's own automobile. This trend is also prevalent on new
media channels, where people seem to be more interested in creating
and sharing their own amateur videos than consuming professionally
produced content. Consumers even enjoy repurposing products beyond
their original designs; for example, Ikeahackers.net provides how-to
guides on reconfiguring standard IKEA products, such as bookshelves,
for less-standard uses, such as artsy hamster cages. In short, consumers
increasingly act as co-creators of goods rather than passive recipients of
them (Firat, Dholakia, & Venkatesh, 1995; Prahalad & Ramaswamy,
2000, 2002; Vargo & Lusch, 2004). Why has the co-creation of products
become so popular among consumers?
Functional fit is one of the most obvious benefits of co-creation.
When consumers are involved in the production of a good and can customize it to their tastes, the good is more likely to meet their needs
(Dellaert & Stremersch, 2005; Franke, Keinz, & Steger, 2009; Franke &
Piller, 2004; Randall, Terwiesch, & Ulrich, 2007; Schreier, 2006;
Simonson, 2005). Additionally, consumers gain utility from truly unique
goods, which often result from co-creation (Franke & Schreier, 2008;
Lynn & Harris, 1997; Michel, Kreuzer, Kuhn, Stringfellow, & Schumann,
⁎ Corresponding author. Tel.: +1 504 862 8067.
E-mail address: dmochon@tulane.edu (D. Mochon).
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.05.001
2009). Finally, customers may derive utility from participating in the design process because they find it to be enjoyable (Csikszentmihalyi,
1990; Dahl & Moreau, 2007; Dellaert & Stremersch, 2005; Franke &
Schreier, 2010).
Although the above-mentioned factors account for some of the benefits of co-creation, recent research has shown that, even after controlling for these factors, consumers overvalue their own creations (Franke,
Schreier, & Kaiser, 2010; Norton, Mochon, & Ariely, 2012). In one experiment, participants were willing to pay significantly more for an IKEA
storage box that they assembled than for an identical box assembled
by someone else. This effect, labeled the “IKEA effect,” shows that people place more value on their own creations, even if they are mundane
products that are not unique, customized, or fun to build (Norton et al.,
2012). Why is it that customers value their own creations more highly
than identical products built by others? Furthermore, what conditions
lead people to seek out the opportunity to create products, and what
are the factors that influence the utility obtained from self-creation?
We suggest that the allure of self-created products stems partly
from their role in fulfilling consumers' desire to signal a competent
identity to themselves and to others. By building things themselves,
people both control and shape their environments, thereby demonstrating their competence to themselves and to others. Indeed, Dahl
and Moreau (2007) found that “feelings of competence” were the
most commonly mentioned motivation for engaging in creative
tasks. Therefore, we propose that the competence consumers associate with self-created products drives the value they attach to their
creations. Consequently, consumers will be relatively more attracted
to opportunities to create products when their feelings of competence
have been threatened, and derive relatively less utility from creating
products when their sense of competence has already been affirmed.
364
D. Mochon et al. / Intern. J. of Research in Marketing 29 (2012) 363–369
1.1. Signaling identities to the self and to others
Consumers learn about their identity from their own actions (Ariely
& Norton, 2008; Bem, 1972), and prefer to act consistently with that
identity because behavior that confirms their identity creates utility
(Akerlof & Kranton, 2000; LeBoeuf, Shafir, & Bayuk, 2010). Recent
research has demonstrated that consumers not only behave in accordance with existing identities, but also actively use their behavior to
send “evidence” to themselves that they possess desired identities
(Benabou & Tirole, 2011; Bodner & Prelec, 2003; Gneezy, Imas,
Nelson, Brown, & Norton, 2012). What important signal about identity
does creating products send to the self? A large body of literature demonstrates a fundamental human need for effectance, the ability to successfully produce desired outcomes in one's environment. One of the
means by which people achieve effectance is by affecting and controlling objects and possessions (Bandura, 1977; Belk, 1988; Furby, 1991;
White, 1959). Therefore, self-created products can be used to signal a
competent identity to the self.
In addition, completed products can signal a competent identity to
others. Indeed, consumers actively use products to signal their identities to others (Belk, 1988; Berger & Heath, 2007; Wernerfelt, 1990)
and use particular products, such as feature-rich electronic products,
to signal their competence (Thompson & Norton, 2011). Therefore, we
predict that the IKEA effect is driven by feelings of competence that
are associated with self-created products. Self-created products function as signals of competence to the self and to others, leading to
improved evaluations of these products and to a greater willingness to
pay to possess them. Indeed, the loss of possessions can deprive people
of the sense of self that is attached to those objects (Belk, 1988).
We provide initial support for this argument by examining whether
feelings of competence associated with self-created products, which we
operationalize as consumers' feelings of pride about their creations, mediate the IKEA effect. Pride is closely linked to experiences of success
and failure across a variety of tasks that involve self-efficacy and competence, from taking math tests to winning sporting events. Pride is also
strongly linked to consumers' evaluations of their own identity (Buss,
2001; Lazarus, 1991; Tracy & Robins, 2004a; Weiner, 1985, 1986;
Weiner, Russell, & Lerman, 1979). Research shows that participants
use words such as “accomplished” when asked to describe feelings of
pride (Tracy & Robins, 2007), and that pride is associated with
dominance among mammals (Tracy & Robins, 2004b), which is further
evidence that pride is closely linked to feelings of success and competence. “Feelings of accomplishment,” a construct similar to pride,
mediates the impact of successfully designing hedonic products (e.g.,
t-shirts) on consumers' increased willingness to pay (Franke et al.,
2010). Therefore, we predict that feelings of competence, measured
by feelings of pride, similarly drive consumers' increased willingness
to pay for self-created products. However, in contrast to previous research and consistent with research suggesting that pride is associated
with successful completion of both boring tests and exciting competitions, we predict that pride drives consumers' increased willingness to
pay even when participants create mundane products (e.g., IKEA storage boxes).
In addition to measuring feelings of competence associated with
self-created products, we test our hypotheses by directly manipulating participants' need to signal competence and examining the
consequences of these manipulations on the value derived from
self-creation. Self-affirmation theory argues that people strive to keep
a positive view of the self (Aronson, Cohen, & Nail, 1999; Sherman &
Cohen, 2006) and suggests that people use a “fluid compensation”
procedure, whereby affirming one important value to the self can
temporarily reduce the weight placed on a different value (Steele,
1988). Therefore, if the IKEA effect is driven by the ability of selfcreated products to signal a valued identity, this effect should be reduced or eliminated if participants are first allowed to affirm the self.
Conversely, threatening people's sense of competence should increase
the value they derive from self-creation and increase their propensity to
engage in such activities. Indeed, affirming the self has been shown to
reduce the likelihood of choosing products that signal important components of the self, whereas threatening the self has been shown to have the
opposite effect (e.g. Gao, Wheeler, & Shiv, 2009; Townsend & Sood, 2012).
1.2. Overview of the studies
We present a series of experiments in which we both manipulate
and measure feelings of competence associated with self-created
products. In Experiment 1, we examine whether feelings of pride associated with one's creation mediate the effect of self-creation on
willingness to pay. In addition, we rule out a mood-based explanation
for the effect. In Experiment 2, we provide additional evidence to support our hypotheses, not only by measuring feelings of competence
associated with self-created products, but also by manipulating the
need to signal competence. If our hypotheses are correct, participants
who have affirmed their identity and do not require further boosts to
their sense of competence will receive less benefit from self-created
products and, consequently, the IKEA effect will be absent among
these participants.
Finally, in Experiments 3A and 3B, we use a different methodology
to demonstrate the importance of feelings of competence for
self-assembly and co-creation. We first shake participants' sense of
competence in one domain and then measure whether this manipulation increases their propensity to engage in self-creation. If people
use self-created products to signal competence, then they may be
more willing to engage in co-creation when their feelings of competence have been threatened. Taken together, these experiments provide converging evidence for the proposition that people value
self-created products because of their ability to signal competence
to themselves and to others.
2. Experiment 1
In Experiment 1, we examine whether the increased value of
self-created products is driven by feelings of competence. In this
study, participants either build a product or are given a finished product
that they are asked to examine. We then elicit participants' willingness
to pay for the product, the feelings of competence associated with the
product, and participants' overall mood. Consistent with previous research (Franke et al., 2010; Norton et al., 2012), we predict that participants are willing to pay more for the same product if they create it
themselves. Importantly, we predict that this effect is mediated by feelings of competence (i.e., pride) associated with the product.
We also distinguish our competence-based account from a moodbased account (Schwarz & Clore, 1983), in which participants' positive
moods, related to successfully creating a product, might directly lead to
their increased willingness to pay for the product. Consistent with previous research on task success and mood (e.g. Weiner et al., 1979), we predict that successful self-creation improves consumers' moods. However,
we predict that mood alone does not mediate the effect of self-creation
on willingness to pay, but that pride associated with the product acts as
a mediator both for increased willingness to pay and elevated mood.
2.1. Method
Participants (N = 79; 33 male) were paid $5 to complete the experiment. We excluded four participants because three failed to answer all of the questions and one was an extreme outlier (more
than 3 SDs from the mean on the willingness to pay measure).
We randomly assigned some participants to a group of “builders,”
who assembled a LEGO car. These participants were given the parts
and instructions that come with the product to assemble the car. The
other participants, the “non-builders,” were given the car already
assembled and were asked to examine it. We next solicited their
D. Mochon et al. / Intern. J. of Research in Marketing 29 (2012) 363–369
reservation price by asking them to bid on the car. We told participants
that we would draw a random price from an unknown distribution. If
participants' bids were equal to or above the random price, they
would pay us the random price and take their LEGO car home, but if
their bid was below the random price, they would not purchase their
LEGO car. This technique is an incentive compatible value elicitation
method and is a variant of the Becker, DeGroot, and Marschak (1964)
procedure. Note that participants were not given the opportunity to
customize this product; thus, participants in both conditions were effectively bidding on the same product.
Participants answered two questions that were designed to measure
feelings of competence associated with the product they created. Specifically, participants were asked to rate, on 7-point scales, the extent to
which they felt proud of the product they created (Franke et al., 2010)
and the extent to which they planned to show off their product to
other people. We averaged these two measures to create a composite
measure of competence associated with one's product (r=.33, pb .01).
Finally, participants completed the Positive and Negative Affect Schedule
(PANAS; Watson, Clark, & Tellegen, 1988), indicating the extent to which
they felt a variety of positive (e.g., happy) and negative (e.g., upset) affective states based on 5-point scales (1: not at all, 5: extremely). We averaged items from the PANAS to form separate indices of participants'
positive and negative affect (Cacioppo, Gardner, & Berntson, 1999).
2.2. Results and discussion
We found that builders were willing to pay significantly more for
their cars (M = $1.20, SD = 1.35) than non-builders (M = $0.57, SD =
.76; t(73) = 2.44, p b .05). Although both groups were given the
chance to buy the same product, those who perceived themselves as
creators of the product imbued it with significantly more value, despite the fact that non-builders could easily have disassembled and
reassembled their cars.
We next examined the competence measure to test the underlying
mechanism of the IKEA effect. As predicted, feelings of competence
were higher for builders (M =4.39, SD= 1.48) than for non-builders
(M = 2.81, SD= 1.34; t(73)= 4.85, p b .001; Table 1). More importantly,
feelings of competence were significantly related to participants' willingness to pay (r = .36, p b .01). Therefore, we examined whether competence mediated the effect of self-creation on willingness to pay
(Baron & Kenny, 1986). While the build condition was a significant
predictor of willingness to pay when entered alone into the regression
(B = .62, SE= .26; t(73)= 2.44, p b .05), this effect became nonsignificant when competence was also included in the model (B = .29,
SE= .28; t(72)= 1.02, p = .30), whereas the mediating effect of competence remained significant (B = .21, SE= .09; t(72)= 2.35, p b .05).
Thus, feelings of competence associated with the product significantly
mediated the effect of the experimental condition on willingness to
pay (Sobel's Z = 2.08, p b .05). These results provide initial support for
our claim that people are willing to pay more for products that they create because of the feelings of competence that arise from successful
self-creation.
We further predicted that feelings of competence associated with
creation would elevate the builders' moods, but that mood itself
would not drive the builders' increased willingness to pay for their
products. Builders and non-builders did not differ in negative affect
Table 1
Means for Experiment 1.
WTP
Competence scale
PANAS — positive
PANAS — negative
Builders
Non-builders
$1.20
4.39
2.80
1.19
$0.57
2.81
2.30
1.19
Note: standard errors are in parentheses.
(.22)
(.24)
(.12)
(.06)
(.13)
(.22)
(.15)
(.06)
365
(t b 1); however, builders reported more positive affect (M = 2.80,
SD = .73) than non-builders (M = 2.30, SD = .91; t(73) = 2.61, p =
.01). Moreover, the competence scale was significantly related to participants' positive affect (r = .36, p b .001), and, similar to willingness
to pay, feelings of competence significantly mediated the effect of
the build condition on their positive affect (Sobel's Z = 2.05, p b .05).
Therefore, feelings of competence not only affected participants' liking of their self-created products, but also their overall happiness. Importantly, the effect of experimental condition on willingness to pay
was not driven by participants' moods. When both variables were entered into the regression equation as potential mediators (Preacher &
Hayes, 2008), competence was a significant mediator (Sobel's Z =
2.01, p b .05), but positive mood was not (Sobel's Z = .39, p = .70),
thus suggesting, as we predicted, that competence plays the critical
role.
3. Experiment 2
The first experiment provides initial support for our theory that
self-created products are valued more because of their influence on
feelings of competence. In Experiment 2, we test our hypotheses by
not only measuring participants' competence (as in the previous experiment) but also by directly manipulating participants' need to feel competent, using a self-affirmation manipulation. Self-affirmation theory
suggests that people use a “fluid compensation” procedure, whereby
affirming one important value reduces the need to affirm another
value (Steele, 1988). If the IKEA effect is driven by the impact of
self-creation on feelings of competence, the effect should be reduced
or eliminated if participants are first allowed to affirm their sense of
self in another way. Unlike existing research, in which consumers
used more hedonic products to reaffirm the self (e.g., Gao et al., 2009;
Townsend & Sood, 2012), in this study we use a mundane product
(IKEA storage boxes) to show that ordinary products can also be used
to increase consumers' feelings of competence and that consumers
can use a wide range of products to affirm their identities.
3.1. Method
Participants (N = 135; 75 male) completed this study to fulfill a
requirement for an introductory undergraduate class. We excluded
eleven participants who did not complete the self-affirmation task
because they failed to rank their values correctly as well as five participants who did not complete all of our dependent measures.
Participants were randomly assigned to one of four conditions of a 2
self-affirmation condition (no affirmation vs. self-affirmation)× 2 build
condition (pre-built vs. build) between-subjects design. Participants in
this experiment were randomized by experimental session to ensure
that they could not see the participants in the other conditions.
Following the procedure used by Sherman, Nelson, and Steele
(2000), participants were first presented with a list of 11 values
that they ranked from most to least important. Next, participants
were asked to write an essay about why a particular value was important to them and to describe a time when that value had been particularly important. Participants in the self-affirmation condition were
asked to write about the value they ranked as being most important,
and participants in the no affirmation condition were asked to write
about their ninth-ranked value.
Following the self-affirmation manipulation, participants were
presented with an IKEA Kassett storage box. Those in the pre-built
condition were presented with a box that was already built and
were asked to examine it. Those in the build condition were given
the necessary parts and IKEA instructions to assemble the box and
were told to assemble it.
Participants were then asked to state the maximum amount that
they would be willing to pay for the box on a scale ranging from $0
to $2 in 10 cent increments (we implemented a BDM procedure in
366
D. Mochon et al. / Intern. J. of Research in Marketing 29 (2012) 363–369
keeping with the previous study). All of the participants were paid $2
for completing this task to ensure that they all had money to spend on
the box. We used a square root transformation to correct for the
skewed distribution of the willingness to pay. Finally, participants
completed the competence measures from Experiment 1 (r = .47,
p b .001).
3.2. Results and discussion
A between-subjects ANOVA showed no significant main effects for
the self-affirmation manipulation (F(1,115)= .44, p = .51) or the build
condition (F(1,115)= 1.06, p = .31). Importantly, we observed an interaction effect between these two variables (F(1,115) = 3.90, p = .05;
Fig. 1). In the no-affirmation condition, we replicated the standard
IKEA effect: builders (M = $0.72, SD= .45) were willing to pay significantly more than non-builders (M =$0.46, SD= .50; t(52) = 1.99, p =
.05). The IKEA effect was eliminated for participants in the
self-affirmation condition, where there was no significant difference between builders (M = $0.49, SD= .46) and non-builders (M = $0.58,
SD= .46; t(63)= .71, p = .48). Therefore, providing participants with
an opportunity for self-affirmation eliminated the IKEA effect.
We next examined how the experimental conditions affected participants' ratings on the competence scale. This analysis showed that the
main effect of self-affirmation was not significant (F(1,115) =1.0, p =
.30), but the main effect of the build condition was significant
(F(1,115) =19.8, p b .001). Consistent with the previous experiment,
the builders reported higher feelings of competence (M = 2.95, SD=
1.50) than the non-builders (M = 1.94, SD= 1.13). This main effect
was qualified by a significant interaction effect (F(1,115) = 5.8,
p b .05). Participants in the no-affirmation condition reported higher
feelings of competence when they built the box (M = 3.39, SD= 1.62)
than when they received it pre-assembled (M= 1.76, SD= 1.01;
t(52) = 4.55, p b .001). However, participants who were first given
an opportunity to self-affirm their values showed no such effect
(Mbuild = 2.57 vs. Mpre-built = 2.09; t(63) = 1.53, p = .13).
We tested a moderated mediation model to examine whether the
indirect effect of competence on willingness to pay depended on
self-affirmation (Preacher, Rucker, & Hayes, 2007). Replicating the
above ANOVA (see Table 2), we found a significant main effect of
build condition on the competence mediator (B = 1.63, p b .001),
which was qualified by a significant interaction between the build and
self-affirmation conditions (B = −1.15, p b .05). Moreover, when the
competence mediator was included in the model predicting willingness
to pay, the effect of the mediator was significant (B =.12, p b .001),
whereas the main effect of build condition (B = .07, p = .59) and its interaction with self-affirmation (B = −.21, p = .22) were no longer significant. Consistent with the previous experiment, these results
suggest that competence mediated the effect of building on willingness
to pay, but that the effect of competence was moderated by the
$1.00
$0.80
$0.60
Pre-Built
Build
$0.40
$0.20
Table 2
Moderated mediation analyses for Experiment 2.
Predictor
B
SE
t
p
Constant
Build condition
Self-Affirmation condition
Build × self-affirmation
Competence
1.76
1.63
.33
−1.15
.23
.35
.31
.48
7.67
4.65
1.08
−2.42
.001
.001
.28
.02
Constant
Competence
Build condition
Self-affirmation condition
Build × self-affirmation
WTP
.26
.12
.07
.08
−.21
.10
.03
.13
.11
.17
2.63
3.58
.54
.70
−1.24
.01
.001
.59
.49
.22
Self-affirmation condition
Indirect effect
SE
Z
p
No affirmation
Self-affirmation
.19
.06
.07
.04
2.80
1.34
.01
.18
self-affirmation manipulation. Indeed, the analyses revealed a significant conditional indirect effect of the competence mediator in the
no-affirmation condition (B =.19, SE= .07; Z = 2.80, p b .01) and a
non-significant conditional indirect effect of the mediator in the
self-affirmation condition (B = .06, SE= .04; Z = 1.34, p = .18).
4. Experiment 3A
The previous experiments demonstrate that the IKEA effect is mediated by feelings of competence associated with assembling products. Experiments 3A and 3B provide further support for our theory.
We examine whether directly manipulating feelings of competence
affects consumers' propensity to engage in self-creation. Since consumers, in part, value self-created products because these items can
bolster feelings of competence, consumers may be more willing to
self-create products when their sense of competence is shaken (Gao
et al., 2009). In both experiments, we give participants the choice to
assemble products or not (see Moreau & Herd, 2010) and examine
whether those participants whose confidence has been threatened
are more likely to choose products that require assembly.
4.1. Method
Participants (N = 75; 42 male) completed an online survey and
were randomly assigned to one of two conditions that manipulated
their sense of competence. Participants in the high-competence condition were presented with four easy math problems (e.g., How likely
is it that a fair coin that is tossed once will come up heads?). Participants
in the low-competence condition were presented with four very difficult math problems (e.g., You have 4 coins. Three of the coins are normal, but one of them is heads on both sides. You pick a coin at random
without looking. The coin you pick has heads on one side. What are the
odds that if you flip the coin over, the other side will be tails?). All of
the questions were presented with four potential answers, and participants were told that they could skip questions if they did not
know the answer. Following the competence manipulation, participants were shown a picture of a bookcase from IKEA and were
asked: “Imagine that you bought the above bookcase from IKEA.
Would you prefer that it came pre-assembled, or would you prefer
to assemble it yourself?”
4.2. Results and discussion
$0.00
No Affirmation
Note: Bars represent standard errors.
Self-Affirmation
Fig. 1. Mean willingness-to-pay in Experiment 2.
We first examined whether our competence manipulation was
successful. Participants in the high-competence condition solved on
average 92% of the questions correctly, whereas participants in the
D. Mochon et al. / Intern. J. of Research in Marketing 29 (2012) 363–369
low-competence condition solved on average only 22% of the questions correctly — no better than chance.
We next examined whether this manipulation affected participants'
propensity to assemble their own products. As predicted, threatening
participants' feelings of competence increased the likelihood that they
chose to assemble their own bookcase (Fig. 2). Only 33% of participants
in the high-competence condition preferred to assemble the bookcase,
whereas 58% of participants in the low-competence condition preferred
to do so (χ2(1) = 4.72, p b .05).
These results further highlight the critical role that competence
plays in the IKEA effect. Feelings of competence not only mediate
the IKEA effect, but threatening consumers' confidence affects their
propensity to engage in self-creation. These results also contribute
to research about self-affirmation in consumer behavior (e.g., Gao et
al., 2009) by demonstrating that consumers can restore their shaken
sense of self even through extremely mundane activities, such as
building a bookcase.
5. Experiment 3B
In a follow-up experiment, we examine whether we can replicate
the effect on choice that we observe in Experiment 3A using a decision that more closely matches one that consumers face in everyday
life.
5.1. Method
Participants (N = 41; 24 male) completed a short online survey that
manipulated their feelings of competence and then were presented
with a consumer choice. Participants were first presented with either
one of the easy or one of the difficult math problem used in the previous
experiment (see the examples above for a description of the exact
problems). Following this manipulation, participants were shown two
tables and were asked which they preferred. One of the tables was a
pre-assembled table from Target. The other table was from IKEA and
required assembly. The image of the table corresponding to the IKEA
versus the Target table was counterbalanced to control for any
preferences for one of the two tables.
5.2. Results and discussion
Ninety-six percent of participants in the high-competence condition correctly answered the math question, whereas 26% of those in
the low-competence condition answered the math question correctly.
More importantly, participants in the low-competence condition
were significantly more likely to choose the IKEA table that required
assembly (74%) than those in the high-competence condition (27%;
χ 2(1) = 8.79, p b .01). These results further support our claim that feelings of competence play a critical role in the value that people derive
from their own creations.
80%
60%
40%
20%
0%
High Competence
Low Competence
Fig. 2. Percent preferring to assemble their own furniture in Experiment 3A.
367
These results also have important practical implications. Despite the
benefits of co-creation for firms (e.g., consumers' willingness to pay
more for products they build, lower costs, etc.) and for consumers
(e.g., increased feelings of competence after successfully assembling
products), consumers are often reluctant to participate in such activities. A different group of consumers were asked (N = 51): “In general,
what would you be willing to pay more for: products that you buy
already assembled, or products that you buy with some assembly required?”. Ninety-two percent of respondents said they would pay
more for preassembled products, (χ2(1) = 36.26, p b .001). The results
from the previous two experiments suggest that consumer participation in co-creation may be increased by appealing to consumers' sense
of competence and their need to signal that competence to others.
Taken together, Experiments 2, 3A and 3B demonstrate that feelings
of competence (the mediator that underlies the IKEA effect) play a critical role in both consumers' propensity to engage in self-creation (with
threats to their competence increasing their desire to build products)
and in the benefits consumers reap from self-creation (with affirmation
attenuating the positive effects of building).
6. General discussion
Prior research has demonstrated that people are willing to pay
more for goods that they create than for identical goods created by
someone else (Franke et al., 2010; Norton et al., 2012). We propose
that this increase in valuation occurs because of the feelings of competence associated with self-created products. Experiment 1 demonstrates how the competence associated with self-created products
mediates consumers' increased willingness to pay for these products
relative to products created by others. Experiments 2, 3A, and 3B manipulate consumers' need to signal competence to further demonstrate the critical role of competence in the IKEA effect. Affirming
consumers' sense of self causes the self-creation of products to be
less rewarding, whereas threatening consumers' feelings of competence leads them to seek out opportunities to build products to restore their sense of competence. Taken together, these experiments
offer evidence regarding the crucial role that competence plays, as
both a mediator and a moderator, in creating consumer interest in
self-created products and in making their efforts feel rewarding.
These results build on prior research that shows that products can
be used to affirm the self (e.g., Gao et al., 2009; Townsend & Sood,
2012) by demonstrating that even the most mundane assembling activities that consumers engage in, such as assembling a storage box,
can have implications for consumers' sense of self.
6.1. Signaling to the self vs. others
Our measure of competence synthesizes two constructs: one
about personal feelings of pride and another about consumers' desire
to show off their creations to others. Although the two constructs are
correlated in each experiment, the relative contribution of these two
types of competence likely varies by context. One factor that likely
determines which of the two constructs is stronger is the type of
product created. The products used in the current paper are less likely
to be displayed to others than products that are, by their nature, explicitly designed to be shown off to others (e.g., clothing). Indeed,
when people design products meant specifically for display, such as
t-shirts and watches, the prospect of showing off these products to
others strongly contributes to their value (Franke et al., 2010). Another important factor is the salience of opportunities to display one's
creation to others. During the current experiments, participants
were not given such an opportunity. However, because the mere
presence of others can increase consumers' concerns about the impressions they make (Puntoni & Tavassoli, 2007), simply making the
social context more salient should increase the impact of the desire
to show off products on the magnitude of the IKEA effect. Finally,
368
D. Mochon et al. / Intern. J. of Research in Marketing 29 (2012) 363–369
the relative strength of these two factors also likely depends on the
intended recipient of the created product (Moreau, Bonney, & Herd,
2011). Most of the self-created content created for new media channels (such as YouTube videos) is made specifically to display to
others. In these cases, it is likely that the utility associated with
one's creations is heavily dependent on people's ability to signal
their competence to others. Future research is needed to examine
the relative contributions of both competence signals to understand
when and how people come to value their own creations.
6.2. Co-creation and customer satisfaction
These results have implications both for firms seeking to maximize
customer satisfaction and for individuals seeking to increase their life
satisfaction. For firms, we note that although our results demonstrate
that involving consumers in creation can lead to higher willingness to
pay, the effect we document here is a retrospective phenomenon
because participants value the product of their labor more highly after
successfully building it. Disrupting consumers' sense of competence increases consumers' desire to engage in co-creation, but absent this
manipulation, they are very unlikely to assemble their own product.
Moreover, involving consumers in co-creation is not without risks.
Although consumers may attribute successful co-creation experiences
to their own efforts, they may attribute co-creation failures to the
firm, which negatively impacts consumers' perceptions of a firm
(Bendapudi & Leone, 2003). Future research should examine the best
way to encourage consumers to co-create.
Results from Experiment 1 demonstrate that self-creation leads to
increases in positive affect among individuals, suggesting that people
may be leaving utility on the table by generally choosing to relax instead of engage in labor. A large body of research highlights the centrality of labor to people's well-being (Blustein, 2008) and shows that
feelings of productivity are important to many people (Hsee, Yang, &
Wang, 2010; Keinan & Kivetz, 2011). For example, unemployment
has lasting psychological consequences; even when people find new
jobs, the adverse impact of past job loss on well-being remains
(Feather, 1990; Lucas, Clark, Georgellis, & Diener, 2004). This perspective dovetails with other research suggesting that effortful activities such as exercising (Mochon, Norton, & Ariely, 2008), making
time to behave pro-socially toward others (Dunn, Aknin, & Norton,
2008), and acquiring life experiences (Van Boven, 2005; Van Boven
& Gilovich, 2003) can lead to lasting changes in people's well-being
(Lyubomirsky, Sheldon, & Schkade, 2005). Consequently, encouraging
people to engage in labor that they would otherwise avoid may lead
to increased life satisfaction. Future research is needed to explore
this hypothesis.
Finally, the overvaluation of products that occurs as a result of the
IKEA effect has broader implications for organizations because overvaluation contributes to two key organizational pitfalls: sunk cost effects (Arkes & Blumer, 1985; Biyalogorsky, Boulding, & Staelin, 2006;
Staw, 1981), which can cause managers to continue to devote resources to failing projects they have previously invested in
(Biyalogorsky et al., 2006), and the “not invented here” syndrome,
in which managers refuse to use perfectly good ideas developed elsewhere in favor of their, sometimes inferior, internally developed
ideas. Our results suggest that managers may persist in pursuing
failed projects and concepts because they truly believe that their
ideas are more valuable; not pursuing their ideas means that money
is left on the table and that using a competitor's ideas would simply
be choosing an inferior option. Not surprisingly, highly innovative
projects are especially likely to generate over-commitment from
managers (Schmidt & Calantone, 1998). While markets may sometimes correct these erroneous overvaluations, the IKEA effect may
be resistant to any intervention, suggesting that the “not invented
here” syndrome may be here to stay.
References
Akerlof, G. A., & Kranton, R. E. (2000). Economics and identity. The Quarterly Journal of
Economics, 115, 715–753.
Ariely, D., & Norton, M. I. (2008). How actions create – not just reveal – preferences.
Trends in Cognitive Sciences, 12(1), 13–16.
Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior
and Human Decision Processes, 35(1), 124–140.
Aronson, J., Cohen, G., & Nail, P. R. (1999). Self-affirmation theory: An update and appraisal. In E. Harmon-Jones, & J. Mills (Eds.), Cognitive dissonance: Progress on a pivotal theory in social psychology (pp. 127–147). Washington, DC: American
Psychological Association.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51(6), 1173–1182.
Becker, G. M., DeGroot, M. H., & Marschak, J. (1964). Measuring utility by a
single-response sequential method. Behavioral Science, IX, 226–232.
Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research, 15,
139–168.
Bem, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology, Vol. 6. (pp. 1–63)New York: Academic Press.
Benabou, R., & Tirole, J. (2011). Identity, morals and taboos: Beliefs as assets. Quarterly
Journal of Economics, 126(2), 805–855.
Bendapudi, N., & Leone, R. P. (2003). Psychological implications of costumer participation in co-production. Journal of Marketing, 67, 14–28.
Berger, J., & Heath, C. (2007). Where consumers diverge from others: Identity signaling
and product domains. Journal of Consumer Research, 34(2), 121–134.
Biyalogorsky, E., Boulding, W., & Staelin, R. (2006). Stuck in the past: Why managers
persist with new product failures. Journal of Marketing, 70, 108–121.
Blustein, D. L. (2008). The role of work in psychological health and well-being: A conceptual, historical, and public policy perspective. American Psychologist, 63(4), 228–240.
Bodner, R., & Prelec, D. (2003). The diagnostic value of actions in a self-signaling model.
In I. Brocas, & J. Carillo (Eds.), The Psychology of Economic Decisions, Vol. 1, Oxford:
Oxford University Press.
Buss, D. M. (2001). Human nature and culture: An evolutionary psychological perspective. Journal of Personality, 69(6), 955–978.
Cacioppo, J. T., Gardner, W. L., & Berntson, G. G. (1999). The affect system has parallel
and integrative processing components: Form follow function. Journal of Personality and Social Psychology, 76(5), 839–855.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York:
Harper and Row.
Dahl, D. W., & Moreau, P. (2007). Thinking inside the box: Why consumers enjoy
constrained creative experiences. Journal of Marketing Research, 44(3), 357–369.
Dellaert, B. G. C., & Stremersch, S. (2005). Marketing mass-customized products: Striking a balance between utility and complexity. Journal of Marketing Research, 42,
219–227.
Dunn, E. W., Aknin, L. B., & Norton, M. I. (2008). Spending money on others promotes
happiness. Science, 319, 1687–1688.
Feather, N. T. (1990). The psychological impact of unemployment. New York:
Springer-Verlag.
Firat, A. F., Dholakia, N., & Venkatesh, A. (1995). Marketing in a postmodern world. European Journal of Marketing, 29(1), 40–56.
Franke, N., Keinz, P., & Steger, C. (2009). Testing the value of customization: When do
customers really prefer products tailored to their preferences? Journal of Marketing, 73(5), 103–121.
Franke, N., & Piller, F. (2004). Value creation by toolkits for user innovation and design:
The case of the watch market. The Journal of Product Innovation Management, 21,
401–415.
Franke, N., & Schreier, M. (2008). Product uniqueness as a driver of customer utility in
mass customization. Marketing Letters, 19, 93–107.
Franke, N., & Schreier, M. (2010). Why customers value self-designed products: The
importance of process effort and enjoyment. Journal of Product Innovation Management, 27, 1020–1031.
Franke, N., Schreier, M., & Kaiser, U. (2010). The “I Designed It Myself” effect in mass
customization. Management Science, 56(1), 125–140.
Furby, L. (1991). Understanding the psychology of possession and ownership: A personal memoir and an appraisal of our progress. Journal of Social Behavior & Personality, 66, 457–463.
Gao, L., Wheeler, S. C., & Shiv, B. (2009). The “shaken self”: Product choices as a means
of restoring self-view confidence. Journal of Consumer Research, 36, 29–38.
Gneezy, A., Imas, A., Nelson, L. D., Brown, A., & Norton, M. I. (2012). Paying to be nice:
Consistency and costly prosocial behavior. Management Science, 58, 179–187.
Hsee, C. K., Yang, A. X., & Wang, L. (2010). Idleness aversion and the need for justified
busyness. Psychological Science, 21, 926–930.
Keinan, A., & Kivetz, R. (2011). Productivity orientation and the consumption of collectable experiences. Journal of Consumer Research, 37(6), 935–950.
Lazarus, R. S. (1991). Progress on a cognitive–motivational–relational theory of emotion. American Psychologist, 46(8), 819–834.
LeBoeuf, R. A., Shafir, E., & Bayuk, J. B. (2010). The conflicting choices of alternating
selves. Organizational Behavior and Human Decision Processes, 111(1), 48–61.
Lucas, R. E., Clark, A. E., Georgellis, Y., & Diener, E. (2004). Unemployment alters the set
point for life satisfaction. Psychological Science, 15(1), 8–13.
Lynn, M., & Harris, J. (1997). The desire for unique consumer products: A new individual
differences scale. Psychology and Marketing, 14(6), 601–616.
D. Mochon et al. / Intern. J. of Research in Marketing 29 (2012) 363–369
Lyubomirsky, S., Sheldon, K. M., & Schkade, D. A. (2005). Pursuing happiness: The architecture of sustainable change. Review of General Psychology, 9(2), 111–131.
Michel, S., Kreuzer, M., Kuhn, R., Stringfellow, A., & Schumann, J. H. (2009). Mass-customized
products: Are they bought for uniqueness or to overcome problems with standard products? Journal of Customer Behavior, 8(4), 307–327.
Mochon, D., Norton, M. I., & Ariely, D. (2008). Getting off the hedonic treadmill, one
step at a time: The impact of regular religious practice and exercise on
well-being. Journal of Economic Psychology, 29(5), 632–642.
Moreau, C. P., Bonney, L., & Herd, K. B. (2011). It's the thought (and the effort) that
counts: How customizing for others differs from customizing for oneself. Journal
of Marketing, 75, 120–133.
Moreau, C. P., & Herd, K. B. (2010). To each his own? How comparisons with others influence consumers' evaluations of their self-designed products. Journal of Consumer Research, 36, 806–819.
Norton, M. I., Mochon, D., & Ariely, D. (2012). The IKEA effect: When labor leads to love.
Journal of Consumer Psychology, 22(3), 453–460.
Prahalad, C. K., & Ramaswamy, V. (2000). Co-opting customer competence. Harvard
Business Review, 78, 79–87.
Prahalad, C. K., & Ramaswamy, V. (2002). The co-creation connection. Strategy and
Business, 27(2), 50–61.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing
and comparing indirect effects in multiple mediator models. Behavior Research
Methods, 40(3), 879–891.
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation
hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research,
42(1), 185–227.
Puntoni, S., & Tavassoli, N. T. (2007). Social context and advertising memory. Journal of
Marketing Research, 44, 284–296.
Randall, T., Terwiesch, C., & Ulrich, K. T. (2007). User design of customized products.
Marketing Science, 26(2), 268–280.
Schmidt, J. B., & Calantone, R. J. (1998). Are really new product development projects
harder to shut down? Journal of Product Innovation Management, 15(2), 111–123.
Schreier, M. (2006). The value increment of mass-customized products: an empirical
assessment. Journal of Consumer Behaviour, 5, 317–327.
Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being:
Informative and directive function of affective states. Journal of Personality and
Social Psychology, 45(3), 513–523.
Sherman, D. K., & Cohen, G. L. (2006). The psychology of self-defense: Self affirmation
theory. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology, Vol. 38.
(pp. 183–242)San Diego: Academic Press.
369
Sherman, D. K., Nelson, L. D., & Steele, C. M. (2000). Do messages about health risks
threaten the self? Increasing the acceptance of threatening health messages via
self-affirmation. Personality and Social Psychology Bulletin, 26(9), 1046–1058.
Simonson, I. (2005). Determinants of customers' responses to customized offers: Conceptual framework and research propositions. Journal of Marketing, 69, 32–45.
Staw, B. M. (1981). The escalation of commitment to a course of action. Academy of
Management Review, 6(4), 577–587.
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of
the self. In L. Berkowitz (Ed.), Advances in experimental social psychology, Vol. 21.
(pp. 261–302)New York: Academic Press.
Thompson, D. V., & Norton, M. I. (2011). The social utility of feature creep. Journal of
Marketing Research, 48, 555–565.
Townsend, C., & Sood, S. (2012). Self-affirmation through the choice of highly aesthetic
products. Journal of Consumer Research, 39.
Tracy, J. L., & Robins, R. W. (2004a). Putting the self Into self-conscious emotions: A theoretical model. Psychological Inquiry, 15(2), 103–125.
Tracy, J. L., & Robins, R. W. (2004b). Show your pride: Evidence for a discrete emotion
expression. Psychological Science, 15(3), 194–197.
Tracy, J. L., & Robins, R. W. (2007). The psychological structure of pride: A tale of two
facets. Journal of Personality and Social Psychology, 92(3), 506–525.
Van Boven, L. (2005). Experientialism, materialism, and the pursuit of happiness.
Review of General Psychology, 9(2), 132–142.
Van Boven, L., & Gilovich, T. (2003). To do or to have? That is the question. Journal of
Personality and Social Psychology, 85(6), 1193–1202.
Vargo, S. L., & Lusch, R. F. (2004). Evolving a new dominant logic for marketing. Journal
of Marketing, 68, 1–17.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and
Social Psychology, 54(6), 1063–1070.
Weiner, B. (1985). An attribution theory of achievement motivation and emotion. Psychological Review, 92, 548–573.
Weiner, B. (1986). An attributional theory of motivation and emotion. New York:
Springer-Verlag.
Weiner, B., Russell, D., & Lerman, D. (1979). The cognition–emotion process in
achievement-related contexts. Journal of Personality and Social Psychology, 37(7),
1211–1220.
Wernerfelt, B. (1990). Advertising content when brand choice is a signal. Journal of
Business, 63(1), 91–98.
White, R. W. (1959). Motivation reconsidered: The concept of competence. Psychological Review, 66(5), 297–333.
Intern. J. of Research in Marketing 29 (2012) 370–379
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
The referral backfire effect: The identity-threatening nature of referral failure
Bart Claus a, b,⁎, Kelly Geyskens c, Kobe Millet d, Siegfried Dewitte a
a
K.U. Leuven, Belgium
Iéseg School of Management, France
c
University of Maastricht, The Netherlands
d
VU University Amsterdam, The Netherlands
b
a r t i c l e
i n f o
Article history:
First received in 2, September 2011 and was
under review for 5 months
Available online 17 September 2012
Keywords:
Word of mouth
Referral failure
Identity threat
Social influence
Self-bolstering
a b s t r a c t
The present paper shows that when a person has the experience of giving advice but that advice is not acted
upon, there is a reduced openness to external information. We call this the “referral backfire effect”. We argue
that this referral backfire effect is due to the identity threatening nature of referral failure: the referral backfire effect is attenuated (1) when the sender perceives oneself as having low expertise in the particular domain of referral and (2) upon self-affirmation. Accordingly, implicit egotism is increased after referral
failure, reflecting the need to bolster the self against threat. Because referral behavior is considered to be
an important predictor of business success, we discuss the implications of our findings for both theory and
practice and sketch future research opportunities.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
“Invite your friends to try Omaha Steaks and we'll throw in a dozen
free burgers. And for every two friends who try us out, we'll send you a
$20 Reward e-Gift Card towards your next purchase of $80 or more.”
This advertising copy shows how Omaha Steaks, a mail order meat
company that is recognized as an innovative marketer, tried to reinforce and incentivize the natural inclination of consumers to refer
others in line with their own preferences. Imagine an enthusiastic
fan of Omaha Steaks participating in this promotion and recommending the service to her friends. Because she will receive a gift
card for every two friends that follow her referral, she is able to
track the extent to which her referral was successful. However,
what happens to the referrer when it becomes clear that almost
none of these friends have followed the recommendation? Alternatively, consider the following example: imagine your new colleague
asking your advice about the best search system for scientific papers.
You recommend your preferred search system. Later on, you notice
that your colleague still works with a less functional system. Again,
what impact does this revelation have on you when your colleague
does not follow your recommendation?
In this paper, we address the question of whether referral failure
has any consequences for subsequent behavior of the referrer. Although consumers refer others on a daily basis and are stimulated
⁎ Corresponding author at: Iéseg School of Management, Socle de la Grande Arche, 1
Parvis de la Défense, F-92044 Paris La Défense Cedex, France. Tel.: +33 155 911 010;
fax: +33 147 759 375.
E-mail address: b.claus@ieseg.fr (B. Claus).
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.06.004
to do so through company rewards (Ryu & Feick, 2007, Schmitt,
Skiera, & Van den Bulte, 2011), we are not aware of any prior research that addresses the effect of referral failure on the referrer.
However, as referral outcomes become increasingly transparent in
online environments, these outcomes represent an issue of growing
importance.
The existing literature indicates that consumption itself can often
be considered a non-verbal form of identity-expression (Belk, Bahn,
& Mayer, 1982; Reed, 2004; Richins, 1994) and that engaging in referrals makes consumption even more publicly visible (Brown, Barry,
Dacin, & Gunst, 2005; Hennig-Thurau, Gwinner, Walsh, & Gremler,
2004). Therefore, we propose that referral failure – the situation in
which one's advice is rejected – may in certain circumstances threaten consumers' identities. Psychology provides ample evidence that
identity threats motivate consumers to bolster their self-concept
(Dunning, Perie, & Story, 1991; Wentura & Greve, 2005). One way
to bolster the self-concept is to act in a self-determined way
(Blanton & Christie, 2003; Deci & Ryan, 1985; Nail, MacDonald, &
Levy, 2000). We argue that referral failure invokes a need to make
self-determined choices, void of external influence. We call this reluctance to comply with external influences – as triggered by referral
failure – the referral backfire effect. Our studies will demonstrate the
effect and test the underlying process in terms of identity threat.
Whereas the proposed effect is relevant to many research domains, such as social psychology, organizational behavior, or communications research, it is useful at this point to highlight this work's
contribution to marketing literature, more specifically, in the Word
of Mouth (WOM) domain. To the best of our knowledge, research
on WOM behavior has focused on the existence and implications of
referrals but has never empirically investigated the outcome of
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
referral behavior on the referrer. Studies on WOM in an offline (East,
Hammond, & Lomax, 2008; Laczniak, DeCarlo, & Ramaswami, 2001)
and online environment (De Bruyn & Lilien, 2008; Hennig-Thurau &
Walsh, 2003) have especially focused on identifying and targeting
consumers with a larger-than-average impact on the spread of information. Spreading consumer information is not reserved for an elite
group of knowledgeable market mavens (Feick & Price, 1987) or influential opinion leaders (Rogers & Cartano, 1962). The ordinary consumer also engages in several conversations about brands and
products every day (Keller, 2007), and her joint impact on spreading
consumer information is, according to several researchers, of no lesser importance than the impacts of the aforementioned elite (Godes &
Mayzlin, 2009; Smith, Coyle, Lightfoot, & Scott, 2007; Watts & Dodds,
2007). The current project opens a new dimension in WOM research
by focusing on the impact of this referral behavior on the sender's
subsequent behavior, rather than on the receiver's.
2. Referral failure as identity threat
Referral behavior has at least two important features that connect
it to consumer identity. Referral behavior reveals information about
the adviser's product preferences and opinions. Because many
consumption decisions reflect a consumer's identity (Escalas &
Bettman, 2005), information about these decisions will also reflect
that identity. Additionally, referral behavior subjects this information
with all of its potential identity implications to public scrutiny because referral behavior offers the listener the opportunity to ignore
or reject the information. Indeed, self-disclosure is perceived to be
risky in general (Olivero & Lunt, 2004). In accordance with this conceptual analysis, the literature suggests that one important driver behind referral behavior is the motivation to maintain and enhance a
positive self-concept, e.g., to demonstrate to others that one is an intelligent shopper or to reduce anxiety stemming from a negative consumer experience (Hennig-Thurau et al., 2004; Sundaram, Mitra, &
Webster, 1998). By conducting in-depth interviews, Dichter (1966)
found that 24% of the 352 investigated (positive) referrals were explicitly driven by self-confirmation motives, including confirmation
of one's judgment capacity or asserting status. Additionally, it is not
unlikely that a substantial number of the referrals explicitly attributed to other reasons (product-, other-, or message-related reasons;
Dichter, 1966), were also partially motivated by self-confirmation.
Furthermore, Von Wangenheim (2005) found that switching brands
might result in a negative WOM review of the initial choice to
self-justify inconsistency demonstrated by the switch. Given that referral behavior is often motivated behavior, referral failure may be
painful, which is a conclusion for which Dichter (1966) found some
anecdotal support. Specifically, the link between consumption and
the self-concept or identity suggests that referral failure may threaten
the referrer's identity.
Identity threat has numerous well-established effects on behavior. For instance, it has been shown that consumers selectively
focus on information that bolsters their self-beliefs when their identity is threatened (Dunning et al., 1991; Wentura & Greve, 2005),
that they also choose products that support their self-concept (Gao,
Wheeler, & Shiv, 2009), and, finally, that they become more motivated to firmly advocate their threatened self-beliefs (Gal & Rucker,
2010). Because referral failure implicitly questions consumers' abilities to make adequate consumer decisions, any such failure will subsequently activate the goal to restore this self-belief. Building on
self-determination theory (Deci & Ryan, 1985), we argue that reducing reliance on external information during decision-making can restore this self-belief. Self-determination theory (Deci & Ryan, 1985)
distinguishes between several motivational states, two of which
are relevant for our argument. First, a behavior is controlled when
others partially control one's intentional behavior (e.g., when someone buys a product because a salesman is pushy or because he or she
371
hopes it will make them blend in with a reference group). In this
state, the actor is subject to external influences. In contrast, a behavior is self-determined when it reflects the self's true preferences and
autonomous decisions (e.g., when someone works hard because they
like the work or when a consumer chooses based on her own needs).
Only in this latter case can decisions be considered unbiased reflections of the self and relevant for one's self view and, therefore, true
choices (Deci & Ryan, 1987). Based on this distinction found in
self-determination theory, we suggest that restoring one's selfconcept can be achieved only when decisions are made autonomously and not when one's decision is (partly) the result of external influence, as is the case in controlled behavior.
We therefore expect that a goal to restore one's self-concept as a
competent decision maker will lead to a decreased likelihood of
complying with external influence. Indeed, existing research mentions that noncompliance is used to defend important self-views
(Nail et al., 2000). People even strategically reduce compliance to
signal their identity to others (Berger & Heath, 2007) or shape it for
themselves (Blanton & Christie, 2003). As a result, we propose that
referral failure motivates consumers to discard external influence
during their decision-making. We predict that
H1. Referral failure reduces the referrer's likelihood to comply with
external influence when making decisions.
The theoretical model we put forward implies that this hypothesized referral backfire effect (i.e., H1) may be explained by the identity threatening nature of referral failure. Prior research showed that
identity threat leads to higher motivation to bolster the self. For example, upon perceiving an identity threat, people show a greater liking for people whose participant number resembles their birthdays
(Jones, Pelham, Carvalho, & Mirenberg, 2004), take up more space
when putting their signature (Rudman, Dohn, & Fairchild, 2007),
and increase their liking for brands and words resembling their
own name (Brendl, Chattopadhyay, Pelham, & Carvallo, 2005;
Jones, Pelham, Mirenberg, & Hetts, 2002). We will measure the
motivation to bolster the self via an implicit egotism measure
(Buhrmester, Blanton, & Swann, 2011; Jones et al., 2002). If referral
failure is threatening to the self, it should invoke the motivation to
bolster the self, which leads to the following hypothesis:
H2. Referral failure increases implicit egotism.
To further investigate the proposed process driving the referral
backfire effect, we turn to moderation designs. We will select two
distinct factors that are believed to have an attenuating impact on
identity threat: first, we focus on consumer knowledge concerning
the specific product domain of the referral, and further below, we
discuss the moderating role of general self-affirmation. Existing literature indicates that the level of knowledge about a topic is related
to the centrality of that topic to the self (Belk, 1988; Sprott, Czellar, &
Spangenberg, 2009). When applied to referrals, this relationship
means that the centrality of the referral outcome to the referrer's
identity correlates with the referrer's level of knowledge in this domain. Furthermore, the personal importance of a specific domain
has been shown to affect the level of self-threat (Boninger,
Krosnick, & Berent, 1995; Sherman & Cohen, 2006). For instance, a
report describing the link between caffeine use and fibrocystic disease triggers defensive mechanisms only for heavy caffeine consumers for whom the topic is of greater importance (Liberman &
Chaiken, 1992). This result indicates that referral failure can occur
without identity threat, as long as the referral was made in a domain
that is not central to one's identity. We use knowledge of the domain
of referral as a proxy for this centrality. Accordingly, we expect that
H3. The referral backfire effect is stronger for consumers who perceive themselves as more knowledgeable about the product domain
than for consumers who do not.
372
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
Additionally, self-affirmation theory proposes that people can
respond to threats by affirming alternative self-resources unrelated
to the identity threat. This “self-affirmation” can be achieved when
people reflect on important aspects of their life or engage in an activity that highlights important values, which causes participants to realize that their own self-worth does not depend solely on the aspect
of their identity that is threatened and thereby reduces the need for
self-bolstering (Sherman & Cohen, 2006).
Accordingly, manipulations of self-affirmation (Steele, 1988) have
been used successfully to reduce various effects of potential identity
threats in different domains (Fein & Spencer, 1997; Sherman,
Nelson, & Steele, 2000; Steele & Liu, 1983). If referral failure leads to
avoiding external input because it increases the need to bolster the
self-concept, external support for the self should satisfy this need
and thereby attenuate the referral backfire effect. Therefore, we propose that
H4. Self-affirmation will attenuate the referral backfire effect.
3. The present studies
The present studies induce referral failure by means of scenario
studies or real decision contexts and measure their effects on subsequent decision-making. In particular, the studies examine the extent
to which consumers incorporate external advice in their decisions. In
study 1, we demonstrate the referral backfire effect, showing that the
awareness that others have not followed the participant's advice reduces his or her willingness to incorporate unrelated consumer reviews in decision making. In studies 2–4, we replicate this effect
via different procedures and further elaborate the role played by
identity threat in the referral backfire effect. In study 2, we show
that referral failure is self-threatening. In study 3, evidence is provided that this effect of referral failure is attenuated for consumers
with low levels of knowledge in the domain of referral. Study 4
shows that the referral backfire effect is also attenuated when the
need for self-bolstering is alleviated by an external manipulation of
self-affirmation subsequent to referral failure. Thus, studies 3 and 4
illustrate important boundary conditions of the effect: the referral backfire effect only occurs when referral failure is identity-threatening. Furthermore, study 3 shows that the referral backfire effect is independent
of the match between domain of the referral and domain of the subsequent product decision.
3.1. Study 1
We had two aims with this study. First, we wanted to show a relation between referral outcome and subsequent compliance with
external influence (H1). After manipulating the referral outcome in
a first scenario, we measured susceptibility to persuasion by a third
party in a second, seemingly unrelated task. Second, we expected
that referral failure influences subsequent decision making, whereas
referral success does not because consumers nourish positively
biased self-concepts: positive feedback regarding one's self-concept
is smoothly incorporated and does not receive considerable attention
(Dawson, Gilovich, & Regan, 2002; Dunning, 2007; Jain & Maheswaran,
2000). Therefore, we added a control condition in which no referral
was made. We did not expect referral success to influence any dependent measure.
3.1.1. Method
One hundred fifty students at a large European university (50
males and 100 females), aged between 18 and 30, participated in
this experiment in exchange for a participation fee.
3.1.1.1. Manipulation of referral outcome. In a first phase, we manipulated referral outcome by means of a scenario. The participants
were asked to imagine the following:
“You are the son/daughter of the manager of a small movie theater in
your town. This gives you the chance to watch all movies for free and
before all the others. Your friends are aware of this and they frequently ask you which movies are worthwhile to see. After a while, you notice that your advice is often or always [seldom or never] followed;
they prefer the movies you liked [you did not like].”
In the control condition, the participants received a neutral filler
task of similar length.
3.1.1.2. Measurement of compliance with external influences. We
adapted a procedure used in Fitzsimons and Lehmann (2004), telling
participants that researchers of a food manufacturer were developing
a new granola bar, and they were interested in consumer impressions.
Subsequently, the participants received the descriptions of four potential granola bar formulations with two different attributes: taste and
calories. Attribute values for each of the four formulations for taste
(1: poor taste, 10: excellent taste) and number of calories were, respectively: A: 7.5, 125; B: 8, 365; C: 9, 220; D: 6, 150. Formulations A and
C are relatively attractive, while formulations B and D are relatively
less attractive (Fitzsimons & Lehmann, 2004). Next, the participants
received reports from an expert magazine (e-health) that strongly
recommended either granola bar A or C (counterbalanced between
participants). After receiving this advice, the participants were asked
to indicate which of 20 different combinations of three granola bars
they would prefer when given the chance to choose three bars
(e.g., “A:1, B:1, C:1, D:0”; “A:2, B:1, C:0, D:0”; “A:0, B:0, C:3, D:0”, …
Fitzsimons & Lehmann, 2004). The participants could either incorporate
this external advice and choose the recommended option or ignore the
external advice by selecting bars that were not recommended.
3.1.2. Results and discussion
To test whether referral outcome (success, failure, control) had an
influence on a participant's compliance with external influence on his
or her choice, we performed an ANOVA with choice of recommended
granola bars (0–3, log-transformed) as the dependent variable and
the Referral Outcome manipulation as the independent variable. The
main effect of the Referral Outcome manipulation was significant
(F(2, 148) = 4.79, p b .01). The choice of recommended granola bars
was lower in the referral failure condition (M = .28, SE = .025) than
in either the referral success (M = .36, SE = .025) (t(148) = 2.14,
p b .05) or the control conditions (M = .38, SE = .019) (t(148) = 3.10,
p b .005). The difference between the latter two conditions was insignificant (t(148) = .67, ns.).
These results imply that experiencing referral failure leads to
lower compliance with external influences in a subsequent choice
compared to referral success or no referral experience at all. Additionally, the referral success condition did not differ from the control condition without a referral.
3.2. Study 2
In the second study, we wanted to replicate our findings with actual behavior. Moreover, we wanted to test our inference that referral
failure is threatening to the self. When referral failure occurred, we
expected that a need to bolster the self against this threat would
arise. Consequently and in accordance with existing literature about
self-threat, we predicted an increased attachment to anything that
is self-relevant after referral failure. Therefore, the aim of this second
study is twofold: to test whether referral failure leads to increased
implicit egotism (H2) and to replicate the referral backfire effect
with a different procedure as in study 1.
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
3.2.1. Method
One hundred sixty seven students at a large European university
(84 males and 83 females), aged between 18 and 25, participated in
this experiment in exchange for course credit.
The participants were told that the experiment was a test case for a
larger project that investigated the viability of a communication network
between two consumer labs via the Internet. The participants were further told that the research concerned the fluency of e-communication
when the participants saw each other's picture. To further this cover
story, the participants took a picture using a web camera. In reality,
there was no interaction partner, and the alleged interaction partner's
behavior was preprogrammed. For the remainder of the description of
this study, we use the word “participant” to refer to the actual person
participating in the experiment, and the words “interaction partner”
for the fictitious, preprogrammed person in the other lab.
The actual experiment consisted of two phases. In each phase, the
participants were connected to a different interaction partner. The
gender of the interaction partner was counterbalanced between the
phases and participant's gender to neutralize order and gender effects. In each round, the participants were asked to choose six product
items (a cell phone, backpack, toothbrush, laptop, ballpoint pen and
yogurt), which had to be selected from sets of three options. All 18
products were displayed as a picture. The participants were told
that to structure the decision, they would be randomly assigned to either an adviser role – in which they could indicate their preference
first – or a decider role – in which they had to make the final choice
(for themselves, not for the interaction partner) after the adviser indicated his preference. The participants were further informed that
they would be teamed up with a different interaction partner and
switch roles in the second round. In reality, all participants were
assigned the role of adviser in the first round, during which a manipulation of referral success took place. In the second round, the participants were all assigned to the role of decider, which allowed us to
assess the referral backfire effect. In between rounds, we gave every
participant the impression that the new interaction partner was not
ready yet, and we asked them to complete some filler questions and
a poll about which name to choose for the new computer system,
which actually included the implicit egotism measure.
3.2.1.1. Manipulation of referral outcome (phase 1). In the control condition, the interaction partner followed the participant's choice in 5
out of 6 times. In the Referral Failure condition, the interaction partner followed the participant's choice in only 1 out of 6 times. The gender of the interaction partner was counterbalanced across conditions
and participant's gender to neutralize gender effects.
3.2.1.2. Implicit egotism measure. Between phases, we told the participants that we still needed to decide on a name for the new software
system. The participants were allowed to indicate by means of a
20-point slider between two possible names how much they preferred one option to another. Importantly, one of the two options
was manipulated such that the first three letters of the name for the
program were the same as those of the first name of the participant.
The suffix of those three letters (“-ano”) and the comparison name
(“Valdamo”) were kept constant. Threats to the self should cause a
higher preference for the name with the same first letters as the
participant's own first name (Jones et al., 2002). Measuring this preference thus allows us to test whether referral outcome affects implicit
egotism.
3.2.1.3. The referral backfire effect. In the second phase, the interaction
partner always gave the advice to choose the least attractive product
out of the three (determined from pre-testing with 46 participants).
We measured referral backfire by the number of times (0–6) the participant did not follow the interaction partner's advice.
373
3.2.2. Results and discussion
Two effects were found for participants whose advice was not
followed. First, we replicated the referral backfire effect found in
study 1. Upon referral failure, the participants were less likely to follow the advice of the interaction partner (M = 2.35, SE = .093) than
in the control condition (M = 2.68, SE = .11) (t(165) = 2.28, p b .05).
Second, the participants in the referral failure condition showed
more implicit egotism: they preferred the name based on their own
name to the control name (M = 10.84, SE = .72) at a higher rate than
participants whose advice had previously been followed (M = 8.32,
SE = .70) (t(165) = 2.51, p b .05). Additionally, we found these two
dependent measures to be significantly correlated (r = − .18, p b .05),
further cross-validating the referral backfire effect as an instance
of self-bolstering. We found no effects stemming from participant's
gender, interaction-partner gender, or any interaction effect between
these factors and our manipulation in an ANOVA (all Fs b 1.5).
The findings of this study provide support for H2 that referral failure leads to an increase in implicit egotism, thereby providing empirical evidence for the idea that referral failure is threatening to the self.
Moreover, we replicated the findings of study 1 by showing that referral failure leads to more self-determined behavior. In combination,
these results support our inference that the referral backfire effect
should be understood as an instance of self-bolstering in reaction to
the identity threatening nature of referral failure. In study 3, we aim
to provide further evidence for the role of identity threat in the referral backfire effect by testing whether the effect is attenuated for consumers with low knowledge levels in the domain of referral.
3.3. Study 3
In this study, we wanted to replicate and combine our previous results and aimed for two additional contributions. First, we wanted to
provide process evidence by showing that the referral backfire effect
is suppressed when referral failure does not provide an identity
threat. In agreement with existing literature that relates knowledge
levels in a domain to centrality in consumers' self-concepts (Sprott
et al., 2009), we used self-perceived knowledge levels in the domain
of referral (e.g., movies) as a proxy of the threatening nature of referral failure. We expected that if the sender were unknowledgeable
about the domain of the referral, then referral failure would threaten
the sender's identity less and therefore reduce the referral backfire effect compared to when the sender is knowledgeable in the domain of
referral. This study effectively provides an important boundary condition for the referral backfire effect.
The second aim of this study is to show that the referral backfire
effect is not dependent on the match between the domain of referral
and the domain of the subsequent product decision, which agrees
with the self-affirmation literature that predicts non-threat specific
effects of identity threat. In the previous two studies, we focused on
the effect of referral failure on product decisions outside of the domain (across domain compensation). However, because there is no
theoretical reason to expect moderation by domain match, this third
study includes data from both within and across domain DVs.
3.3.1. Method
One hundred thirty-seven students at a large European university
(57 males and 80 females), aged between 18 and 29, participated in
this experiment in exchange for a participation fee. For this study, restaurants and movies were used as both the domains of referral and
domains within which an external influence was provided. A scenario
similar to the one used in study 1 manipulated referral outcome in an
initial domain (either movies or restaurants), after which popularity
information for items within and outside of this initial domain was
given (for both movies and restaurants). This popularity information
was based on the preferences of a large group of consumers and can
be considered a subtle type of social influence (Nail et al., 2000).
374
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
Therefore, we obtain two measures of compliance with external
influence — one within a domain (e.g., preferences for the most popular movie after recommending movies), and one across a different
domain (e.g., preferences for the most popular restaurant after
recommending movies).
3.3.1.1. Manipulation of referral outcome. This study used a scenario
similar to the one employed in study 1. Importantly, in addition to
manipulating the referral outcome between participants, we also
manipulated the domain of referral to be either restaurants or
movies. This factor was manipulated between participants and
counterbalanced across referral outcome conditions. First, we asked
participants to think of an acquaintance and to imagine this person
asking them – occasionally – for advice about restaurants [movies].
We then presented participants with a list of six restaurants [movies]
and asked them to recommend one. Next, we presented the remainder of this scenario, containing the referral failure manipulation (between participants):
“One week after recommending this restaurant, you coincidentally
meet your acquaintance at the doorstep of the restaurant you
recommended [a different restaurant than the one you recommended]. From the small conversation that follows, it turns out that
this is not a coincidence: your acquaintance had indeed chosen to
follow up [not follow up] on your referral.”
We asked the participants to imagine this in a vivid way, after
which they answered several filler questions meant to provide a rationale for this scenario. A similar text was used in the movie condition.
3.3.1.2. Measurement of referral backfire. In the measurement phase,
the participants were provided with lists of both restaurants and
movies used in the recommendation phase and were asked to rank
order them according to personal preference. Crucially, prior to ranking, they were given popularity information as a source of external influence. The participants were told that one restaurant and one movie
had been chosen as best by panels of visitors on relevant review
websites. This approach allowed us to test their compliance with
this external influence by looking at the extent to which these most
popular items were ranked highly. Because the participants had
made a recommendation in one of these two domains before, this
resulted in two DVs, which allowed us to test the referral backfire effect both within and across domains (e.g., a participant who
recommended a movie in phase one and ranks both movies and restaurants in phase two provides a within domain measure for movies
and an across domain measure for restaurants). Measurement of
these two DVs was counterbalanced for order between conditions.
Importantly, because of the rank ordering task, a higher number indicates a weaker attitude towards an item, implying lower compliance
with the external influence. Therefore, a higher number implies a
stronger backfire effect.
3.3.1.3. Knowledgeability. Additionally, we asked the participants to
indicate to what extent they knew about the items of each set they
had been asked to rank order in the measurement phase. A single
item scale, which we will further refer to as “knowledgeability,” between “not at all” (−5) and “very well” (5) was used to measure this
(Bergkvist & Rossiter, 2007; Mobley, Bearden, & Teel, 1988). For analysis
of knowledgeability, it is important to understand that this measure is
matched with the participants' domain of referral: participants that
had recommended a restaurant [movie] in the referral phase, rated
their own knowledgeability about the restaurants [movies] presented.
3.3.2. Results and discussion
We investigated whether the main effect of referral failure on
subsequent compliance with external influence depended on the
knowledgeability of the domain of referral (as a potential moderator
of the threatening nature of referral failure) or match with the
domain of referral failure (i.e., within or across domain evaluation
of items). We conducted a linear mixed model with referral outcome
as a between participants categorical independent variable, domain
knowledgeability as a continuous independent measure, and domain
match (i.e., the rank of the externally recommended item within
domain and rank of the externally recommended item across
domain) as within-participants dependent measures. 1 We focus on
three aspects of the results: the main effect of referral outcome
(F(1,133) = 12,02, p b .005), which replicates the referral backfire effect; the two-way interaction between knowledgeability and referral
outcome (F(1,133) = 5.08, p b .05), which tests the attenuating role
of low knowledgeability on the referral backfire effect; and the
absence of the three-way interaction, (F(1,133) = 0.38, ns.), which
shows that the referral backfire effect is insensitive to domain match.
To elaborate on the main effect of referral outcome on rank across
and within domain, we conducted contrast tests. First, we found a significant effect of referral outcome on ranking across domain: after
referral failure, the participants ranked the recommended item
lower (M = 3.39, SE = .19) than after referral success (M = 2.84,
SE = .19; t(136) = − 2.07, p b .05), which replicates the referral backfire effect found in studies 1 and 2. Second, a similar significant effect
occurred within a domain: after referral failure, the participants
ranked the recommended item lower (M = 2.99, SE = .18) than after
referral success (M = 2.37, SE = .18; t(136) = − 2.45, p b .05).
To examine in more detail the two-way interaction between
knowledgeability and referral outcome (see Fig. 1), we conducted
spotlight analysis (Fitzsimons, 2008; Irwin & McClelland, 2001).
At one standard deviation below the mean of knowledgeability,
no significant difference appeared in the ranking of the externally
recommended item between participants in the referral success
condition and those in the referral failure condition (M = 2.54 and
M = 2.74, respectively; β = .10, SE = .12, t = .84, ns). As hypothesized,
low knowledgeability regarding the domain of referral attenuates the
referral backfire effect. Interestingly, however, the referral backfire
effect occurred at the mean level of knowledgeability (Msuccess = 2.61,
Mfailure = 3.19, β = .29, SE= .084, t = 3.47, p b .001) and at a high
level (one standard deviation above the mean) of knowledgeability
(Msuccess = 2.68, Mfailure = 3.64, β = .48, SE= .12, t = 4.05, p b .001).
For sake of completeness, the mixed model also showed a positive
(r = .248) main effect of knowledgeability (F(1,133) = 9.69, p b .005 ),
i.e., knowledgeability generally leads to lower ranking of the recommended item. Additionally, and unrelated to the full model analyzed
above, we found that the participants were less knowledgeable
about local restaurants (M = − 1.88, SE = .25) than about movies
(M = 1.48, SE = .23; t(136) = − 10.50, p b .001).
In previous studies, we took measures to avoid potential alternative explanations that explained the referral backfire effect in ways
other than the identity threatening nature of referral failure. These
measures largely consisted of separating the phase where referral
failure takes place as much as possible from the phase where external
information is incorporated into a decision or not. Failing to do this
could lead to a game of “tit for tat”, where participants ignore advice
because theirs was also ignored. This failure could also lead to an alternative explanation in terms of experimental demand. On a more conceptual level, one could also argue that experiencing referral failure
induces a social norm of independence. All of these arguments make
1
Ranks are not on an interval scale. We conducted a generalized mixed model with
ordinal probit dependent variables, which led to the same conclusions: the main effect
of referral outcome was significant (Wald χ2 (1) = 9.00, pb .005), as was the two-way
interaction between knowledgeability and referral outcome (Wald χ2 (1) = 4.91,
p b .05). Again, a three-way interaction was absent (Wald χ2 (1) = .24, ns). We preferred parametric analysis because of its greater flexibility in testing contrasts and
spotlight analyses and because the violations are typically inconsequential (Velleman
& Leland, 1993).
375
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
4
Rank or Recommended Item
3,5
3
2,5
Referral
Success
2
1,5
Referral
Failure
1
0,5
0
Low Domain
Knowledge
Mean
High Domain
Knowledge
Fig. 1. Effect of referral outcome x knowledgeability on rank of recommended item.
predictions similar to the referral backfire effect. However, these arguments do not account for the moderation of this effect by knowledgeability. Thus, this study not only provides further support for our
hypotheses, but also rules out these potential alternative explanations.
Furthermore, this study provides several preliminary insights into
the nature of threat from referral failure. Because the effect is not
observed under low knowledgeability, it is clear that referral failure
does not call into question the relationship between the sender and
receiver nor the sender's power to influence others. These two explanations would not be eliminated under low knowledgeability. Referral
failure threatens self-image at the level of competence in decisionmaking in the domain of referral.
3.4. Study 4
In study 4, we aimed to provide further evidence for self-bolstering
as the underlying mechanism for the referral backfire effect. This study
again uses a moderation design. The current study aims to eliminate
threat to self-identity after referral failure and before the subsequent
decision. This study design is in contrast to study 3, which prevented
the threat altogether. If the referral backfire effect relies on a need
to resolve identity-threat, the effect of referral failure should be attenuated when an alternative way to bolster the ego is provided. We
asked the participants to evaluate the helpfulness of reviews for a
product in an online store. Based on existing literature (Deutsch &
Gerard, 1955; Nail et al., 2000), we use these reviews as sources of influence. This approach is in line with the fact that consumers rely
heavily on the opinion of others when making consumer decisions
(Chevalier & Mayzlin, 2006). In accordance with previous results, we
expected these evaluations of helpfulness to decrease after referral
failure, as the participants would be more inclined to rely on their
own preferences to make choices and discard external information.
However, in accordance with H4, we expected this effect to disappear
when self-threat comprised by referral failure had been compensated
for by a self-bolstering, positive experience.
3.4.1. Method
Ninety-three participants (62 males and 31 females), aged between 19 and 28 years, participated in this experiment in exchange
for a participation fee. The participants came to the lab for a
one-hour session that included several studies, of which this study
was first. Four participants reported extreme evaluations of the reviews and were removed as outliers based on the 3SD criterion (3%).
3.4.1.1. Manipulation of referral outcome. The participants were asked
to choose their favorite out of six movie posters. In the control condition, the participants were told that the study concerned what
factors people take into account when making this sort of choice
and were asked to write a short justification for why they chose
that particular poster. Participants in the referral failure condition
were asked to write the same justification but were also told that
their arguments would be shown to another participant (called
the reviewing participant, which was preprogrammed). The actual
participants were told that the reviewing participant would indicate the degree to which he or she considered switching their favorite poster to the favorite poster of the participant after reading
the justification. For this condition, no information about the identity of the reviewing participant was revealed. After writing and
submitting their justification, participants in both conditions saw
the favorite poster and the argument of a third, different preprogrammed participant, who justified his choice for a randomly
picked poster by means of generic arguments. For example, one
such argument was that the poster represented the movie accurately. At that point, the participants were asked to indicate how
likely they were to change their opinion on a 7-point scale (1: “certainly not” to 7: “certainly”). Because participants in the referral
failure condition had not received feedback on their own preferences yet, this likelihood did not serve as a dependent variable
but was rather a part of the procedure aimed at making the advice
phase believable. We found no correlation between this switching
likelihood, and subsequent measures (all Fs b 1). Afterwards, participants in the referral failure condition always received a response
from the (preprogrammed) reviewing participant that he would
be “very unlikely” (2 out of 7) to switch to the participant's choice
after having read the participant's argumentation. Participants in
the control condition did not receive any such feedback. Afterwards,
all participants were told that this task was completed and were
asked to start the next task.
3.4.1.2. Self-affirmation manipulation. After the first phase of the experiment, the participants were asked to perform a writing task that
seemed to be completely unrelated but included the self-affirming
manipulation. We adopted a procedure frequently used in previous
research (e.g., Fein & Spencer, 1997; Sherman et al., 2000). The participants were asked to rank a list of 11 values and personal characteristics (e.g., physical attractiveness, being a good friend, financial
success) in order of personal importance. In the self-affirmation condition, the participants wrote a brief essay explaining why their
376
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
top-ranked value was important to them and described a time in their
lives when this value had been particularly useful. In the no-affirmation
condition, the participants wrote about why and when the value they
had ranked seventh in importance could be of importance to the average college student.
3.4.1.3. Measurement of referral backfire. In this study, the participants
rated the helpfulness of online evaluations in a way similar to what
actually takes place in such online stores as Amazon.com. The participants indicated the helpfulness of negative and positive comments
made about a product that – at the time of this study – was little
known among our participants (i.e., the Amazon kindle) by awarding
1 to 5 stars for each specific comment. Consumer reviews should be
considered less helpful when participants are less willing to comply
with external influence.
3.4.2. Results and discussion
A 2 (referral outcome) by 2 (self-affirmation) dimensional design
showed that referral outcome had a significant main effect on the perceived helpfulness of consumer reviews: referral failure (M = 3.45,
SE = .25) led to a lower perceived helpfulness than the control condition (M = 4.14, SE = .24; F(1,85) = 4.03, p b .05). Importantly, this
effect was qualified by a significant interaction effect between referral outcome and the self-affirmation manipulation (F(1,85) = 6.78,
p b .05). As depicted in Fig. 2, only participants in the no-affirmation
condition rated the consumer reviews as less helpful when having
encountered referral failure (M = 2.77, SE= .34) compared to the
control condition (M = 4.35, SE= .39; t(85)= 3.07, p b .005). No referral
backfire effect was observed in the self-affirmation condition (M = 4.13,
SE= .33 in the referral failure condition vs. M = 3.93, SE=.31 in the
control condition; t(85)= −.45, ns.). Additionally, participants in the
referral failure condition rated the consumer reviews as more
helpful after self-affirmation (M = 4.13, SE= .33) compared to the noaffirmation condition (M = 2.77, SE= .34; t(85)= 2.85, p b .01). In the
control condition, self-affirmation had no effect on perceived helpfulness when comparing the neutral task condition (M = 4.35,
SE = .39) with the self-affirmation condition (M = 3.93, SE = .31;
t(85) = .86, ns.).
Study 4 conceptually replicated the referral backfire effect with a
different procedure, thereby providing additional evidence for the
process causing the effect. Moreover, study 4 supported H4; that
self-affirming information attenuates the effect of referral failure on
the evaluations of consumer reviews.
4. General discussion
Addressing the effect of referral outcome on the person making
the referral is of growing importance because consumers engage in
referral behavior on a daily basis and are even stimulated by companies to do so (Ryu & Feick, 2007, Schmitt et al., 2011). Making a referral is often driven by motivation to maintain and enhance a positive
self-concept (e.g., Hennig-Thurau et al., 2004). Moreover, engaging
in a referral makes consumer's choices, and therefore her identity, notably visible. This publicity is even more relevant in the online environment. Referral failure can thus bring into question not only
consumer's choices but also the view of oneself as a capable choice
maker. We reasoned that consumers are motivated to re-establish
this self-image by making choices that are a direct reflection of
one's capability as a decision maker. Based on self-determination theory, this motivation implies a relative shift from controlled to
self-determined behavior (Deci & Ryan, 1985), which we predicted
would make consumers less likely to comply with external sources
of influence. The fact that reduced compliance can be used to bolster
the self is in line with existing literature (Blanton & Christie, 2003;
Nail et al., 2000). Four studies showed that referral failure indeed
reduces the referrer's likelihood to comply with external influence
in subsequent choices. Furthermore, based on the idea that referral
failure is identity-threatening, we provided process evidence and
boundary conditions for this referral backfire effect. Implicit egotism
(reflecting a need to bolster the self) is increased after referral failure
(study 2). Furthermore, the referral backfire effect is attenuated when
the referral topic is not central to the identity of the referrer (study 3)
and is eliminated when the threat is compensated by self-affirmation
(study 4). In addition, the referral backfire effect is independent of the
match between referral domain and the domain of the subsequent
product decision (study 3). Throughout our studies, we rule out
several potential alternative explanations for the referral backfire
effect. Moreover, the results of study 3 allow us to make inferences
concerning the nature of the threat of referral failure. Because our
findings show that the referral backfire effect is attenuated in participants with low self-perceived knowledgeability in the domain of
referral, we can conclude that what is under threat is not the relationship between the sender and receiver of information, nor the sender's
power to influence others. Rather, referral failure threatens the idea
that consumers have of themselves as competent decision makers
in the domain of referral. In combination, these studies suggest that
referral failure poses a threat to consumers' identity, which leads
5
4,35
4,5
Helpfulness of Reviews
4
3,93
4,13
3,5
3
2,77
Self-Affirmation
2,5
Neutral
2
1,5
1
0,5
0
Control
Referral Failure
Fig. 2. Effect of referral outcome x self-affirmation on perceived helpfulness of product reviews.
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
them to discard external influence in an attempt to restore their
self-images as competent in making consumer choices.
4.1. Theoretical and managerial implications
We believe that our findings extend previous research in several
important ways. First, to the best of our knowledge, we are the first
researchers to acknowledge that the outcome of giving advice to
others might be consequential to the sender of advice, which is a
finding that has implications in many research domains other than
marketing. Although several authors have identified the motivations
involved in giving advice (Dichter, 1966; Hennig-Thurau et al., 2004;
Sundaram et al., 1998), no research has investigated what happens
when the needs behind those motivations are not met. Whereas
dominant models of information diffusion often model referral behavior (e.g., literature using the Bass diffusion model; Bass, 1969),
they do not seem to assume feedback loops involving the referrers
based on referral outcomes. Our results demonstrate the importance
of considering the potential effects of the referral outcome on the referrer. These effects are particularly important in models of information diffusion or communication models in general. Second, we
provide evidence that the outcome of a referral can be identitythreatening. Accordingly, we can expect many other behavioral effects because the existing literature indicates that self-affirmation
in general can be used to cope with identity threats (Steele, 1988).
For instance, it has been shown that identity threat leads to “compensatory” consumption of status goods (Sivanathan & Pettit,
2010). Therefore, senders who incur a referral failure may be subsequently more interested in offers that promise to bolster the self,
such as compensatory status consumption or consumption in other
unrelated domains, as long as this consumption helps protect against
or compensate for the identity threat. Third, while previous consumer literature has identified reactions to identity threat in terms of
what consumer decisions are made (Gao et al., 2009), we show that
threats can also lead to reactions in how decisions are made. Consumers comply less with external influences on their decision processes after experiencing a referral failure.
Fourth, we arrived at our hypothesis about the nature of referral
backfire based on a distinction made in the self-determination literature between controlled and self-determined behavior (Deci &
Ryan, 1985). Whereas this literature explicitly states that selfdetermined choices do not preclude complying with external influence, as long as one fully identifies with this influence (Chirkov,
Ryan, Kim, & Kaplan, 2003; Ryan, 1993), our results highlight a potential limitation to the independence of self-determination and
compliance. If choice serves to establish competence in choice making and thereby restore a self-image, then external influences prohibit choices to be a valid proof of this competence. However, one
could argue that behavior aimed at restoring a self-image after a
threat is caused or controlled by this threat, and therefore such behavior is not actually self-determined behavior. Future research in
this domain should determine whether and when self-bolstering is
self-determined or controlled behavior.
We also see important managerial implications of our findings.
First, the potential effects of referral failure on subsequent behavior
calls into question the much-heralded efficacy of WOM (Chevalier
& Mayzlin, 2006; Reichheld, 2003). Considering that consumers
who make referrals are usually the company's most satisfied customers (Anderson, 1998), our results reveal the undesirable possibility that an ill-considered campaign that stimulates WOM might
induce unintended behavioral changes in exactly those most valuable customers. It is difficult to predict a priori whether consumers'
discarding of external influence might make them abandon the
product because of referral failure or cling to the product even
more vigorously. Theoretically, both reactions are possible, and further research could find circumstances under which either condition
377
prevails. However, customer development efforts, such as up-selling
or cross-selling, are likely to be compromised in customers less willing to comply with external influences due to prior referral failure.
Such attempts at persuasion might be particularly ineffective if
they are not somehow framed as a support for an initial choice
made by the customer. In other words, a good WOM campaign is
more complicated than merely increasing the Net Promoter Score
(Reichheld, 2003). To arrive at a more complete view of behavioral
changes associated with stimulating word of mouth, a good WOM
campaign should also consider the likelihood of referral failure and
estimate its impact. We may therefore additionally recommend
monitoring not only referral success in referral reward programs
(Ryu & Feick, 2007, Schmitt et al., 2011) but also referral failure.
This approach is potentially even more important because referral
failure might – as previously mentioned – affect the company's
best customers. Interestingly, our findings not only reveal potential
risks in the current practice of stimulating consumer referral, but
also offer methods of addressing consumers facing referral failure.
More specifically, self-affirmation in an unrelated domain can compensate for identity threat caused by referral failure (study 4). Indeed, highlighting a customer's value to the company, their
contribution or significance in general (e.g., by offering them rewards for loyalty) or affirming his or her lifestyle aspects relevant
to the company's offer (e.g., a perfume brand promoting a luxurious
lifestyle) might be exactly the type of self-affirmation such a customer needs.
A final practical observation is that stimulating referrals is often
mentioned as the solution to the declining effectiveness of traditional
advertising methods (Kumar, 2010; Van den Bulte & Wuyts, 2009).
Ironically, referral failures might contribute to that very decline because consumers facing referral failure might be impelled to make
their own decisions and discard external information in the form of
marketing messages.
4.2. Suggestions for future research
Our studies constitute a first step in answering the question of
how referral outcome affects consumer behavior. Nevertheless, our
research also evokes several questions for future enquiry. Most of
those questions are based on our main contribution: referral outcome
may be consequential for the referrer. However, it is essential that additional boundary conditions and both theoretical and practical implications be investigated. First, in our studies, the referrer is always
aware of the failure of his/her referral. It might be an interesting avenue for future research to find out not only (a) when consumers notice referral failure but also (b) when they experience it as a failure.
Second, in our studies, participants made recommendations to others
that were moderately distant to them (fellow students). Research
shows that recommendations travel through weak ties, where they
have a bridging function, and through strong ties, where they are
most influential (Brown & Reingen, 1987). Future research could
focus on the moderating effect of the strength of these ties on the referral backfire effect. On the one hand, one may argue that strong ties
buffer against referral failure because they are usually the result of
rich and diverse relationships. Consequentially, referral failure
would not necessarily lead to self-threats when the receiver and the
referrer have strong ties. On the other hand, strong ties are characterized by a high degree of association, which could lead to an assumption of similarity. In that case, referral failure could violate that
assumption and therefore cause even stronger effects in strong ties.
Third, although our findings in study 3 already suggest that the
threat invoked by referral failure is on the level of specific consumer
decisions or decision making capability, future research could also
provide more detailed insights into the nature of the threat of referral
failure. A priori, referral failure could call into question one's personality at large, the relationship between receiver and referrer, one's
378
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
consumer choices, one's decision capability, one's capability as an
advisor, or other identity aspects. Additionally, the circumstances
under which a referral occurs could affect different identity aspects,
leading to varying effects of referral failure. Making a referral to a
close friend or an unknown audience in the internet, as an
established expert or as a lay person, after lengthy research or off
the top of one's head, all of these different conditions could lead to
different inferences concerning the significance of referral failure
for one's identity with different possible behavioral patterns.
Fourth, it remains an empirical question whether consumers are
on a certain level aware of the potential negative consequences of
making a referral and whether they take this risk into consideration
when deciding to make a referral. If this awareness is the case, it
should bias these decisions toward more “safe” instances of referrals.
Referrals would subsequently become more likely for brands with a
high-perceived social support or under other circumstances that minimize the risk of referral failure.
Finally, we identified in our Introduction the need for selfpresentation or self-enhancement of identity as the core drivers behind referral behavior (De Angelis, Bonezzi, Peluso, Rucker, &
Costabile, in press; Hennig-Thurau et al., 2004). Because our data
show that the self is indeed involved in referral behavior, this notion
might offer insights into the behavior of key actors in the diffusion of
information. The existing literature has identified hubs (Goldenberg,
Han, Lehmann, & Hong, 2009) and market mavens (Feick & Price,
1987) as important actors in information proliferation. Because
hubs and market mavens share the attribute that they are active referrers, as well as avid information harvesters – and therefore very
open to external influences – one might wonder how they accommodate referral failure. Does the centrality of being a hub or market
maven in these consumers' self-identities and the ensuing desire to
accumulate information override the threat of referral failure? Do
they rely on sources of self-affirmation that are intrinsic to their
role? If they do so, would these sources differ between types of key
actors? Hubs, defined in terms of their social connectedness, may
prefer to intensify the spreading of (other) information as a way to
bolster their identity. Mavens may spread knowledge that is more
technical and objective, which would be less open to opinion and
therefore less discounted by referral outcomes. Further research
might investigate how different key actors deal with the experience
of referral failure because this knowledge might provide insight into
the development of such hubs and mavens.
In conclusion, we emphasize the need for more research into the
effects of referral outcome on the referrer, especially in large mass
of ordinary consumers because the implications could be substantial
for both consumer theory and the market place.
Acknowledgments
We would like to acknowledge the help of colleagues of the consumer behavior group at KU Leuven for comments on earlier versions
of this paper. In addition, the authors would like to thank Peeter
Verlegh two anonymous reviewers and the editorial team for a number of helpful suggestions and comments which have significantly
improved the quality of the finished manuscript.
References
Anderson, E. W. (1998). Customer satisfaction and word of mouth. Journal of Service
Research, 1(1), 5–17.
Bass, F. (1969). A new product growth model for consumer durables. Management
Science, 15(5), 215–227.
Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research,
15(2), 139–168.
Belk, R. W., Bahn, K. D., & Mayer, R. N. (1982). Developmental recognition of consumption symbolism. Journal of Consumer Research, 9(1), 4–17.
Berger, J., & Heath, C. (2007). Where consumers diverge from others: Identity signaling
and product domains. Journal of Consumer Research, 34(2), 121–134.
Bergkvist, L., & Rossiter, J. R. (2007). The predictive validity of multiple-item versus
single-item measures of the same constructs. Journal of Marketing Research,
44(2), 175–184.
Blanton, H., & Christie, C. (2003). Deviance regulation: A theory of action and identity.
Review of General Psychology, 7(2), 115–149.
Boninger, D. S., Krosnick, J. A., & Berent, M. K. (1995). Origins of attitude importance:
Self-interest, social identification, and value relevance. Journal of Personality and
Social Psychology, 68(1), 61–80.
Brendl, C. M., Chattopadhyay, A., Pelham, B. W., & Carvallo, M. (2005). Name letter
branding: Valence transfers when product specific needs are active. Journal of Consumer Research, 32(3), 405–415.
Brown, T. J., Barry, T. E., Dacin, P. A., & Gunst, R. F. (2005). Spreading the word: Investigating antecedents of consumers' positive word-of-mouth intentions and behaviors in a
retailing context. Journal of the Academy of Marketing Science, 33(2), 123–138.
Brown, J. J., & Reingen, P. H. (1987). Social ties and word-of-mouth referral behavior.
Journal of Consumer Research, 14(3), 350–362.
Buhrmester, M. D., Blanton, H., & Swann, W. B., Jr. (2011). Implicit self-esteem: Nature,
measurement, and a new way forward. Journal of Personality and Social Psychology,
100(2), 365–385.
Chevalier, J. A., & Mayzlin, D. (2006). The effect of word of mouth on sales: Online book
reviews. Journal of Marketing Research, 43(3), 345–354.
Chirkov, V., Ryan, R. M., Kim, Y., & Kaplan, U. (2003). Differentiating autonomy from individualism and independence: A self-determination theory perspective on internalization of cultural orientations and well-being. Journal of Personality and Social
Psychology, 84(1), 97–110.
Dawson, E., Gilovich, T., & Regan, D. T. (2002). Motivated reasoning and performance
on the Wason selection task. Personality and Social Psychology Bulletin, 28(10),
1379–1387.
De Angelis, M., Bonezzi, A., Peluso, A. M., Rucker, D. D., & Costabile, M. (2011). On braggarts and gossips: A self-enhancement account of word-of-mouth generation and
transmission. Journal of Marketing Research, 49(4), 551–563.
De Bruyn, A., & Lilien, G. L. (2008). A multi-stage model of word-of-mouth influence
through viral marketing. International Journal of Research in Marketing, 25(3),
151–163.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human
behavior. New York: Plenum.
Deci, E. L., & Ryan, R. M. (1987). The support of autonomy and the control of behavior.
Journal of Personality and Social Psychology, 53(6), 1024–1037.
Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational social influences upon individual judgment. Journal of Abnormal and Social Psychology,
51(3), 629–636.
Dichter, E. (1966). How word-of-mouth advertising works. Harvard Business Review,
44(6), 147–157.
Dunning, D. (2007). Self-image motives and consumer behavior: How sacrosanct
self-beliefs sway preferences in the marketplace. Journal of Consumer Psychology,
17(4), 237–249.
Dunning, D., Perie, M., & Story, A. L. (1991). Self-serving prototypes of social categories.
Journal of Personality and Social Psychology, 61(6), 957–968.
East, R., Hammond, K., & Lomax, W. (2008). Measuring the impact of positive and negative word of mouth on brand purchase probability. International Journal of
Research in Marketing, 25(3), 215–224.
Escalas, J. E., & Bettman, J. R. (2005). Self‐construal, reference groups, and brand meaning. Journal of Consumer Research, 32(3), 378–389.
Feick, L. F., & Price, L. L. (1987). The market maven: A diffuser of marketplace information. The Journal of Marketing, 51(1), 83–97.
Fein, S., & Spencer, S. J. (1997). Prejudice as self-image maintenance: Affirming the self
through derogating others. Journal of Personality and Social Psychology, 73(1), 31–44.
Fitzsimons, G. J. (2008). Death to dichotomizing. Journal of Consumer Research, 35(1), 5–8.
Fitzsimons, G. J., & Lehmann, D. R. (2004). Reactance to recommendations: When
unsolicited advice yields contrary responses. Marketing Science, 23(1), 82–94.
Gal, D., & Rucker, D. D. (2010). When in doubt, shout! Paradoxical influences of doubt
on proselytizing. Psychological Science, 21(11), 1701–1707.
Gao, L., Wheeler, S. C., & Shiv, B. (2009). The “shaken self”: Product choices as a means
of restoring self‐view confidence. The Journal of Consumer Research, 36(1), 29–38.
Godes, D., & Mayzlin, D. (2009). Firm-created word-of-mouth communication:
Evidence from a field test. Marketing Science, 28(4), 721–739.
Goldenberg, J., Han, S., Lehmann, D. R., & Hong, J. W. (2009). The role of hubs in the
adoption process. Journal of Marketing, 73(2), 1–13.
Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronic
word-of-mouth via consumer-opinion platforms: What motivates consumers to articulate themselves on the Internet? Journal of Interactive Marketing, 18(1), 38–52.
Hennig-Thurau, T., & Walsh, G. (2003). Electronic word-of-mouth: Motives for and
consequences of reading customer articulations on the Internet. International Journal of Electronic Commerce, 8(2), 51–74.
Irwin, J. R., & McClelland, G. H. (2001). Misleading heuristics and moderated multiple
regression models. Journal of Marketing Research, 38(1), 100–109.
Jain, S. P., & Maheswaran, D. (2000). Motivated reasoning: A depth-of‐processing
perspective. Journal of Consumer Research, 26(4), 358–371.
Jones, J. T., Pelham, B. W., Carvalho, M., & Mirenberg, M. C. (2004). How do I love thee?
Let me count the Js: Implicit egotism and interpersonal attraction. Journal of
Personality and Social Psychology, 87(5), 665–683.
Jones, J. T., Pelham, B. W., Mirenberg, M. C., & Hetts, J. J. (2002). Name letter preferences
are not merely mere exposure: Implicit egotism as self-regulation. Journal of Experimental Social Psychology, 38(2), 170–177.
Keller, E. (2007). Unleashing the power of word of mouth: Creating brand advocacy to
drive growth. Journal of Advertising Research, 47(4), 448–452.
B. Claus et al. / Intern. J. of Research in Marketing 29 (2012) 370–379
Kumar, V. (2010). A customer lifetime value-based approach to marketing in the
multichannel, multimedia retailing environment. Journal of Interactive Marketing,
24(2), 71–85.
Laczniak, R. N., DeCarlo, T. E., & Ramaswami, S. N. (2001). Consumers' responses to negative word-of-mouth communication: An attribution theory perspective. Journal of
Consumer Psychology, 11(1), 57–73.
Liberman, A., & Chaiken, S. (1992). Defensive processing of personally relevant health
messages. Personality and Social Psychology Bulletin, 18(6), 669–679.
Mobley, M. F., Bearden, W. O., & Teel, J. E. (1988). An investigation of individual responses to tensile price claims. Journal of Consumer Research, 15(2), 273–279.
Nail, P. R., MacDonald, G., & Levy, D. A. (2000). Proposal of a four-dimensional model of
social response. Psychological Bulletin, 126(3), 454–470.
Olivero, N., & Lunt, P. (2004). Privacy versus willingness to disclose in e-commerce
exchanges: The effect of risk awareness on the relative role of trust and control.
Journal of Economic Psychology, 25(2), 243–262.
Reed, A. (2004). Activating the self-importance of consumer selves: Exploring identity
salience effects on judgments. Journal of Consumer Research, 31(2), 286–295.
Reichheld, F. F. (2003). The one number you need to grow. Harvard Business Review,
81(12), 46–54.
Richins, M. L. (1994). Valuing things: The public and private meanings of possessions.
The Journal of Consumer Research, 21(3), 504–521.
Rogers, E. M., & Cartano, D. G. (1962). Methods of measuring opinion leadership. Public
Opinion Quarterly, 26(3), 435–441.
Rudman, L. A., Dohn, M. C., & Fairchild, K. (2007). Implicit self-esteem compensation:
Automatic threat defense. Journal of Personality and Social Psychology, 93(5),
798–813.
Ryan, R. M. (1993). Agency and organization: Intrinsic motivation, autonomy, and the
self in psychological development. In J. E. Jacobs (Ed.), Nebraska symposium on motivation, 1992: Developmental perspectives on motivation (pp. 1–56). Lincoln, NE,
US: University of Nebraska Press.
Ryu, G., & Feick, L. (2007). A penny for your thoughts: Referral reward programs and
referral likelihood. Journal of Marketing, 71(1), 84–94.
Schmitt, P., Skiera, B., & Van den Bulte, C. (2011). Referral programs and customer
value. Journal of Marketing, 75(1), 46–59.
379
Sherman, D. K., & Cohen, G. L. (2006). The psychology of self-defense: Self-affirmation
theory. In M. P. Zanna (Ed.), Advances in experimental social psychology, Vol. 38.
(pp. 183–242)San Diego, CA: Academic Press.
Sherman, D. A. K., Nelson, L. D., & Steele, C. M. (2000). Do messages about health risks
threaten the self? Increasing the acceptance of threatening health messages via
self-affirmation. Personality and Social Psychology Bulletin, 26(9), 1046–1058.
Sivanathan, N., & Pettit, N. C. (2010). Protecting the self through consumption: Status
goods as affirmational commodities. Journal of Experimental Social Psychology,
46(3), 564–570.
Smith, T., Coyle, J. R., Lightfoot, E., & Scott, A. (2007). Reconsidering models of influence:
The relationship between consumer social networks and word-of-mouth effectiveness. Journal of Advertising Research, 47(4), 387–397.
Sprott, D., Czellar, S., & Spangenberg, E. (2009). The importance of a general measure of
brand engagement on market behavior: Development and validation of a scale.
Journal of Marketing Research, 46(1), 92–104.
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the
self. Advances in Experimental Social Psychology, 21, 261–302.
Steele, C. M., & Liu, T. J. (1983). Dissonance processes as self-affirmation. Journal of
Personality and Social Psychology, 45(1), 5–19.
Sundaram, D. S., Mitra, K., & Webster, C. (1998). Word-of-mouth communications:
A motivational analysis. Advances in Consumer Research, 25, 527–531.
Van den Bulte, C., & Wuyts, S. (2009). Leveraging customer networks. In P. R.
Kleindorfer, Y. Wind, & R. E. Gunther (Eds.), The network challenge: Strategy, profit,
and risk in an interlinked world (pp. 243–258). Upper Saddle River, NJ: Wharton
School Publishing.
Velleman, P. F., & Leland, W. (1993). Nominal, ordinal, interval, and ratio typologies are
misleading. The American Statistician, 47(1), 65–72.
Von Wangenheim, F. (2005). Postswitching negative word of mouth. Journal of Service
Research, 8(1), 67–78.
Watts, D. J., & Dodds, P. S. (2007). Influentials, networks, and public opinion formation.
Journal of Consumer Research, 34(4), 441–458.
Wentura, D., & Greve, W. (2005). Assessing the structure of self-concept: Evidence for
self-defensive processes by using a sentence priming task. Self & Identity, 4(3),
193–211.
Intern. J. of Research in Marketing 29 (2012) 380–389
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
On the importance of social integration for minority targeting effectiveness
Michael Antioco a,⁎, Joëlle Vanhamme b, Anaïk Hardy c, Lidwine Bernardin d
a
EMLYON Business School, 23 Avenue Guy de Collongue, 69134 Ecully Cedex, France
EDHEC Business School, 24 Avenue Gustave Delory, CS 50411, 59057 Roubaix, France
c
UNHYCOS, ZI Les Marcots, 325 chemin du parc, 95480 Pierrelaye, France
d
Interface Tourisme, 11bis rue Blanche, 75009 Paris, France
b
a r t i c l e
i n f o
Article history:
First received in 29, June 2011 and was under
review for 7 months
Available online 15 September 2012
Keywords:
Advertising
Ethnic minority
Credibility
Similarity
a b s t r a c t
Research into the influence of model ethnicity on advertising effectiveness usually compares a dominant or
mainstream group with a single, ethnically distinctive minority group. Such two-group studies implicitly assume that all minorities react similarly to distinctive and non-distinctive models portrayed in ads. However,
minorities differ in their level of integration into the dominant societal group, which likely influences their
responses. This research studies target/non-target market effects for minorities whose levels of integration
into the host society vary. In contrast with relatively less integrated minorities, minority groups better integrated into the dominant society exhibit target/non-target market effects unlike those previously reported.
This study also contrasts two mediation routes for the observed effects of ethnic models: ad model–
self-perceived similarity and perceived ad model credibility. The results show that the two routes are confounded for less integrated minorities but dissociated for better integrated minorities. The findings offer
clear applications for the advertising industry.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction: Cultural diversity, identities, and advertising
Research has recognized the influence of social identities on
consumer responses to marketing stimuli (e.g., Deshpandé & Stayman,
1994; Reed, 2004). Social identities are “mental representations that
can become a basic part of how consumers view themselves;” they
are based on aspects as diverse as traits, social roles, moral values,
nationality, gender, sports team membership, or ethnicity (Reed,
2004, p. 286). Not only do these identities vary in their degree of importance and salience, but they also differ in the extent to which they guide
individual beliefs about how to think, feel, and behave (Hogg, Terry, &
White, 1995).
This study addresses a specific, prominent type of social identity
(or in-group identity), namely, ethnicity (McGuire, McGuire, Child,
& Fujioka, 1978; Phinney, 1990), which often refers to minority
populations (Rosenbaum & Montoya, 2007). The effects of ethnicity
are pertinent for a wide range of marketing variables, such as consumption patterns, shopping orientations, responses to promotions
and advertising, purchase decisions, media usage, and brand loyalty
(e.g., Aaker, Brumbaugh, & Grier, 2000; Lee, Fernandez, & Martin,
2002).
Yet prior advertising research that focuses on the effects of ethnicity by studying models who represent an ethnic minority tends
⁎ Corresponding author. Tel.: +33 478 337 936.
E-mail addresses: antioco@em-lyon.com (M. Antioco), Joelle.Vanhamme@edhec.edu
(J. Vanhamme), annaik-hardy@unhycos.com (A. Hardy), lidwine.bernardin@hotmail.fr
(L. Bernardin).
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.08.001
to oversimplify the influence of ethnicity in that it ignores the impact of the minority's level of socio-cultural integration into the
host society. With a few exceptions (Brumbaugh & Grier, 2006;
Grier & Brumbaugh, 2007), research on the influence of ethnicity
on advertising effectiveness compares only two groups: usually
an ethnically non-distinctive dominant group and an ethnically
distinctive group (e.g., White vs. Black Americans Brumbaugh,
2002; White vs. Asian New Zealanders Martin, Lee, & Yang, 2004;
Anglo vs. Hispanic Americans Deshpandé & Stayman, 1994).
These studies implicitly assume that all distinctive people display
the same pattern of responses when exposed to ads designed to
appeal to them, as well as to ads that portray non-distinctive
models (Brumbaugh & Grier, 2006).
We argue instead that the assumption of ethnic similarity—which
largely reflects the difficulty of collecting sufficient data from minority
respondents (Brumbaugh & Grier, 2006; Hunt, 2000)—is unwarranted.
Ethnic minorities have unique historical experiences and statuses in societies (Hunt, 2000) related to their relative integration into the host society
(or dominant group). For example, cultural integration, or the acquisition
of traits and behaviors of the host or mainstream culture (Laroche, Kim, &
Tomiuk, 1998), influences the enduring salience of cultural identity and
identity-related attitudes and behaviors (Chattaraman, Lennon, & Rudd,
2010).
We challenge the assumption that all minority groups should
react in a similar way when exposed to ads designed to appeal to
them, as well as to ads that portray non-distinctive models, and
thereby aim to make two main contributions. First, we show that
not all minorities react similarly to minority-targeted ads, and the
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
differences in responses can be explained by the level of integration
of different minority groups into the host society. Second, we show
that the primary driver of model ethnicity effects in advertising is
not viewers' perceived similarity to the model in the advertisement
per se, as previous studies have proposed (e.g., Grier & Brumbaugh,
1999), but rather the viewers' perceptions of the credibility of that
model. Specifically, we show that for ads portraying an ethnic
model, model–self-perceived similarity and model perceived credibility have a similar pattern of influence on advertising effectiveness
for minorities who are relatively less integrated into the host society,
but this result no longer holds for more closely integrated minorities.
By highlighting this dissociation and confirming credibility's superior explanatory power, we make a case for credibility as the “true”
causal mediator of the effect of ethnic model portrayal on advertising
effectiveness.
Furthermore, this study offers clear managerial implications for companies addressing an ethnically diverse audience. Our results show that
the use of models from ethnically distinctive groups may be less effective than portraying models from a dominant group (i.e., mainstream
models).
2. Theory
2.1. Viewer ethnicity and target/non-target market effects
Ethnicity reflects the frequent patterns of association and identification with common national or cultural origins of a subgroup in a larger
society (Arnould, Price, & Zinkhan, 2004). For an ethnic minority group,
ethnic membership often has great prominence, particularly in an
ethnically mixed, rather than uniform, society (McGuire et al., 1978).
In addition, ethnically distinctive consumers tend to prefer ads with
ethnically distinctive models because they interpret the ad according
to its ethnicity depictions (Deshpandé & Stayman, 1994). In contrast,
non-distinctive ethnic groups usually do not react differently to ads
portraying a distinctive or a non-distinctive model (Appiah, 2004;
Brumbaugh, 2002) because ethnicity is less salient to non-distinctive
viewers, and thus they prefer ads they perceive to be targeting them
through a broader set of cues (e.g., age, gender, familiar lifestyle, ad
scenario) (Sudbury & Wilberforce, 2006). In the vast majority of studies,
White consumers respond equally favorably to ads featuring Black
models as White models (e.g., Aaker et al., 2000 [Experiment 1];
Sudbury & Wilberforce, 2006; Whittler, 1991; Whittler & Spira, 2002).
However, enhanced preferences for the ad with the Black model
(Appiah, 2004; Brumbaugh, 2002), as well as decreased preferences
for such ads (e.g., Aaker et al., 2000 [Experiment 2]), also occasionally
appear in the literature. Decreased preference for an ad with a distinctive model mostly occurs among non-distinctive consumers when the
model portrayed in the ad belongs to a controversial minority, such as
gay men (Grier & Brumbaugh, 1999; Puntoni, Vanhamme, & Visscher,
2011). Thus, an ethnically distinctive model should have positive effects
on ethnically distinctive consumers, but model ethnicity appears less
relevant to the dominant group of consumers who identify with advertisements on a larger set of cues.
However, Brumbaugh and Grier's (2006) findings challenge the
established pattern for distinctive consumers. In their study, Asian minority consumers in the United States, unlike Black and Indian participants, did not prefer an ad portraying an Asian model to an ad with a
White model. Brumbaugh and Grier speculate an influence of social status or visual salience of group membership. In U.S. society, Asian and
White people tend to display a higher status than other groups, and
Asians are a more visually ambiguous group than the highly visually
distinctive Black minority (Brumbaugh & Grier, 2006). The speculation
that status differences might account for these unexpected results also
echoes Grier and Deshpandé's (2001, p. 223) suggestion that “Asian
Americans may respond relatively less favorably to ethnically targeted
381
advertisements than other minority groups that are perceived as having
lower social status.”
Following these suggestions, we investigate whether the relative
socio-cultural integration of minorities into the host culture, which affects their social status (Lenski, 1954), explains differences in target/
non-target market effects for minorities. In this study, we define target
(non-target) market effects as the increased (decreased) preference for
an ad by people (not) in the target audience (Aaker et al., 2000).
2.2. Socio-cultural integration and target/non-target market effects
Socio-cultural integration refers to the knowledge of the host
country's language, the understanding of its society, participation
in its institutions, the development of interpersonal contacts with
members of other ethnic groups (e.g., inter-ethnic friendships), and
respect for or adoption of different norms and patterns of behavior
(Hamberger, 2009; Junger-Tas, 2001). Most studies investigating
the target/non-target market effects of ads that portray ethnic
models contrast the attitudes of White Americans with those of
Black Americans (e.g., Aaker et al., 2000; Brumbaugh, 2002; Bush,
Smith, & Martin, 1999; Grier & Brumbaugh, 1999; Whittler & Spira,
2002). Yet Black Americans represent a specific case because of
their persistent segregation, an institutional mechanism that creates
and reinforces inequality and prevents socio-cultural integration
(Junger-Tas, 2001). No other U.S. minority group has experienced
as much segregation as Black Americans (Collins & Williams, 1999).
Extant findings thus might hold true only for minorities with a relatively minimal level of integration into the dominant society. Grier
and Deshpandé (2001) show, for example, that a lower status and
perceived status differences, which are directly related to the concept of integration (Lenski, 1954), increase the positive effect of ethnic spokespeople on advertising effectiveness; this effect occurs
through the heightening of ethnic salience.
Groups with a higher level of integration instead may tend to be more
receptive to standardized marketing strategies and exhibit response patterns that are more similar to those of the majority (Webster, 1992). In
particular, the level of immigrant socio-cultural integration should affect
attitudes and responses to advertising and promotion (Wallendorf &
Reilly, 1983; Webster, 1992), such that members of a better integrated
minority most likely define themselves more in accordance with the majority than with a less integrated minority (Wood, Pool, Leck, & Purvis,
1996). Groups undergoing a socio-cultural integration process learn to
think and behave similarly to the dominant group (Hamberger, 2009;
Junger-Tas, 2001).
Because they also may observe how the dominant group rejects the
less integrated minority (as signaled by a lack of models who represent
the less integrated minority in media or country regulations and asylum
agreements regarding specific minority groups; Diplômatie Française,
2011), members of a better integrated minority may display negative
attitudes toward less integrated minority groups (Allen & Wilder,
1980; Oakes, 1987). For example, people avoid consuming products
that could have negative symbolic implications or project a negative
image of them (Banister & Hogg, 2004). White and Dahl (2006) show
that men have more negative attitudes toward and therefore are less inclined to choose a product that would associate them with femaleness
(e.g., ladies-cut steak). They further show that groups with which people prefer not to be associated have more influence on attitudes than
out-groups in general.
Thus, for well-integrated minorities, ads that portray a model from
their own or the majority group should be more effective than ads
portraying a model from a less integrated ethnic group. In contrast, because their lack of socio-cultural integration into the host society implies
some rejection of that host society's culture and norms (Junger-Tas,
2001), less integrated minorities may be motivated to avoid associations
with the dominant group. That is, for less integrated minorities, ads that
portray a model from an ethnic minority group should be more effective
382
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
than ads portraying a model from the dominant group, in line with previous findings regarding target/non-target market effects.
We expect to replicate the extant results pertaining to the target/
non-target market effects for the dominant group and minorities with a
relatively lower level of integration, but we predict a different pattern of
results for more integrated minorities. Thus, respondent ethnicity should
moderate the influence of model ethnicity on advertising effectiveness:
H1a. A more integrated minority group displays a more positive attitude toward an ad that portrays a model from its own ethnic group or
from the majority group rather than a model from a less integrated ethnic group.
H1b. A less integrated minority group displays a more positive attitude
toward an ad that portrays a distinctive model from its own group or
from a more integrated minority group rather than a non-distinctive
model.
For the dominant group, we do not expect any difference in reactions to ads because of model ethnicity. Therefore, we do not specify a
separate hypothesis for this group.
2.3. Target/non-target market effects: the credibility versus similarity
mediation routes
A model portrayed in an ad is a type of presenter. Literature on
presenters shows that more credible presenters appear more persuasive (Pornpitakpan, 2004). Credibility encompasses elements such as
source expertise (e.g., an expert user or technical expert) and trustworthiness (Rossiter & Smidts, 2011). Rossiter and Smidts (2011)
also show that source expertise is a necessary condition for persuasion by celebrity presenters and posit that the same requirement
holds for non-celebrity presenters, though these “real people” also
possess some form of positive trustworthiness that grants them
high source credibility as well. Klucharev, Smidts, and Fernandez's
(2008) study on brain imaging also confirms the positive effect of celebrity presenters' perceived expertise on persuasion. Thus, from the
literature on presenters, perceived model credibility seems to drive
persuasion.
In the literature on the advertising effectiveness of ethnic models,
some studies also propose source credibility as a mediator of the effect
of model ethnicity on advertising effectiveness (e.g., Deshpandé &
Stayman, 1994). How does the viewer's socio-cultural integration affect
this relationship? Similar to social status (see Grier & Deshpandé, 2001),
a lower level of integration into the dominant society should contribute
to heightened ethnic salience. We find evidence that ethnic salience
moderates the influence of model ethnicity on perceived credibility
(Deshpandé & Stayman, 1994; Williams & Qualls, 1989) and that
model credibility mediates the joint influence of ethnic salience and
model ethnicity on brand attitudes (Deshpandé & Stayman, 1994).
Thus, credibility should mediate the joint influence of viewers' integration level and model ethnicity on advertising effectiveness (see Fig. 1).
Beyond such credibility investigations, the majority of studies address the advertising effectiveness of ethnic models by considering
two other mediators of the effect of model ethnicity on advertising effectiveness, depending on the viewers' group considered. We discuss
these two mediators hereinafter and highlight their relationship to
credibility. Among minority groups, studies show that perceived similarity to the ad model drives advertising persuasion (e.g., Aaker et al.,
2000; Brumbaugh & Grier, 2006; Grier & Brumbaugh, 1999; Whittler,
1991). These consumers interpret the ad according to its ethnicity
depictions (Deshpandé & Stayman, 1994) and are more persuaded by
an ad because of the congruence between the ad model and their
self-identity (Johnson & Grier, 2011). For this group of viewers, ad
model–self-perceived similarity and perceived model credibility mediation routes are likely confounded because similar sources tend to be perceived as credible by the targeted audience (O'Keefe, 1991; Simons,
Berkowitz, & Moyer, 1970), and both similarity and credibility drive advertising effectiveness.
For viewers in the dominant group, ad model–self-perceived similarity is not the driver of advertising effectiveness; rather, it is the feeling of being targeted (e.g., Aaker et al., 2000; Brumbaugh & Grier, 2006;
Grier & Brumbaugh, 1999; Whittler, 1991). Consumers are persuaded
by an ad because of the congruence between the targeting attempt
and their schematic knowledge about advertising stimuli (e.g., type of
model that seems suitable for a certain type of product advertising)
(Dimofte, Forehand, & Deshpandé, 2003; Johnson & Grier, 2011) and
evaluate the ad on a broad set of cues (e.g., age, gender, occupation, representation of a familiar lifestyle or scenario) (Sudbury & Wilberforce,
2006). This type of judgment is related to credibility judgments: schematic knowledge about advertising forms through multiple previous
exposures to advertising examples, and research has established that
stimuli credibility increases with multiple exposures to the stimuli
(Brown, Brown, & Zoccoli, 2002). In other words, if congruence exists
between the ad model and the schematic knowledge, high perceived
credibility ensues and subsequently drives advertising effectiveness.
Thus, for the dominant group of viewers, perceived model credibility,
not ad model–self-perceived similarity, likely mediates the advertising
effectiveness of ethnic models.
Because better integrated minorities learn to think and behave like
the dominant group (Hamberger, 2009; Junger-Tas, 2001), they likely
display a response pattern to advertising more similar to that of the
dominant group than to that of the less integrated minority (Webster,
1992; see also H1a and H1b). Thus, we expect that model perceived
credibility, rather than ad model–self-perceived similarity, mediates
advertising effectiveness for this group.
To summarize, we predict that (1) perceived credibility mediates
the joint effect of the respondent's level of integration and model
ethnicity on advertising effectiveness for all ethnic groups (including
the dominant group) and (2) among better integrated minorities,
the mediation route through credibility is not confounded with the
pattern of influence found for similarity as is the case for less integrated minorities.
H2a. Credibility, rather than similarity, mediates the effect of the interaction between respondent ethnicity and model ethnicity on the
attitude toward the ad.
Fig. 1. Proposed model.
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
H2b. Similarity and credibility display the same pattern of influence on
advertising effectiveness for less integrated minorities, but this does not
hold true for better integrated minorities.
3. Method
To test our hypotheses, we conducted a quasi experiment in which
we asked a sample of French female consumers belonging to the
White majority, the North African minority, or the Black minority to
evaluate ads portraying models from all three groups.1 In France, the
North African minority experiences a relatively low level of integration
into the White French group, whereas the Black minority has achieved a
relatively higher level of integration (Geddes, 2003; Marthaler, 2008;
Ouellet, 2007). Before the experiment, we conducted a pilot study and
ran six pretests to calibrate the stimuli and check the manipulations,
as well as a general pretest of our questionnaire.
3.1. Study setting
France serves as the study context, which is both theoretically and
substantively interesting. France is the most multi-cultural country in
the European Union (Tréguer & Ségati, 2003), comprising three main
ethnic groups: a White majority; a Black population consisting of
French people with a sub-Saharan background (approximately 1.08 million people in 2009) or from overseas territories (approximately 757,000
in 2009); and North African French people (approximately 3.26 million
in 2009) with roots in countries such as Algeria, Morocco, and Tunisia
(Independent, 2009). These minority groups have different historical
and contemporary experiences, which influence their degree of integration into the host society. The first three points in Table 1 depict the main
socio-cultural reasons that the North African French tend to be less integrated than the Black French, including language use, the French model
of integration, and the difference between “imposed” and “chosen” immigration. Table 1's remaining points pertain to the economic and
politico-legal integration, both of which influence the socio-cultural integration of these minority groups (Grosjean, 2011; Lacoste, 2010).
This study setting enables us to assess the effects of integration and
to avoid the potential confound that marks much prior research, namely, the degree of visual salience of group membership. In studies comparing American Black and White groups, the group for which
ethnicity is most visible is also the group that has the relatively lower
level of integration into the host society (Collins & Williams, 1999).
The comparison of the White dominant group with both the Black
French minority, characterized by relatively higher integration and
higher visual salience, and the North African French minority, characterized by relatively lower integration and lower visual salience, circumvents this potential confound.
Ethnic minorities' representation remains minimal in French advertising, especially on television. Non-European models appeared in only
3.1% of advertising in 2005 (Autorité de Régulation Professionnelle de la
Publicité, 2006), though recent laws and agreements mandate a minimum presence of visible minorities on French television (Frachon &
Sassoon, 2008). Therefore, it is substantively interesting to study the effect of the presence of visible minorities on advertising effectiveness
among the multi-ethnic French population (Spears & Singh, 2004).
3.2. Design and participants
The study used a 3 (respondent ethnic group: more integrated
Black French vs. less integrated North African French vs. White
French)× 3 (model ethnicity: Black vs. North African vs. White model)
between-subjects design. In addition to the evidence provided in
1
We label two groups by race and one group by origin (North African) because these
are the most commonly used and simplest grouping terminologies.
383
Table 11, we measured ethnicity and the level of socio-cultural integraTable
Socio-cultural
integration
markers.
tion
into French
society,
using self-reports, as manipulation checks. Be• Approximately 75% of French immigrants speak a foreign language at home.
This is especially true for North African French, which suggests their poorer
socio-cultural integration (Clanché, 2002).
• The perception that Muslim immigrants, mainly of North African origin, can no
longer assimilate into French culture is an important issue (Geddes, 2003). For example, 30% of White French are reluctant to marry people from a North African minority, whereas only 21% express reluctance about marrying a person from the
Black minority (Le Figaro, 2010).
• Many French politicians and White French voters openly favor “chosen immigration” that welcomes only laborers or students (Marthaler, 2008). However, 70% of
new immigrants, mainly from North Africa, move for family reasons, which is a
form of “imposed immigration” (Independent, 2009).
• The North African minority's integration has been slowed on various fronts: economic integration, measured as its access to the labor market and employment;
education; income; housing; and politico-legal integration, which refers to
the right to vote and be elected (Hamberger, 2009; Junger-Tas, 2001). Reasons
include:
• France's postcolonial behavior toward North African French, including poor housing and working conditions and low-quality education, negatively affected this
group's integration into the dominant society (Lacoste, 2010), while overseas territories were being integrated into the French State. North African French have the
highest unemployment rates (27% in 2005; INSEE, 2005), and differences in their
literacy scores compared with non-migrants are among the highest in Europe
(Entorf & Minoiu, 2005).
• In contrast with North African French, Black French from overseas territories have
French nationality and participate in French politics (Lacoste, 2010).
• Most illegal non-EU immigrants in France are from Morocco and Algeria. Several
countries such as Haiti, Congo, and Senegal have signed concerted agreements to
repatriate their illegal immigrants, but Morocco and Algeria have not (République
Française, 2011).
• The European Union policy of immigration implies actions against ‘unwanted’
North African immigration but disincentives for Black immigrants (Lindstrom,
2005).
fore the study, we also ran a pretest with a convenience sample of
French respondents recruited from the staff of a large French business
school (21 White, 6 Black, and 6 North African), which confirmed the
higher perceived level of integration of the Black minority: 54.55% of
the respondents perceived Black French people as better integrated in
France than North African French people; 18.18% believed the opposite
(z = 3.77, p > .01), and the remaining respondents (27.27%) considered
Black and North African French equally well-integrated (z = 2.45,
p > .05).
All participants were randomly assigned to complete one of three
different versions of the survey. With this design, we can assess our
hypotheses using an analysis of variance (ANOVA) of the effect of
the ethnic group × model ethnicity interaction on attitude toward
the ad (with similarity and/or credibility as covariate(s) to test H2a
and H2b), followed by planned contrast analyses.
Four hundred fifty-five French women from White (N= 210), North
African (N= 123), and Black (N= 122) ethnic groups, all between 18
and 35 years of age, participated in the study. Thirty-four percent of
White respondents saw a White model, 27% saw a North African
model, and 39% saw a Black model; in addition, 33% of Black and 34%
of North African respondents saw a White model, 32% and 33% saw a
Black model, and 35% and 33% saw a North African model, respectively.
Invitations to participate were posted in various women's magazines
and website forums. The link on the websites changed to ensure the random assignment of respondents to various conditions. Our data collection procedure targeted female consumers, who a priori have a higher
probability of being exposed to such print ads. Moreover, women are
more prone to react to the visual characteristics of people in advertising
and more vulnerable to those images than men (Baird & Grieve, 2006).
The selected age bracket also suggests that the respondents mostly represent second- or third-generation immigrants who had been born in
France. Thus, the gender and age restrictions we applied created a homogeneous sample that generally matched the gender and age category of
the models depicted in the ads.
384
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
3.3. Stimuli
Three ads for a fictitious mass-market cosmetic product (Sanydra)
served as the study stimuli (see Fig. 2). We use advertising for a personal
care product for three reasons. First, physical characteristics exert an important influence on purchase intentions for these items (Solomon,
2007). Because of differences in skin and/or hair type among different
ethnic groups, personal care products belong to a category of products
for which consumers likely formulate ethnic-specific quality requirements. In other words, this is a category for which ethnic social identity
and, therefore, similarity to the model truly matters. Personal care thus allows for a strong test of our hypotheses and, particularly, of the dissociative pattern between model similarity and model credibility (H2b).
Second, the prominent display of a (female) model is a well-established
advertising format for this product category (Puntoni et al., 2011).
Third, personal care products are an important product category in
terms of sales (Xerfi, 2010) and represent a large and growing part of
advertising expenditures in France (Kantar Media, 2011).
We ran a pilot study by reviewing two issues of each of three popular magazines: Elle, Marie-Claire, and Grazia (those published the last
week of May and the second week of June 2011). This pilot study
aimed to verify the dominance of White models in ads for personal
care products in France. Of the 1398 pages in these six issues, 108 ads
for personal care products featured lipstick (7.4%), skin care (38%),
hair spray (2.8%), nail polish (3.7%), tanning products (2%), perfume
(36.1%), and shampoo (8.3%), as well as multiple products together in
one ad (1.7%). The ads contained 113 models (i.e., 5 ads depicted 2
models; 13.3% celebrities), 97% of who was White, and 3% of who represented an ethnic minority.
To create the main study stimuli, we cut and pasted photographs
from a cosmetics magazine. Because the position and body language
of the model, as well as her body mass index, needed to be identical
across conditions, we kept one body shape and altered the face and
skin color to display her ethnic heritage. We ran a first pretest with
nine models (three per ethnicity) and 45 respondents (from a French
business school's administrative staff and students) to confirm unambiguous recognition of the model's ethnicity. Respondents identified
the ethnic membership of all models except one correctly. We ran another pretest with the eight remaining models to select the best set of
three models, one from each ethnicity. For this pretest, we created
eight cards and asked 18 business school staff members to pick a set
of three models, one from each ethnicity, that they perceived as most
similar. We selected the set of three models most frequently chosen
(61.1%). During the pretest stage, we also asked 34 graduate students
to evaluate the realism and professionalism of the ad (“Does it look
like a real ad?”). All respondents answered affirmatively for the three
ads we selected.
Finally, we ran a pretest to (1) confirm recognition of the models'
ethnicity and (2) check that the North African model's ethnicity was
perceived as visually less salient than the Black model's ethnicity (to ensure that visual salience and integration are not confounded). Three
versions of a questionnaire, depicting the selected set of three models
in different sequences, were distributed to 24 female graduate students
from the dominant White group (average age = 22.25 years, SD=
1.92). We asked them to identify the ethnicity of the models, among
five possibilities (Caucasian, North African, Asian, Black, or other), to indicate which physical traits/characteristics led them to that conclusion,
and to rate the extent to which these traits were visible and salient,
using a seven-point Likert scale. All respondents correctly assigned
the models to their ethnic groups. They mainly relied on two physical
criteria: skin color and hair type. The average salience of criteria mentioned in position 1 (M Black = 6.65, SD= .42; M North African = 5.58,
SD= .86; F(2, 69) = 11.42, p b .01) and position 2 (M Black= 5.86,
SD= 1.26; M North African= 4.77, SD= .92; F(2, 65) = 11.14, p b .01)
differed significantly across models and indicated that the Black model's
ethnicity appeared significantly more visually salient than that of the
North African model (the most often used criteria were skin color and
hair type, followed by eye shape and face shape). The experiment ran
for four weeks, and the questionnaire took 10 min to complete.
3.4. Measures
After viewing one of the ads in Fig. 2, respondents answered questions pertaining to the dependent variables (attitude toward the ad,
model credibility, and perceived model similarity; see Table 2). Attitude
toward the brand was also measured, but given the fictitious nature of
the product, as well as the absence of detailed product information, the
results for this measure were the same as those for attitude toward the
ad, and therefore we do not report them. As a proxy for their level of integration, respondents then indicated (1) if and how frequently they
spoke a foreign language at home (five-point Likert scale), (2) whether
and how frequently they watched foreign television channels (five-point
Likert scale), and (3) the number of same-ethnicity friends among their
five closest friends (e.g., Garrett & Pichette, 2000; Ryder, Alden, &
Paulhus, 2000). These measures tap socio-cultural integration, which
complements the established findings in Table 1. Then, the questionnaire
included questions about respondents' ethnic background and age, to
confirm the comparability of the three subsamples. Because of the sensitive nature of questions about ethnic background, especially for a French
population, many minority respondents hesitated or simply refused to
Fig. 2. Stimuli.
385
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
of integration. North African French respondents were significantly
less integrated on each criterion (Fs ≥ 6.51, ps b .01).
Table 2
Measurement items.
Scale
Items
Attitude toward the Ad (Holbrook &
Batra, 1987), semantic seven-point
scale
1. I dislike/like the advertisement.
2. I react unfavorably/favorably to
the advertisement.
3. I feel negative/positive toward
the advertisement.
4. The advertisement is bad/good.
The model portrayed in the ad is:
1. Not credible/credible.
2. Novice/expert.
3. Not likable/likable.
4. Not trustworthy/trustworthy.
1. I can easily find similarities
between myself and the model in
the ad.
2. The model in the ad belongs to a
group of which I am a part.
3. I can easily relate myself to the
model in the ad.
Model credibility (Lee et al., 2002),
semantic seven-point scale
Model similarity (Aaker et al., 2000;
Gwinner & Eaton, 1999), seven-point
Likert scale (1 = totally disagree;
7 = totally agree)
answer questions related to other socio-demographic or economic variables when the questionnaire was pretested. As a proxy for education,
we asked respondents whether they were aware (yes/no) of four different news headlines and the extent to which they were concerned about
those headlines (not at all/very). We did not use the second measure in
our analysis but employed it to prevent respondents from discovering
that the first question was designed as an educational test. All respondents answered the questions in French. The scales were professionally
translated into French and back-translated into English (Brislin, 1980).
4. Results
4.1. Sample and measure checks
Because respondent ethnicity is not a random variable, we checked
the comparability of the three samples in terms of age and found no significant differences (M Black=24.32, SD=4.40; M North African=
23.58, SD=4.55; M White=23.49, SD=2.76; F(2, 452)=2.03, p>.14;
medians for the three groups are 23, 23, and 22, respectively; χ2[2] =
4.35, p>.11). News awareness across the ethnic groups also did not differ significantly (F(2, 452)=.42, p>.66).
Using Amos 19.0, we conducted a confirmatory factor analysis
(maximum likelihood) of attitude toward the ad (Aad), model credibility, and model similarity (χ2[41]= 152.507, p b .001; NFI = .95; CFI =
.96; RMSEA = .06). The Jöreskog rho values range from .82 for similarity
to .93 for Aad; the average variance extracted ranges from .58 for credibility to .77 for Aad. In support of discriminant validity, the square roots
of the average variance extracted are superior to any correlations between latent variables. As additional checks of discriminant validity,
we performed systematic comparisons of the three-variable model
with models containing one or two variables (e.g., we merged credibility and similarity into a single latent variable). The three-dimension
model performs best (p b .001), in further support of the discriminant
validity of our three constructs. To aggregate each variable, we took
the mean of its respective items. The correlations between variables
range between .39 (Aad and similarity) and .63 (Aad and credibility).
4.2. Manipulation check
To confirm the difference in level of integration between our two
minorities, we analyzed respondents' answers to three questions:
use of a foreign language at home (M Black = 3.08, SD = 1.76;
M North African =3.83, SD= 1.47) foreign television channels watched
(M Black= 2.27, SD= 1.55; M North African = 3.14, SD= 1.36), and
number of own-ethnicity friends (M Black = 2.60, SD= 1.53; M North
African= 3.03, SD= 1.11), where higher scores indicate a lower level
4.3. Level of minority integration and advertising responses
In H1a and H1b, we predicted an interaction effect between respondent ethnicity and model ethnicity on Aad, such that a planned contrast
analysis should show different response patterns for the two minorities,
with only the less integrated minority displaying the target/non-target
market effect indicated in prior research. Therefore, we conducted an
ANOVA on Aad, with respondent ethnicity and model ethnicity as
between-subjects factors.
For Aad, the significant main effects of respondent ethnicity
(F(2, 446) = 14.72, p b .001) and model ethnicity (F(2, 446) = 8.40,
p b .001) were subsumed by the respondent ethnicity × model ethnicity interaction (F(4, 446) = 3.70, p b .01). As we expected, the
planned contrast analyses (Table 3) showed no significant differences across ads for White French respondents and asymmetrical
response patterns for the two minorities. In support of H1a and
H1b, the Black minority liked the Black and White model ads more
than the North African ad (H1a), whereas the North African minority
liked both minority ads more than the White model ad (H1b). Thus,
the dominant group and the least integrated minority displayed the
patterns of target and non-target market effects predicted by prior research, but the results for the more integrated minority deviated from
conventional predictions.
4.4. Driver of model ethnicity effects in advertising
In H2a and H2b, we predicted that model credibility, rather than
similarity, should mediate the interaction effect of respondent ethnicity and model ethnicity on attitude toward the ad (H2a) and
that for the better integrated minority, the credibility mediation
route should be dissociated from the effect of similarity (H2b). To
test the mediated moderation hypothesized in H2a, we estimated
two models (Muller, Judd, & Yzerbyt, 2005) and determined whether
the indirect path from the respondent ethnicity × model ethnicity interaction to Aad, through the proposed mediator, differed significantly
from zero (Zhao, Lynch, & Chen, 2010). Thus, we first regressed the proposed mediator on respondent ethnicity, model ethnicity, and respondent ethnicity× model ethnicity (Model 1) and then regressed Aad on
Table 3
Cell means and planned contrasts.
Less integrated minority
(North African)
• Attitude toward the ad
• Credibility
• Similarity
More integrated minority
(Black)
• Attitude toward the ad
• Credibility
• Similarity
Dominant group (White)
• Attitude toward the ad
• Credibility
• Similarity
Distinctive models
Non-distinctive model
North African
Black
White
4.38 (1.81)
4.49 (1.70)
2.48 (1.31)
4.54 (1.64)
4.46 (1.86)
2.17 (1.06)
3.47 (1.60)a
3.75 (1.62)a⁎
1.88 (.93)a
3.21 (1.86)b
3.16 (1.33)b
1.69 (.92)b
4.38 (2.00)
4.84 (1.56)
3.15 (1.18)
4.11 (1.61)
4.21 (1.60)
1.91 (.81) b
2.81 (1.32)
3.30 (1.23)
1.96 (.93)
3.64 (1.37)
4.66 (1.25)c
2.01 (.86) c
3.22 (1.31)
3.62 (1.31)
1.73 (.82)
Notes: All p values are less than .05, unless indicated otherwise. Significant differences
with the standard of comparison are identified by a, b, or c. Numbers in bold are the
standards for comparison, such that for each respondent's ethnic group, planned contrast
analyses compare the model that matches the respondents' ethnic group with the two
other models.
a
Significantly different from the North African model (a*: p= .056).
b
Significantly different from the Black model.
c
Significantly different from the White model.
386
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
respondent ethnicity, model ethnicity, respondent ethnicity× model
ethnicity, and the proposed mediator (Model 2).
When we estimated the Model 1 with credibility as a mediator, we
found that the respondent ethnicity × model ethnicity interaction significantly influenced perceived credibility (F(4, 446) = 5.47, p b .001).
With Model 2, we determined that the respondent ethnicity × model
ethnicity interaction was non-significant (F(4, 445)= 1.86, p > .10),
whereas the influence of credibility was positive and significant
(F(1, 445) = 259.47, p b .001). To test the significance of the indirect
effect, we used Preacher and Hayes's (2008) bootstrapping mediation
script (as recommended by Zhao et al., 2010), after recoding respondent
ethnicity and model ethnicity into two dummy variables each, such that
we created four two-way interaction terms with the products of the
dummy variables (see Table 4 for the coding). Preacher and Hayes's
(2008) SPSS macro can estimate the indirect effects of the respondent
ethnicity × model ethnicity interaction on Aad and the 95% confidence
interval around these estimates (5000 bootstraps). The results show
that the indirect effect for all four interaction terms was significant
(see Table 4), and thus the conditions for mediation were met (Zhao
et al., 2010). Credibility mediated the respondent ethnicity × model ethnicity interaction on Aad, in preliminary support of H2a.
To assess the mediating role of credibility versus similarity, we ran
the same analyses (Model 1 and Model 2 estimations) with similarity
and then pit the two competing mediators against each other in a single
model, using Preacher and Hayes's (2008) approach. In the Model 1 estimate, the respondent ethnicity × model ethnicity interaction significantly influenced perceived similarity (F(4, 446)= 10.223, p b .001). As
expected, exposure to models of one's own ethnic group translated
into improved judgments of similarity to the model among both minority groups (F(2, 242)= 10.57, p b .001). Black respondents felt significantly more similar (F(2, 119) = 26.09, p b .001) to the Black model
(M = 3.15, SD= 1.18) than to the North African (M = 1.69, SD= .92;
p b .001) and White (M = 1.91, SD= .81; p b .001) models. The North
African respondents felt significantly more similar (F(2, 120) = 2.95,
p b .05) to the North African (M = 2.48, SD= 1.30) and Black (M =
2.17, SD= 1.06; p >.05) models than to the White model (M = 1.88,
SD= .93; p b .05). For Black respondents though, the patterns for similarity and for credibility with Aad clearly varied (see Table 3). Black
respondents found the Black and White models more credible and
displayed more positive attitudes toward ads depicting Black and
White models. In contrast, they did not perceive similarity between
themselves and the White model. In addition, and in contrast with extant claims, Black respondents did not indicate feeling less similar to
the non-distinctive (White) model than to the distinctive (North African)
model, whereas North African respondents expressed more similarity to
distinctive (North African and Black) models than to the non-distinctive
models. These results offer preliminary support for H2b.
With the estimation of Model 2, we determined that the influence of
similarity was positive and significant (F(1, 445)= 62.104, p b .001), as
was the respondent ethnicity × model ethnicity interaction effect
(F(4, 445) = 3.399, p b .01). This latter significant influence implies
an incomplete theoretical framework (Zhao et al., 2010). Preacher
and Hayes's (2008) SPSS macro estimates for the indirect effect of
the respondent ethnicity × model ethnicity interaction on Aad show
that the indirect effects of two interaction terms were non-significant
(see Table 4), in further support of H2a. Similarity is not a “one-sizefits-all” mediator; that is, it does not work for all types of respondents
or models.
With Preacher and Hayes's (2008) approach, we can unravel the
unique ability of each putative mediator to mediate the respondent
ethnicity × model ethnicity interaction effect on Aad; this approach
compares the strength of specific indirect effects to determine which
Table 4
Mediation of the effect of respondent ethnicity × model ethnicity on attitudes toward the ad.
Mediator
Independent variable (respondent
ethnicity × model ethnicity)
Indirect path point estimate
Standard error
Confidence lower bound
Interval upper bound
1. Credibility (C)
Black respondent × Black model
Black respondent × White model
White respondent × Black model
White respondent × White model
Black respondent × Black model
Black respondent × White model
White respondent × Black model
White respondent × White model
1.112
1.166
.904
.690
1.006
.464
.201
.207
.311
.309
.277
.273
.214
.182
.158
.159
1.645a
1.675
1.369
1.141
1.377
.772
.467
.481
3. Contrast analysis C–S
Total indirect effect
Black respondent × Black model
Black respondent × White model
White respondent × Black model
White respondent × White model
Credibility
Black respondent × Black model
Black respondent × White model
White respondent × Black model
White respondent × White model
Similarity
Black respondent × Black model
Black respondent × White model
White respondent × Black model
White respondent × White model
Contrast C–S
Black respondent × Black model
Black respondent × White model
White respondent × Black model
White respondent × White model
.618
.673
.457
.246
.666
.169
−.048
−.044
1.430
1.258
.902
.711
1.018
1.070
.821
.629
.413
.188
.082
.082
.605
.882
.739
.547
.363
.335
.307
.291
.325
.304
.276
.256
.146
.090
.077
.073
.349
.298
.265
.240
.710
.619
.326
.145
.429
.492
.324
.132
.161
.047
−.042
−.038
−.028
.327
.273
.100
2.142
1.942
1.538
1.284
1.702
1.697
1.409
1.149
.748
.405
.272
.262
1.350b
1.491b
1.328b
1.052b
2. Similarity (S)
Notes: Confidence intervals for significant estimates are in bold. The standards of reference for both model and respondent ethnicity are North African model/respondents: for the
model ethnicity factor, Black model stands for Black model vs. North African model and White model stands for White model vs. North African model; for the respondent ethnicity
factor, Black respondent stands for Black respondent vs. North African respondent and White respondent stands for White respondent vs. North African respondent.
a
This significant indirect effect indicates that the effect of portraying a Black model vs. a North African model differs depending on whether the respondent is Black vs. North
African, and this effect is mediated by credibility.
b
The contrast results can be interpreted as follows: the effect of portraying a White model vs. a North African model differs depending on whether the respondent is Black vs.
North African; the effect of portraying a Black model vs. a North African model differs depending on whether the respondent is White vs. North African; and the effect of portraying
a White model vs. a North African model differs depending on whether the respondent is White vs. North African. These effects are mediated by credibility rather than similarity.
The effect of portraying a Black model vs. a North African model differs depending on whether the respondent is Black vs. North African, and this effect is equally well mediated by
credibility or by similarity.
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
mediation theory is dominant. We used Preacher and Hayes's SPSS
macro to estimate the total and specific (i.e., through credibility versus
through similarity) indirect effects of the respondent ethnicity × model
ethnicity interaction on Aad, as well as the contrast value (i.e., indirect
effect through credibility versus indirect effect through similarity) and
the 95% confidence interval around these estimates (5000 bootstraps).
The contrasts compared the two mediators' unique abilities to mediate
the respondent ethnicity × model ethnicity interaction effect on Aad,
beyond any other mediators or covariates in the model. The results,
which we summarize in Table 4, show that credibility was a significantly better mediator of three interaction terms and equally good for the
fourth (“contrast C–S”). Therefore, credibility has more credence than
similarity as a mediator of the respondent ethnicity× model ethnicity
interaction effect on Aad, in support of H2a.
Also in accordance with H2b, the Black respondent × White model
indirect effect, mediated by credibility rather than similarity, highlights the dissociation between similarity and credibility for better integrated minorities; if integration did not matter, this effect would
have been equally well-mediated by credibility and similarity. This
is because similarity and credibility mediation routes are confounded
for less integrated minorities, whereas credibility and similarity show
a clear dissociation in their pattern of influence for better integrated
minorities.
5. Discussion
We put forth a fine-grained account of differences across minority
groups to provide a better theoretical understanding of the effect of
ethnic model portrayals on advertising effectiveness. This issue is
important from a practical standpoint because reliance on faulty
assumptions about how minorities perceive ethnic and mainstream
models could lead advertisers to select suboptimal models for their
campaigns. We used a 3 (respondent ethnicity) × 3 (model ethnicity)
between-subject design to raise and examine multi-ethnic advertising issues and to contribute on both theoretical and managerial levels.
In particular, we show that ethnic minorities who differ in their
level of integration into the host society do not display the same response patterns when exposed to ads that portray distinctive and
non-distinctive models. Compared with relatively less integrated minorities, ethnic minorities who are well embedded within the host
culture respond in ways that indicate their shift toward the responses
of the majority. For less integrated minorities, ads portraying a distinctive model, regardless of the model's ethnicity, are the most effective, in line with prior target marketing literature. In contrast, for
more integrated minorities, ads exhibiting a distinctive model are
least effective if the model belongs to a less integrated minority.
These differences offer an important theoretical contribution to
existing advertising literature, which so far has implicitly assumed that
all minorities react similarly when exposed to advertising models. This
study offers a first step toward engaging the complexity of consumer
identities beyond single categorical labels. The next step is to consider individual differences among members of minorities; for example, how do
better integrated members of less integrated minorities react to distinctive models? Although we do not explore this within-group difference, it
remains an important conceptual issue worthy of further research.
Because our study also highlights how perceived model credibility
mediates the effect of ethnic models (majority and minority) on attitudes toward the ad, we offer two additional contributions. First, we
show that the level of integration of ethnic minorities into the host
society truly matters. We empirically demonstrate that the common
“minority versus majority” categorization (e.g., Bush et al., 1999) is an
oversimplification of reality. Treating African Americans and Asians in
the United States or South Americans and North Africans in Spain as
undifferentiated minority groups likely has negative consequences for
advertisers. In France, exposing members of the Black minority to ads
portraying North African models prompted more negative responses.
387
Therefore, a deeper understanding of the integration levels of ethnic minorities in any geographical area is an important prerequisite before
approaching specific markets for segmentation, communication, or advertising efforts. Further research should increase the precision of measures of integration over time (e.g., Arnould et al., 2004), especially as
the number of people of mixed heritage continues to rise (Coleman,
2008).
Second, our study significantly increases the theoretical precision of
the drivers of ad model effectiveness that result from exposure to minority models (e.g., Brumbaugh, 2002; Bush et al., 1999). A sole consideration of similarity to the model, as observed in current literature, is the
wrong approach when a minority member is better integrated into the
dominant society. We challenge the importance of perceived similarity;
model credibility better explains the persuasive capability of ethnic
models and allows for a more parsimonious modeling of ethnic models'
influence in advertising. It might be challenging for advertisers to select
and portray credible models rather than merely building on similarity.
However, because credibility is the true mediation route traveled by all
ethnic (majority or minority) groups, it also enables advertisers to comprehend a market's multi-ethnic reality more parsimoniously.
In this study, we did not formulate hypotheses about the specific
pattern of means for White respondents. However, as Table 3 shows,
White viewers give higher similarity and credibility ratings to the
Black model than to the other models (White and North African).
At least two other studies also have found unexpected higher scores
for ads portraying a Black model by White viewers (Appiah, 2004;
Brumbaugh, 2002). Although White viewers may not perceive themselves as similar to the Black model on the basis of ethnicity, they
may identify with the Black model on stereotypical characteristics
associated with Black individuals that fascinate them (e.g., urban,
cool, musical, athletic, dance, hip-hop fashion) (Appiah, 2004). This
explanation would be worthwhile to test in further research.
The process of social comparison related to body shape is another
process that might be worth examining in further research. Models
depicted in ads may have a lower body mass index than that of the
viewers, which would negatively affect their self-esteem (Smeesters,
Mussweiler, & Mandel, 2010) and, in turn, their attitude toward the
ad. Therefore, introducing factors such as the body mass index might
be valuable, especially if ethnic differences in actual or ideal body
mass exist.
We also acknowledge some limitations. In particular, our study includes only French women, a single product category, and ads portraying
only one ethnic group; further studies should try to replicate our findings
using different settings.
Our study fits into a larger stream of research studying how individual differences associated with the viewer might moderate the
impact of identity portrayals (e.g., Reed, 2004). Specifically, it adds
to a growing body of work that investigates how individual differences influence an ethnic spokesperson's effectiveness among different ethnic groups (e.g., Grier & Deshpandé, 2001). This stream of
research could be expanded to incorporate identity markers other
than ethnicity, as well as the potential interaction effects between
these markers. Further research could study how bases of identification,
such as social roles (e.g., being a mother), personal traits or characteristics (e.g., being extraverted, successful, or attractive), or moral values
(e.g., religion) moderate the relative effectiveness of ethnic models, in
addition to the level of integration. For example, portraying a model
from a minority enjoying a low level of integration might not be more
effective among the minority group than portraying a model from the
dominant group if the role the model portrays (e.g., a self-centered professional woman versus a caring mother) is at odds with the values of
the minority group. In this case, it might be preferable to portray a
model in that role from another minority (because less integrated minority viewers prefer distinctive models, from their minority or another
minority, to models from the dominant group). Viewers' religious beliefs might offer another interesting avenue for further investigation.
388
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
For example, the display of (provocative) nudity or explicit or implicit
encouragement of unwelcome sentiments, such as dissatisfaction with
one's body or vanity (Fam, Waller, & Erdogan, 2004), could offend
viewers with more entrenched religious faith and affect their attitudes
toward ethnic advertising. This effect may be further strengthened by
the minority's relative lack of integration, especially if such displays
are well-accepted by the dominant group.
In summary, by adopting a more fine-grained and detailed perspective, the present study offers new insights into the effects of ethnicity in
advertising. These insights are both socially relevant and managerially
significant. Despite some challenges that are inherent to the study of
sensitive topics such as ethnicity and self-identity, this article demonstrates an effective means to broach such subjects, as well as offering
some ideas for further research.
Acknowledgments
The authors thank Hua Li and Gilles Marion for their comments on
previous drafts of this manuscript. We are also very grateful to
Stefano Puntoni and Luk Warlop, as well as the reviewers, for their
helpful and valuable suggestions.
References
Aaker, J., Brumbaugh, A., & Grier, S. (2000). Non-target markets and viewer distinctiveness:
The impact of target marketing on processing and attitudes. Journal of Consumer
Psychology, 9(3), 127–140.
Allen, V. L., & Wilder, D. A. (1980). Impact of group consensus and social support on
stimulus meaning: Mediation of conformity by cognitive restructuring. Journal of
Personal and Social Psychology, 39(6), 1116–1125.
Appiah, O. (2004). It must be the cues: Racial differences in adolescents' responses
to culturally embedded ads. In J. D. Williams, W. -N. Lee, & C. P. Haugtvedt (Eds.),
Diversity in advertising (pp. 319–339). Mahwah, NJ: Erlbaum.
Arnould, E., Price, L., & Zinkhan, G. (2004). Consumers (2d ed.). New York: McGraw-Hill.
Autorité de Régulation Professionnelle de la Publicité (2006). Diversité ethnique en
publicité—Etude sur l'année publicitaire 2005. Retrieved from. http://www.calameo.
com/read/0003132918abd24f57968?authid=VFeWVFqrhJC1
Baird, A. L., & Grieve, F. G. (2006). Exposure to male models in advertisements leads to a decrease in men's body satisfaction. North American Journal of Psychology, 8(1), 115–121.
Banister, E. N., & Hogg, M. K. (2004). Negative symbolic consumption and consumers'
drive for self-esteem. European Journal of Marketing, 38(7), 850–868.
Brislin, R. W. (1980). Translation and content analysis of oral and written materials. In
H. C. Trandis, & J. W. Berry (Eds.), Handbook of cross-cultural psychology. Boston,
MA: Allyn & Bacon.
Brown, A. S., Brown, L. A., & Zoccoli, S. L. (2002). Repetition-based credibility enhancement
of unfamiliar faces. The American Journal of Psychology, 115(2), 199–209.
Brumbaugh, A. M. (2002). Source and nonsource cues in advertising and their effects
on the activation of cultural and subcultural knowledge on the route to persuasion.
Journal of Consumer Research, 29(2), 258–269.
Brumbaugh, A. M., & Grier, S. A. (2006). Insights from a “failed” experiment: Directions
for pluralistic, multiethnic advertising research. Journal of Advertising, 35(3), 35–47.
Bush, A. J., Smith, R., & Martin, C. (1999). The influence of consumer socialization variables
on attitude toward advertising: A comparison of African-Americans and Caucasians.
Journal of Advertising, 28(3), 13–24.
Chattaraman, V., Lennon, S. J., & Rudd, N. A. (2010). Social identity salience: Effects on
identity-based brand choice of Hispanic consumers. Psychology and Marketing,
27(3), 263–284.
Clanché, F. (2002). Langues régionales, langues étrangères : De l'héritage à la pratique »,
Insee première, n° 830 (21/02/2002).
Coleman, D. A. (2008). New Europe, new diversity. Population Studies, 62(1), 113–120.
Collins, C. A., & Williams, D. R. (1999). Segregation and mortality: The deadly effects of
racism. Sociological Forum, 14(3), 493–521.
Deshpandé, R., & Stayman, D. M. (1994). A tale of two cities: Distinctiveness theory and
advertising effectiveness. Journal of Marketing Research, 31(1), 57–64.
Dimofte, C. V., Forehand, M. R., & Deshpandé, R. (2003). Ad schema incongruity as elicitor
self-awareness and differential advertising response. Journal of Advertising, 32(4),
246–254.
Diplômatie Française (2011). Document cadre de partenariat France—Haïti-DCP—
(2008–2012). Available from. http://www.diplomatie.gouv.fr/fr/pays-zones-geo_
833/haiti_513/france-haiti_1218/evenements_4655/document-cadre-partenariatfrance-haiti-dcp-2008-2012_59439.html
Entorf, H., & Minoiu, N. (2005). What a difference immigration policy makes: A comparison of PISA Scores in Europe and traditional countries of immigration. German
Economic Review, 6(3), 355–376.
Fam, K. S., Waller, D. S., & Erdogan, B. Z. (2004). The influence of religion on attitudes
towards the advertising of controversial products. European Journal of Marketing,
38(5/6), 537–555.
Le Figaro (2010). Pourquoi la France compte 27% de mariages mixtes (21/06/2010).
Retrieved from. http://www.lefigaro.fr/actualite-france/2010/01/21/0101620100121ARTFIG00021-pourquoi-la-france-compte-27-de-mariages-mixtes-.php
Frachon, C., & Sassoon, V. (2008). Médias et diversité: De la visibilité aux contenus: etats
des lieux en France, au Royaume-Uni, en Allemagne, et aux Etats-Unis. Paris: Panos.
Garrett, M., & Pichette, E. F. (2000). Red as an apple: Native American acculturation and
counseling with or without reservation. Journal of Counseling and Development,
78(1), 3–13.
Geddes, A. (2003). France: Still the one and indivisible republic? In A. Geddes (Ed.), The
politics of migration and immigration in Europe (pp. 52–78). London: Sage.
Grier, S. A., & Brumbaugh, A. M. (1999). Noticing cultural differences: Ad meanings created
by target and non-target markets. Journal of Advertising, 18(1), 79–93.
Grier, S. A., & Brumbaugh, A. M. (2007). Compared to whom? The impact of status on
third person effects in advertising persuasion in a South African context. Journal
of Consumer Behavior, 6(5), 5–18.
Grier, S. A., & Deshpandé, R. (2001). Social dimensions of consumer distinctiveness:
The influence of social status on group identity and advertising persuasion. Journal
of Marketing Research, 38(2), 216–224.
Grosjean, P. (2011). The weight of history on European cultural integration: A gravity
approach. American Economic Review, 101(3), 504–508.
Gwinner, K. P., & Eaton, J. (1999). Building brand image through sponsorship: The role
of image transfer. Journal of Advertising, 28(4), 47–57.
Hamberger, A. (2009). Immigrant integration: Acculturation and social integration.
Journal of Identity and Migration Studies, 3(2), 2–21.
Hogg, M. A., Terry, D. J., & White, K. M. (1995). A tale of two theories: A critical comparison
of identity theory with social identity theory. Social Psychology Quarterly, 58(1),
255–269.
Holbrook, M. B., & Batra, R. (1987). Assessing the role of emotions as mediators of consumer
responses to advertising. Journal of Consumer Research, 14(3), 404–420.
Hunt, M. (2000). Status, religion, and the “belief in a just world”: Comparing African
Americans, Latinos, and whites. Social Science Quarterly, 81(1), 326–343.
Independent (2009). France's crisis of national identity. Retrieved from. http://www.
independent.co.uk/news/world/europe/frances-crisis-of-national-identity-1826942.
html
INSEE (2005). Les immigrés en France: Une situation qui évolue. Retrieved from. http://
www.insee.fr/fr/ffc/docs_ffc/ip1042.pdf
Johnson, G. D., & Grier, S. A. (2011). Targeting without alienating: Multicultural advertising and the subtleties of targeted advertising. International Journal of Advertising,
3(2), 233–258.
Junger-Tas, J. (2001). Ethnic minorities, social integration and crime. European Journal
on Criminal Policy and Research, 9(1), 5–29.
Kantar Media (2011). Barométre des investissements publicitaires en 2010. ADEX Report.
Klucharev, V., Smidts, A., & Fernandez, G. (2008). Brain mechanisms of persuasive
communication: How “expert power” modulates memory and attitudes. Social
Cognitive and Affective Neuroscience, 3(4), 353–366.
Lacoste, Y. (2010). La question post-coloniale: Une analyse géopolitique. : Editions Fayard.
Laroche, M., Kim, C., & Tomiuk, M. A. (1998). Italian ethnic identity and its relative impact on
the consumption of convenience and traditional foods. Journal of Consumer Marketing,
15(2), 125–151.
Lee, C. K. C., Fernandez, N., & Martin, B. A. S. (2002). Using self-referencing to explain
the effectiveness of ethnic minority models in advertising. International Journal of
Advertising, 21(3), 363–375.
Lenski, G. E. (1954). Status cristallization: A non-vertical dimension of social status.
American Sociological Review, 19(4), 405–413.
Lindstrom, C. (2005). European Union policy on asylum and immigration. Addressing
the root causes of forced migration: A justice and home affairs policy of freedom,
security and justice? Social Policy & Administration, 39(6), 587–605.
Marthaler, S. (2008). Nicolas Sarkozy and the politics of French immigration policy.
Journal of European Public Policy, 15(3), 382–397.
Martin, B. A. S., Lee, C. K. -C., & Yang, F. (2004). The influence of ad model ethnicity and
self-referencing upon attitudes. Journal of Advertising, 33(4), 27–37.
McGuire, W. J., McGuire, C. V., Child, P., & Fujioka, T. (1978). Salience of ethnicity in the
spontaneous self-concept as a function of one's ethnic distinctiveness in the social
environment. Journal of Personality and Social Psychology, 36, 511–520.
Muller, D., Judd, C. M., & Yzerbyt, V. Y. (2005). When moderation is mediated and mediation is moderated. Journal of Personality and Social Psychology, 89(6), 852–863.
O'Keefe, D. J. (1991). Persuasion: Theory and research. Newbury Park, CA: Sage.
Oakes, P. J. (1987). The salience of social categories. In J. C. Turner, M. A. Hogg, P. J.
Oakes, S. D. Reicher, & M. S. Wetherell (Eds.), Rediscovering the social group
(a self-categorization theory) (pp. 117–141). Oxford, England: Basil Blackwell.
Ouellet, J. -F. (2007). Consumer racism and its effects on domestic cross-ethnic product
purchase: An empirical test in the United States, Canada, and France. Journal of
Marketing, 71(1), 113–128.
Phinney, J. S. (1990). Ethnic identity in adolescents and adults: Review of research. Psychological Bulletin, 4(3), 499–514.
Pornpitakpan, C. (2004). The persuasiveness of source credibility: A critical review of
five decades' evidence. Journal of Applied Social Psychology, 34(2), 243–281.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing
and comparing indirect effects in multiple mediator models. Behavior Research
Methods, 40(3), 879–891.
Puntoni, S., Vanhamme, J., & Visscher, R. (2011). Two birds and one stone: Strategic
ambiguity in advertising targeting and advertising evaluations. Journal of Advertising,
40(1), 25–42.
Reed, Americus, II (2004). Activating the self-importance of consumer selves: Exploring identity salience effects on judgments. Journal of Consumer Research,
31(2), 286–295.
M. Antioco et al. / Intern. J. of Research in Marketing 29 (2012) 380–389
République Française (2011). Les accords de gestion concertée des flux migratoires et de
développement solidaire. Retrieved from. http://www.immigration.gouv.fr/spip.
php?page=dossiers_det_dev&numrubrique=216&numarticle=1359 (accessed
November 24, 2011).
Rosenbaum, M. S., & Montoya, D. (2007). Exploring the role of ethnicity in place avoidance
and approach decisions. Journal of Business Research, 60, 206–213.
Rossiter, J. R., & Smidts, A. (2011). Print advertising: Celebrity presenters. Journal of
Business Research, 65, 874–879.
Ryder, A., Alden, L., & Paulhus, D. (2000). Is acculturation unidimensional or bidimensional?
A head-to-head comparison in the prediction of personality, self-Identity, and adjustment. Journal of Personality and Social Psychology, 79(1), 49–65.
Simons, H. W., Berkowitz, N. N., & Moyer, J. (1970). Similarity, credibility, and attitude
change: A review and theory. Psychological Bulletin, 73(1), 1–16.
Smeesters, D., Mussweiler, T., & Mandel, N. (2010). The effects of thin and heavy media
images on overweight and underweight consumers: Social comparison processes
and behavioral implications. Journal of Consumer Research, 36(6), 930–949.
Solomon, M. (2007). Consumer behavior (7th ed.). Upper Saddle River, NJ: Prentice Hall.
Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase
intentions. Journal of Current Issues and Research in Advertising, 26(2), 53–66.
Sudbury, L., & Wilberforce, F. (2006). The portrayal of black people in UK television
advertising: Perception and reality. Journal of Consumer Behaviour, 5(5), 465–476.
389
Tréguer, J. P., & Ségati, J. M. (2003). Les nouveaux marketings: Marketing générationnel et
autres marketing catégoriels. Paris: Dunod.
Wallendorf, M., & Reilly, M. D. (1983). Ethnic migration, assimilation, and consumption. Journal of Consumer Research, 10(3), 292–302.
Webster, C. (1992). The effects of Hispanic subcultural identification on information
search behavior. Journal of Advertising Research, 32(5), 54–62.
White, K., & Dahl, D. W. (2006). To be or not be: The influence of dissociative reference
groups on consumer preferences. Journal of Consumer Psychology, 16(4), 404–413.
Whittler, T. E. (1991). The effect of actors' race in commercial advertising: Review and
extension. Journal of Advertising, 20(1), 54–60.
Whittler, T. E., & Spira, J. S. (2002). Model's race: A peripheral cue in advertising messages?
Journal of Consumer Psychology, 12(4), 291–301.
Williams, J. D., & Qualls, W. J. (1989). Middle-class black consumers and intensity of
ethnic identification. Psychology and Marketing, 6(4), 263–286.
Wood, W., Pool, G. J., Leck, K., & Purvis, D. (1996). Self-definition, defensive processing,
and influence: The normative impact of majority and minority groups. Journal of
Personality and Social Psychology, 71(6), 1181–1193.
Xerfi (2010). Parfums, savons, et produits d'entretiens: Prévision 2011.
Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and
truths about mediation analysis. Journal of Consumer Research, 37(2), 197–205.
Intern. J. of Research in Marketing 29 (2012) 390–394
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
How to say “no”: Conviction and identity attributions in persuasive refusal
Vanessa M. Patrick a,⁎, Henrik Hagtvedt b
a
b
C. T. Bauer College of Business, University of Houston, Houston, TX 77204-6021, USA
Carroll School of Management, Boston College, MA 02467, USA
a r t i c l e
i n f o
Article history:
First received in 29, June 2011 and was under
review for 6 months
Available online 14 September 2012
Keywords:
Consumer identity
Language
Persuasion
Interpersonal communication
a b s t r a c t
This research investigates the influence of refusal frames on persuasiveness in an interpersonal context. Specifically, the refusal frame “I don't” is more persuasive than the refusal frame “I can't” because the former connotes conviction to a higher degree. This perceived conviction is tied to the identity-signaling function of the
refusal frame. Two studies demonstrate that 1) the “don't” frame is more persuasive than the “can't” frame,
2) perceived conviction mediates the influence of the refusal frame on persuasiveness, and 3) attributions to
the refuser's identity explain the perceived conviction.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
No man is an island. To function in society, we must interact,
negotiate, and communicate with other people. A large part of that
interaction involves conversing effectively, not only to get what we
want but to avoid what we do not want.
One of the key factors that determines whether a refusal is effective is compliance from others (Cialdini & Goldstein, 2004). For
instance, anecdotal evidence from attendees of the Weight Watchers
program reveals that one of the key reasons why individuals gain
weight is because their friends, family, and spouses tempt them
with unhealthy food and they do not know how to refuse when
confronted with this type of social pressure. From weight loss programs to Nancy Reagan's famous “Just Say No” campaign against
drugs, individuals are often advised to say no to temptation. However,
little research focuses on exactly how to do so effectively.
In the current research, we propose that the persuasiveness of a
refusal depends on how it is framed. Recent research demonstrates
that saying “I don't” is more empowering than saying “I can't” in the
context of self-talk (Patrick & Hagtvedt, 2012). In the present
research, we investigate the effectiveness of these two refusal frames
in interpersonal communication. We theorize that when a refusal
evokes one's identity, it speaks of one's traits, characteristics, goals,
and beliefs (Oyserman, 2009). This attribution increases the perceived conviction of the refuser and consequently enhances the
persuasiveness of the refusal. Specifically, we demonstrate that
when refusing others 1) the “don't” frame is more persuasive than
the “can't” frame, 2) this persuasiveness stems from the conviction
⁎ Corresponding author. Tel.: +1 713 743 3661; fax: +1 713 743 4572.
E-mail addresses: vpatrick@uh.edu (V.M. Patrick), hagtvedt@bc.edu (H. Hagtvedt).
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.04.004
conveyed by the frame, and 3) the perceiver attributes this conviction
to the refuser's identity.
For example, if a hostess offers a rich slice of cake to a guest who
declines by saying “I don't eat cake,” we argue that the refusal would
engender the hostess’ compliance and not typically invite negotiation.
It reflects who the person is and what he or she stands for. However, if
the guest declines by saying “I can't eat cake,” this statement reflects
an impediment or inability, and thus, it connotes less force or determination. In fact, it may connote external concerns that are often relied
upon in excuse-making (Snyder & Higgins, 1988) and therefore is
more likely to provoke discussion, debate, and negotiation.
This research contributes to theory in five key ways: 1) It contributes
to the extant literature on consumer identity by illustrating how
language, in terms of the actual words used, can signal consumer identity; 2) It contributes to the literature on language in social interactions
to demonstrate how subtle differences in language can have a profound
influence on the outcome of the interaction; 3) It contributes to the
literature on compliance-gaining and negotiation by demonstrating
the effectiveness of refusal frames in an interpersonal context; 4) It contributes to the emerging literature on semantic framing (as distinct
from logically equivalent framing effects in the context of choice
under uncertainty; Levin & Gaeth, 1988), in which two words used
interchangeably in conversation can have substantially different effects;
5) Because the current work is prescriptive, it provides recommendations for how individuals can adopt refusal frames to more effectively
navigate interpersonal interactions in everyday life.
2. The power of words: a brief review of the literature
Whorf (1956) proposed that there is no thought without language
and suggested that individuals' perception of the world and their
V.M. Patrick, H. Hagtvedt / Intern. J. of Research in Marketing 29 (2012) 390–394
interaction with it are based on the linguistic system that they use.
Indeed, the semantic meaning of words depends on when and how
the words are used (Chafe, 1970; Grice, 1975). In the context of social
interactions, Pennebaker, Mehl, and Niederhoffer (2003, p. 566)
assert that “remarkably little research has been conducted on the
ways the interactants use words with each other.” However, a distinction between linguistic content (what is said) and linguistic style
(how it is said) can be seen in, for instance, two ways to make a
simple request. “Would it be possible for you to pass me the butter?”
and “Pass the butter” both express the speaker's desire for butter
and direct the listener's action, but the two utterances also reveal
different features of the interactants' relationship, the speaker's personality, and perhaps the way the speaker understands himself.
Indeed, a growing body of literature has demonstrated that listeners
make inferences about the attitudes and characteristics of speakers
based on the words those speakers employ. For instance, Douglas and
Sutton (2005) find that describers who use relatively abstract language
are perceived by others to have biased attitudes and motives compared
with describers who use more concrete language. In the context of
favorable (unfavorable) word of mouth, Schellekens, Verlegh, and
Smidts (2010) demonstrate that abstract language leads receivers to
infer that the sender has a more (less) favorable product attitude and
a higher (lower) buying intention for the product under consideration.
It does not seem unreasonable, then, that listeners can also infer aspects
such as conviction and identity based on language use.
2.1. Semantic framing
Extant research has examined the role played by message framing in
persuasion processes (Mayer & Tormala, 2010), especially in
health-related communications (Rothman, Salovey, Antone, Kelli, &
Drake, 1993; Tykocinski, Higgins, & Chaiken, 1994). Language use has
been shown to impact domains as disparate as sensory perception
(Hoegg & Alba, 2007), categorical perception (Schmitt & Zhang, 1998),
perceived durations of time intervals (Cheema & Patrick, 2008), health
outcomes (Campbell & Pennebaker, 2003), and even the ability to distinguish colors (Athanasopoulos, Damjanovic, Krajciova, & Sasaki,
2011). In the research on duration estimates, for example, framing
deadlines one way (“only three months to finish”) made individuals
feel subjectively closer to that deadline and highlighted the difficulty
of completion compared to framing the deadline a different way (“still
have three months to finish”), keeping the objective amount of time
constant (Sanna, Parks, Chang, & Carter, 2005). Similarly, Cheema and
Patrick (2008) demonstrate that the framing of a time interval for coupon redemption as “anytime between” or “only between” significantly
influences coupon redemption behavior. The current research contributes to an emerging area of work that investigates semantic framing
in which words that are used interchangeably “may not always be psychologically interchangeable” (LeBoeuf, 2006, 60) and can have profoundly different influences on behavioral outcomes.
2.2. The role of conviction in persuasive refusal
The present research is concerned with the effective framing of a refusal in an interpersonal context, the assumption being that effectiveness is indicated by compliance with the refusal. In other words, if
X says “no” and Y complies, the refusal is deemed effective. A great
deal of prior research on compliance-gaining strategies has focused on
how to influence the actions of others. Overarching themes such as
“promise” (If you do this, I will reward you) or “threat” (If you do not
do this, you will be punished) reflect common compliance-seeking
strategies (Marwell & Schmitt, 1967; Schenck-Hamlin, Wiseman, &
Georgacarakos, 1982). For a summary of compliance-gaining and
compliance-resisting strategies, see Wilson (2002). While broad
themes have been identified, the use of specific words to gain compliance represents a gap in the extant literature.
391
Patrick and Hagtvedt (2012) identify the words “don't” and “can't”
as differentially effective in resisting temptation and motivating goaldirected behavior because the former provides a sense of psychological
empowerment. For instance, study participants concerned with healthy
eating were more likely to choose granola health bars (vs. chocolate
candy bars) if they engaged in self-talk by saying they “don't”
(vs. “can't”) eat unhealthy foods. In the current context of interpersonal
communication, we theorize that the “don't” frame is more persuasive
than the “can't” frame because the former conveys conviction to a higher
degree. As a speech act, “don't” conveys more force or determination and
is therefore more final and non-arguable (Austin, 1975; Searle, 1969/
1999). While expressed commitment to a cause, attitude, or behavior
encourages individuals to be swayed by requests in line with that
commitment, the same commitment signals the unlikelihood of being
swayed by requests that are incompatible with it (Cialdini & Goldstein,
2004). Indeed, a determined stance thus tends to discourage argument
or negotiation and increases compliance with the refusal (Cialdini &
Goldstein, 2004; Patrick & Hagtvedt, 2012; Searle, 1969/1999).
2.3. The role of identity in perceived conviction
One way to convey conviction is to provide a reason for the refusal. As
Cialdini (2006) suggests, sometimes providing a reason that makes no
sense is more effective than not providing a reason at all. Reasons for refusal typically fall into two basic categories: those stemming from within
oneself and those stemming from outside oneself (Ross & Nisbett, 2011).
We propose that the former tends to convey more conviction.
A great deal of research has shown that consumers are likely to behave in accordance with their identities, even if the identity is primed
by something as simple as a signature (Kettle & Häubl, 2011). They
may even distance themselves from others who do not resonate with
their own identity (Berger & Heath, 2007; White & Dahl, 2007). One's
identity is deeply entrenched; it is not something one changes on a
whim. If a refusal is tied to one's identity, then it reflects conviction:
a committed, unwavering stance that does not invite negotiation. Conversely, when external causes are invoked, they come across as excuses
(Snyder & Higgins, 1988), thereby inviting discussion and negotiation.
Much extant research in sociolinguistics supports the notion that our
choice of language may be used to construct and signal identity (Eckert,
2000). This stream of research has roots in early work that revealed that
even seemingly trivial indicators such as sound changes could serve as
symbols of local identity (Labov, 1963). In research on voting behavior,
Bryan, Walton, Rogers, and Dweck (2011) demonstrate that invoking
the self is sufficient to significantly influence whether individuals vote
in presidential and gubernatorial elections. In their case, the manipulation of identity relied on framing questions about voting behavior as a
verb or a noun (e.g., “How important is it to you to vote (be a voter) in
the upcoming election?”). Nouns were expected to invoke identity to
a higher degree, and indeed, they encouraged voting to a higher degree.
In the current context, we suggest that attributions to the refuser's
identity result in “don't” conveying a strong sense of conviction. The
“I don't” refusal frame emphasizes that one simply does not do something or is not the kind of person to make that choice. It thus reflects a
permanent state. Conversely, the “I can't” refusal frame emphasizes
impediments or inabilities that may be temporary and not permanently linked to one's identity. We propose the following hypotheses
based on the above arguments:
H1. The “don't” refusal frame is more persuasive than the “can't”
refusal frame.
H2. Perceived conviction mediates the influence of refusal frame on
persuasiveness.
H3. Attributions to identity mediate the influence of refusal frame on
perceived conviction.
392
V.M. Patrick, H. Hagtvedt / Intern. J. of Research in Marketing 29 (2012) 390–394
3. Study 1: persuasive refusal and the role of conviction
3.1. Method and procedure
Table 1
Studies 1 and 2: ANOVA results for the influence of refusal frames on dependent
variables.
Study 1
The objective of this experiment was to investigate the effectiveness of identity-based refusal framing in an actual interpersonal context. The study was focused on the domain of personal finance.
Forty-four undergraduate students (male = 63%) participated in this
role-playing experiment as part of a lecture on professional selling.
All of the participants were told that they would be participating in a
role-playing exercise in which they were to make a sales pitch to a
prospective customer. Their task was to sell an annual subscription
of the local newspaper. They were provided with information about
the newspaper deal (frequency, price, discount rates, etc.). The participants were told that the “first few minutes of the negotiation determines the success of the sale,” so they had approximately 3 min to
make their pitch. Each participant individually made a sales pitch to
a prospective customer. The prospective customers were confederates
of the experimenter. There were five of these confederates (two male
and three female), who were told that they had a goal of saving money
and that they were to use the given strategies to refuse any attempt by
the salesperson to sell them the subscription. Each confederate was
given a script that acted as a rough guideline regarding how the refusals were to be handled during the actual sales negotiation experiment. Each confederate was individually trained by one of the
authors to rehearse their refusals. The script included suggestions
such as “I don't/can't make spontaneous decisions about money,” or
“I don't/can't spend money without checking my budget first.”
Each participant was randomly assigned to one of five confederates who refused the salesperson's attempts with either the “don't”
or the “can't” framing. Neither party was informed of the goals of
the study. After the dyad completed the sales negotiation, the respondents were given a feedback questionnaire in which they evaluated
the prospective customer. Specifically, they reported how convinced
they were by the prospective customer's response (1 = not at
all, 9 = extremely), the degree of effectiveness of the response
(1 = low degree, 9 = high degree), and how likely they would be
to respect the customer's response (1 = not at all, 9 = extremely);
these items were later combined in a persuasion index: α = .75. The
participants also evaluated the customer in terms of how determined,
committed, and confident they were (1 = not at all, 9 = very much;
later combined in a conviction index: α = .88).
3.2. Results
A one-way ANOVA with the persuasion index as the dependent
variable revealed the predicted main effect of framing (Mdon't = 7.05,
SD= 1.32 vs. Mcan't = 5.80, SD= 1.70, F(1, 42) = 7.33, p b .05, η2 = .15).
This result supports H1.
A similar ANOVA with the conviction index as the dependent variable revealed a similar pattern of results (Mdon't = 7.06, SD = 1.64 vs.
Mcan't = 5.52, SD = 1.54, F(1, 42) = 10.38, p b .05, η 2 = .19); see also
Table 1. Bootstrap estimation (Preacher & Hayes, 2004) with 5000
resamples, as well as a Sobel test, confirmed that conviction mediates
the influence of framing on persuasion (M = − .97, SE = .40,
95% CI = − 1.84, − .29; Sobel test: z = −2.71, p b .01). This result
supports H2.
4. Study 2: persuasive refusal and the role of identity
This study replicates Study 1 in a laboratory experiment in the
health domain. An additional goal was to demonstrate that the influence of the “don't” framing on conviction stems from attributions to
the refuser's identity.
Study 2
Don't
M
Identity attrib.
Conviction
Persuasiveness
7.06⁎
7.05⁎
Can't
SD
1.64
1.32
M
5.52⁎
5.80⁎
Don't
Can't
SD
M
SD
M
SD
1.54
1.70
6.21⁎
6.80†
6.76⁎
1.72
1.61
1.43
5.17⁎
5.63†
5.89⁎
1.74
2.08
1.55
All means within each study significantly different across rows.
⁎ p b .05.
†
p b .001.
4.1. Pretest: refusal frames as identity signals
As a pretest, 60 participants completed a survey designed to test
whether the words “I don't” are a stronger signal of identity in the
current context than the words “I can't.” The participants were told
that one can use different words or phrases when refusing to do
something and that the purpose of this study was to understand the
language of refusal. They were asked to indicate the extent to which
the two phrases “I don't” and “I can't” connote identity. They completed 3 sets of items with both the “don't” phrase and the “can't”
phrase within-subjects on 9-point scales anchored by 1 = not at all
and 9 = definitely. The items were as follows: “Saying “I don't/
can't” do something reflects who one is as a person,” “Saying “I
don't/can't” do something reflects one's identity,” and “Saying “I
don't/can't” do something reflects one's beliefs as an individual.”
The three “don't” items were combined to form an index (α = .89),
and the three “can't” items were combined to form an index (α =
.86). A within-subjects means analysis revealed that the phrase
“I don't” is a strong signal of identity compared to the phrase
“I can't” (Mdon't = 7.12, SD = 1.55 vs. Mcan't = 4.29, SD = 1.94, F(1,
59) = 59.11, p b .05, η 2 = .48).
4.2. Method and procedure
Fifty-five individuals participated in this between-subjects experiment regarding the use of refusal framing in a health domain. The participants were asked to imagine that they were hosting a party. As
they were serving some rich chocolate cake for dessert, one of their
guests refused the cake, saying either “I don't eat chocolate cake” or
“I can't eat chocolate cake,” depending on the condition. The participants reported the persuasion index (α = .69) and the conviction
index (α = .82) on the same scales as in Study 1. Additionally, they
reported the degree to which the guest's response conveyed his/her
identity (1 = low degree; 9 = high degree) and stance as an individual (1 = low degree; 9 = high degree); the items were later combined
in an identity attribution index (r = .77).
4.3. Results
A one-way ANOVA with the persuasion index as the dependent
variable revealed the expected main effect of framing (Mdon't = 6.76,
SD= 1.43 vs. Mcan't = 5.89, SD= 1.55, F(1, 53) = 4.73, p b .05, η 2 = .08).
This result supports H1.
A similar ANOVA with the conviction index as the dependent variable also revealed the expected main effect of framing (Mdon't = 6.80,
SD= 1.61 vs. Mcan't = 5.63 SD= 2.08, F(1, 53) = 5.43, p b .001, η2 = .09).
A similar ANOVA with identity attribution as the dependent variable revealed a main effect of framing (Mdon't = 6.21, SD = 1.72 vs.
Mcan't = 5.17, SD = 1.74, F(1, 53) = 5.05, p b .05, η 2 = .09), thereby corroborating the pretest; see also Table 1.
Taken together, H2 and H3 predict a three-path mediation with
identity attribution and conviction as successive mediators of the
V.M. Patrick, H. Hagtvedt / Intern. J. of Research in Marketing 29 (2012) 390–394
Table 2
Study 2: regressions for the mediation of the influence of refusal frames on
persuasiveness.
Frame
Identity
Conviction
R2
393
framing as well the role of language-based identity attributions in
an interpersonal context.
5.1. Implications and directions for future research
Identity
Conviction
(1)
(2)
(3)
(4)
Persuasiveness
(5)
(6)
.52⁎
.58⁎
.26
.61†
.44⁎
.19
.46†
.09
.09
.39
.08
.35
.06
.15
.52†
.61
Note: Frame is a binary variable (1 = don't, −1 = can't). Each column (numbered in
second row) is a separate regression. Regressions (1), (3), and (6) correspond to
steps (a), (b), and (c), respectively. Regressions (2), (4), and (5) are included for
completeness. The first row indicates DV. The first column (rows 3–5) indicates
predictors. The table reports unstandardized regression coefficients.
⁎ p b .05.
†
p b .001.
effect of refusal frame on persuasiveness (refusal frame → identity →
conviction → persuasiveness). We tested this with the approach
developed by Baron and Kenny (1986), as adapted to three-path
mediation by Taylor, MacKinnon, and Tein (2008) (see Hamilton,
Puntoni, & Tavassoli, 2010 for an example). Following this procedure,
we estimated a regression for each of the dependent variables (identity attribution, conviction, and persuasiveness) with the preceding
variables in the proposed mediation chain as predictors. In other
words, mediation was assessed by estimating the regression of
(a) identity attribution on the refusal frame, (b) conviction on the refusal frame and identity attribution, and (c) persuasiveness on the refusal frame, identity attribution, and conviction; see Table 2 for the
results. In step (a), the refusal frame is a significant predictor of identity attribution (b = .52, t (53) = 2.25, p b .05). In step (b), identity attribution is a significant predictor of conviction (b = .61, t (52) = 4.99,
p b .001), but the effect of the refusal frame is not significant. In step
(c), conviction is a significant predictor of persuasiveness (b = .52,
t (51) = 5.76, p b .001), but the effects of the refusal frame and identity
attribution are not significant. The results of these regressions are
consistent with the hypothesized three-path mediation.
We next estimated the magnitude of the indirect (i.e., mediated)
effect by using the bias-corrected bootstrap test developed by
Taylor et al. (2008), drawing 1000 random samples with replacement
from the full sample (Hamilton et al., 2010). This bootstrap sample
served as the basis to construct a bias-corrected confidence interval
and thus test whether the indirect effect is significantly different
from 0. The indirect effect from the original sample is the product of
the effects revealed in the regressions from steps (a), (b), and
(c); in other words, .52 × .61 × .52 = .17 (allowing for more than two
decimals in the rounding). The 95% confidence interval (.03, .44)
excludes 0, indicating a significant indirect effect. The results thus
support the three-path mediation model predicted by hypotheses 2
and 3.
5. General discussion
This research is concerned with how to effectively say no. Empirical
findings from two experiments support the notion that the “I don't”
refusal frame is more persuasive than the “I can't” refusal frame and
conveys greater conviction on the part of speaker. This conviction
mediates the influence of the refusal frame on perceived persuasiveness. Furthermore, identity attributions explain the perceived
conviction of the refuser. A three-path mediation is therefore
hypothesized and supported, in which conviction mediates the
influence of framing on persuasiveness and identity attributions
mediate the influence of framing on conviction. Study 1 is a
role-playing experiment, whereas Study 2 is a scenario-based
experiment. Together, these studies provide insights into refusal
The current research contributes to the identity literature by demonstrating not only how specific words may serve as identity signals
but also how an identity signal may be used as a tool in persuasion.
As previously discussed, this research also contributes to the literature on compliance-gaining and on language in social interactions.
Furthermore, the practical implications are notable, especially given
that communication strategies based on the use of specific words
such as “don't” or “can't” may be easily implemented. However,
future research could investigate other word pairs of this kind to
broaden the relevance of the current findings. Additionally, a limitation of the current research is that conviction is arguably conceptually
close to persuasiveness; therefore, future research could provide both
increased precision and increased richness to our understanding of
the underlying process. Further, although the word “don't” conveys
conviction and represents the firm stance of the speaker and consequently works well in a negotiation, there is a fine line between firmness and rudeness. In adopting this refusal frame, we suggest that the
manner in which the words are spoken is as important as the words
themselves. Firm but polite refusals are presumably the most effective.
Future research should investigate linguistic differences, not only in
terms of content but also in terms of style. It is also likely that additional
words presented alongside the refusal frames could influence the
extent to which they convey conviction and identity. For instance, providing a time expression such as “always” may reduce the difference in
impact resulting from the “don't” and “can't” framing because “I always
can't” implies a permanent rather than temporary stance.
As a general observation, the intersection between language and
identity is a fruitful domain of investigation. Although research has
suggested that in order to be successful, brands need to tailor themselves to fit a consumer's identity (Belk, 1988; Berger & Heath, 2007),
the language used to forge this bond needs to be further investigated.
How does one effectively communicate identification with a consumer
or another brand?
Recent research on identity signals in the health domain (Berger &
Rand, 2008) also presents some interesting avenues for future investigation. Expanding this work to understand how best to communicate health risks that implicate and make accessible a consumer's
identity is a critical issue that remains relatively unexamined. A
recent research article by Puntoni, Sweldens, and Tavassoli (2011)
investigates how communication cues can make gender identity
salient and consequently trigger defense mechanisms that prevent
consumers from responding to the information communicated.
Further research of this nature will help close the gap in our knowledge
about how identity salience can facilitate or interfere with consumer
information processing.
A related question pertains to how linguistic framing effects might
interact with other identity cues such as sentence structure, tone of
voice, gestures, and facial expressions. In general, future research
could investigate how such cues signal identity and perhaps also
serve as a feedback mechanism for the individual exhibiting the cues.
Language is a tool that is so broadly used and has such farreaching applications and consequences that it is difficult to
overestimate the importance of mapping out the influences it has
on consumers. Importantly, these influences may not be universal
and might depend on dimensions of accessible identity, cultural differences, or even the specific circumstances in which they are used.
Furthermore, there may be linguistic influences that are common, or
that diverge markedly, across different languages. Undoubtedly, language use in communication is a complex issue, as is the concept of
identity, and much research remains to be conducted to increase
our understanding in both these fields of inquiry.
394
V.M. Patrick, H. Hagtvedt / Intern. J. of Research in Marketing 29 (2012) 390–394
References
Athanasopoulos, P., Damjanovic, L., Krajciova, A., & Sasaki, M. (2011). Representation of
colour concepts in bilingual cognition: The case of Japanese blues. Bilingualism:
Language and Cognition, 14, 9–17.
Austin, J. L. (1975). How to do things with words. Oxford: Oxford University Press.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic and statistical considerations.
Journal of Personality and Social Psychology, 51(6), 1173–1182.
Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research,
15(2), 139–168.
Berger, J., & Heath, C. (2007). Where consumers diverge from others: Identity-signaling
and product domains. Journal of Consumer Research, 34(2), 121–134.
Berger, J., & Rand, L. (2008). Shifting signals to help health: Using identity-signaling to
reduce risky health behaviors. Journal of Consumer Research, 35(2), 509–518.
Bryan, C. J., Walton, G. M., Rogers, T., & Dweck, C. S. (2011). Motivating voter turnout
by invoking the self. Proceedings of the National Academy of Sciences, 108(31),
12653–12656.
Campbell, R. S., & Pennebaker, W. J. (2003). The secret life of pronouns: Flexibility in
writing style and physical health. Psychological Science, 14(1), 60–65.
Chafe, W. (1970). Meaning and the structure of language. Chicago: University of Chicago
Press.
Cheema, A., & Patrick, M. V. (2008). Anytime versus only: Mindsets moderate the effect of
expansive versus restrictive frames on promotion evaluation. Journal of Marketing
Research, 45, 462–472.
Cialdini, B. R. (2006). Influence: The psychology of persuasion. New York: Harper Collins.
Cialdini, B. R., & Goldstein, J. N. (2004). Social influence: Compliance and conformity.
Annual Review of Psychology, 55, 591–621.
Douglas, K. M., & Sutton, R. M. (2005). When what you say about others says something
about you: Language abstraction and inferences about describers' attitudes and
goals. Journal of Experimental Social Psychology, 42, 500–508.
Eckert, P. (2000). Linguistic variation as social practice: The linguistic construction of
identity in Belten High. Oxford: Blackwell.
Grice, P. H. (1975). In Peter Cole, & J. L. Morgan (Eds.), Logic and conversation in syntax
and semantics: Speech acts (pp. 41–58). (3rd ed.). New York: Academic.
Hamilton, R. W., Puntoni, S., & Tavassoli, N. T. (2010). Categorization by groups and
individuals. Organizational Behavior and Human Decision Processes, 112, 70–81.
Hoegg, J., & Alba, W. J. (2007). Taste perception: More than meets the tongue. Journal of
Consumer Research, 33, 490–498.
Kettle, K. L., & Häubl, G. (2011). The signature effect: Signing influences consumptionrelated behavior by priming self-identity. Journal of Consumer Research, 38,
474–489.
Labov, W. (1963). The social motivation of a sound change. Word, 18, 1–42.
LeBoeuf, A. R. (2006). Discount rates for time versus dates: The sensitivity of
discounting to time-interval description. Journal of Marketing Research, 43(1),
59–72.
Levin, P. I., & Gaeth, J. G. (1988). How consumers are affected by the framing of attribute
information before and after consuming the product. Journal of Consumer Research,
15, 374–378.
Marwell, G., & Schmitt, R. D. (1967). Dimensions of compliance-gaining behavior: An
empirical analysis. Sociometry, 30(4), 350–364.
Mayer, D. N., & Tormala, L. Z. (2010). “Think” versus “feel” framing effects in persuasion.
Personality and Social Psychology Bulletin, 36(4), 443–454.
Oyserman, D. (2009). Identity-based motivation: Implications for action-readiness,
procedural-readiness, and consumer behavior. Journal of Consumer Psychology,
19(3), 250–260.
Patrick, V. M., & Hagtvedt, H. (2012). "I don't" versus "I can't": When empowered refusal
motivates goal-directed behavior. Journal of Consumer Research, 39(2), 371–381.
Pennebaker, W. J., Mehl, R. M., & Niederhoffer, G. K. (2003). Psychological aspects of
natural language use: Our words, our selves. Annual Review of Psychology, 54,
547–577.
Preacher, J. K., & Hayes, F. A. (2004). SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behavior Research Methods, Instruments, &
Computers, 36(4), 717–731.
Puntoni, S., Sweldens, S., & Tavassoli, N. (2011). Gender identity salience and perceived
vulnerability to breast cancer. Journal of Marketing Research, 48(3), 413–424.
Ross, L., & Nisbett, R. E. (2011). The person and the situation: Perspectives of social
psychology (2nd edition). Pinter & Martin.
Rothman, J. A., Salovey, P., Antone, C., Kelli, K., & Drake, C. M. (1993). The influence of
message framing on intentions to perform health behaviors. Journal of Experimental
Social Psychology, 29(5), 408–433.
Sanna, L. J., Parks, C. D., Chang, E. C., & Carter, S. E. (2005). The hourglass is half full or
half empty: Temporal framing and the group planning fallacy. Group Dynamics:
Theory, Research, and Practice, 9(3), 173–188.
Schellekens, G. A. C., Verlegh, P. W. J., & Smidts, A. (2010). Language abstraction in
word of mouth. Journal of Consumer Research, 37, 207–223.
Schenck-Hamlin, J. W., Wiseman, R. L., & Georgacarakos, G. N. (1982). A model of properties of compliance gaining strategies. Communication Monographs, 30(2), 92–100.
Schmitt, B., & Zhang, S. (1998). Language structure and categorization: A study of
classifiers in consumer cognition, judgment and choice. Journal of Consumer
Research, 25, 108–122.
Searle, J. R. (1969/1999). Speech acts: An essay in the philosophy of language. Cambridge,
UK: Cambridge University Press.
Snyder, C. R., & Higgins, L. R. (1988). Excuses: Their effective role in the negotiation of
reality. Psychological Bulletin, 104(1), 23–35.
Taylor, A. B., MacKinnon, D. P., & Tein, J. -Y. (2008). Tests of the three-path mediated
effect. Organizational Research Methods, 11(2), 241–269.
Tykocinski, O., Higgins, T. E., & Chaiken, S. (1994). Message framing, self-discrepancies,
and yielding to persuasive messages: The motivational significance of psychological
situations. Personality and Social Psychology Bulletin, 20(1), 107–115.
White, K., & Dahl, D. W. (2007). Are all out-groups created equal? Consumer identity
and dissociative influence. Journal of Consumer Research, 34, 525–536.
Whorf, L. B. (1956). Language, thought, and reality. Boston: MIT Press.
Wilson, R. S. (2002). Seeking and resisting compliance: Why people say what they do
when trying to influence others. Sage Publications.
International Journal of Research in Marketing 29 (2012) 395–405
Contents lists available at SciVerse ScienceDirect
International Journal of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
Consumers' use of brands to reflect their actual and ideal selves on Facebook
Candice R. Hollenbeck a,⁎, 1, Andrew M. Kaikati b, 1, 2
a
b
Terry College of Business, University of Georgia, 148 Brooks Hall, Athens, GA 30602, United States
Terry College of Business, University of Georgia, 111 Brooks Hall, Athens, GA 30602, United States
a r t i c l e
i n f o
Article history:
First received in 5, July 2011 and was under
review for 7 months
Available online 16 September 2012
Keywords:
Consumer identity
Self-concept
Self-expression
Actual self
Ideal self
Social networking
Social media
Facebook
a b s t r a c t
How do consumers represent their identities on online social media platforms? In this article, we focus on
consumers' use of brands on their Facebook pages as subtle cues to represent their selves. Although recent
research suggests that consumers present an actual, not an ideal self, our data reveal that veridical presentations of the actual self through brands rarely exist. Furthermore, we contribute to an understanding of how
multiple selves interact to inform brand connections. We offer insights into how and why consumers either
blend or integrate their actual and ideal selves or choose one of them exclusively when these selves conflict.
Additional contributions and future directions in the areas of self, branding, and social media are discussed.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
Scholars have investigated the presentation of the self online since
the early days of the internet (e.g., McKenna & Bargh, 1998, 2000).
One key question of interest in this stream of research has concerned
whether people present an accurate version of themselves online, a
more idealized version of themselves, or both (Bargh, McKenna, &
Fitzsimons, 2002). Early studies of online impression formation focused on relatively anonymous online environments, such as chat
rooms, bulletin boards, product discussion forums, and gaming
websites. These studies generally concluded that, due to the anonymity of these environments, users tend to construct idealized versions
of themselves without fear of disapproval and social sanction from
those in their off-line social circles (e.g., Bargh et al., 2002; Walther,
1996). For instance, a study of a prominent online gaming site
shows that players create virtual, alternate selves who embody aspects of the players' ideal selves (Bessière, Seay, & Kiesler, 2007).
As the media landscape has changed dramatically in recent years,
new online platforms have altered the ways in which people interact
with one another. At least 75% of adults who use the internet use social
media (Stephen & Galak, 2010; Urstadt, 2008). Social networking sites
(SNSs), which fall under the umbrella of social media, have recently become prominent. SNSs such as Facebook, MySpace, and Twitter, attract
⁎ Corresponding author. Tel.: +1 706 542 3762; fax: +1 706 542 3738.
E-mail addresses: chollenbeck@terry.uga.edu (C.R. Hollenbeck), akaikati@uga.edu
(A.M. Kaikati).
1
Both authors contributed equally to this work.
2
Tel.: +1 706 542 9385; fax: +1 706 542 3738.
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.06.002
more than 90% of young adults and teens, and represent over a quarter
of all internet traffic (Trusov, Bodapati, & Bucklin, 2010).
Consistent with the growing research on human brands
(Close, Moulard, & Monroe, 2011; Thomson, 2006), there is a renewed
interest in how people present themselves online, especially in the
context of these SNSs. A great deal of interest is centered on Facebook,
the largest SNS, boasting 640 million members worldwide, with
165 million of these members residing in the United States (Stone,
2010). There are key characteristics of Facebook that distinguish it
from other forms of social media, and even from other social networking sites such as Twitter and MySpace. Facebook users are “primarily
communicating with people who are already part of their extended
social network” (Boyd & Ellison, 2009, p. 210), and these users, or
“friends,” are all visible within their networks. In other words,
Facebook represents a means for individuals to continue their offline
relationships and conversations in an online medium. In addition,
within each network lies a certain degree of visibility through features
including wall posts and public displays of connections. Indeed,
Facebook use is shown to be significantly associated with the maintenance and creation of social capital (Ellison, Steinfield, & Lampe, 2007).
Recent research on Facebook examined the conveyed personalities of users' profiles to test the validities of the idealized virtual identity hypothesis (that Facebook users' profiles display idealized
information that does not reflect their personalities) and the extended
real-life hypothesis (that Facebook users' profiles display information
to communicate their real personalities) (Back et al., 2010). The research concluded that consumers present an actual, but not an ideal,
version of themselves on this social networking platform. The authors' proposed explanation for this finding that an idealized version
396
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
of the self is not reflected in users' profiles is that creating idealized
identities should be hard to accomplish, primarily because “friends
provide accountability and subtle feedback on one's profile” (Back
et al., 2010, p. 372). In other words, Facebook “friends” might question the validity of information if it does not reflect their perceptions
of the person.
With the current research, we aim to build on this literature
studying consumers' expressions of self via Facebook to complement
and extend these findings. Specifically, we focus on consumers' use of
brands as subtle cues to represent their selves. Brand mentions are arguably the most relevant and important aspect of this medium to
marketers, and can include “liking” a brand by publicly linking it to
one's profile, as well as mentions of brands in other subtle ways,
such as through narratives, photographs, and profile activities and interests. While it has been noted that consumers may use brands to
identify themselves with specific subcultures and/or identities online
(Stern, 2004), we have a limited understanding of what purpose
these brand linkages serve in the expression of consumer identities
in SNSs. Thus, we aim to shed light on two important questions.
First, how do users present their identities through brands on
Facebook—do they represent the actual self, the ideal self, or both?
As a limiting condition to recent research concluding that Facebook
profiles reflect actual but not ideal selves, we propose and show
that consumers may present both actual and ideal versions of themselves through the brands that they publicly associate with on
Facebook. Second, and perhaps more importantly, how do consumers
use brands to blend their actual and ideal selves when these identities
are congruous, or cope when these identities are incongruous? As we
will discuss later in more detail, Facebook has a number of characteristics (e.g., its ubiquitous nature, high visibility, direct connection to a
sizable and heterogeneous network of known individuals) that provide unique and interesting conditions for investigating the interaction of multiple selves and the incorporation of brands in consumer
self-expression.
We aim to make at least three important contributions with this
article. First, we respond to a call for further research on identity and
brands. Kirmani (2009, p. 274) notes that research is needed “to pursue
issues dealing with the intersection of identity and brands” to offer theoretical and substantive insights in this area. According to congruity
theory, consumers tend to prefer brands that are congruent with certain aspects of their identities (Sirgy, 1985). However, this literature
is relatively silent on how multiple identities interact to inform brand
preference. We aim to extend this theory by shedding light not only
on how consumers choose brands that are congruent with their selves
but also how and why consumers publicly link themselves to brands to
resolve conflicts engendered by different salient aspects of the self.
Second, we extend the current knowledge on how consumers use
brands as cues to represent themselves, in the context of the most
well-known and most used social networking platform: Facebook.
Whereas most prior research has examined brand-self congruence
in the context of offline measures, such as brand perceptions and purchase intent, we examine the ways in which consumers directly and
publicly link themselves to a brand to present their selves on a prominent social networking platform. This is important as it has been
noted that the particular self that consumers choose to express may
be dependent on contextual factors (Schenk & Holman, 1980).
Third, this research builds on the recent literature by providing an
improved understanding of consumers' self-presentations on SNSs.
To complement recent research showing that Facebook profiles reflect only actual, and not ideal, versions of the self (Back et al.,
2010), we show that by using brands as subtle cues, consumers do indeed communicate idealized versions of their selves, and even more
so in aggregate than their actual selves. Overall, this research builds
on a strong foundation of work employing qualitative methods to explore new media (e.g., Brown, Broderick, & Lee, 2007; Kozinets, Valck,
Wojnicki, & Winer, 2010).
The remainder of this manuscript is organized as follows. Next, we
review the literatures on brands and the self and self-expression with
regard to different aspects of the self. Then, we introduce our methodology and findings. Finally, we conclude by discussing the implications for firms and future research directions.
2. Conceptual background
2.1. Brands and the self-concept
Self-concept is defined as a person's perception of him or herself
(Shavelson, Hubner, & Stanton, 1976). Importantly, the self consists
of multiple aspects, or dimensions (James, 1890; Markus & Kunda,
1986). Although the self-concept is relatively stable over time, the
self is also malleable in that it may be influenced by social roles and
cues, causing people to behave differently in varying situations.
Thus, any of these dimensions can be activated at any time due to a
number of factors that may become salient as a function of a situation
(Aaker, 1999).
According to self-concept theory, people behave in ways that
maintain and enhance their sense of self. One way to do this is
through the use of brands (Strizhakova, Coulter, & Price, 2008,
2011), which are imbued with symbolic meanings that develop as
early as middle childhood (Chaplin & John, 2005). Through these
meanings, brands can serve as a means for a consumer to express different aspects of his or her self (Aaker, 1997; Escalas & Bettman,
2005; Levy, 1959; Torelli, Monga, & Kaikati, 2012). Indeed, the congruence between brand and self-image is shown to be positively related to consumers' evaluations of products (Graeff, 1996; Sirgy,
1982). For instance, self-congruity has been shown to significantly influence promotion effectiveness (Close, Krishnen, & LaTour, 2009),
quality perceptions of stores in shopping malls (Chebat, Sirgy, &
St-James, 2006), and brand loyalty (Kressmann et al., 2006). Furthermore, consumers choose brands with appealing personalities to enhance their selves (Swaminathan, Stilley, & Ahluwalia, 2009) and
may develop more positive self-perceptions in line with the brand's
personality even after using a brand for only a brief amount of time
(Park & John, 2010).
With the current research, our aim is gain a deeper understanding of
how and why consumers choose to link themselves to brands on
Facebook. Our first question is: Do consumers link to brands on this
platform to reflect their actual selves, their ideal selves, or both? Recent
research concluded that only actual, and not ideal, selves are reflected in
Facebook profiles (Back et al., 2010). However, that research was
conducted in the context of Facebook users' general profiles, which contain more explicit statements about the self, including descriptions of
one's activities and behaviors, and not on Facebook users' utilization
of brands as more subtle cues about their selves. In the next section,
we discuss brands as reflections of self.
2.2. Brands as reflections of actual and ideal selves
While there are numerous conceptualizations of the self, the prominent facets of the self-concept that we focus on here are the actual and
ideal selves. Scholars have long debated the relative influence of the actual versus ideal selves on consumers' evaluations of and preferences
for branded products (Grubb & Grathwohl, 1967; Landon, 1974). One
important aspect of brands is that they can be used by a consumer to express to others their actual (i.e., who they are) and ideal selves (Belk,
1988; Dolich, 1969; Holt, 2002; Landon, 1974), with the ideal self
representing either an expansion of self (self-representation; Belk,
1988) or a contraction of self (self-presentation; Leary & Kowalski,
1990; more information on this point will be provided in the findings
section).
In the context of Facebook, there is reason to expect that consumers will reflect their actual selves by linking themselves to brands.
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
Users are linked to the people in their network and they tend to know
each other to some degree. As such, users may be less likely to deviate
from others' perceptions of themselves on Facebook because “friends”
might consequently question the user's authenticity (Back et al.,
2010). One reason for using this platform is to keep in touch with
friends (Ellison et al., 2007), such that traditional offline relationships
are facilitated by this technology in nontraditional ways. Thus, because Facebook friendships may be grounded in offline relationships,
social sanctions could result from grossly misrepresenting oneself.
At the same time, there is reason to expect that consumers may
also reflect their ideal selves by linking themselves to brands on
Facebook. Although Facebook users tend to use this platform to
keep in touch with friends, most of these relationships can be characterized as “weak ties” (Ellison et al., 2007; Granovetter, 1973; Trusov
et al., 2010) and part of a person's “extended social network” (Boyd &
Ellison, 2009, p. 210). Thus, the meaning of “friend” on Facebook does
not carry the traditional connotation. Consider that the mean number
of Facebook friends is approximately 250 (Walther, Heide, Kim,
Westerman, & Tong, 2008), whereas the typical number of close relationships offline is approximately 10–20 (Parks, 2007). Consequently,
a user's “friends” may not know him very well, making it possible to
present a more idealized version of oneself using subtle cues, with a
relatively low risk of social sanction. Furthermore, following from a
symbolic interactionist perspective (Schenk & Holman, 1980), extant
research suggests that consumers' publicly observable behaviors may
be driven to a greater extent by their ideal (vs. actual) selves (Alpert
& Kamins, 1995; Dolich, 1969; Graeff, 1996; Ross, 1971). Most recently, the findings of Swaminathan et al. (2009) suggest a signaling role
of brands in public (vs. private) consumption settings. These authors
find that individuals with anxious attachment styles tend to associate
themselves with brands with favorable personality characteristics
only when the product is consumed in a public setting, where they
can manage others' impressions of themselves.
Building on prior research, we expect that many users will connect
with a combination of brands on Facebook, representing both their
actual and ideal selves. Furthermore, as consumers appear to be especially driven by their ideal selves in public circumstances, and because
Facebook enables them to deliberately and publicly link themselves to
brands, one might expect the ideal to outweigh actual self, regardless
of whether the branded product or service is publicly or privately
consumed.
2.3. Expressing different aspects of the self
Our second question is as follows: How do users make decisions
when expressing multiple selves by linking to brands? As implied in
the previous discussion, the self develops not in isolation but through
the process of social experience (Grubb & Grathwohl, 1967). Thus,
feedback and reactions from others affect the growth of the self, and
consumers strive to elicit positive reactions from others. Leary and
Kowalski's (1990) two-factor model posits that impression management consists of both impression motivation, which is the extent to
which the desire to control the self-image that is projected to others
is activated, and impression construction, which involves the selection of the appropriate impression to convey to others and how to
go about doing so.
As Facebook users are linked to other people whom they know,
their desire to control the image projected to others, or impression
motivation, tends to be high in this context. Indeed, Facebook use is
shown to be significantly associated with the maintenance and creation of social capital (Ellison et al., 2007). Linking oneself to brands on
Facebook serves as a means of impression construction, which involves determining the type of impression to make on others and
how to go about it (Leary & Kowalski, 1990). For instance, in linking
oneself to a brand, others' perceptions of a brand owner's personality
397
may be affected by carry over from the personality of a brand (Fennis,
Pruyn, & Maasland, 2005).
Importantly, however, it is rare that individuals who are constructing
impressions will have a single persona that they use in virtually all situations and with virtually all targets (Leary & Allen, 2011). One of the
major research programs in consumer culture theory (CCT) focuses on
the means by which consumers achieve balance among their multiple
selves (Arnould & Thompson, 2005). Some of this research focuses on
how individuals in a liminal state may generate both positive and negative possible selves and weigh these against one another in a mental balance (Markus & Nurius, 1986; Schouten, 1991) or develop addictions as
a means of escaping an inauthentic self (Hirschman, 1992).
Most central to our main arguments is research from both the CCT
and experimental consumer research literatures demonstrating that
individual consumers tailor their public images to different targets
(i.e., display different selves in different situations) as a means of coping in creating their identities. For instance, Aaker (1999) notes that
conflicting personality traits may be present in an individual's selfconcept (e.g., sociability and intelligence). To reconcile these traits,
an individual may express sociability at a party and express intelligence when focusing on his/her work, which is consistent with Sirgy's
contention that the consumption of brands may be strongly related to
an individual's self-image in one situation but not in another (Sirgy,
1982; Solomon, 1983). Similarly, James (1890) and Mead (1934) contend that individuals have a separate self (or “me”) for each of their
social roles.
Building on this theory, research has shown that as they transition
from role to role, consumers use varying product cues to signify their
occupation of each new role (e.g., political, religious; Fournier, 1998;
Solomon, 1983). In some cases, consumers may even use different elements of the same consumption object as a means of representing
varying aspects of the self. For instance, it has been suggested that a
home's interior décor may reflect the true self (representing family),
with the home's exterior representing a different aspect of the self
(Belk, 1988). Notably, each of these examples from the literature illustrate that consumers tailor their public images to different targets
in an effort to achieve balance among multiple selves (Leary & Allen,
2011; Otnes, Lowrey, & Shrum, 1997).
There is increasing interest in how individuals reveal multiple
identities in the digital space, such as through blogs (Zhao & Belk,
2007) and personal websites (Schau & Gilly, 2003). As will be further
elaborated in the findings section, Facebook has a number of distinct
properties (e.g., direct connections to multiple audiences) that distinguish it from offline and other online venues. Importantly, these
properties increase the likelihood of conflict between the actual and
ideal selves of Facebook users who may wish to construct several
identities simultaneously. While congruity theory demonstrates that
consumers tend to choose brands that are congruent with some aspect of their selves, we aim to shed light on how the congruity (or
lack thereof) between selves (actual and ideal) affects conspicuous
displays of brand preference.
Next, we describe our methodology for studying how consumers
use brands on Facebook as reflections of their identities.
3. Method
To investigate how brands are used on Facebook to depict aspects of
the self, our study employed multiple methods over a two-year
research period. To gain a perspective of action (Belk, Sherry, &
Wallendorf, 1988) we utilized observational research, diaries, focus
groups, in-depth interviews and electronic journal entries. The researchers observed the activity on 84 volunteer participants' Facebook
pages (123 observational sessions lasting 45 min resulting in over 200
pages of notes). All volunteers gave consent to be observed, and volunteers were not “Facebook friends” with the researchers. During the observations, the researchers looked for subtle uses of brands on the
398
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
participants' walls and in their photos and narratives. Special attention
was given to how participants integrated brands in their interactions
with other Facebook users. Table 1 demonstrates how the researchers
categorized the brands observed on these Facebook pages. Participants
expressed themselves using “brand” cues via various Facebook tools
(e.g., profile activities, likes, ordinary routines). The participants were
male and female undergraduate college students, ranging in age from
20 to 28. Eighty percent of them were Caucasian, while 10% were
African-American, 5% were Asian, and another 5% were of Hispanic
descent.
To triangulate across data points (Belk et al., 1988), all 84 of the volunteer participants were asked to keep a consecutive seven-day diary
with the goal of identifying brands on their Facebook pages and
explaining why particular brands were on their pages, describing the
image of each brand, and comparing the brand's image to his/her personality, goals, and aspirations. At the end of the week, participants created two collages depicting the following: brands on their Facebook
pages that represent their actual selves (qualities a person feels describe
him/her at the present time) and brands on their Facebook pages that
represent their ideal selves (an expansion or limitation of their actual
selves, see Table 4). The collages were used as projective vehicles to elicit participant commentary during eight focus group sessions with approximately 10–12 participants in each session. Each focus group
session lasted approximately 60 min. During the focus groups, the participants discussed why they have certain brands on their Facebook
Table 1
Observational data illustrative of brands displayed on participants' Facebook pages.
Facebook tools
Profile activities (22%)
Brand cues used to express the self
Brands are displayed through all hedonic activities
illustrative of fun, adventure, or volunteerism such
as golf, tennis, wine class, cooking class, shopping,
and participating in fund raisers or mission trips.
Brands in this category include: Callaway golf
clubs, William Sonoma, Polo, Brooks Brothers, Saks
Fifth Avenue, Chanel, and Chateau Margaux, Red
Cross, American Cancer Society, and religious
brands.
Profile interests (21%)
Brands are displayed through all personal interests
and goals such as eating right, exercising, movies,
socializing with friends over coffee, and traveling.
Brands in this category include: Nabisco 100
calorie snacks, Nike, Star Wars, Starbucks,
Whoopsie Daisy Designs, and Hampton Inns.
Likes (24%)
Brands are displayed through the symbolic act of
“liking” a brand. Consumers are primarily
motivated by the need for expressing a particular
style. Facebook Ads work best when their appeal
(profile picture) matches the dominant type of
self-concept held by the consumer. Brands in this
category include: retailers, restaurants, and
organizations.
Attitudes displayed in
Brands are integrated throughout discussions
narratives form (5%)
posted on participants' walls, newsfeeds, and
personal messages. Participants who openly
promoted or dissuaded brands are highly
motivated by expressing their opinions and are
appreciative of retail aesthetics and technical
innovations. Brands in this category include: retail
brands, technology brands, and iconic brands.
Ordinary routines displayed in Brands are expressed through story-telling.
narrative form (17%)
Consumers discuss brands and how brands are
part of their everyday life. Brands in this category
include: Old Spice deodorant, Gillette razors,
Allergy Guard hypo-allergenic pillows, and Rembrandt Canker Sore Whitening Mint Toothpaste.
Photos of consumption (11%) Brands are displayed through the act of consuming
the brand. Many of these brands are displayed in
Wall posts as “in-the-moment” photos. Brands in
this category include: beer brands, branded
vacation destinations, University brands, and
retail brands.
pages and how each brand is related to their self-concepts. Participants
discussed brands that represent both their actual and ideal selves. The
projective task with their collages engaged participants through storytelling and allowed for more personal and meaningful insights during
the focus groups to further elaborate on why and how brands are
used to reflect aspects of the self.
The findings from the first phase revealed that consumers use
brands on Facebook to represent both their actual and ideal selves.
We found evidence of iconic brands (Holt, 2004), relational brands
(Fournier, 1998), and mundane brands (Coupland, 2005). In particular, participants were probed to elaborate on their linkages with mundane brands, such as athletic brands (e.g., a university sports team);
functional brands (e.g., Gillette); or not-for-profit brands (e.g., the
American Red Cross). The most easily identifiable brands were iconic,
well-known brands (e.g., Nike, Starbucks). Hence, focus groups began
with a discussion of iconic brands and then proceeded to discuss the
less obvious brands (from the perspectives of the participants). The
mundane brands provided equally insightful data and were typically
used to express aspects of the actual self.
In the second phase of our research, 23 in-depth interviews were
conducted with volunteer graduate students. All of the participants
in the study were assured of anonymity. Table 2 provides a list of participants' pseudonyms and a brief description of their backgrounds.
The participants were male and female, ranging in age from 22 to 36.
Before the interviews, participants were asked to maintain an electronic journal for two weeks and given instructions to record their use
of brands, the number of times they encountered brands (on their
own Facebook pages and those of their friends), and their observations
of their Facebook friends' uses of brands. Table 2 highlights the most
prominent brands recorded in the participants' electronic journals.
After the two-week period, interviews were conducted in a private setting with access to a computer (e.g., library, office).
During the interviews, participants were shown a figure demonstrating the actual and ideal selves on opposite ends and were asked
to place the brands mentioned in their journals on the figure (to represent more actual, more ideal, or equal). Table 2 demonstrates whether
participants felt the majority of brands on their pages were more actual
or ideal. The interviewer followed a semi-structured guide beginning
with the brands that represent the actual self and then proceeded to
discuss the brands that represent the ideal self (see Table 3); however,
the interviews were informal in nature and characterized by a conversational quality in which the discussions were largely driven by the participants. The interviews concluded with an analysis of the participant's
Facebook page. Each participant logged onto his/her Facebook page and
walked the interviewer through his/her page(s) discussing points
where brands were present. Participants were asked to elaborate on
how they expressed their online identities and how they coped if/
when their actual and ideal selves conflicted.
Our data analysis utilized the extended case method (Burawoy,
1991). Our analysis of verbatim interview and focus group transcripts
involved an iterative, part-to-whole strategy in which the researchers
aimed to develop a holistic understanding of how brands are used to
represent the self. Aided by Atlas software, we used prior theories to
guide our analysis of the diaries, focus groups, interviews, journal entries, and observational data. Following Burawoy's (1991) suggestion
to utilize existing bodies of literature in search of theories to explain
behavior, the data analysis process involved going back and forth between the data and existing theories on actual and ideal selves and,
more generally, self-concept.
4. Findings
The findings are organized as follows. First, we will briefly discuss
participants' reflections of their actual and ideal selves through selfbrand linkages on Facebook. Next, we will review how they view
Facebook as a means of communicating identity through brands. Finally,
399
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
Table 2
Demographics of interview participants.
Name
Gender
Age
Brands publically viewed on their Facebook page
Self-reported their
Facebook brands as
more actual or ideal
Jeff
Callie
Donna
John
Bella
Chet
Amber
Deb
Alisa
Derek
Matt
Valarie
Sarah
Jody
Emma
Daniel
Daisy
Mark
Eric
Brooke
Shea
Diane
Tom
M
F
F
M
F
M
F
F
F
M
M
F
F
M
F
M
F
M
M
F
F
F
M
28
22
36
29
22
24
32
35
26
34
30
28
22
29
22
31
27
32
23
23
25
31
24
Red Socks, Nike, Nacoochee Village Tavern and Pizzeria, Great Clips, Slack Auto Parts, local library
Chanel, Lancome, Serena Williams, Local Newspaper, Menchie's Frozen Yogurt, Five A Day Café, Justin Bieber
New York Yankees, Hershey, Salty Dog Café, Harry Potter, Publics, Quelf, Red Hot Chili Peppers
Audi, Patek Philippe, Barnes and Noble, Adidas, Amazon, Sports Authority
Starbucks, Easy Bistro and Bar, Shoedazzle.com
BMW, Georgio Armani, Levi's jeans
New York City Ballet, Ann Taylor Loft, Lifeway Christian Stores, Rite Aid
Prada, Guess jeans, Versace, Akif Hakan Celebi Photography, Porterhouse Grill, Chili's Bar and Grill, Red Lobster
Ann Taylor Loft, Mad Men, Chick-fil-A, Grumpy's Pizza
Great Big Bertha golf bag, John Lobb golf shoes, Oakley's Romeo 2 sunglasses
Brioni, Columbus Blue Jackets, Landon Donovan, Italy, Umbro, Amazon, Chicago's Pizza, Franklin Covey
Katie Couric, Pete Sampras, Rick Steves, Clinique makeup, Bob Marley, Keri lotion, Papa John's Pizza Internet Movie Database
Food Network, Oprah, The Galaxy Concert Theater, Wholly Cow, Grocery Coupon Network, 4 Walls
Dogfish Head beer, Human Rights Campaign, Gillette, Anchor White Toothpaste,
Scrubs, Grey's Anatomy, Coca-Cola, American Cancer Society Relay for Life, March of Dimes, Churches in Action
Nintendo, Kappa Total Cereal, Gatorade, Megan Fox, Two and a Half Men, Delta, Groupon, Taco Bell
Clinique, Diet Coke, Dasani, YouTube, Twlight, Blue Dragon Thai Food, FAGE Total Greek Yogurt
KCM, North Face, REI, Jeep, Flixster, Family Guy, South Park, Netflix
World Wrestling Entertainment, Derek Truck Band, Nike, Notre Dame, Smoken' Joe's BBQ & Blues, Fox News, Texas Holdem
Wilsons Leather, Jersey Shore, BP, Naked Juice, Papa John's
Lil Wayne, Costa Del Mar, Ann Taylor, Banana Republic, Jared, Athleta, Duck Life 3, Nayan Cat
Atlanta Braves, Lady Gaga, Outback Steakhouse, Domino's Pizza, Jack & Daisy, Disney World
Armani, Corvette, Usher, Imo's pizza, Swiss Army, Raze
A
I
A
I
I
I
A
I
I
I
I
A
I
A
A
A
I
I
I
I
I
I
I
we will elaborate on how these Facebook users make decisions in expressing multiple salient identities by linking to brands as cues. We
identify conditions that lead to greater actual- or ideal- self-congruity,
exclusive ideal, or neither (see Fig. 1). A holistic representation of our
findings is visually demonstrated in the conceptual model in Fig. 1. As
shown in Fig. 1, participants' situational contexts will influence their
motivations in terms of presenting aspects of their actual or ideal selves.
As such, participants may experience a decision conflict among various
ways of presenting the self. The identity decision is a background for
which participants use brands as a means to express aspects of the self.
4.1. Brand linkages to reflect the actual and ideal selves
The self-concept, as suggested earlier, is a multidimensional construct reflecting more than one type of self-perspective (Sirgy,
1985). As prior research has suggested that only one aspect of the
self-concept (actual, and not ideal) is represented in Facebook profiles
(Back et al., 2010), we sought first to examine whether this is also the
case specifically for brands, which serve as more subtle cues of identity due to their symbolic meanings. As a limiting condition to prior
research, our findings revealed that Facebook fosters expressions of
both actual and ideal aspects of the self, each of which can be made
salient by individual traits and situational factors. We found evidence
of users linking to brands that represent the actual self, as well as two
subcomponents of the ideal self: ideal self-representation and ideal
self-presentation (see Table 4).
As noted in Table 4, ideal self-representation and ideal selfpresentation stem from the actual self, but the three have different motivations in terms of self-expression. When expressing aspects of the actual self, participants typically selected brands that maintained the
self-concept. As Brooke demonstrates below, Facebook users utilize
brands to clarify personality ambiguities and highlight individual characteristics that maintain the self-concept.
I publicize brands that I like. I want you to know more about me,
so I will use brands to tell others things about me and who I am
now (Brooke, maintain the self-concept).
When expressing aspects of the ideal self, participants described two
distinct motives associated with linking to brands: to enhance the self
and to protect the self. Self-representation refers to an expansion of
the self-concept that relates to the many possible selves an individual
could be, while self-presentation refers to a purposeful, restrictive
version of the actual self, with strategic implications aimed at influencing others' perceptions. The underlying motive for ideal selfrepresentation is self-enhancement, whereas the motive behind ideal
self-presentation is protection. Alicke and Sedikides (2009) elucidate
the motives underpinning the two concepts by explaining that people
tend to exaggerate their virtues (self-enhancement) and minimize
their shortcomings (self-protection). The narratives below are examples of ideal self-representation and ideal self-presentation.
One day I want to be involved in research that benefits society as a
whole. I want to contribute to something bigger…I can express my
desires to others by friending these brands (not for profit brands)
on Facebook and I'm able to connect with other people (who also
like the brand) who have similar goals…Facebook streamlines my
efforts (Emma, ideal self- representation exemplar).
I really have a reserved personality because I like being alone,
reading books, and spending time in the library. But, on Facebook
all you see is my social side. If you just looked at my Facebook
page, you'd probably think I'm a social butterfly…the brands I associate with are popular brands and all seems like I'm this dynamic, outgoing person (Diane, ideal self- presentation exemplar).
Table 3
Classification of in-depth interview questions.
Guiding research questions used to
understand the actual self concept
Guiding research questions used to
understand the ideal self concept
1. How do you actually see yourself?
2. What brands displayed on your
Facebook page represent your real or
actual self?
3. How do you think others actually see
you?
4. What brands on your Facebook page
would your friends use to describe
you?
1. How would you like to see yourself?
2. What brands displayed on your
Facebook page would you like to see
yourself associated with?
3. How would you like others to view
you?
4. What brands would you like for
others to use to describe you?
The self-concept is the totality of the consumer's thoughts and feelings toward him or
herself.
400
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
Fig. 1. Self-expression on Facebook.
Emma connects with brands that are associated with social improvements (e.g., the American Cancer Society, Relay for Life, the
March of Dimes). In the future, Emma desires to engage in meaningful
research to provide functional knowledge that could improve the quality of life for the broader society. She is involved with non-profit brands
on her Facebook page, which allows her to connect with others who
have similar goals. In addition, Emma wishes to convey an idealized version of herself by signifying to others that she not only wants to support
non-profit organizations with her time but also with her talents.
Emma's motive for connecting with non-profit brands is to enhance
the self-concept, and this represents an expansion of her actual self
(the person she could be and wants to become). Turning to the second
exemplar, Diane presents an ideal self that is a restriction of her actual
self where the underlying motive is to protect the self-concept. Although Diane enjoys socializing with others, she confessed that she prefers being alone. When faced with the option of either spending time at
home reading a good book or meeting up with her friends at a restaurant, she derives more enjoyment spending time alone at home. However, she shows the social side of herself and identifies with socially
oriented brands (e.g., the Atlanta Braves, Disney World) on Facebook
because this is the part of herself that she wants to show others.
Diane suggests, “people like social people” and “popular people are
not loners;” therefore, she presents a strategically selected self, motivated by public perception (how I want to be perceived). Diane feels that
she will be more popular and well-liked if she presents the outgoing
side of herself on Facebook, and this is an attempt to protect her
self-concept.
Importantly, Diane and Emma demonstrate distinct motives in relation to expressing the ideal self. Diane aims to protect the self, while
Emma desires to enhance the self. Both are aspects of the ideal self.
However, motives associated with the ideal self are in contrast to motives associated with linking to brands that represent the actual self,
which serve to maintain the self-concept (see Table 4).
4.2. Facebook as a means of communicating identity through brands
Facebook profiles allow for self-expression through brands, which
are a subtle but powerful means of non-verbal communication. Our participants are aware of using Facebook for their own purposes in punctiliously managing their self-identities. They discussed the ability to link
to brands instantaneously and deliberately on this platform, with this
linkage being public and observable by all members of a user's social
network. The linkage can be more powerful than offline brand identifiers because brands on Facebook can be continuously showcased
using Facebook tools (see Table 1).
If you like a brand just for yourself, then you know you like it and
you don't have to remind yourself of that. You want other people
to see a brand when you “friend” a brand on Facebook because
you want people to characterize you that way. If you don't, then
Table 4
The self concept.
Dimensions of the self
Definition
Relation to the self-concept
Motive for linking to brands
Actual self
The authentic self related to who I am now
(who I am)
The aspirationally-defined self related to the many selves a person could be
(who I could be)
The strategically-chosen self related to concerns about public perception
(how I want to be perceived)
How I actually see myself
To maintain the self-concept
An expansion of my actual self
To enhance the self-concept
A limitation of my actual self
To protect the self-concept
Ideal self-representation
Ideal self-presentation
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
I don't know why you would click “like” anything or even be on
Facebook… brands show your true self and the self you want to
be (Tom, M, 24).
As Tom notes, brands can send powerful messages in online social
networking sites such as Facebook, and his linkage to brands is more deliberate and perceptible than his offline means of communication. Our
participants suggest that they are able to present aspects of their
self-concepts, and brands allow them to highlight certain characteristics
to the larger network. Tom elaborates on this idea by stating: “if you buy
a new shirt with a popular brand, only a few people will take notice of
the brand on that given day.” However, on Facebook, consumers can
publicize brands to the broader network, and the level of visibility is
controlled by the individual. For example, the consumer can simply
“like” a brand, or the consumer may elevate the visibility by adding
brand-related photos and narratives (see Table 1). In addition, the ubiquitous nature of Facebook allows for ongoing visibility and immediate
recognition among friends. For instance, when an participant updates
or makes changes to his or her profile, other users can check in at any
point during the day to view these updates. Facebook allows for a unified awareness among people in a network. Because it is just as easy
to “unfriend” a brand as it is to “friend” a brand, users have the autonomy to manage their self-expressions by selecting brands with high
self-brand congruity. As such, brands can be used to communicate aspects of the self to the larger network instantaneously.
Importantly, our participants noted that Facebook allows greater opportunity (than offline self-brand expressions, for example) for the simultaneous expression of multiple salient identities. Consequently,
this simultaneous expression provides an environment for conflicting
identities. As discussed by Shea, users are perplexed when they want
to show different aspects of the self to different audiences.
Facebook has its downsides. It is like you are on stage for the masses
to see. You can't just be casual about the brands you have. It's like
you have to rehearse for a play. What part of myself am I going to
show today? (Shea, F, 25)
Shea uses the metaphor of a play. In essence, Facebook is the stage,
brands are the props, and the users are the actors. The users must utilize
brands to present themselves to their audiences. As Shea explains,
Facebook users must consciously and calculatingly consider the person
that they want to portray. Actors (users) are easily able to engage their
audiences (the network) through the use of props (brands). The specific
reflection of the self-concept through brands on Facebook depends precisely on the level of congruity among participants' multiple selves.
Next, we will discuss how linking with brands represent aspects of
the self.
4.3. Identity decision and brand linkages
Our data revealed that Facebook users often have multiple salient
identities and must choose how to present aspects of their actual and
ideal selves on a single platform to widespread viewers with whom
they are connected in a network. Importantly, these identities could
be either congruous or incongruous with one another, depending largely on current goals elicited by situational contexts and individual traits.
A high level of congruity means that consumers are able to find harmony between their actual and ideal selves and, thus, present a blending of
the two identities on their Facebook profiles. Overall, this congruity led
to a subjective blending of the two identities. A high level of incongruity,
however, implies an inharmonious relationship between the actual and
ideal selves. We first discuss the case of congruity, and its manifestations in terms of greater actual or ideal self-expression through brands.
Then, we review incongruity and its related tradeoffs and coping
mechanisms: exclusive ideal self-expression or the avoidance of selfexpression.
401
4.4. Congruous actual and ideal selves
When the actual and ideal identities were congruous, participants
expressed both their actual and ideal selves through brand linkages.
The relative strengths of these selves depended on their relative accessibility; that is, participants described situations when they would want
to show more of their actual selves and other times when they would
want to show more of their ideal selves. Notice that the situational context influences self-expression. For example, Callie subjectively rates
her Facebook identity as 20% actual and 80% ideal (meaning that 20%
of the brands on her profile represent her actual self while 80% represent her ideal self), while John rates his Facebook profile as 45% actual
and 55% ideal. As described here, both of the participants feel that
they show more of their ideal selves, but they selected different proportions to demonstrate their own perceived blending of the two identities
through brand linkages. In a follow-up interview, Callie was asked to
rate her Facebook profile again, and during the second interview Callie
rated her Facebook identity as 40% actual and 60% ideal. Thus, the
“liked” brands had changed slightly between the first and the second interviews. When probed to explain why she identified with more brands
representing the actual self in the second interview, Callie explained
that her situation had changed.
I think I was in a new place and was meeting new friends so I
wanted to show more of my ideal self, but now I've just had some
good things happen to me recently…I got a new car and I've got a
new boyfriend… I want to show more of my actual self now
(Callie, F, 22).
As demonstrated by Callie, we found evidence among our participants that their identities are continually evolving and, as a result, the
subjective calculation of the actual and ideal selves can change from
day-to-day and minute-to-minute. Facebook offers a platform that allows consumers to change their identities to reflect personal goals
and interests that are current and relevant to the situational contexts
of their own lives. In other words, situational context influences
self-expression motives on Facebook. This platform facilitates the redefinition of identity over time, as users are easily able to link to and
update their identifications with brands, regardless of current ownership or prior use. Next, we describe two manifestations of brand linkages when actual and ideal selves are congruous: greater actual
self-expression and greater ideal self-expression.
4.4.1. Greater actual self expression
During the interviews, 30% of participants self-reported that more
brands represented their actual selves (vs. ideal selves). When these
participants were asked about the brands that represent their actual
selves, they described inconspicuous brands that have utilitarian
meanings. For example, narratives describing brands that represent
the actual self typically followed this script: “I use this brand and
therefore it represents me.” In the following exemplar, Matt discusses
pizza as a utilitarian brand that expresses his actual self.
The brands that I am friends with are similar to my own interests
and activities…I have brands on my page that I use every day, like
Chicago's Pizza and Papa John's Pizza. My friends know that I eat
pizza every week (Matt, actual exemplar).
As conveyed by Matt, Papa John's Facebook friends receive discounts and promotions. Matt suggests that eating pizza is a part of
his day-to-day life routine. Therefore, Matt connects with various
brands of pizza. The pizza brands do not serve as a means for selfenhancement or self-protection. Rather, the pizza brands represent
his actual self (who I am now).
Participants who identified with more brands representing their
actual selves seemed less concerned about the opinions of others and
402
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
tended to seek out brands that represent who they are in the present
moment (e.g., links with brands to maintain the self-concept). Brands
representing the actual self tended to be currently owned or used. The
participants that linked with more brands to maintain the self-concept
expressed higher self evaluations (e.g., more confidence, higher selfesteem), and, interestingly, we found evidence that several of the participants were prompted to link to actual-self-brands by distinctiveness
motives. Distinctiveness motives refer to the ways in which a particular
individual is willing to deviate from the normative image that most people attempt to convey (see Leary & Allen, 2011). In other words, participants expressing more of their actual selves desired to be distinct and
aimed to use brands as a means for setting themselves apart from
their peers, and, in some cases, they linked with brands that many of
their peers would shun. For instance, Jeff associates with low cost brands
such as Great Clips and Wal-Mart, whereas most people in his peer
group would not publicly link to such economical brands.
I don't really care about what people think about the brands I like.
I like them because I like them, not because other people are going
to accept me more. I think some people are too concerned with
other people's opinion…I know most people would not friend
Wal-Mart, but it doesn't bother me (Jeff, higher self-evaluation
exemplar).
During his interview, Jeff expressed a high level of self-confidence
and conveyed that he enjoys deviating from the norm by associating
with economical brands because it reinforced his own sense of selfconfidence and self-sufficiency. Jeff suggests that being economical
is part of his own personality and he likes displaying this distinctiveness trait to his network of friends.
4.4.2. Greater ideal self expression
During the interviews, 70% of participants self-reported that more
brands represented aspects of their ideal selves (vs. actual selves).
When these participants were asked to describe the brands that represent their ideal selves, they typically discussed iconic, luxury brands.
Furthermore, they expressed more concerns about either promoting
or expanding parts of the self that they felt good about (ideal selfrepresentation) or altering the opinions and feelings of others to obviate
the risk of rejection or criticism (ideal self-presentation). Both selfenhancement and self-protection are instrumental in maintaining, promoting, and safeguarding pivotal self-interests (Alicke & Sedikides,
2009). Participants expressing more of their ideal selves demonstrated
lower self evaluations (e.g., seemed less confident, expressed lower
self-esteem). When probed to explain why they selected brands to represent the ideal self, participants discussed linking with brands for the
sake of highlighting certain characteristics or goals to others in their
networks (ideal self-representation) or only showing certain parts of
the self thus restricting the information available to others in their networks (ideal self-presentation). These participants described instances
when they would link with brands to receive higher normative evaluations from the acquaintances in their networks. Normative motives
refer to the degree to which the individual's image resembles the
types of impressions that most people attempt to make (see Leary &
Allen, 2011). Our participants were inspired to link with iconic brands
because these brands are recognizable, and most people associate desirable traits with such brands. Sirgy (1985) suggests that conspicuous
brands embody more congruity with the ideal self and inconspicuous
brands are more consistent with the actual self. Our findings revealed
similar themes. The following exemplar relates to Derek's use of conspicuous, iconic brands to demonstrate his ideal self on Facebook.
I want people to know that I play golf, so my page shows different
golf brands. But, I really want people to think I am better than I really
am at golf, so I intentionally pick golf brands that are reputable and
expensive. Like I talk about my Great Big Bertha golf bag and my
dream John Lobb golf shoes, and my Oakley's Romeo 2 sunglasses
(Derek, ideal self-representation exemplar).
Derek believes that iconic brands such as John Lobb and Big Bertha
convey to others his desire to become a proficient golf player. This is
an expansion of his actual self. Derek plays golf currently; he wishes
to become a better player and suggests, “I think most people want
to be really good at something.” Derek's connection with iconic golf
brands is driven by normative motives. Although Derek does not
own all the golf brands displayed on his page, he identifies with
recognizable brands to express an idealized image of himself that is
consistent with the normative expectations of superior golf brands
(e.g., high performance).
With the noted examples above (greater actual self or greater ideal
self), the actual and ideal selves are congruous because the multiple identities (actual self, ideal self-representation, and ideal self-presentation)
can be blended on a unified platform. Although participants may present
more of the actual selves, ideal self-representation, or ideal selfpresentation, the identities are still harmonious. For instance, Matt believes that he presents more of his actual self on his Facebook profile
but, during his interview, he also discussed brands that represent his
ideal self such as Umbro that mirrors his dream of becoming a pro soccer
player.
4.5. Incongruous actual and ideal selves
In the following two sections, we discuss situations in which aspects the actual and ideal selves conflict or, in other words, are incongruous. In such instances, participants conveyed that they used two
very different coping strategies: an exclusive ideal strategy (to enhance and protect) or an avoidance strategy.
4.5.1. Exclusive ideal self expression
As previously noted, Facebook provides greater opportunity (vs.
offline environments or other online venues) for the actual and ideal
selves to conflict. In fact, our findings reveal that the social networking
platform may elevate conflict between the actual and ideal selves because
there is only one means for communicating the self to a large network of
friends, and there is a continual opportunity to do so. In contrast, during
face-to-face interactions, consumers may show many selves to various
acquaintances at different points in time (Leary & Allen, 2011). For example, Leary and Allen suggest that the persona consumers may adopt in interactions with a minister may be very different than in interactions with
a person to whom they are romantically attracted.
When the actual and ideal identities conflict on a networking platform such as Facebook (e.g., the participant does not want to show
both the actual and ideal selves to his/her network), consumers must
find a means to cope. Chet describes the increased likelihood of conflict
between his actual and ideal selves on Facebook and his efforts to cope.
I like for people to know about the brands that I like, but you have to
be careful because you have so many viewers on Facebook. Sometimes I may really like a brand but don't want everyone to see it… like
Wal-Mart because I want my professional acquaintances to see me as
professional and in-charge and I don't think Wal-Mart conveys that
kind of image. I'm careful about my selection of brands and I know
that I'm creating a certain image. I guess there is more room to show
different aspects of myself, but also more variance or inconsistency
with what parts of myself to show different people. I'm always on display, so that means I've got to really watch what I say and do online…
so sometimes I just show one side of myself (Chet, M, 24).
As noted by Chet, Facebook allows for self-expression, but at the same
time there is more opportunity for incongruity between the actual and
ideal selves. A popular strategy among our participants when their actual
and ideal selves are incongruous is exclusive ideal self-expression. The
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
primary goal of participants using this strategy is to construe their
Facebook presence in such a way that brand linkages place their personal
attributes in the most favorable light. Using the exclusive ideal strategy,
participants seek credibility from others in their network, and this strategy includes both self-enhancement and self-protection. According to
Alicke and Sedikides (2009), a commonality between self-enhancement
and self-protection is that both concepts are based upon the notion of
viewing oneself more positively, or less negatively, than objective circumstances warrant, and both concepts can be used to advocate a desired
self-view (ideal self). As demonstrated below, when identities conflict,
users may choose to solely display aspects of their ideal selves.
When I'm not sure, I go all ideal from brands to pictures…When I
first came to college I think that I was all ideal, but now I am a little
of both…I feel like I am at a point in my life where I feel more comfortable with my actual self (Deb, exclusive ideal self strategy).
Facebook has become a place where people go to promote themselves as models. My cousin does this. She posts certain brands on
her News Feed all the time. She thinks that people are going to form
different opinions about her based on the brands she identifies with.
She wants to become a movie star so she associates with movie star
brands (Daisy, cousin uses exclusive ideal self strategy).
As illustrated above, Deb explains that she used an exclusive
ideal self strategy as she transitioned from high school to college.
She described her actual self during this transition period as immature
and insecure, but her ideal self desired to become mature and confident.
During the transition period, Deb linked with brands solely to
either enhance the self or to protect the self. As her situation changed
(e.g., “made new friends,” “made good grades,” “joined university
clubs”), she gained more confidence and began to use a blended strategy of the actual and ideal selves. The exclusive ideal self was simply
a coping strategy Deb used during a transitional period in her life.
“I was very insecure, so I didn't really want to reveal my insecure self”
(Deb). Similarly, Daisy provides an example where her cousin uses an
exclusive ideal self strategy by posting pictures of herself as a model.
“If the pictures are not model worthy, she's not gonna post it” (Daisy).
Linking with brands provides one means for deriving success with this
strategy. For instance, Daisy's cousin uses brands to embellish her desired model image and to protect her sense of self in terms of public perception. “She's far from model material in my opinion…but the brands
she friends makes people think she's the real thing” (Daisy). Daisy's
cousin is consumed by her ambition to become a model, and she uses
an exclusive ideal self strategy to cope with conflicting identities (she
lacks model-like attributes but desires to become a model).
4.5.2. Avoidance of self-expression
A second coping strategy when the actual and ideal identities conflict
is the avoidance of self-expression. Mick and Fournier (1998) describe
how consumers use an avoidance strategy to cope with the paradoxes
of new technology. Similarly, several of our participants suggested that
as they formerly engaged in job search initiatives, they closed their
Facebook accounts and avoided self-expression altogether. Daniel's narrative illustrates this point.
I know employers use the internet to get information about people
interviewing so I closed my Facebook account so that my personal
information couldn't be used against me. My character shown on
Facebook is not who I really am. I have the ability to be really professional, but you wouldn't know that by looking at my Facebook page
(Daniel, avoidance strategy).
Driven by the situational context of the job search, Daniel's actual
and ideal selves conflicted. As a result, Daniel coped by using the
avoidance strategy. When Daniel was asked to describe his actual
403
self, he used descriptors such as “party animal,” “ladies' man,” and
“get this party started kinda person.” When asked to talk about his
ideal self, Daniel explained that he wanted to be “professional and
clean cut,” “a successful corporate American man,” and “ethical.”
When Daniel was faced with searching for a summer internship, he
avoided using Facebook by suspending his account temporarily. The
avoidance strategy provided a viable means for coping when his actual and ideal selves conflicted.
Participants realize that brands communicate characteristics
about the self, and when the actual and ideal identities conflict, they
may choose to avoid self-expression altogether. Donna explains this
idea well.
Brands speak identity and if I don't want to show myself to others
for whatever reason, I'm not gonna like a brand. I'm gonna stay
away from saying anything about myself online. Whatever you
put on Facebook, it's there for the world to see…sometimes it's
just best to say nothing (Donna, avoidance strategy).
As suggested by Leary and Allen (2011), it is rare that individuals
have a single persona that they use in virtually all situations and with
virtually all targets. Therefore, SNSs such as Facebook present more opportunity for conflict between the actual and ideal selves. Facebook, due
to its connected nature, forces consumers to be more consistent and authentic in their self-expressions compared to offline and other online
personas. Therefore, consumers must make trade-offs when their actual
and ideal selves are incongruous. This means that consumers may solely
link with brands that represent aspects of their ideal selves or avoid
linking to brands altogether.
5. Discussion
Individuals represent aspects of their selves to others in their online
social networks. Even in the absence of the nonverbal cues that characterize offline communications, people use whatever information is
available online to form impressions of others (Walther & Parks,
2002). In the context of Facebook, most prior impression formation research focuses on sociometric data regarding the user's “friends” (Tong,
Heide, Langwell, & Walther, 2008; Walther et al., 2008). Less research,
however, is focused on impressions formed based on information
posted directly by a user (for an exception, see Back et al., 2010). Importantly, the information that is directly created or posted by a Facebook
user (e.g., brands “liked” by the user) can also be used to mold others'
impressions of him or herself. Consumers are networked to other
users on this platform, and within each network lies a certain degree
of visibility through features including wall posts and public displays
of connections. Indeed, the online environment allows for a transcendence of physical and material constraints, thus facilitating consumers'
extensions of self, even in the absence of material possessions (Zhao &
Belk, 2007).
Whereas earlier research suggests that Facebook users' profiles reflect only the actual self (Back et al., 2010), our findings present evidence of the use of brands to express both the actual and ideal selves.
Back et al.'s (2010) study of self-expression on Facebook was conducted
at a relatively high level of inquiry, examining users' Facebook profiles
as a whole. Our unique methodology and tighter focus on brands
allow for a deeper understanding and discovery of more subtle
ideal-self expressions that slipped “under the radar” in earlier research.
Indeed, we find that most people edit their presented selves in some
motivated way. Over half of our participants self-reported expressing
the ideal self to a greater extent than the actual self in their linkages
to brands. Furthermore, participants revealed two subcomponents of
the ideal self: ideal self-representation and ideal self-presentation.
Ideal self-representation involves an expansion of the actual self
(self-enhancement), while ideal self-presentation is a limitation of the
404
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
actual self (self-protection). Thus, we identify reflections of self in this
novel context that appear to have gone undetected in prior work.
An interesting finding of this research is that the ideal and actual
selves often conflict on Facebook. Consequently, we contribute to the
literature on how multiple selves interact to inform brand connections.
Due to the virtual nature of Facebook as a unified platform for networking, consumers are forced to present a single persona to a wide range of
acquaintances. In deciding how to express the self on Facebook, we find
that participants' actual and ideals selves either blend (congruous) or
conflict (incongruous). Users express their identities through brand
linkages depending on the nature of the congruity (or lack thereof).
When the identities are congruous, participants present either the actual or ideal self to a greater extent by linking with brands; however, aspects of both the actual and ideal exist. Therefore, we found evidence
of a blending of the two identities. In other words, participants use a
mix of brands representing both the actual and ideal selves, but typically more brands will represent either the actual or ideal self. When the
identities are incongruous, participants take one of two routes in an effort to resolve the incongruity: linking with brands to enhance and protect the self-concept (i.e., ideal only strategy), or abstaining from linking
to brands altogether. The avoidance of some brands and the embrace
of others is a means through which users express aspects of their
self-concepts. As the usage of social media platforms such as Facebook
continues to increase, it is important for marketers to understand
whyconsumers link with certain brands and avoid others as a form of
self-expression.
Our study sheds light on the importance that marketers need to
place on their social media promotional efforts. Participants in our
study conveyed that brands can send powerful virtual messages. Furthermore, our findings demonstrate that a brand's online image is just
as important as its offline image, and possibly more important. Consistent with self-concept theory, consumers behave in ways that maintain,
enhance, and protect their senses of self (Cheema & Kaikati, 2010), and
one way consumers do this is by linking with brands on Facebook to
portray a certain image to others.
6. Further research opportunities
There are several avenues for extending this research. First, future
research could explore how individual difference factors moderate consumers' uses of brands to express their selves on Facebook. One individual difference factor is a person's attachment style. Recent research
shows that anxiously attached individuals use brands as a way to signal
their ideal self-concepts to others in a public setting (Swaminathan et
al., 2009). Thus, these individuals might be especially likely to link
themselves to brands on Facebook to communicate their ideal (vs. actual) selves. A second individual difference factor of interest is selfmonitoring, which reflects the extent to which a person observes and
controls expressive behavior and self-presentation in accordance with
social cues (Snyder, 1979). As they are highly concerned with the social
environment, high self-monitors might also have a strong tendency to
express themselves with brands on Facebook. In addition, future research should examine the relationship between image congruence
and brand evaluations among high self-monitors compared to low
self-monitors within the context of a consumer's online social network.
Second, future research should investigate how “liking” a brand on
Facebook affects the way in which the user perceives him or herself. In essence, future research could attempt to apply the notions of self extensions (Belk, 1988) to the social networking medium. Recent research
shows that using a brand for only a short time leads consumers to
alter their own self-perceptions (Park & John, 2010), and a similar effect
may be found for consumers who link themselves to brands on
Facebook. Our preliminary findings suggest that participants who
linked with more brands to maintain the self-concept expressed higher
self-evaluations (e.g., more confidence, higher self-esteem). However,
expressing an ideal self may also have positive implications for
consumers' self-esteem. As self-discrepancy theory (Higgins, 1987)
notes that large ideal-actual self-discrepancies are related to lower
self-esteem, it has been suggested that consumers can increase their
well-being (i.e., reduce this discrepancy) by presenting an ideal self online (McKenna & Bargh, 1998, 2000).
Third, future research can examine the effectiveness of using brands
as subtle signals for impression construction. Recent research shows
that consumers' use of brands as signals of one's characteristics can
backfire if perceived by others as inauthentic (Ferraro, Kirmani, &
Matherly, 2011). Thus, using brands to represent the self on Facebook
could possibly have unintended, adverse consequences for users' impression formation attempts. This seems especially relevant in the context of expressions of one's ideal self-image (which might be perceived
by others as inauthentic) by linking oneself to brands on Facebook.
Fourth, our research could be leveraged to develop a number of new
theories. For instance, one new area of study could investigate how consumers use brands in social media to simply ‘hint’ at an underlying
identity, tease, or manage an impression of modesty or conservatism.
Another possibility would be to develop a theory that moves beyond a
polarized view of presentation as private versus public to suggest
more interactive concepts. Yet another area for theory-building in relation to our work is to explore how and why consumers choose to
“unfriend” brands.
Finally, future quantitative research may be useful for complementing
and validating our findings. Such research could explore the effects of social norms bias on individuals' responses, as well as the effects of other
users of the brand on self-congruity. Additional research could investigate
these and other possibilities within Facebook and other platforms in the
growing social media landscape.
References
Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34,
347–356.
Aaker, J. A. (1999). The malleable self: The role of self-expression in persuasion. Journal
of Marketing Research, 36(1), 45–57.
Alicke, Mark D., & Sedikides, Constantine (2009). Self-enhancement and self-protection:
What they are and what they do. European Review of Social Psychology, 20, 1–48.
Alpert, F., & Kamins, M. (1995). An empirical investigation of consumer memory, attitude,
and perceptions toward pioneer and follower brands. Journal of Marketing, 59(4), 34–45.
Arnould, Eric J., & Thompson, C. J. (2005). Consumer culture theory: Twenty years of research. Journal of Consumer Research, 31(4), 868–882.
Back, M. D., Stopfer, J. M., Vazire, S., Gaddis, S., Schmukle, S. C., Egloff, B., et al. (2010).
Facebook profiles reflect actual personality, not self-idealization. Psychological Science,
21(3), 372–374.
Bargh, J., McKenna, K. Y. A., & Fitzsimons, G. M. (2002). Can you see the real me? Activation and self-expression of the ‘true self’ on the internet. Journal of Social Issues,
58(1), 33–48.
Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research, 15,
139–168.
Belk, R. W., Sherry, J. F. J., & Wallendorf, M. (1988). A naturalistic inquiry into buyer and
seller behavior at a swap meet. Journal of Consumer Research, 14(4), 449–470.
Bessière, K., Seay, A. F., & Kiesler, S. (2007). The ideal elf: Identity exploration in World
of Warcraft. Cyberpsychology & Behavior, 10(4), 530–535.
Boyd, D. M., & Ellison, N. B. (2009). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13, 210–230.
Brown, J., Broderick, A. J., & Lee, N. (2007). Word of mouth communication within online communities: Conceptualizing the online social network. Journal of Interactive
Marketing, 21(3), 2–20.
Burawoy, M. (1991). Ethnography unbound. Berkeley and Los Angeles: University of
California Press.
Chaplin, L. N., & John, D. R. (2005). The development of self-brand connections in children and adolescents. Journal of Consumer Research, 32, 119–129.
Chebat, J., Sirgy, M. J., & St-James, V. (2006). Upscale image transfer from malls to stores:
A self-image congruence explanation. Journal of Business Research, 59, 1288–1296.
Cheema, A., & Kaikati, A. M. (2010). The effect of need for uniqueness on word of mouth.
Journal of Marketing Research, 47, 553–563.
Close, A. G., Krishnen, A., & LaTour, M. S. (2009). This event is me!: How consumer-event
congruity leverages sponsorship. Journal of Advertising Research, 49(3), 271–284.
Close, A. G., Moulard, J. G., & Monroe, K. B. (2011). Establishing human brands: Determinants of placement success for first faculty positions in marketing. Journal of the
Academy of Marketing Science, 39, 922–941.
Coupland, J. C. (2005). Invisible brands: An ethnography of households and the brands
in their kitchen pantries. Journal of Consumer Research, 32(1), 106–118.
Dolich, I. (1969). Congruence relationships between self images and product brands.
Journal of Marketing Research, 6(1), 80–84.
C.R. Hollenbeck, A.M. Kaikati / International Journal of Research in Marketing 29 (2012) 395–405
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook ‘friends:’ Social
capital and college students' use of online social network sites. Journal of ComputerMediated Communication, 12(4), 1143–1168.
Escalas, J. E., & Bettman, J. R. (2005). Self-construal, reference groups, and brand meaning. Journal of Consumer Research, 32(3), 378–389.
Fennis, B., Pruyn, A. T. H., & Maasland, M. (2005). Revisiting the malleable self: Brand
effects on consumer self-perceptions of personality traits. Advances in Consumer
Research, 32(1), 371–377.
Ferraro, R., Kirmani, A., & Matherly, M. (2011). Are they who they claim? Intentionality and
authenticity in identity signaling with brandsWorking paper. University of Maryland.
Fournier, S. (1998). Consumers and their brands: Developing relationship theory in
consumer research. Journal of Consumer Research, 24(4), 343–373.
Graeff, T. R. (1996). Image congruence effects on product evaluations: The role of selfmonitoring and public/private consumption. Psychology and Marketing, 13(5), 481–499.
Granovetter, M. S. (1973). The strength of weak ties. The American Journal of Sociology,
78(6), 1360–1380.
Grubb, E. L., & Grathwohl, H. L. (1967). Consumer self-concept, symbolism, and market
behavior: A theoretical approach. Journal of Marketing, 31, 22–27.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological
Review, 94(3), 319–340.
Hirschman, E. C. (1992). The consciousness of addiction: Toward a general theory of
compulsive consumption. Journal of Consumer Research, 19(2), 155–179.
Holt, D. B. (2002). Why do brands cause trouble? A dialectical theory of consumer culture and branding. Journal of Consumer Research, 29, 70–90.
Holt, D. B. (2004). How brands become icons: The principles of cultural branding. Boston:
Harvard Business School Press.
James, W. (1890). The principles of psychology. New York: Henry Holt.
Kirmani, A. (2009). The self and the brand. Journal of Consumer Psychology, 19(3), 271–275.
Kozinets, R. V., Valck, K., Wojnicki, A. C., & Winer, S. J. S. (2010). Networked narratives:
Understanding word-of-mouth marketing in online communities. Journal of Marketing, 74, 71–89.
Kressmann, F., Sirgy, M. J., Herrmann, A., Huber, F., Huber, S., & Lee, D. (2006). Direct
and indirect effects of self-image congruence on brand loyalty. Journal of Business
Research, 59, 955–964.
Landon, L. (1974). Self-concept, ideal self-concept, and consumer purchase intentions.
Journal of Consumer Research, 1(2), 44–51.
Leary, M. R., & Allen, A. B. (2011). Self-presentational persona: Simultaneous management of multiple impressions. Journal of Personality and Social Psychology, 101(5),
1033–1049.
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review
and two-component model. Psychological Bulletin, 107, 34–47.
Levy, S. J. (1959). Symbols for sales. Harvard Business Review, 37(4), 117–124.
Markus, H., & Kunda, Z. (1986). Stability and malleability of the self-concept. Journal of
Personality and Social Psychology, 15, 858–866.
Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41(9), 954–969.
McKenna, K. Y. A., & Bargh, J. (1998). Coming out in the age of the internet: Identity
“demarginalization” through virtual group participation. Journal of Personality and Social Psychology, 75, 681–694.
McKenna, K. Y. A., & Bargh, J. (2000). Plan 9 from cyberspace: The implications of
the internet for personality and social psychology. Personality and Social Psychology
Review, 4, 57–75.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Mick, D. G., & Fournier, S. (1998). Paradoxes of technology: Consumer cognizance, emotions,
and coping strategies. Journal of Consumer Research, 25(2), 123–143.
Otnes, C., Lowrey, T. M., & Shrum, L. J. (1997). Toward an understanding of consumer
ambivalence. Journal of Consumer Research, 24, 80–93.
Park, J. K., & John, D. R. (2010). Got to get you into my life: Do brand personalities rub
off on consumers. Journal of Consumer Research, 37(4), 655–669.
405
Parks, M. R. (2007). Personal networks and personal relationships. Mahwah, NJ: Lawrence Erlbaum Associates.
Ross, I. (1971). Self-concept and brand preference. Journal of Business, 44(1), 38–50.
Schau, H. J., & Gilly, M. C. (2003). We are what we post? Self-presentation in personal
web space. Journal of Consumer Research, 30(3), 385–404.
Schenk, C. T., & Holman, R. H. (1980). A sociological approach to brand choice: The concept of situational self-image. Advances in Consumer Research, 7(1), 610–614.
Schouten, J. W. (1991). Selves in transition: Consumption in personal rites of passage
and identity reconstruction. Journal of Consumer Research, 17(4), 412–425.
Shavelson, R. J., Hubner, J. J., & Stanton, G. C. (1976). Self-concept: Validation of construct interpretations. Review of Educational Research, 46(3), 407–441.
Sirgy, M. J. (1982). Self-concept in consumer behavior: A critical review. Journal of
Consumer Research, 9(3), 287–300.
Sirgy, M. J. (1985). Using self-congruity and ideal congruity to predict purchase motivation. Journal of Business Research, 13, 195–206.
Snyder, M. (1979). Self-monitoring processes. Advances in Experimental Social Psychology,
12, 85–128.
Solomon, M. R. (1983). The role of products as social stimuli: A symbolic interactionism
perspective. Journal of Consumer Research, 10(3), 319–329.
Stephen, A. T., & Galak, J. (2010). The complementary roles of traditional and social media
publicity in driving marketing performance. INSEAD working paper 2010/97/MKT.
Stern, S. R. (2004). Expressions of identity online: Prominent features and gender differences in adolescents' World wide web home pages. Journal of Broadcasting &
Electronic Media, 48(2), 218–243.
Stone, B. (2010). How Facebook sells you. Bloomberg Businessweek, 27, 64–72.
Strizhakova, Y., Coulter, R. A., & Price, L. L. (2008). The meanings of branded products: A
cross-national scale development and meaning assessment. International Journal of
Research in Marketing, 25(2), 82–93.
Strizhakova, Y., Coulter, R. A., & Price, L. L. (2011). Branding in a global marketplace:
The mediating effects of quality and self-identity brand signals. International
Journal of Research in Marketing, 28(4), 342–351.
Swaminathan, V., Stilley, K. M., & Ahluwalia, R. (2009). When brand personality matters: The moderating role of attachment styles. Journal of Consumer Research, 35,
985–1002.
Thomson, M. (2006). Human brands: Investigating antecedents to consumers' strong
attachments to celebrities. Journal of Marketing, 70, 104–119.
Tong, S. T., Heide, B. V. D., Langwell, L., & Walther, J. B. (2008). Too much of a good thing?
The relationship between number of friends and interpersonal impressions on
Facebook. Journal of Computer-Mediated Communication, 13, 531–549.
Torelli, C. J., Monga, A., & Kaikati, A. M. (2012). Doing poorly by doing good: Corporate
social responsibility and brand concepts. Journal of Consumer Research, 38(5),
948–963.
Trusov, M., Bodapati, A. V., & Bucklin, R. E. (2010). Determining influential users in social networks. Journal of Marketing Research, 57, 643–658.
Urstadt, B. (2008). Social networking is not a business. MIT Technology Review, July/August.
Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal,
and hyperpersonal interaction. Communication Research, 23(1), 1–43.
Walther, J. B., Heide, B. V. D., Kim, S. Y., Westerman, D., & Tong, S. T. (2008). The role
of friends' appearance and behavior on evaluations of individuals on Facebook:
Are we known by the company we keep? Human Communication Research, 34,
28–49.
Walther, J. B., & Parks, M. R. (2002). Cues filtered out, cues filtered in: Computer-mediated
communication and relationships. In M. L. Knapp, & J. A. Daly (Eds.), Handbook of Interpersonal Communication (pp. 529–563). (3rd ed.). Thousand Oaks, CA: Sage.
Zhao, Xin, & Belk, R. W. (2007). Live from shopping malls: Blogs and Chinese consumer
desire. Advances in Consumer Research, 34, 131–137.
Intern. J. of Research in Marketing 29 (2012) 406–418
Contents lists available at SciVerse ScienceDirect
Intern. J. of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar
Drivers of consumer–brand identification
Nicola Stokburger-Sauer a, 1, S. Ratneshwar b, 2, Sankar Sen c,⁎
a
b
c
Department for Strategic Management, Marketing and Tourism, University of Innsbruck, Universitaetsstr. 15, A-6020 Innsbruck, Austria
Trulaske College of Business, University of Missouri, Columbia, MO 65203, United States
Zicklin School of Business, Baruch College/CUNY, One Bernard Baruch Way, New York, NY 10010, United States
a r t i c l e
i n f o
Article history:
First received in 18, July 2011 and was under
review for 6 months
Available online 23 September 2012
Keywords:
Consumer–brand identification
Consumer self-identity
Brand relationships
Product category involvement
a b s t r a c t
The concept of consumer–brand identification (CBI) is central to our understanding of how, when, and why
brands help consumers articulate their identities. This paper proposes and tests an integrative theoretical
framework of the antecedents of CBI. Six drivers of CBI, a moderator, and two consequences are posited
and tested with survey data from a large sample of German household consumers. The results confirm the
influence of five of the six drivers, namely, brand–self similarity, brand distinctiveness, brand social benefits,
brand warmth, and memorable brand experiences. Further, we find that all five of these antecedents have
stronger causal relationships with CBI when consumers have higher involvement with the brand's product
category. Finally, CBI is tied to two important pro-company consequences, brand loyalty and brand advocacy.
Theoretical and managerial significance of the findings are discussed.
© 2012 Elsevier B.V. All rights reserved.
“Choices are made more easily—either more routinely or more impulsively, seemingly—because one object is symbolically more
harmonious with our goals, feelings, and self-definitions than
another.”Sidney J. Levy (1959, p. 120)
“Why has the Toyota Prius enjoyed such success … when most
other hybrid models struggle to find buyers? One answer may be
that buyers of the Prius want everyone to know they are driving
a hybrid …. In fact, more than half the Prius buyers surveyed this
spring … said the main reason they purchased their car was that ‘it
makes a statement about me.’”Micheline Maynard (2007)
1. Introduction
Striving for a sense of self (i.e., answering the question, “Who am I?”)
is a fundamental aspect of the human condition (Belk, 1988; Berger &
Heath, 2007; Brewer, 1991; Freud, 1922; Kleine, Kleine, & Kernan,
1993; Tajfel & Turner, 1985). Further, as succinctly put by Belk (1988, p.
160), “we are what we have”—what we buy, own, and consume define
us to others as well as to ourselves. In this context, it is widely recognized
⁎ Corresponding author. Tel.: +1 646 312 3302.
E-mail addresses: nicola.stokburger-sauer@uibk.ac.at (N. Stokburger-Sauer),
ratneshwar@missouri.edu (S. Ratneshwar), sankar.sen@baruch.cuny.edu (S. Sen).
1
Tel.: +43 512 507 7080.
2
Tel.: +1 573 882 3748.
0167-8116/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijresmar.2012.06.001
that brands have the ability to embody, inform, and communicate desirable consumer identities (Bhattacharya & Sen, 2003; Escalas, 2004;
Escalas & Bettman, 2003, 2009; Fournier, 1998, 2009; Lam, Ahearne, Hu,
& Schillewaert, 2010; Levy, 1959; Strizhakova, Coulter, & Price, 2008;
Tsai, 2005). Not surprisingly then, a growing body of research has focused
on what it means for consumers to identify with brands and the implications of such consumer–brand identification (CBI) for both consumer behavior and effective brand management (e.g., Chernev, Hamilton, & Gal,
2011; Escalas & Bettman, 2003, 2009; Lam et al., 2010).
Much less is understood, however, about the drivers of CBI—what
factors cause it, when, and why. While a comprehensive sense for
what produces CBI is of considerable importance to both marketing
academics and practitioners, these issues have been examined from
many diverse perspectives, causing our understanding to be rather
fragmented. For instance, in their work on consumer–company identification, Bhattacharya and Sen (2003) draw on social identity theory
to proffer consumers' self-defining and enhancing motives as the
main drivers of identification (see also Ahearne, Bhattacharya, & Gruen,
2005; Bhattacharya, Rao, & Glynn, 1995). The work of Escalas and
Bettman (2003, 2009), on the other hand, locates such identity-based
bonds in the broader social context, suggesting that consumers bond
with brands whose identities are consonant with desirable reference
groups and celebrity endorsers. The communal consumption of brands
and its role in the construction of identity narratives by consumers is
stressed in the work of Fournier (2009), McAlexander, Schouten, and
Koenig (2002), Muniz and O'Guinn (2001), and O'Guinn and Muniz
(2009). Sociocultural factors such as the circulation of brand stories and
myths among consumers are highlighted by Brown, Kozinets, and
Sherry (2003), Diamond et al. (2009), Holt (2005), and Thompson,
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
Rindfleisch, and Arsel (2006). Thomson, MacInnis, and Whan Park (2005)
and Park, MacInnis, Priester, Eisingerich, and Iacobucci (2010), on the
other hand, emphasize the role of emotional reactions to the brand in
the formation of consumer–brand connections. It is also worth noting
that while it is assumed that CBI can occur in a wide range of categories,
empirical research in this domain has usually been restricted to single category studies (e.g., Lam et al., 2010), thereby precluding a deeper understanding of the category-specific determinants, if any, of CBI.
Given this backdrop of fragmented insights into the drivers of CBI, the
present paper makes three key contributions. First, it synthesizes a wide
range of ideas pertaining to identity construction, identification, and
brand relationships to provide a comprehensive conceptual framework
for the determinants of CBI. The outcome of our conceptual synthesis is
a set of six antecedents of CBI that includes three primarily cognitive variables (brand–self similarity, brand distinctiveness, and brand prestige) as
well as three affectively rich brand-related factors (brand social benefits,
brand warmth, and memorable brand experiences). Second, to strengthen the validity of our conceptual framework and to go beyond extant
single-category examinations of CBI (e.g., Lam et al., 2010), we implicate
an important category-level variable—product category involvement—as
a moderator of the relationships between CBI and its various drivers. Additionally, we relate CBI to two key consequences: brand loyalty and
brand advocacy. Finally, in the process of establishing this nomological
network for CBI, we develop a valid, parsimonious measure of this
focal construct that attempts to assess the state of CBI more independently of its antecedents and consequences (e.g., social rewards, negative affective states produced by discontinued brand usage) than
extant measures (e.g., Lam et al., 2010).
Next, we develop our conceptual framework culminating in a set of
predictions pertaining to the antecedents and consequences of CBI. We
then test our hypotheses with survey data from a large sample of German household consumers. The paper ends with a discussion of the theoretical and managerial significance of our findings.
2. Conceptual framework
2.1. The concept of consumer–brand identification
Brands, as carriers of symbolic meanings (Levy, 1959), can help consumers achieve their fundamental identity goals and projects (Belk,
407
1988; Escalas & Bettman, 2009; Fournier, 2009; Holt, 2005; Huffman,
Ratneshwar, & Mick, 2000). Therefore, consumer–brand identification,
defined here as a consumer's perceived state of oneness with a brand,
is a valid and potent expression of our quest for identity-fulfilling meaning in the marketplace of brands. This definition is consistent with the organizational behavior literature, wherein identification typically has
been defined as a perception of oneness with or belongingness to some
human aggregate, such as employees with their companies or students
with their alma maters (Ashforth & Mael, 1989; Bergami & Bagozzi,
2000; Bhattacharya et al., 1995; Mael & Ashforth, 1992; Stuart, 2002).
Note that consistent with the theory in this domain (Bergami &
Bagozzi, 2000; Bhattacharya & Sen, 2003), we assume that the state of
CBI is distinct from the process of comparison of self traits with brand
traits that may contribute to CBI.
Table 1 presents an overview of constructs related to CBI and their
definitions from prior literature. While our conceptualization of CBI is
rooted in organizational identity, it is related to the construct of
self-brand connections proposed by Escalas and Bettman (2003,
2009), defined as the extent to which an individual has incorporated a
brand into his or her self-concept. However, the construct of CBI is, notably, narrower in that it excludes the potential motivations guiding
such self-brand connections, such as communicating one's identity to
others and achieving one's desired self (both part of Escalas and
Bettman's self-brand connection measure). CBI is similarly akin to the
broader notions of brand-self connection in the work of Park et al.
(2010) and the component called self-connection in Fournier's (2009)
Brand Relationship Quality (BRQ) scale. As in the case of self-brand connections, we regard the construct of CBI as narrower than, but potentially overlapping with, constructs such as BRQ (Fournier, 2009) and brand
attachment (Park et al., 2010).
Lam et al. (2010) take a somewhat different approach by defining
CBI as “a customer's psychological state of perceiving, feeling, and valuing his or her belongingness with a brand” (p. 130). In doing so, these
authors view CBI as a formative construct composed of three dimensions. The cognitive dimension of their construct is similar to the notion
of cognitive organizational identification in the work of Bergami and
Bagozzi (2000). The emotional consequences of brand usage serve as
the second dimension of Lam et al.'s CBI construct, and “evaluative
CBI” is its final dimension, defined as “whether the consumer thinks
the psychological oneness with the brand is valuable to him or her
Table 1
Constructs related to consumer–brand identification in prior literature.
Construct name
Construct definition
Representative literature
Organizational identification
Perceived oneness with an organization and the experience of the organization's
successes and failures as one's own.
Cognitive connection between the definition of an organization and the definition
a person applies to himself or herself, thereby viewing identification as a process
of self-definition.
Perceived oneness or belongingness to an organization of which the person is a member.
Cognitive state of self-categorization that reflects self-awareness of one's membership
in an organization.
Identification with a company as an active, selective, and volitional act motivated by
the satisfaction of one or more self-definitional needs.
Customer's psychological state of perceiving, feeling, and valuing his or her
belongingness with a brand.
Individual's judgment of the extent to which a brand is good or bad.
Degree of passionate emotional attachment a satisfied consumer has for a
particular trade name.
Different cognitions (e.g., about self-identity), feelings, sense of connectedness and
fit, and behaviors (e.g., frequent interactions, and resource investments).
Strength of the bond connecting a brand with the self.
Bond that connects a consumer with a specific brand and involves feelings
(i.e., affection, passion, and connection) toward the brand.
Salience of the cognitive and affective bond that connects a brand to the self.
Extent to which an individual has incorporated a brand into his or her self-concept.
Ashforth and Mael (1989), Mael and
Ashforth (1992)
Dutton, Dukerich, and Harquail (1994),
and Dukerich, Golden, and Shortell (2002)
Cognitive organizational
identification
Consumer–company
identification
Customer–brand
identification
Brand attitude
Brand love
Brand attachment
Emotional brand attachment
Brand prominence
Self-brand connection,
Brand-self connection
Consumer–brand identification Consumer's perceived state of oneness with a brand.
Bhattacharya et al. (1995)
Bergami and Bagozzi (2000)
Bhattacharya and Sen (2003)
Lam et al. (2010)
Park et al. (2010)
Carroll and Ahuvia (2006)
Batra, Ahuvia, and Bagozzi (2012)
Park et al. (2010)
Malär, Krohmer, Hoyer, and Nyffenegger (2011)
Park et al. (2010)
Escalas and Bettman (2003, 2009), Fournier (2009),
and Park et al. (2010)
Present research
408
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
individually and socially” (p. 137). Unlike Lam et al. (2010), we do not
regard brand partner value as part of the construct of identification. Indeed, as we argue later, the social benefits of a brand can actually influence brand identification, thereby serving as an antecedent to the
construct rather than being an integral part of it. We further depart
from Lam et al.'s (2010) conceptualization in that, consistent with the
theory of Park et al. (2010), we view CBI primarily as a cognitive representation, albeit one that can have an abundance of emotional associations. Our view is consistent with that of Bergami and Bagozzi (2000),
who argue that the emotional consequences of identification must be
kept separate from the state of identification.
More generally, we regard CBI as different from the pure emotional bond that is embodied in the concepts of emotional brand attachment (Malär, Krohmer, Hoyer, & Nyffenegger, 2011) and brand love
(Batra, Ahuvia, & Bagozzi, 2012; Carroll & Ahuvia, 2006). Batra et al.
(2012), in particular, include not only the dimensions of positive
emotional connection and self-brand integration in their construct
of brand love, but also positive brand evaluations. In contrast, we regard positive brand evaluations (i.e., brand attitudes) as conceptually
different from CBI because while the former references just the brand
as a target of evaluation, CBI hinges on both the perceived brand identity and the self-identity of the consumer. Accordingly, brand evaluations are likely to be either inputs or outputs of CBI rather than a part
of CBI. Similarly, Allen and Meyer's (1990) measure of affective commitment goes beyond the individual's perceived psychological state
to include certain consequences of that state (e.g., “I enjoy discussing
my organization with people outside it”).
2.2. Antecedents of CBI
The need for identification is thought to be motivated by one or
more higher-order self-definitional needs (Brewer, 1991; Kunda,
1999; Tajfel & Turner, 1985). Specifically, we need to (1) know ourselves, (2) feel relatively unique, and (3) feel good about ourselves.
Thus, three key needs that are likely to drive identification in the consumption domain are that for self-continuity or self-verification,
self-distinctiveness, and self-enhancement (Berger & Heath, 2007;
Bhattacharya & Sen, 2003; Chernev et al., 2011). Based on these need
drivers, we argue that identification with a brand is likely to be related
to the extent to which a person perceives the brand (1) to have a personality that is similar to his or her own (i.e., brand–self similarity),
(2) to be unique or distinctive, and (3) and to be prestigious.
Note, however, that many theories of social identity and identification based on self-definitional needs (e.g., those in the social cognition
and organizational behavior domains) mainly focus on the cognitive
construction of the self. As such, they do not fully account for the nature
of the ties that bind consumers to brands. Most fundamentally, brands
are things that we consume, often over time and repeatedly, thereby implicating aspects of the consumption experience itself as integral to why
we identify with some brands and not many others (Escalas, 2004;
Escalas & Bettman, 2003; Fournier, 1998; Holt, 2005; McAlexander et
al., 2002; Thompson et al., 2006; Thomson et al., 2005). Accordingly,
we consider the phenomenology of consumers' interactions with
brands and proffer three additional antecedent factors as predictive of
consumer–brand identification. These include the extent to which consumers (1) feel that their interactions with a brand help them connect
with important social others, (2) perceive a brand in warm, emotional
terms rather than cold, rational ones, and (3) have fond memories of
brand consumption experiences (see Fig. 1). Notably, these three antecedents are, by virtue of their experiential nature, more affect-laden
than the three previously described cognitively driven antecedents of
identification.
2.2.1. Brand–self similarity
Much research (e.g., Kunda, 1999) attests to self-continuity or
self-verification as a key motive for why people desire to maintain a
clear and functional sense of who they are. Moreover, this need for a stable and consistent sense of self is increasingly met in today's consumer
culture through assessments of congruity or similarity between one's
sense of self and one's sense of commercial entities, such as companies
or brands (see Bhattacharya & Sen, 2003; Escalas & Bettman, 2003;
Kleine et al., 1993). In the brand domain, several researchers have
noted the important role of the perceived congruity between brand
and self personalities in consumers' affiliations with brands (Aaker,
1997; Grubb & Grathwohl, 1967; Levy, 1959; Sirgy, 1982). Based on
this prior literature, we formally define brand–self similarity as the degree of overlap between a consumer's perception of his or her own personality traits and that of the brand, and we propose that this construct
is a determinant of CBI.
H1. The more the brand–self similarity, the more the consumer will
identify with that brand.
2.2.2. Brand distinctiveness
It has long been recognized that people strive to distinguish themselves from others in social contexts (e.g., Tajfel & Turner, 1985);
Snyder and Fromkin's (1977) theory of uniqueness positions this need
as a key component of people's drive to feel good about themselves
(i.e., self-esteem). This theme is developed further in Brewer's (1991)
theory of optimal distinctiveness, which suggests that people attempt
to resolve the fundamental tension between their need to be similar
to others and their need to be unique by identifying with groups that
satisfy both needs. The expression of such needs for distinctiveness in
the consumption realm is perhaps best reflected in the construct labeled as consumer's “need for uniqueness” (Tepper Tian, Bearden, &
Hunter, 2001), defined as “an individual's pursuit of differentness relative to others that is achieved through the acquisition, utilization, and
disposition of consumer goods for the purpose of developing and enhancing one's personal and social identity” (p. 50).
We therefore posit that the distinctiveness of a brand is a key precursor to a consumer's desire to identify with that brand (see also
Berger & Heath, 2007). Further support for this argument comes from
research that has documented that consumers often seek to affirm
their identities via consumption of brands that are perceived as being
the polar opposites of mass-production, mass-consumption brands
(Thompson et al., 2006). All else being equal, brands with images or
identities that set them apart from their competitors will be more likely
to be identified with, provided, of course, that the basis of this distinctiveness is not perceived as entirely undesirable or negative. Formally,
we define brand distinctiveness as the perceived uniqueness of a
brand's identity in relation to its competitors, and we implicate it as a
driver of CBI.
H2. The more a consumer perceives a brand to be distinctive, the
more the person will identify with that brand.
2.2.3. Brand prestige
People like to see themselves in a positive light. Self-concept research
(Kunda, 1999) indicates that people's need for self-continuity goes
hand-in-hand with their need for self-enhancement, which entails the
maintenance and affirmation of positive self-views, which lead to
increased levels of self-esteem. Thus, it is not surprising that this
identity-related need is also met, in part, through people's identification
with prestigious social entities such as organizations (Bhattacharya &
Sen, 2003; Dutton, Dukerich, & Harquail, 1994; Mael & Ashforth, 1992).
This aspect is paralleled in the notion of the extended self in the domain
of consumer behavior, which refers to the incorporation of products and
services that reflect positively on the owner into the person's sense of self
(Belk, 1988; Kleine et al., 1993). More broadly, much consumer research
attests to the driving role of self-enhancement in consumers' affinities toward brands (Escalas & Bettman, 2003; Fournier, 1998; Rindfleisch,
Burroughs, & Wong, 2009; Thomson et al., 2005). Drawing on this
409
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
Conceptual framework
Product
Involvement
Brand-Self
Similarity
H1(+)
H7a-f(+)
Brand
Distinctiveness
Brand
Prestige
Brand Social
Benefits
Brand
Warmth
H2(+)
H3(+)
H8(+)
Brand
Loyalty
H9(+)
Brand
Advocacy
ConsumerBrand Identification
H4(+)
H5(+)
H6(+)
Memorable Brand
Experiences
Fig. 1. Conceptual framework.
research, we hypothesize that brand prestige, defined as the status or esteem associated with a brand, is a driver of CBI.
H3. The more a consumer perceives a brand to be prestigious, the
more the person will identify with that brand.
2.2.4. Brand social benefits
Self-definition is, per se, a social endeavor that involves locating
oneself in reference to one's social environment (Berger & Heath,
2007; Brewer, 1991; Laverie, Kleine, & Kleine, 2002; Solomon, 1983;
Tajfel & Turner, 1985). Further, several studies suggest that brands
are major carriers of social and cultural meaning (Diamond et al.,
2009; Holt, 2005; Thompson et al., 2006). Three broad streams of research point to the social benefits provided by certain brands.
First, research on reference groups, defined as social groups that are
important to a consumer, suggests that people often consume brands
used by their reference groups to gain or strengthen their membership
in such groups (see Escalas & Bettman, 2003 for a review). Second, the
growing literature on brand communities portrays the brand as an essential device for connecting people to one another (Muniz &
O'Guinn, 2001; O'Guinn & Muniz, 2009; Stokburger-Sauer, 2010). A
brand community can be conceptualized as a specialized, nongeographically bound community that is based on a structured set of social relationships among the admirers of a brand (Muniz & O'Guinn,
2001). Membership in such communities can result not only in identification with the community, but also with the brand that is its raison
d'etre (Bagozzi, Bergami, Marzocchi, & Morandin, 2012; Bagozzi &
Dholakia, 2006; O'Guinn & Muniz, 2009). Finally, and most broadly,
scholars who have investigated subcultures of consumption suggest
that consumers sometimes coalesce into distinct subgroups of society
on the basis of a shared commitment to a brand (Schouten &
McAlexander, 1995; Thompson et al., 2006). As in the case of brand
communities, such groups are identifiable. That is, they can have a hierarchical social structure, a set of shared beliefs and values, and a unique
set of rituals and jargon. Based on these various lines of research, we
argue that consumers are more likely to identify with brands that help
them to connect with important others, groups, communities, or
subcultures. Formally defined as the social interaction opportunities
and gains afforded by a brand, we expect brand social benefits to be a
driver of CBI.
H4. The more social benefits a consumer perceives in a brand, the
more the person will identify with that brand.
2.2.5. Brand warmth
Drawing on Hirschman and Holbrook's (1982) notion of hedonic
consumption, a large and growing body of research distinguishes between hedonic and utilitarian benefits, where the former refer to intensely experiential or emotional benefits (e.g., fun, pleasure, and
excitement) and the latter refer to instrumental or functional benefits
(e.g., Dhar & Wertenbroch, 2000; Okada, 2005). Interestingly, research in the person's perception area provides a similar dichotomy
by theorizing that the content of people's stereotypes can be organized with two key perceptual dimensions, “warmth” and “competence” (Fiske, Cuddy, Glick, & Xu, 2002). Thus, a social or ethnic
category in which a person belongs may be seen by others in relatively warm or cold terms, independent of perceptions of effectiveness,
capability, or competence.
We suggest that a similar warm versus cold distinction can be applied to brands. Based on a brand's product category (e.g., clothing vs.
coat hangers), its salient or differentiating attributes (e.g., visually
pleasing vs. boring esthetics in the brand's product designs), and its
positioning via marketing communications (e.g., Apple vs. Dell;
Rathnayake, 2008), the brand's personality can come across as relatively warm or cold (Aaker, 1997; Fournier, 1998; Keller, 2004). Further, as in the case of person perception (Fiske et al., 2002), we posit
that the warm–cold distinction as applied to brands is relatively independent of perceptions of brand quality, reliability, and functionality.
We further propose that the extent to which a brand is perceived
in warm, emotional terms rather than in cold, rational ways is a key
determinant of CBI. Given that identity construction and maintenance
are inherently affective processes, warm, lovable brands are likely to
be viewed as more suitable candidates for such important
life-projects than cold brands. This notion is underscored by research
410
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
on brand love, the passionate attachments that some consumers form
with brands, which is associated primarily with hedonic brands
(Carroll & Ahuvia, 2006). Indeed, much research points to the integral
role of emotions in the construction of consumption-based identities
(see Laverie et al., 2002 for a recent review), thus suggesting that
warm, emotional brands are stronger candidates for identification
than their cold counterparts. More specifically, warm brands are
more likely, ceteris paribus, to carry more abstract, higher-order,
identity-related brand meanings that pertain, for instance, to the
brand's values and ethics, rather than to concrete, lower-order meanings that pertain to the brand's concrete features. Accordingly, warm
brands are likely to be stronger and more meaningful candidates for
identification than cold brands, and consumers will have more intense feelings about warmer brands and the role of those brands in
their lives than they will about colder brands (Fournier, 1998; Park
et al., 2010).
H5. The more a consumer perceives a brand to have a warm (vs.
cold) personality, the more the person will identify with that brand.
2.2.6. Memorable brand experiences
Brands vary in the extent to which they provide their consumers
with memorable experiences. Some brands do not occupy a salient
position in memory in spite of frequent usage (Park et al., 2010).
Other brands, even when infrequently used, can leave an indelible, affectively charged mark on the consumer's consciousness; thus, the
consumer periodically relives the positive experience. Arnould and
Price (1993) document the nature of such emotional as well as memorable brand experiences stemming from an extraordinary consumption activity (i.e., river-rafting), but suggest that such experiences
may also be tied, under certain conditions, to more mundane consumption activities involving everyday brands.
The role of memorable brand experiences is further supported by
consumer research on autobiographical memories and narrative processing (Escalas, 2004; Sujan, Bettman, & Baumgartner, 1993). This
line of research suggests that the consumption of certain brands is associated with greater self-referencing and the construction of
brand-related stories or narratives. Such a self-referencing process,
in turn, produces more affect-laden as well as easily retrievable memories (Escalas, 2004). Over time, such memories can even become imbued with strong feelings of nostalgia (Brown et al., 2003; Holbrook,
1993; Holt, 2005; Moore & Wilkie, 2005; Muehling & Sprott, 2004).
Drawing on these ideas, we propose that the final antecedent of
CBI is a construct we term memorable brand experiences, defined as
the extent to which consumers have positive, affectively charged
memories of prior brand experiences. Such brands are more likely
to play a defining role in a person's sense of self due to increased
co-mingling of brand-related thoughts with self-related thoughts
(Davis, 1979; Moore & Wilkie, 2005). Additionally, such experiences
are often likely to result from narrative rather than discursive processing, as the former has been found to build stronger connections
between the consumer and the brand (Escalas, 2004).
H6. The more memorable brand experiences a consumer has, the
more the person will identify with that brand.
2.3. Product category involvement as a moderator
Knowing who they are (i.e., self-definition) and feeling good about
themselves are twin goals of utmost importance to most individuals.
Thus, the efforts people make to meet these objectives are likely to be
both considered and considerable. In particular, to the extent that
these goals are met, in part, through consumers' identification with
certain brands, these brands must belong, almost by definition, to
product categories that the consumer actually cares about (Malär et
al., 2011; Reimann & Aron, 2009). It is only in such categories that
individuals are likely to find brands that can actually meet their
self-definitional needs, that is, brands that are eligible and worthy
of identification.
Product category involvement (PI) is generally understood as the perceived relevance of a product category to an individual consumer based
on his or her inherent values, needs, and interests (Zaichkowsky, 1985).
We expect PI to moderate the relationship between the various antecedents and CBI for at least two reasons. First, as the definition suggests,
product categories are more enticing to people when they associate
them with important higher-order goals, such as value satisfaction
(Bloch & Richins, 1983). In turn, such categories are likely to be more
closely associated with individuals' self-concepts and viewed as selfdefining, making these product categories stronger candidates for CBI
(Reimann & Aron, 2009). Second, people are more motivated to systematically process information pertaining to categories with which they are
more involved (Chen & Chaiken, 1999). Thus, all else being equal, consumers' knowledge structures regarding high (vs. low) involvement
product categories are more likely to contain deeply processed and highly
elaborated beliefs regarding brands' abilities to meet self-definitional
needs.
To the extent that judgments of a brand's standing with regard to
each of the six posited antecedent variables, and an assessment of the
implications of these perceptions for identification, require cognitive
resources and effort, these influences are more likely to materialize
in product categories where the consumer is highly involved. That
is, high (vs. low) PI should enhance not only a brand's perceived ability to meet a consumer's self-definitional needs but also a person's
motivation to process relevant information in that regard.
H7. The higher a consumer's involvement in the product category in
which a brand belongs, the stronger the relationship between (a)
brand–self similarity and CBI, (b) brand distinctiveness and CBI, (c)
brand prestige and CBI, (d) brand social value and CBI, (e) brand
warmth and CBI, and (f) memorable brand experiences and CBI.
2.4. Consequences of CBI
2.4.1. Brand loyalty
The conceptual and practical significance of examining the antecedents of CBI is predicated on the extent to which CBI, in turn, produces valuable pro-brand outcomes. Drawing on prior research, we
relate CBI to two valuable consequences—brand loyalty and brand advocacy—that contribute to a company's financial performance and, ultimately, to its long term success (for a review, see Gupta & Zeithaml,
2006).
The marketing literature provides ample support for the notion
that identification is linked to a sustained, long-term preference for
the identified-with company's products (Bhattacharya & Sen, 2003).
Homburg, Wieseke, and Hoyer (2009), for instance, report a strong
influence of customer–company identification on customer loyalty.
Additionally, Lam et al. (2010) show that CBI inhibits consumers
from switching brands. Finally, Park et al. (2010) report a positive relationship between a construct related to CBI, namely, brand attachment, and actual purchase behavior (i.e., total purchases in a set
time frame). Thus, it can be argued that CBI is a predictor of loyal
brand behavior, which we define as a deeply held intent to rebuy or
re-patronize a preferred brand in the future (Oliver, 1999).
H8. The more a consumer identifies with a brand, the more loyal that
person will be to that brand.
2.4.2. Brand advocacy
One of the key consequences of identification is the promotion of
the identified-with organization or company (e.g., Ashforth & Mael,
1989; Bhattacharya & Sen, 2003). Such promotion or advocacy can
take place both socially and physically. Socially, advocacy includes
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
the recommendation of the company (and its offerings, or as an employer) to others or the defense of the company when it is attacked
by others. Physically, advocacy can involve buying and using company
merchandise that displays the company logo or name, collecting memorabilia, apparel, or even acquiring tattoos (Katz, 1994). In organizational research, Mael and Ashforth (1992), for instance, have found
that there is a strong positive relationship between identification of
alumni with their alma mater and both social (e.g., advising others to attend the school) and physical promotion (e.g., attending college functions). Accordingly, we suggest that CBI will produce brand advocacy,
at least in the social sense of the brand's promotion to social others.
This is consistent with the research of Park et al. (2010), who report a
strong influence of brand attachment on such promotion behavior. Additionally, Ahearne et al. (2005) show that customer–company identification affects customer extra-role behaviors, which was measured in
part through company recommendation. All in all, then, we expect CBI
to be positively related to brand advocacy.
H9. The more a consumer identifies with a brand, the more the person
will advocate that brand.
Next, we describe three studies, two pilot studies and one main
study that together provide tests of the nomological network for CBI
as presented above.
3. Pilot studies
3.1. Pilot study 1
Our goals in this study were to develop a better understanding of the
phenomenon of CBI, generate items to measure CBI, and conduct a reality check on whether we were generally on the right path with the variables we had identified as the drivers of CBI. One-on-one in-depth
interviews that lasted between 30 and 80 min were conducted with
eight (three females and five males) German consumers who ranged
in age from 26 to 57. The interviews were conducted in German by
the first author and a research assistant. A semi-structured approach
was used in these interviews. The interviews commenced with a very
broad question on the topic of identification, “Is there anything with
which you identify?” Subsequently, more focused questions were
asked, e.g., “With whom do you identify? Which companies? Which
brands?” The interview questions then built on the participants' initial
responses by exploring the salient attributes and characteristics of particular companies and brands named by the participant and the brands'
and companies' links to the participant's sense of self. Several of the
follow-up questions also probed the thoughts and feelings that were
prompted by the brands mentioned by the participants.
The data from these depth interviews furthered our phenomenological understanding of the concept of CBI and the specific role of
brands in identity construction. The comments of the participants regarding what identification meant to them also enabled us to generate 16 potential items for our measure of CBI. The participants'
comments and responses additionally provided us some initial validation for the antecedent variables postulated in our framework.
For example, many of the comments related to similarities between
what a brand personified to a participant and the participant's construal of his or her own personality. A second group of responses
delved into unique or distinguishing characteristics of particular
brands, including aspects such as brand quality and prestige, and
then tied these specific aspects to self-image and the projection of
identity. Another set of responses involved explicit references to
friends, significant others, and other consumers who used the same
brand. Finally, a sense of deep intimacy with a brand, emotional connections to a brand, and vivid brand consumption experiences were
also frequently mentioned by the participants.
411
3.2. Pilot study 2
Our goal in this study was to develop a parsimonious, valid measure
of CBI that captured the psychological state of identification and distinguished it from its antecedents and consequences. We started with the
16 items from pilot study 1 and added to it 16 additional items that
were adapted from prior literature on identification (e.g., Bergami &
Bagozzi, 2000; Mael & Ashforth, 1992; Smidts, Pruyn, & van Riel,
2001). The study was conducted online with German consumers and
using a very large mail order retailer as the target brand. We chose
this retailer as the target brand for two reasons. First, due to the brand's
long history and tradition, not only is virtually every respondent likely
to be aware of the target brand, but many are likely to have or to have
had a relationship with this brand. Relatedly, because of its market presence and full-range of products, this retailer has a very large customer
base, making it more likely that our respondents are quite knowledgeable about the target brand. Indeed, we found that of the 382 participants, as many as 148 participants (39%) were customers; the average
age of the respondents was 36.7 years, and 43% were females. As in
the main study, the participants were asked to rate their level of agreement or disagreement (using 7-point scales) with the statements
pertaining to the various items.
Analytical methods such as an exploratory factor analysis (EFA), reliability analyses (Cronbach's alpha and item to total correlation, ITTC),
and a confirmatory factor analysis (CFA) were employed. Statistical
criteria for item retention involved an average factor loading above
.70 and an average corrected ITTC above .50. The normal distribution assumption associated with the maximum likelihood (ML) method of estimation in the CFA did not hold for the responses to the scale. The ML
procedure, however, is known to be fairly robust against moderate violations of normality (e.g., Browne, 1984; McDonald & Ho, 2002). The
global fit criteria applied in the CFA were the Chi-square test (χ2), the
Comparative Fit Index (CFI), Root Mean Squared Error of Approximation (RMSEA), and the Tucker–Lewis Index (TLI). Additionally, we
used the following local fit measures: indicator reliability (IR), factor
loading (FL), t-value of factor loading (t-value), composite reliability
(CR), and average variance explained (AVE). The MPLUS program was
employed for the CFA (Muthén & Muthén, 2006). The results of a multitude of iterative analyses showed that 15 of the 32 items were the best
candidates for assessing CBI. Of these, ten emerged from the first pilot
study, while five were based on measures used in prior research.
Next, we describe our main study, the goals of which were
two-fold: to (a) develop our final measure of CBI based on the 15
items obtained in this pilot study and (b) use this measure to test
our predictions regarding the antecedents and consequences of CBI.
4. Main study
4.1. Method
4.1.1. Participant demographics and selected product categories
We tested the hypothesized model of CBI with data from a panel of
German household consumers (796 participants). A few of the participants either did not follow the instructions or provided incomplete responses and thus were dropped from the data set (final N = 781). The
participants had an average age of 40.6 years, and nearly half (46.9%)
were females. Table 2 provides a demographic profile of the survey participants broken down by product category.
The study involved four product categories, that is, athletic shoes,
mobile phones, soft drinks, and grocery stores. While we selected product categories that vary considerably in the needs that they serve, all are
widely consumed and are often consumed in public. Further, to ensure
that the data would have sufficient variance with regard to key variables
such as CBI and product category involvement, we avoided products
where people may have responded uniformly at the upper end of the
scales for those variables (e.g., automobiles). Note that the four product
412
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
Table 2
Main study: product categories and profile of respondents.
Product category involvement mean (s.d.)
Number of test persons
Age
Gender (female)
Profession (pooled sample)
Pooled sample
Cell phones
Athletic shoes
Soft drinks
Grocery stores
2.81 (1.51)
796
40.6
46.9%
High school student: .6%
Student: 11.1%
Worker: 5.8%
(State) employee: 55.5%
Self-employed: 14.7%
3.06 (1.57)
198
39.9
44.9%
2.92 (1.55)
200
38.7
43.7%
Homemaker: 4.6%
Retiree: 5.9%
Unemployed: 4.3%
Other: 2.8%
2.90 (1.61)
199
40.1
52.3%
2.36 (1.17)
199
43.8
45.2%
Product category involvement was measured on a 1–7 scale.
categories included one frequently purchased consumable (soft drinks),
two shopping goods (athletic shoes and mobile phones), and one
service-oriented category (grocery stores).
4.1.2. Data collection procedure
Data for this study were collected through an online survey of a German household panel maintained by a commercial market-research
firm. Panel members were contacted via email and provided the URL
of a password-protected website for participating anonymously in the
online survey. A lottery-based monetary incentive system was used to
motivate participation. Given the length of the instrument, each respondent was asked to take the survey with respect to only one product category, which the survey software randomly assigned to each
respondent. In effect, the final sample sizes did not vary much across
the target product categories, ranging from 198 to 212 in each case.
The participants were informed that they would be participating in a
short consumer behavior survey. The survey first assessed whether a respondent was a user in the assigned product category. If the answer was
in the negative, a different product category was randomly assigned.
Next, depending on the assigned product category, the respondent
was asked to name the brand that he or she (a) had last consumed
(soft drinks), (b) had last visited (grocery stores), or (c) currently
used most often (athletic shoes and mobile phones). After eliciting the
target brand in this manner, the survey then presented the respondent
with items designed to measure the variables in the theorized model.
The sequence of the items was intentionally designed to be the reverse
of the causal direction of the hypotheses to minimize possible demand
effects. Specifically, the instrument first measured brand loyalty and
brand advocacy. Next, the survey included items that assessed CBI (described in Section 4.1.4, below); then, the hypothesized antecedents
and moderating variables were presented. Note that to better evoke
the thoughts and feelings associated with the target brand, the survey
software automatically inserted the respondent-specific brand name
in each and every one of the brand-related scale items that were
presented to the respondents. The survey closed with a few demographic questions.
4.1.3. Overview of study measures
All items used in the survey involved 7-point agree–disagree
Likert scales. These items were initially developed in English by the
research team and then translated into German and finally translated
back into English to verify vocabulary, idiomatic, grammatical, and
syntactical equivalence (Steenkamp, ter Hofstede, & Wedel, 1999).
The questionnaire was pre-tested with a small sample of respondents
(N = 12), and based on their feedback, some minor changes were
made in the wording of the items. For developing the items used in
our measures, we drew on existing scales wherever possible and
adapted them as necessary to fit the context of our research. The details are discussed in subsequent sections and the full scale items are
presented along with reliability and validity statistics in Appendix A.
4.1.4. CBI measure
As previously discussed, prior literature and our two pilot studies
yielded a battery of 15 potential items to assess CBI. The results of a
multitude of iterative analyses (including alpha, ITTC, EFA, and CFA)
revealed that of these 15 items, five items best represented the state
of CBI (see Table 3). This scale is highly reliable with an alpha of .94
and ITTCs ranging from .82 to .89. Furthermore, the EFA explained
82% of variance, and second-generation fit indices were also satisfactory. The study, for instance, produced the following goodness-offit-measures: χ 2 (5) = 54.12, RMSEA = .10, SRMR = .03, CFI = .99,
and TLI = .98.
We tested the robustness of the scale in four additional ways. First,
we assessed the model fit separately for each of the four product categories and found it to be satisfactory in every case. Second, using a
bootstrapping procedure (Fitzgerald Bone, Sharma, & Shimp, 1989),
we serially drew 100 bootstrap samples (with replacement) of the
total sample with a 50% fraction of the base sample and then tested
the measurement model on the basis of the covariance matrices produced by these samples. The results showed that the scale was stable
with respect to all fit indices. Third, in an effort to further evaluate the
candidate scale, three alternative factor structures (with 15, 8, and 6
items, respectively) were estimated. These scales were compared to
the five-item scale using χ 2-tests in CFA. To ensure that no important
content was lost in developing a parsimonious scale, a measure
containing all 15 items was used as a null model. The five-item
scale strongly outperformed the null model (Δχ 2 (85) = 911.6,
p b .01) and the six-item scale (Δχ 2 (4) = 19.61, p b .01). Additionally,
the five-item scale slightly outperformed the six-item scale with respect to global fit (CFI = .990 vs. .989; TLI = .985 vs. .979; SRMR =
Table 3
Test statistics for the 5-item CBI scale.
Construct/item
Consumer–brand identification scale
I feel a strong sense of belonging to brand X.
I identify strongly with brand X.
Brand X embodies what I believe in.
Brand X is like a part of me.
Brand X has a great deal of personal meaning for me.
ITTC
Alpha
Factor loading EFA
.94
.89
.82
.86
.83
.85
.93
.88
.91
.90
.91
AVE EFA
CR CFA
.82
.88
Indicator reliability CFA
AVE CFA
.61
.86
.71
.78
.76
.77
AVE = average variance explained, CFA = confirmatory factor analysis, CR = composite reliability, EFA = exploratory factor analysis, ITTC = item-to-total-correlation.
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
Table 4
Test of model fit of competing CBI models.
Model
χ2
df
χ2-differencea
Null model (15 items, 1 factor)
Model one (8 items, 1 factor)
Model two (6 items, 1 factor)
Model three (5 items, 1 factor)
965.72
243.73
73.73
54.12
90
20
9
5
721.99 (Δdf = 70)⁎⁎⁎
170.00 (Δdf = 11)⁎⁎⁎
19.61 (Δdf = 4)⁎⁎⁎
a
The χ2-differences represent comparisons of model one versus the null model,
model two versus model one, and model three versus model two.
⁎⁎⁎ Significant at p b .01.
.029 vs. .031). Overall, the five-item scale was found to be the
best-fitting CBI measure (see Table 4).
Finally, to examine the robustness of the CBI scale, we conducted a
validation study in another country with respondents who varied
considerably in their demographics compared to the main study respondents. This study involved a survey of a sample of U.S. college
students from a large mid-western university (415 participants;
final N = 414) and included two of the four product categories in
the main study, namely, athletic shoes (N = 212) and soft drinks
(N = 203). The study was conducted using a computerized survey
and administered in a university laboratory setting. Students signed
up for participation to earn extra class credit. The average age of the
student participants was 20.6 years, and 53.9% of the participants
were females. Again, with this sample, the scale was found to be highly reliable, with an alpha of .93 and ITTCs ranging from .78 to .84. The
EFA explained 78.3% of variance, and the CFA global and local fit indices were also found to be satisfactory.
A test of discriminant validity of all study constructs including CBI
is described in Section 4.2.2. However, we also assessed discriminant
validity for the CBI measure with regard to another related construct,
namely, brand commitment. As evident in the definition of Allen and
Meyer (1996, p. 253), who view affective commitment as “identification with, involvement in, and emotional attachment to the organization,” identification and commitment are closely related constructs.
Notwithstanding, in our conceptualization, we do not view identification as part of commitment or, conversely, commitment as part of
identification. Instead, we assume that brand commitment is a psychological consequence of CBI. Evidence of discriminant validity for
our CBI measure vis a vis brand commitment thus helps to establish
the credentials of the former.
To operationalize brand commitment, we adapted items from the
commitment scales of Coulter, Price, and Feick (2003) and Beatty,
Kahle, and Homer (1988). The resulting 3-item scale showed acceptable reliability and fit measures in our data with an alpha value of .68
and 62.3% of variance extracted by the EFA. The Fornell and Larcker
(1981) test was used to analyze discriminant validity for CBI versus
commitment. Pairwise correlations between factors obtained from a
two-factor model were compared with the variance extracted estimates (AVE) for the two constructs. Discriminant validity exists
when both variance extracted estimates exceed the square of the correlation between the factors. The AVE for CBI was .78, and the AVE for
commitment was .43; in contrast, the squared correlation between
CBI and commitment was .42. Thus, even though there is a positive
relationship between the two constructs, there is also evidence of discriminant validity between them.
4.1.5. Other study measures
The measure for brand–self similarity was derived from separate
assessments of the brand's personality and the respondent's own personality using identical scales (Sirgy & Danes, 1982). These scales
drew on Aaker's (1997) work on brand personality, which shows
that brand personality can be conceptualized as having five different
dimensions, which, in turn, can be represented with a total of 15
413
“facets” (Aaker, 1997, pp. 351–352). Therefore, we assessed both
brand personality and the respondent's own personality with 15
items (7-point, agree–disagree scales), with one item for each of the
facets proposed by Aaker (1997). The brand–self similarity measure
was then constructed as follows. First, the Euclidean distance D between the two personality assessments was computed. Next, this distance measure was rescaled to a similarity measure S with a
theoretical range of 1 (minimum) to 7 (maximum) by applying an appropriate linear transformation (S = 7 − .2582D).
The three items used to measure brand distinctiveness were
taken from existing scales (e.g., Bhattacharya & Sen, 2003). Brand
prestige was also measured with three items drawn from prior literature (e.g., Bhattacharya & Sen, 2003; Mael & Ashforth, 1992). The
four items used for brand social benefits were based on the literature
on brand communities (e.g., McAlexander et al., 2002). Brand warmth
was operationalized with three items adapted from Moore, Ratneshwar,
and Moore (2012). Memorable brand experiences were measured with
three items adapted from Gladden and Funk (2001). Product category involvement was assessed with four items drawn from prior research (e.g.,
Mittal & Lee, 1988). Brand loyalty was measured with three items taken
from the literature (Chaudhuri & Holbrook, 2001; Coulter et al., 2003),
and finally, brand advocacy was assessed with three items (e.g., Brown,
Barry, Dacin, & Gunst, 2005).
4.2. Results
4.2.1. Preliminary analyses
All measures were found to be highly reliable and valid. The psychometric properties of the scales were assessed through Cronbach's
α, CR, ITTC, EFA, and CFA. Descriptive statistics and the correlations
between the various measures are shown in Table 5 with the data
pooled across the four product categories. Before pooling the data,
tests of measurement invariance were conducted using MPLUS, and
Chi-square difference tests in the CFA (Steenkamp & Baumgartner,
1998) confirmed that the measures were equivalent across the four
product categories.
4.2.2. Test for discriminant validity
We assessed discriminant validity using the Fornell and Larcker
(1981) test. Table 5 shows the correlations between all constructs.
As the squared correlations are less than the AVE for every construct,
there is evidence of discriminant validity.
4.2.3. Tests for common method bias
Because we measured all constructs using respondents' selfreports, it is important to rule out common method variance as a
source of bias in the results (Podsakoff, MacKenzie, Jeong-Yeon, &
Podsakoff, 2003). Accordingly, we conducted the following tests.
First, we included a common method factor (CMF) into the structural equation model (SEM). Similar to Homburg, Mueller, and
Klarmann (2011), who included a common method factor in the
SEM to test H1 only, the CMF in our SEM was constructed to load on
a selection of items used to measure the antecedents and CBI. Inclusion of such a factor helps to control for common method bias in
our tests of hypotheses. We specified the loadings of the CMF to be
of the same size to reflect the notion that common method variance
influences all items equally and to achieve model convergence (e.g.,
Homburg et al., 2011; Rindfleisch, Malter, Ganesan, & Moorman,
2008). This procedure of including a method factor did not affect
the findings regarding our hypotheses (see Section 4.2.4), which indicates that common method variance did not bias the results.
Second, we relied on the findings of simulation studies which have
shown that the risk of common method variance is strongly reduced
in models involving non-linear relationships (e.g., Evans, 1985;
Siemsen, Roth, & Oliveira, 2010). Multiple group analyses of our
data (see Section 4.2.5) confirmed that the relationships between
414
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
Table 5
Descriptive statistics and correlations between measures.
Variable
Mean
SD
CR
AVE
1
2
3
4
5
6
7
8
9
1. Consumer–brand identification
2. Brand–self similarity
3. Brand distinctiveness
4. Brand prestige
5. Brand social benefits
6. Brand warmth
7. Memorable brand experiences
8. Product category involvement
9. Brand loyalty
10. Brand advocacy
1.98
4.50
3.36
4.32
1.82
2.43
2.75
2.81
4.08
3.03
1.32
1.07
1.78
1.66
1.22
1.44
1.72
1.51
1.45
1.60
.88
N.A.
.91
.92
.93
.85
.93
.85
.82
.84
.61
N.A.
.78
.80
.78
.66
.81
.59
.61
.63
.46
.48
.41
.68
.69
.59
.47
.42
.68
.32
.35
.43
.55
.45
.30
.25
.31
.68
.43
.60
.53
.40
.27
.33
.31
.58
.52
.37
.23
.28
.69
.56
.53
.37
.46
.68
.57
.38
.47
.51
.32
.40
.20
.32
.60
The theoretical scale range is 1–7 for all variables. For more details of the measures, see Section 4.1 and Appendix A. The statistics reported above are based on the data pooled across
the product categories in the main study.
AVE = average variance extracted based on CFA, CR = composite reliability.
the antecedent variables and CBI differ in a predictable manner in different subgroups of the sample. In addition to the multiple group
analysis, we tested the nonlinearity in our model for selected variables using latent interaction terms (e.g., Homburg, Klarmann, &
Schmitt, 2010). The analyses, for instance, showed that a latent interaction model produces significant (p b .01) and stable results for the
interactions of product category involvement (PI) with brand–self
similarity and brand distinctiveness.
4.2.4. Tests of main effects
H1 to H6 and H8 to H9 were tested with structural equation modeling
using MPLUS (Muthén & Muthén, 2006). In addition to the hypothesized antecedents of brand–self similarity, brand distinctiveness,
brand prestige, brand social benefits, brand warmth, and memorable
brand experiences, we added three dummy variables as independent
variables to account for product category main effects on CBI. This set
of predictor variables explained over half of the variance in CBI (squared
multiple correlation, SMC= .59). Similarly, we achieved a good SMC for
brand advocacy (.47) and a satisfying SMC for brand loyalty (.16). The
fit statistics indicate a good fit for our model: χ2 (389) = 2.21,
RMSEA= .08, SRMR = .07, CFI= .91, TLI = .90.
The standardized parameter estimates for the main effects model
are shown in the second column of Table 6. The results supported
the predictions for brand–self similarity (H1, γ = .05, p b .10), brand
distinctiveness (H2, γ = .08, p b .05), brand social benefits (H4, γ =
.34, p b .01), brand warmth (H5, γ = .30, p b .01), and memorable
brand experiences (H6, γ = .15, p b .01). Regarding brand prestige
(H3), our study did not yield a statistically significant result. With respect to the downstream relationships between CBI and brand loyalty
(H8), and between CBI and brand advocacy (H9), the model showed
strong and significant results (β = .55 and .68, respectively, both
p b .01).
The main effects model was further validated with the U.S. sample
previously mentioned. The results were generally replicated in that
sample, except that the relationship between brand distinctiveness
and CBI was statistically non-significant, whereas the path from
brand prestige to CBI was marginally significant (p b .10).
4.2.5. Tests of moderated relationships
To test the moderated relationships implied by H7a to H7f for product category involvement (PI), we employed multiple group structural equation modeling. We first performed a median split for PI in each
of the four product categories to create two subgroups, one with low
Table 6
Results of the structural equation models: main model and moderated model.
Moderated model
Relationships
H1: brand–self similarity →CBI
H2: brand distinctiveness → CBI
H3: brand prestige → CBI
H4: brand social benefits→ CBI
H5: brand warmth →CBI
H6: memorable brand experiences → CBI
Dummy variable #1
Dummy variable #2
Dummy variable #3
H8: CBI →brand loyalty
H9: CBI →brand advocacy
Brand loyalty ↔ brand advocacy
Main model
(N = 796)
.05⁎
.08⁎⁎
.01
.34⁎⁎⁎
.30⁎⁎⁎
.15⁎⁎⁎
−.05⁎
−.05
−.05
.55⁎⁎⁎
.68⁎⁎⁎
.48⁎⁎⁎
χ2-difference
Product involvement
Low
(N = 391)
High
(N = 405)
.04
−.04
−.10
.19⁎⁎
.32⁎⁎⁎
.14⁎⁎
.07⁎
.11⁎
−.03
.35⁎⁎⁎
.37⁎⁎⁎
.15⁎⁎⁎
10.73⁎⁎⁎
17.53⁎⁎⁎
–
15.24⁎⁎⁎
18.69⁎⁎⁎
13.69⁎⁎⁎
−.06
−.07
−.03
–
–
–
−.01
−.10⁎⁎
.00
–
–
–
–
–
–
–
–
–
Standardized path coefficients from structural equation models are reported in columns 2, 3, and 4. The χ2-differences for comparisons of the unconstrained versus the constrained
model are shown in column 5.
The dummy variables represent main effects for the four product categories included in the study.
⁎⁎⁎ Significant at p b .01.
⁎⁎ Significant at p b .05.
⁎ Significant at p b .10.
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
PI and one with high PI. Next, we tested the model for CBI (excluding
loyalty and advocacy) simultaneously in both subgroups using
MPLUS. The model fit was satisfactory as χ 2 (498) = 1.43, RMSEA =
.07, SRMR = .05, CFI = .93, TLI = .92.
The standardized parameter estimates for the two subgroups are
shown in columns three and four of Table 6. The relationship between
brand prestige and CBI was statistically non-significant in both subgroups. The coefficients for the paths between brand–self similarity
and CBI and between brand distinctiveness and CBI were statistically
significant only in the high PI subgroup. The coefficients for the paths
between the other three antecedents and CBI were statistically significant in both PI groups. However, in accord with our prediction, in all
three cases, the coefficient values were larger in the high PI subgroup
when compared to the low PI subgroup.
As a follow up to the above pattern of results, a χ 2-difference test
was used to assess the statistical significance of the difference in the
path coefficients between the subgroups. Here, an unconstrained
model was compared with a constrained model wherein the two
path coefficients were set to be equal. The results confirmed that
the higher the PI, the stronger the relationship between brand–self
similarity and CBI (H7a: Δχ 2 (1) = 10.73, p b .01), brand distinctiveness and CBI (H7b: Δχ 2 (1) = 17.53, p b .01), brand social benefits
and CBI (H7d: Δχ 2 (1) = 15.24, p b .01), brand warmth and CBI (H7e:
Δχ 2 (1) = 18.69, p b .01), and memorable brand experiences and CBI
(H7f: Δχ 2 (1) = 13.69, p b .01). 3
4.2.6. Test of an alternative model
Given that our hypothesis tests entail causal modeling of survey
data, the question naturally arises as to whether an alternative
model may fit the data equally well. One way to examine this issue
is to test an alternative model where the roles of the antecedent variables and the focal dependent variable (CBI) are reversed. Our approach follows the path of Morgan and Hunt (1994) and
Hennig-Thurau, Gwinner, and Gremler (2002).
Specifically, we earlier offered theoretical arguments as to why
brand–self similarity, brand distinctiveness, brand prestige, brand social benefits, brand warmth, and memorable brand experiences
should influence CBI, which, in turn, should influence brand loyalty
and brand advocacy. It is far more difficult to argue that CBI could
play a similar causal role in driving the hypothesized antecedent variables. Accordingly, if our theory has merit, the causal model we
employed earlier for H1 to H6 and H8 to H9 should fare much better
empirically compared to an alternative model wherein CBI is treated
as an independent (i.e., exogenous) variable and brand–self similarity,
brand distinctiveness, brand prestige, brand social benefits, brand
warmth, and memorable brand experiences become mediating variables in the relationship between CBI and the two downstream variables of brand loyalty and brand advocacy.
Following this logic, we tested the aforementioned alternative
model with SEM. The model included the dummy variables for
product category and the nine variables mentioned above. Note that
the alternative model has ten more paths than the hypothesized
model (i.e., 18 vs. 8 hypothesized paths, thus more paths by a factor
of 2.25). We first compared the overall fit of the two models. The
results showed that the global fit of the alternative model was much
worse than that of the hypothesized model: CFIAlternative Model =.86 vs.
CFI = .91; SRMRAlternative Model = .11 vs. SRMR = .07; and χ 2
(413)Alternative Model = 3195 vs. χ 2 (389) = 2206. Next, as the
two models are not nested in one another, it is appropriate to rely on
3
As in the case of the main effects model, we attempted to validate the tests of the
moderated relationships with the U.S. data. However, on account of the much smaller
sample size and the complex nature of the multi-group model, we were unable to obtain any plausible results.
415
Akaike's Information Criterion (AIC; Akaike, 1987) for model comparison (Rust, Lee, & Valente, 1995). Smaller values of AIC indicate a better
model fit. Again, the alternative model fared worse with respect to this
fit statistic, AICAlternative Model = 48,645 vs. AIC= 47,694. Finally, it is
worth comparing the proportions of the two models' hypothesized parameters that turned out to be statistically significant. Again, the alternative model performed worse than our hypothesized model. As
noted earlier, as many as seven out of eight hypothesized paths (88%)
in the hypothesized model were found to be statistically significant. In
contrast, in the alternative model, only 12 out of 18 paths between
the variables (67%) were statistically significant. The paths that were
not supported by the data in the alternative model were those from
brand–self similarity, brand warmth, and memorable brand experiences to brand loyalty and the paths from brand–self similarity, brand
distinctiveness, and memorable brand experiences to brand advocacy.
Therefore, all of the comparisons confirm that the hypothesized
model is indeed superior to the alternative model.
5. General discussion
The idea that brands can play a crucial role in the construction and
maintenance of consumers' identities is not new (Keller, 1993; Levy,
1959). Nonetheless, it is only recently that the concept of consumer–
brand identification is finally receiving the conceptual and empirical attention it deserves (Escalas & Bettman, 2003, 2009; Fournier, 2009;
Lam et al., 2010). This research adds to the growing body of knowledge
on this topic by proposing and testing an integrated framework for the
drivers, moderators, and consequences of consumer–brand identification. Based on a synthesis of a variety of literatures, the framework
includes six antecedent factors for CBI, of which three are mainly cognitive in nature (brand–self similarity, brand distinctiveness, and brand
prestige), while the other three are more affect-based (brand social
benefits, brand warmth, and memorable brand experiences). We also
theorize that the drivers will display stronger relationships with CBI
when a consumer's involvement with a brand's product category is relatively high.
The results of the two pilot studies and a main survey study provided convincing empirical support for our framework. The six antecedents together accounted for over half of the variance in CBI in
the main study, emphasizing their collective efficacy as determinants
of our focal construct. Interestingly, very little support was obtained
for the predictive role of brand prestige. Aside from the most obvious
possibility that brand prestige does not, in general, influence CBI, it is
plausible, given that there was sufficient variance on the brand prestige measure, that CBI is less sensitive to brand prestige in categories
such as supermarkets, soft drinks, and even athletic shoes than in the
more conventionally status or luxury product categories. Clearly, further research is needed to better understand the precise role of brand
prestige. The results also provided reliable and consistent support for
the hypothesized moderating role of product category involvement.
Additional analyses revealed no evidence of a common method bias
or the plausibility of alternative nomological networks (e.g., with reversed causality). In providing these insights, this paper makes several
contributions, both theoretical and managerial, which are discussed
next.
5.1. Contributions to branding and consumer behavior theory
This paper advances our understanding of the relationship between brands and consumer identity in several key ways. First, this
research provides an integrative understanding of the antecedents
of CBI, bringing together drivers that have, thus far, been examined
only in isolation. In particular, this paper is the first to propose and
test an overarching framework for CBI that contains both cognitive
and experiential (i.e., affect-rich) drivers of identification. Second,
by providing evidence for the role of multiple drivers of CBI in a single
416
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
model, we demonstrate that each of these drivers has an influence
even when controlling for the effects of the others. Third, in contrast
to the prominence bestowed on the cognitive drivers of identification in
the company domain (e.g., Bhattacharya & Sen, 2003), our findings highlight a generally stronger role of the affective drivers (see, e.g., Table 6). It
may be the case that a consumer's idiosyncratic experiences with the object of identification matter more than reflective, mindful factors when
the object is concrete and palpable, such as a brand, than when it is
more abstract and amorphous, such as a company. Given that we did
not contrast, in our studies, consumer identification with a brand versus
a company, future research that does so would help to more definitively
establish whether and to what extent the drivers of the two differ.
Further, this research is the first, to the best of our knowledge, to
empirically examine the category-specific moderators of the relationship between CBI and a number of diverse drivers. By documenting
the moderating effect of product category involvement, the present
research provides a more nuanced, contingent theoretical picture of
the forces underlying CBI (see also Escalas & Bettman, 2009;
Reimann & Aron, 2009). Future research may be able to expand our
nomological network even further, providing a fuller explication of
the various contingencies that regulate the influence of the key
drivers of CBI. Such research may also clarify whether the same
drivers and moderating variables are at work in determining the intensity and durability of consumer–brand relationships (Fournier,
2009; Park et al., 2010). For instance, while we conceptualized
brand–self similarity as the perceived congruence between consumers'
sense of their current or actual self and that of the brand, it is possible
that under certain contingencies, such as low category involvement,
low self-esteem, or high self-consciousness (Malär et al., 2011), it is
the consumers' sense of who they ideally would be (i.e., the ideal self)
that may be more relevant in driving CBI.
5.2. Managerial implications
Our research corroborates the positive link between CBI and its
pro-brand consequences, such as loyalty and advocacy. Thus, it
would seem that marketers would want to maximize CBI. At the
same time, however, some recent research points to the potential
limits and drawbacks of encouraging CBI (Chernev et al., 2011),
suggesting that doing so may, in fact, make the consumer–brand relationship more rather than less susceptible, at least in the short-run, to
competing forces that satiate consumers' need for identityexpression. While marketers would be wise to take this eventuality
into account when planning their brand strategies, it is contingent
on future research that establishes the long-term, rather than the
short-term, effect of identity saturation on brand preferences.
That said, this research provides brand managers with some actionable insights into the why of CBI within their specific contexts.
Managers of all product categories, regardless of consumer involvement levels, must focus on and better understand the more idiosyncratic and affect-rich experiences their consumers have with their
brands to harness these in the service of greater CBI. Additionally,
managers who offer products and services to highly involved consumers
should pay special attention to consumers' perceptions of a brand's personality and distinctiveness. Our research also suggests that to enhance
CBI, managers must ensure that their brands have high social value and
thus serve consumers' interpersonal goals (see also Muniz & O'Guinn,
2001; O'Guinn & Muniz, 2009; Schouten & McAlexander, 1995). This
can occur not only through the fostering of interactions between the
brand and the consumer through a myriad of approaches, from event
marketing to product co-creation, but also through interactions among
consumers around a brand, through brand communities, both physical
and virtual. Additionally, marketers can enhance or at least highlight
the emotional appeal of their brand, albeit within the constraints imposed by the brand's functionality and overall positioning, with the objective of convincing consumers to think of the brand in a warm and
emotional way, rather than as a cold and distant entity (see also Aaker,
1997; Diamond et al., 2009; Fournier, 1998; Park et al., 2010). This can
be achieved not only through conventional communication strategies
but also through a coherent, well thought-out CSR (i.e., corporate social
responsibility) platform that, implemented correctly and genuinely,
may contribute greatly to a brand's perceived warmth.
Our findings also suggest that marketers can increase CBI by creating distinctive and memorable out-of-the-ordinary brand experiences
(e.g., cruise ship gala banquets, Napa Valley winery wine tastings, and
extraordinary personal touches at Kimpton Hotels). Clearly, this is
easier said than done; affectively-charged memories reside ultimately
in the consumer's mind and are, thus, not under complete control of
the marketer. At a minimum, then, marketers can attempt to provide
the types of unique and vivid experiences that engender indelible
memories, and they can then assess the extent to which these experiences and memories enhance CBI.
5.3. Limitations and future research directions
While this paper represents a significant advance in our understanding of the antecedents of CBI, it is not without limitations. The most important of these limitations is the fact that to test our framework, we
had to infer causal relationships from cross-sectional survey data.
Given the lengthy gestation period that is likely involved in a consumer
identifying with a brand, experimental and longitudinal study designs
seem unrealistic. Nonetheless, future research may be able to test at
least parts of our framework with alternative methodologies.
Additionally, our results reveal a great deal of consonance between the European respondents of our main survey and the U.S. respondents of the validating survey in terms of the key relationships in
our model. However, an important question remains: Might the nomological network for CBI be different, particularly in radically different cultures, for instance, in parts of Asia, Africa, and the Middle-East?
Future tests of the generalizability of the relationships to a broader
cultural context are needed. In particular, given the rising economic,
political, and cultural influences of the BRIC nations, it is important
to validate our model with populations from these countries.
Similarly, while our framework was tested with research participants
spanning a wide range of demographics, there is room for greater
generalizability through tests among demographic groups that
we did not examine. More specifically, some secondary analyses
suggest that the relationships between CBI and its antecedents are
directionally stronger for the men in our sample than for the
women. A deeper investigation of such gender differences may be of
both conceptual and practical value.
Further, our study categories were restricted largely to products
rather than to services, with supermarkets perhaps being the only exception. One could argue that this restriction results in a particularly
stringent test of our theory, as the posited relationships, particularly
between CBI and its experience-based determinants, are likely to be
stronger for services than for products. At the same time, we obtained
no meaningful differences across the four categories we examined in
terms of the predicted relationships. Regardless, it is essential that
our model be examined in a wider set of categories and settings, including services that have experiences as their core offerings. Such research and others of its type may be able to provide many more
valuable insights on the fascinating topic of why consumers identify
with some brands and not with others.
Appendix A. Measures for variables
Unless otherwise noted, the measures were assessed on 7-point
Likert scales where 1 = “completely disagree” and 7 = “completely
agree.” Cronbach's α values and explained variance from exploratory
factor analysis (EFA) are noted for each construct.
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
Consumer–brand identification (α = .94; EFA explained variance =
82.0%)
1.
2.
3.
4.
5.
I feel a strong sense of belonging to brand X.
I identify strongly with brand X.
Brand X embodies what I believe in.
Brand X is like a part of me.
Brand X has a great deal of personal meaning for me.
Brand–self similarity (α = .94; EFA explained variance = 72.3%; the
measure, however, is entered in aggregated form into the SEM, see
Section 4.1.5. for more details of this measure)
Brand X is …
I am …
1. Down-to-earth.
2. Honest.
3. Wholesome.
4. Cheerful.
5. Daring.
6. Spirited.
7. Imaginative.
8. Up-to-date.
9.
10.
11.
12.
13.
14.
15.
Reliable.
Intelligent.
Successful.
Upper class.
Charming.
Outdoorsy.
Tough.
Brand distinctiveness (α = .91; EFA explained variance = 84.6%)
1. Brand X has a distinctive identity.
2. Brand X is unique.
3. Brand X stands out from its competitors.
Brand prestige (α = .92; EFA explained variance = 86.6%)
1. Brand X is very prestigious.
2. Brand X is one of the best brands of [athletic shoes].
3. Brand X is a first-class, high-quality brand.
Brand social benefits (α = .93; EFA explained variance = 83.0%)
1.
2.
3.
4.
Brand X offers me the opportunity to socialize.
I feel a sense of kinship with other people who use brand X.
I gain a lot from interactions with other customers/users of brand X.
Being a customer of brand X makes me feel like I belong to a special group.
Brand warmth (α = .85; EFA explained variance = 77.4%)
1. Brand X creates warm feelings among its users.
2. Brand X is very loveable.
3. Brand X is emotional rather than rational.
Memorable brand experiences (α=.93; EFA explained variance=
87.3%)
1. I have had a lot of memorable experiences with brand X.
2. Thinking of brand X brings back good memories.
3. I have fond memories of brand X.
Product category involvement (α = .84, EFA explained variance =
86.2%)
1. I am very interested in anything related to [product category, e.g.,
athletic shoes].
2. Which brand of [athletic shoes] I buy matters a lot.
3. I value [athletic shoes] as an important part of my life.
4. [Athletic shoes] mean a lot to me.
Brand loyalty (α = .80; EFA explained variance = 72.0%)
1. I stick with brand X because I know it is the best for me.
2. I will buy brand X the next time I buy [athletic shoes].
3. I intend to keep purchasing brand X.
417
Brand advocacy (α = .83; EFA explained variance = 75.0%)
1. I like recommending brand X to other consumers.
2. I love to talk about the good points of brand X to people I know.
3. I have managed to convince other people to buy brand X.
Brand commitment (α = .68; EFA explained variance = 62.3%)
1. I am very fond of brand X.
2. I am very committed to brand X.
3. I consider myself to be very loyal to brand X.
References
Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research,
34(3), 347–356.
Ahearne, M., Bhattacharya, C. B., & Gruen, T. (2005). Antecedents and consequences of
customer–company identification: Expanding the role of relationship marketing.
Journal of Applied Psychology, 90(3), 574–585.
Akaike, H. (1987). Factor analysis and AIC. Psychometrika, 52, 317–332.
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance and normative commitment to the organization. Journal of Occupational
Psychology, 63, 1–18.
Allen, N. J., & Meyer, J. P. (1996). Affective, continuance and normative commitment to
the organization: An examination of construct validity. Journal of Vocational Behavior, 49, 252–276.
Arnould, E. J., & Price, L. L. (1993). River magic: Extraordinary experience and the extended service encounter. Journal of Consumer Research, 20(1), 24–45.
Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy
of Management Review, 14(1), 20–39.
Bagozzi, R. P., Bergami, M., Marzocchi, G., & Morandin, G. (2012). Customer–organization
relationships: Development and test of a theory of extended identities. Journal of Applied Psychology, 97(1), 63–76.
Bagozzi, R. P., & Dholakia, U. M. (2006). Antecedents and purchase consequences of
customer participation in small group communities. International Journal of Research in Marketing, 23(1), 45–61.
Batra, R., Ahuvia, A., & Bagozzi, R. P. (2012). Brand love. Journal of Marketing, 76(2), 1–16.
Beatty, S. E., Kahle, L. R., & Homer, P. (1988). The involvement–commitment model:
Theory and implications. Journal of Business Research, 16(2), 149–167.
Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research,
15(2), 139–168.
Bergami, M., & Bagozzi, R. P. (2000). Self-categorization, affective commitment and
group self-esteem as distinct aspects of social identity in the organization. British
Journal of Social Psychology, 39(4), 555–577.
Berger, J., & Heath, C. (2007). Where consumers diverge from others: Identity signaling
and product domains. Journal of Consumer Research, 34(2), 121–134.
Bhattacharya, C. B., Rao, H., & Glynn, M. A. (1995). Understanding the bond of identification: An investigation of its correlates among art museum members. Journal of
Marketing, 59(4), 46–57.
Bhattacharya, C. B., & Sen, S. (2003). Consumer–company identification: A framework
for understanding consumers' relationships with companies. Journal of Marketing,
67(2), 76–88.
Bloch, P. H., & Richins, M. L. (1983). A theoretical model for the study of product importance perceptions. Journal of Marketing, 47(3), 69–81.
Brewer, M. B. (1991). The social self: On being the same and different at the same time.
Personality and Social Psychology Bulletin, 17(5), 475–482.
Brown, T. J., Barry, T. E., Dacin, P. A., & Gunst, R. F. (2005). Spreading the word: Investigating
antecedents of consumers' positive word-of-mouth intentions and behaviors in a retailing context. Journal of the Academy of Marketing Science, 33(2), 123–138.
Brown, S., Kozinets, R. V., & Sherry, J. F., Jr. (2003). Teaching old brands new tricks: Retro
branding and the revival of brand meaning. Journal of Marketing, 67(3), 19–33.
Browne, M. W. (1984). Asymptotically distribution-free methods for the analysis of covariance structures. British Journal of Mathematical and Statistical Psychology, 37, 62–83.
Carroll, B. A., & Ahuvia, A. C. (2006). Some antecedents and outcomes of brand love.
Marketing Letters, 17(2), 79–89.
Chaudhuri, A., & Holbrook, M. B. (2001). The chain effects from brand trust and brand effect
to brand performance: The role of brand loyalty. Journal of Marketing, 65(2), 81–93.
Chen, S., & Chaiken, S. (1999). The heuristic-systematic model in its broader context. In
S. Chaiken, & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 73–96).
New York, NY: Guilford Press.
Chernev, A., Hamilton, R., & Gal, D. (2011). Competing for consumer identity: Limits to
self-expression and the perils of lifestyle branding. Journal of Marketing, 75(3), 66–82.
Coulter, R. A., Price, L. L., & Feick, L. (2003). Rethinking the origins of involvement and
brand commitment: Insights from postsocialist Central Europe. Journal of Consumer
Research, 30(2), 151–169.
Davis, F. (1979). Yearning for yesterday: A sociology of nostalgia. New York, NY: Free Press.
Dhar, R., & Wertenbroch, K. (2000). Consumer choice between hedonic and utilitarian
goods. Journal of Marketing Research, 37(1), 60–71.
Diamond, N., Sherry, J. F., Jr., Muniz, A. M., Jr., McGrath, M. A., Kozinets, R. V., & Borghini,
S. (2009). American girl and the brand Gestalt: Closing the loop on sociocultural
branding research. Journal of Marketing, 73(3), 118–134.
Dukerich, J., Golden, B., & Shortell, S. (2002). Beauty is in the eye of the beholder: the
impact of organizational identification, identity, and image on the cooperative behaviors of physicians. Administrative Science Quarterly, 47(3), 507–533.
418
N. Stokburger-Sauer et al. / Intern. J. of Research in Marketing 29 (2012) 406–418
Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organizational images and member identification. Administrative Science Quarterly, 39(2), 239–263.
Escalas, J. E. (2004). Narrative processing: Building consumer connections to brands.
Journal of Consumer Psychology, 14(1/2), 168–180.
Escalas, J. E., & Bettman, J. R. (2003). You are what they eat: The influence of reference
groups on consumers' connections to brands. Journal of Consumer Psychology, 13(3),
339–348.
Escalas, J. E., & Bettman, J. R. (2009). Self-brand connections: The role of reference groups
and celebrity endorsers in the creation of brand meaning. In D. J. MacInnis, C. W. Park,
& J. R. Priester (Eds.), Handbook of brand relationships (pp. 107–123). Armonk, NY:
M.E. Sharpe.
Evans, M. G. (1985). A Monte Carlo study of the effects of correlated method variance
in moderated multiple regression analysis. Organizational Behavior and Human Decision Processes, 36, 305–323.
Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype
content: Competence and warmth respectively follow from perceived status and
competition. Journal of Personality and Social Psychology, 82(6), 878–902.
Fitzgerald Bone, P., Sharma, S., & Shimp, T. A. (1989). A bootstrap procedure for evaluating goodness-of-fit indices of structural equation and confirmatory factor
models. Journal of Marketing Research, 26(1), 105–111.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable
variables and measurement error. Journal of Marketing Research, 18(1), 39–50.
Fournier, S. M. (1998). Consumers and their brands: Developing relationship theory in
consumer research. Journal of Consumer Research, 24(4), 343–373.
Fournier, S. M. (2009). Lessons learned about consumers' relationships with their
brands. In D. J. MacInnis, C. W. Park, & J. R. Priester (Eds.), Handbook of brand relationships (pp. 5–23). Armonk, NY: M.E. Sharpe.
Freud, S. (1922). Group psychology and the analysis of the ego. London: Hogarth.
Gladden, J. M., & Funk, D. C. (2001). Understanding brand loyalty in professional sport:
Examining the link between brand associations and brand loyalty. International
Journal of Sports Marketing & Sponsorship, 3(2), 67–91.
Grubb, E., & Grathwohl, H. (1967). Consumer self-concept, symbolism and market behavior: A theoretical approach. Journal of Marketing, 31(4), 22–27.
Gupta, S., & Zeithaml, V. (2006). Customer metrics and their impact on financial performance. Marketing Science, 25(6), 718–739.
Hennig-Thurau, T., Gwinner, K. P., & Gremler, D. D. (2002). Understanding relationship
marketing outcomes: An integration of relational benefits and relationship quality.
Journal of Service Research, 4(3), 230–247.
Hirschman, E. C., & Holbrook, M. (1982). Hedonic consumption: Emerging concepts,
methods and propositions. Journal of Marketing, 46(3), 92–101.
Holbrook, M. B. (1993). Nostalgia and consumption preferences: Some emerging patterns of consumer tastes. Journal of Consumer Research, 20(2), 245–256.
Holt, D. B. (2005). How societies desire brands: Using cultural theory to explain brand
symbolism. In S. Ratneshwar, & D. G. Mick (Eds.), Inside consumption (pp. 273–291).
London, New York: Routledge.
Homburg, C., Klarmann, M., & Schmitt (2010). Brand awareness in business markets:
when is it related to firm performance? International Journal of Research in Marketing, 27(3), 201–212.
Homburg, C., Mueller, M., & Klarmann, M. (2011). When should the customer really be
king? On the optimum level of salesperson customer orientation in sales encounters. Journal of Marketing, 75(2), 55–74.
Homburg, C., Wieseke, J., & Hoyer, W. D. (2009). Social identity and the service-profit
chain. Journal of Marketing, 73(2), 38–54.
Huffman, C., Ratneshwar, S., & Mick, D. G. (2000). Consumer goal structures and goal
determination processes: An integrative framework. In S. Ratneshwar, D. G. Mick,
& C. Huffman (Eds.), The why of consumption: Contemporary perspectives on consumer motives, goals, and desires (pp. 9–35). London: Routledge.
Katz, D. R. (1994). Just do it: The Nike spirit in the corporate world. New York: Random House.
Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand
equity. Journal of Marketing, 57(1), 1–22.
Keller, K. L. (2004). Strategic brand management: Building, measuring, and managing
brand equity. Upper Saddle River, NJ: Prentice-Hall.
Kleine, R. E., III, Kleine, S. S., & Kernan, J. B. (1993). Mundane consumption and the self:
A social identity perspective. Journal of Consumer Psychology, 2(3), 209–235.
Kunda, Z. (1999). Social cognition: Making sense of people. Cambridge, MA: MIT Press.
Lam, S. K., Ahearne, M., Hu, Y., & Schillewaert, N. (2010). Resistance to brand switching
when a radically new brand is introduced: A social identity theory perspective.
Journal of Marketing, 74(6), 128–146.
Laverie, D. A., Kleine, R. E., III, & Kleine, S. S. (2002). Reexamination and extension of Kleine,
Kleine, and Kernan's social identity model of mundane consumption: The mediating
role of the appraisal process. Journal of Consumer Research, 28(4), 659–669.
Levy, S. J. (1959). Symbols for sale. Harvard Business Review, 37(4), 117–124.
Mael, F., & Ashforth, B. E. (1992). Alumni and their alma mater: A partial test of the
reformulated model of organizational identification. Journal of Organizational Behavior, 13(2), 103–123.
Malär, L., Krohmer, H., Hoyer, W. D., & Nyffenegger, B. (2011). Emotional brand attachment and brand personality: The relative importance of the actual and ideal self.
Journal of Marketing, 75(4), 35–52.
Maynard, M. (2007). Toyota's hybrid car more than a mixed success. (accessed February 12, 2011), available at , http://www.nytimes.com/2007/07/04/business/
worldbusiness/04iht-hybrid.4.6491389.html
McAlexander, J. H., Schouten, J. W., & Koenig, H. F. (2002). Building brand community.
Journal of Marketing, 66(1), 38–54.
McDonald, R. P., & Ho, M. -H. R. (2002). Principles and practice in reporting structural
equation analysis. Psychological Methods, 7(1), 64–82.
Mittal, B., & Lee, M. -S. (1988). Separating brand-choice involvement from product involvement via consumer involvement profiles. Advances in Consumer Research, 15, 43–49.
Moore, M. L., Ratneshwar, S., & Moore, R. S. (2012). Understanding loyalty bonds and
their impact on relationship strength: A service firm perspective. Journal of Services
Marketing, 26(4), 253–264.
Moore, E. S., & Wilkie, W. L. (2005). We are who we were: Intergenerational influences
in consumer behavior. In S. Ratneshwar, & D. G. Mick (Eds.), Inside consumption
(pp. 208–232). London, New York: Routledge.
Morgan, R. M., & Hunt, S. D. (1994). The commitment–trust theory of relationship marketing. Journal of Marketing, 58(3), 20–38.
Muehling, D. D., & Sprott, D. E. (2004). The power of reflection. Journal of Advertising,
33(3), 25–35.
Muniz, A. M., Jr., & O'Guinn, T. C. (2001). Brand community. Journal of Consumer Research, 27(4), 412–432.
Muthén, L. K., & Muthén, B. O. (2006). MPLUS. Los Angeles: Muthén and Muthén.
O'Guinn, T. C., & Muniz, A. M., Jr. (2009). Collective brand relationships. In D. J.
MacInnis, C. W. Park, & J. R. Priester (Eds.), Handbook of brand relationships
(pp. 173–194). Armonk, NY: M.E. Sharpe.
Okada, E. M. (2005). Justification effects of consumer choice of hedonic and utilitarian
goods. Journal of Marketing Research, 42(1), 43–53.
Oliver, R. L. (1999). Whence consumer loyalty? Journal of Marketing, 63(4), 33–44.
Park, C. W., MacInnis, D. J., Priester, J., Eisingerich, A. B., & Iacobucci, D. (2010). Brand
attachment and brand attitude strength: Conceptual and empirical differentiation
of two critical brand equity drivers. Journal of Marketing, 74(6), 1–17.
Podsakoff, P. M., MacKenzie, S. B., Jeong-Yeon, L., & Podsakoff, N. P. (2003). Common
method biases in behavioral research: A critical review of the literature and
recommended remedies. Journal of Applied Psychology, 88(5), 879–903.
Rathnayake, C. V. (2008). Brand personality and its impact on brand feelings: A study
among young television viewers. South Asian Journal of Management, 15(2), 7–24.
Reimann, M., & Aron, A. (2009). Self-expansion motivation and inclusion of bands in self: Toward a theory of brand relationships. In D. J. MacInnis, C. W. Park, & J. R. Priester (Eds.),
Handbook of brand relationships (pp. 65–81). Armonk, NY: M.E. Sharpe.
Rindfleisch, A., Burroughs, J. E., & Wong, N. (2009). The safety of objects: Materialism, existential insecurity, and brand connection. Journal of Consumer Research, 36(1), 1–16.
Rindfleisch, A., Malter, A. J., Ganesan, S., & Moorman, C. (2008). Cross-sectional versus
longitudinal survey research: Concepts, findings, and guidelines. Journal of Marketing Research, 45(3), 261–279.
Rust, R. T., Lee, C., & Valente, E., Jr. (1995). Comparing covariance structure models: A general
methodology. International Journal of Research in Marketing, 12(4), 279–291.
Schouten, J. W., & McAlexander, J. H. (1995). Subcultures of consumption: An ethnography of the new bikers. Journal of Consumer Research, 22(1), 43–61.
Siemsen, E., Roth, A., & Oliveira, P. (2010). Common method bias in regression models
with linear, quadratic, and interaction effects. Organizational Research Methods,
13(3), 456–476.
Sirgy, M. J. (1982). Self-concept in consumer behavior: A critical review. Journal of Consumer Research, 9(3), 287–300.
Sirgy, M. J., & Danes, J. E. (1982). Self-image/product-image congruence models. Advances in Consumer Research, 9, 556–561.
Smidts, A., Pruyn, A. Th. H., & van Riel, C. B. M. (2001). The impact of employee communication and perceived external prestige on organizational identification. Academy
of Management Journal, 49(5), 1051–1062.
Snyder, C. R., & Fromkin, H. L. (1977). Abnormality as a positive characteristic: The development and validation of a scale measuring need for uniqueness. Journal of Abnormal Psychology, 86(5), 518–527.
Solomon, M. R. (1983). The role of products as social stimuli: A symbolic interactionism
perspective. Journal of Consumer Research, 10(3), 319–329.
Steenkamp, J. -B. E. M., & Baumgartner, H. (1998). Assessing measurement invariance
in cross-national consumer research. Journal of Consumer Research, 25(1), 78–90.
Steenkamp, J. -B. E. M., ter Hofstede, F., & Wedel, M. (1999). A cross-national investigation into the individual and national cultural antecedents of consumer innovativeness. Journal of Marketing, 63(2), 55–69.
Stokburger-Sauer, N. E. (2010). Brand community-drivers and outcomes. Psychology
and Marketing, 27(4), 347–368.
Strizhakova, Y., Coulter, R. A., & Price, L. L. (2008). The meanings of branded products: A
cross-national scale development and meaning assessment. International Journal of
Research in Marketing, 25(2), 82–93.
Stuart, H. (2002). Employee identification with the corporate identity. International
Studies of Management and Organization, 32(3), 28–44.
Sujan, M., Bettman, J. R., & Baumgartner, H. (1993). Influencing consumer judgments
using autobiographical memories: A self-referencing perspective. Journal of Marketing Research, 30(4), 422–435.
Tajfel, H., & Turner, J. C. (1985). The social identity theory of intergroup behavior. In S.
Worchel, & W. G. Austin (Eds.), Psychology of intergroup relations (pp. 7–24). (2nd
ed.). Chicago, IL: Nelson Hall.
Tepper Tian, K., Bearden, W. O., & Hunter, G. L. (2001). Consumers' need for uniqueness:
Scale development and validation. Journal of Consumer Research, 28(1), 50–66.
Thompson, C. J., Rindfleisch, A., & Arsel, Z. (2006). Emotional branding and the strategic
value of the Doppelgänger brand image. Journal of Marketing, 70(1), 50–64.
Thomson, M., MacInnis, D. J., & Whan Park, C. (2005). The ties that bind: Measuring the
strength of consumers' emotional attachment to brands. Journal of Consumer Psychology, 15(1), 77–91.
Tsai, S. (2005). Utility, cultural symbolism and emotion: A comprehensive model of brand
purchase value. International Journal of Research in Marketing, 22(3), 277–291.
Zaichkowsky, J. L. (1985). Measuring the involvement construct. Journal of Consumer
Research, 12(3), 341–352.