Trudel Argo Meng Recycled Self 2016

Full text

The Recycled Self: Consumers’ Disposal
Decisions of Identity-Linked Products
REMI TRUDEL
JENNIFER J. ARGO
MATTHEW D. MENG
It has been known for some time that consumers’ identities influence purchasing
decisions and people form strong identity connections, or “links,” with products
and brands. However, research has yet to determine whether identity-linked prod-
ucts are differentially treated at disposal in comparison to products that are not
identity linked. Across seven studies, the current research shows that when an
everyday product (e.g., paper, cups, aluminum cans) is linked to a consumer’s
identity, it is less likely to be trashed and more likely to be recycled. Further, the
tendency to recycle an identity-linked product increases with the strength and
positivity of the connection between the consumer and product (or brand). Finally,
the disposal behavior can be explained by consumers’ motivation to avoid trashing
a product that is linked to the self because it is viewed as an identity threat. In
sum, consumers will be more likely to recycle (rather than trash) a product if the
product is linked to a consumer’s identity. This occurs because placing an identity-
linked product in the trash is symbolically similar to trashing a part of the self, a
situation consumers are motivated to avoid.
Keywords: recycling, sustainability, identity, disposal, decision making
There is great public concern about our environment
(Gallup 2015;Hoornweg and Bhada-Tat 2012) and, in
particular, how consumption affects large-scale environ-
mental issues such as greenhouse gas emissions and its
subsequent impact on climate change (Bostrom et al. 1994;
Reynolds et al. 2010). Fortunately, research has shown that
individual consumers can positively drive large-scale
environmental changes by engaging in socially responsible
behaviors such as recycling (Environmental Protection
Agency [EPA] 2012). Success of any recycling program
largely depends on consumer participation, and therefore
understanding what drives consumers’ disposal decisions is
of critical importance. Prior research has found that con-
sumers’ decisions to recycle versus trash an object is sub-
ject to many influences including cost (Jenkins et al.
2003), market incentives (Reschovsky and Stone 1994),
convenience (Halvorsen 2008;Schultz and Oskamp 1996),
infrastructure (Ebreo and Vining 2001), environmental
awareness (Vining and Ebreo 1992), attitudes (Taylor and
Todd 1995), individual and cultural differences (McCarty
and Shrum 2001), social norms (Ewing 2001;Goldstein,
Cialdini, and Griskevicius 2008), and marketing promo-
tions (Kidwell, Farmer, and Hardesty 2013;White,
MacDonnell, and Dahl 2011). More recently, Trudel and
Argo (2013; also Trudel, Argo, and Meng 2015) found that
a product’s tangible characteristics (i.e., size and form)
also have implications for consumers’ disposal behaviors.
In the present research we seek to build on Trudel and
Argo (2013) by exploring how another product-related
Remi Trudel is assistant professor of marketing at the Questrom
School of Business, Boston University, 595 Commonwealth Avenue,
Boston, MA 02214. Phone: 617-358-3316. Email: rtrudel@bu.edu.
Jennifer J. Argo is Carthy Professor of Marketing at the School of
Business, University of Alberta, Edmonton, Alberta T6G 2R6. Phone:
780-492-3900. Email: jennifer.argo@ualberta.ca. Matthew D. Meng is a
PhD candidate at the Questrom School of Business, Boston University,
595 Commonwealth Avenue, Boston, MA 02214. Phone: 617-353-4149.
Email: mdmeng@bu.edu. Address all correspondence to Remi Trudel.
This work was partially funded by an Association for Consumer Research
Transformative Consumer Research Grant as well as a grant from the
Social Sciences and Humanities Research Council.
Gita Johar served as editor, and Zeynep Gu¨rhan-Canli served as associ-
ate editor for this article.
Advance Access publication March 25, 2016
V
CThe Author 2016. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc.
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com Vol. 43 2016
DOI: 10.1093/jcr/ucw014
246

characteristic, whether the product is linked to the focal
consumer’s identity, predicts disposal decisions to recycle
or trash the product.
It is clear from the literature investigating identity and
consumption that consumers have a preference for purchas-
ing products that fit or reflect their desired self (Belk 1988;
Chan, Berger, and Van Boven 2012;Escalas and Bettman
2005;Ferraro, Escalas, and Bettman 2011;Forehand and
Deshpande´ 2001;Kettle and H€
aubl 2011). Through owner-
ship and use of these products, consumers signal their iden-
tity to themselves and others. But what happens when
these products are no longer needed or wanted? Do con-
sumers dispose of identity-linked products differently than
products not linked to their identities? These unanswered
questions are the impetus behind this research project. One
possibility is that consumers dispose of products in the
same way regardless of whether or not there is an identity
link. Indeed, even prior to disposal, consumers have
decided that the product is no longer useful because they
no longer want or need it for functional or identity pur-
poses. Another possibility, and the one we demonstrate in
the current research, is that the presence (vs. absence) of an
identity link between the self and a product will influence
consumers’ disposal decisions even when they have
decided that they no longer want the product.
Across seven studies we provide robust evidence for the
effect of identity on consumers’ disposal decisions using a
variety of everyday products (i.e., paper, plastic cups,
paper cups, and aluminum cans) and identities (i.e., self-
identity at the individual and brand level, as well as social
identity at the university and national level). In doing so
we make notable contributions to theory and practice.
Foremost, we contribute to a growing consumer disposal
literature by demonstrating that a novel product-related
characteristic (i.e., the presence, strength, and valence of
an identity link) can predictably influence disposal behav-
ior. In particular, this research extends the work of Trudel
and Argo (2013) that examines the tangible properties of a
product, by demonstrating that intangible product charac-
teristics can also differentially influence consumers’ deci-
sions to trash versus recycle products. In this research, we
provide evidence that everyday products, such as paper and
plastic cups, can be (and often are) intangibly linked to
consumers’ identities and how people dispose of these
types of products can have a profound impact on our envir-
onment. The US EPA (2012) has identified that nearly
40% (paper and paperboard account for approximately
27%; plastics account for approximately 13%) of what
ends up in the landfill are everyday products that could
have been recycled. Understanding why individuals treat
functionally similar items differently when disposing of
them can allow us to increase overall recycle yield and de-
crease waste through design changes and marketing
promotions.
Our research also contributes to the identity literature by
investigating how consumers dispose of products that
while still linked to their identity are no longer wanted or
needed. We show that consumers’ identity links with prod-
ucts persist even when the products are no longer useful to
themselves or others, and that this connection to products
profoundly influences disposal decisions. In addition, we
provide evidence that a consumer’s decision to recycle an
identity-linked product arises because the act of throwing
such a product into the trash poses an identity threat. Since
consumers are motivated to avoid or minimize identity
threats (Ethier and Deaux 1994;Steele 1988), we demon-
strate that when a relevant (and positively valenced) iden-
tity is linked to a product, they dispose of the product in
the least threatening way (i.e., recycle). In the next section,
we review relevant disposal and identity literature to pro-
vide theoretical support for our predictions.
THEORETICAL BACKGROUND
Consumer Recycling Behavior
While the majority of consumer research has focused on
the acquisition phase of consumption, there is also an
emerging body of research investigating consumer disposal
behaviors and, in particular, consumers’ decision to trash
or recycle (Catlin and Wang 2012;Kidwell et al. 2013;
Trudel and Argo 2013;Trudel et al. 2015). Prior research
has found that decisions to recycle (as compared to trash) a
product are subject to a variety of different influences. For
instance, Schultz and Oskamp (1996) found that con-
sumers’ concerns for the environment predicted how much
effort they would be willing to exert to recycle. The au-
thors also found that offering monetary incentives in-
creases recycling intentions for those consumers with a
low concern for the environment but has little impact on
recycling intentions for those with a high concern for the
environment. As another example, Kidwell et al. (2013)
found that promotion-related strategies that match persua-
sive appeals to political ideologies positively influence
recycling behaviors. Specifically, conservatives’ recycling
increases when exposed to persuasive messages that are
consistent with their “binding moral” foundations, whereas
liberals’ recycling increases in response to persuasive mes-
sages consistent with their “fairness and caring” founda-
tions (Kidwell et al. 2013). As a final example, Fullerton
and Kinnaman (1996) found that municipalities are able to
reduce the number of trash bags collected and increase
recycling rates when they charge consumers for each bag
of trash collected. In sum, prior research shows that con-
sumers are susceptible to external influences when making
decisions between recycling and trashing, however little re-
search has investigated the impact of factors related to the
product itself.
TRUDEL, ARGO, AND MENG 247

One product-related factor that has been shown to play
an influential role in consumers’ recycling decisions is
whether or not the product has been distorted (Trudel and
Argo 2013;Trudel et al. 2015). In particular, when a prod-
uct departs from its prototypical form (e.g., a full size sheet
of paper is cut into smaller pieces or an aluminum can is
dented), there is an increased tendency for the product to
be trashed versus recycled. This change in disposal behav-
ior arises due to the belief that as products deviate from
their prototypical form they become less useful, and the lay
belief that useless items belong in the trash. In the current
work we explore the impact of a different type of product-
related characteristic on a consumer’s disposal behavior:
whether the product possesses an identity link.
Identity-Linked Products
Over the past two decades an influx of research has
focused on the relationship between identity and attitudes
and behaviors (Forehand and Deshpande´, 2001;Mannetti,
Pierro, and Livi 2004;Pelham, Mirenberg, and Jones 2002;
Reed 2004;Winterich, Mittal, and Ross 2009). One con-
clusion that has been consistently demonstrated is that situ-
ational demands can make salient a specific aspect of
identity that will in turn influence an individual to respond
in an identity-congruent way (Brewer 1991;Deaux 1991;
Deshpande´ and Stayman 1994;Kettle and H€
aubl 2011).
Further, the more salient an identity is, the more influence
that identity has on identity-congruent behavior (Laverie
and Arnett 2000). Indeed, linking consumer identities to
brands and products has been shown to be effective in ap-
pealing to consumers (Bolton and Reed 2004), and that a
consumer’s desire to signal a specific identity can influ-
ence brand choice (Argo, Dahl, and Manchanda 2005;
Chan et al. 2012). Although the literature clearly shows
that identity can influence brand and product choice and
that consumers can form strong identity links with prod-
ucts, research has yet to provide insight into how con-
sumers dispose of identity-linked products.
Some insight into the role that identity links may play in
disposal decisions can be garnered from research that
focuses on how consumers sell or give away meaningful
(i.e., special) possessions to others. Meaningful posses-
sions, such as a childhood teddy bear or a grandfather’s
watch, often receive special treatment when it comes time
to their divestiture because of their strong link to one’s
identity (Belk 1988;Belk, Sherry, and Wallendorf 1988;
Belk, Wallendorf, and Sherry 1989;Burris and Rempel
2004;Ferraro et al. 2011). For instance, sellers of such
products often seek owners with whom they share an iden-
tity so that the meaning of the possession can continue to
transfer with the product (Price, Arnould, and Curasi
2000). Sellers also prefer to sell special possessions to re-
cipients who know and appreciate the product’s meaning
and may even decide not to sell the product in situations
where they deem a buyer’s usage intentions as inappropri-
ate for the possession (Brough and Isaac 2012). In sum,
special possessions are generally disposed of in purposeful
ways so as to preserve their meaningfulness and the iden-
tity with which they are imbued (Belk 1988).
Extending this notion into the current work, we suggest
that linking everyday products to consumers’ identities
may similarly enhance the product’s meaningfulness and
subsequently impact consumers’ disposal behaviors.
However, unlike the meaningful possessions studied in pre-
vious research, everyday products such as paper or plastic
cups are rarely kept, sold, or given away to others after
consumption; thus disposal options for such products typic-
ally involve a decision between either throwing the product
in the trash or in the recycling bin. Given that past research
has found that products are trashed when they are deemed
worthless (i.e., useless; Trudel and Argo 2013;Trudel et al.
2015), we argue that a decision to throw an identity-linked
product in the trash symbolically represents an identity
threat, as in essence you would be throwing a piece of your
“self” in the trash and by extension signaling to yourself
that you must be worthless. As such, when an identity link
is made salient and because consumers are motivated to
maintain a positive self-concept (Steele 1988), and by ex-
tension their identity is tied to the product, we expect that
they will opt for the least threatening means to dispose of
an identity-linked product (i.e., recycle as compared to
trash). However, not all product identity links are of equal
strength and valance, and as such we expect these two
identity elements will moderate our effects.
Strength of Connection
The decision of how to dispose of identity-linked prod-
ucts is expected to be moderated by the strength of connec-
tion between a consumer’s identity and a product.
Consistent with prior work (White and Argo 2009), we ex-
pect that the strength of the connection between a con-
sumer’s identity and a product (i.e., strength of identity
link) will lead to different representations of identity threat
when disposing of products. This is consistent with identity
strength research demonstrating that consumers are more
likely to choose a brand when the brand name starts with
letters from their names when they need to self-enhance
(Brendl et al. 2005). Additional research in the consump-
tion domain has found that consumers’ strength of connec-
tion to brands can substantially influence brand attitudes
(Aaker, Fournier, and Brasel 2004;Escalas and Bettman
2005;Fournier 1998), purchase intentions (Paharia et al.
2011;Park et al. 2010), and brand loyalty (Sprott, Czellar,
and Spangenberg 2009). For example, Paharia et al. (2011)
found that consumers are more likely to purchase underdog
brands when they strongly (vs. weakly) identify with them.
As a final example, White and Argo (2009) used collective
self-esteem to demonstrate that consumers who strongly
248 JOURNAL OF CONSUMER RESEARCH

identify with a group make product choices that strengthen
their connection to that social identity, even under identity
threat. In sum, utilizing this prior research investigating
identity to extend predictions in the context of consumers’
disposal decisions, we expect that consumers who have a
strong product identity link, that is, the product is linked to
a strongly held identity, will be less likely to trash (more
likely to recycle) the product than consumers who have a
weak product identity link or no connection with the
product.
Valence
Our focus in this article is primarily on positive product
identity links, where a product is linked to a positive com-
ponent of a consumer’s identity for which there are posi-
tive associations. In these situations, we argue that a
decision to throw the identity-linked product in the trash
will represent a threat to one’s identity, and as such we ex-
pect that consumers will opt to recycle (vs. trash) because
recycling products that are linked to positive components
of a consumer’s identity allows them to maintain a more
positive self-view than if they trash the product. In con-
trast, when consumers possess a product that is linked to a
negative identity, we predict that they will be more likely
to trash the identity-linked product because they will want
to separate the self from this undesirable identity. This pre-
diction is supported by research that finds that people avoid
situations that reinforce negative associations of them-
selves or of the groups to which they belong (Steele and
Aronson 1995). For instance, White and Argo (2009) found
that when women were given a gender threat (told that
women generally had lower grade point averages, were
more likely to drop out of university, etc.), they were more
likely to evaluate gender films more negatively. Similarly,
work on stereotype threat shows that people prefer to avoid
identity-consistent behaviors that confirm negative stereo-
type associations about one’s group (Lee, Kim, and Vohs
2011;Spencer, Steele, and Quinn 1999;Steele and
Aronson 1995). In these situations the research suggests
that consumers prefer to disassociate from negative identi-
ties (Steele and Aronson 1995,White and Argo 2009).
Accordingly, we expect that trashing a product with a
negative identity link may serve as a symbolic act of dis-
sociation from that identity, and therefore we expect that
consumers will be more likely to dispose of a negative
identity-linked product in the trash (vs. the recycling).
To summarize, prior research has consistently demon-
strated the effects of identity on the consumption of prod-
ucts. Drawing on this research we hypothesize that when
deciding how to dispose of an everyday product that is no
longer wanted or needed, consumers will be more likely to
recycle (rather than trash) it if the product is linked to a
consumer’s identity. This will occur because placing an
identity-linked product in the trash would be symbolically
similar to trashing a part of the self, a situation consumers
will be motivated to avoid. We also hypothesize that the
tendency to recycle identity-linked products will be great-
est among consumers for which the product is linked to a
strongly (vs. weakly) held identity. Lastly, we predict that
our effects are contingent on the valence of the identity
link such that consumers will be more likely to trash nega-
tive identity-linked products than positive identity-linked
or neutral products. Next, we test our predictions across
seven studies. We conclude with a general discussion of
our findings and implications for theory and practice.
STUDY 1
The goal of this study is to establish initial support for
our prediction that products are more likely to be recycled
as compared to trashed if they are identity linked. We do
so using actual behavior for an everyday product (paper)
containing a self-identity-link: their name. Prior research
has demonstrated that individuals strongly associate their
names with their self-identities (Jones et al. 2002). In line
with our theorizing, we expect that consumers will be more
likely to recycle a piece of paper with their name on it, as
compared to a piece of paper with someone else’s name on
it, because one’s name forms a self-identity and product
link (Kettle and H€
aubl 2011).
Method
Participants and Design. A total of 206 (105 females,
M
age
¼19.77) at a large northeastern US university partici-
pated in a single factorial between-subjects design, result-
ing in two conditions (self-identity: linked vs. not linked).
Procedure. Participants were seated at a computer sta-
tion in which a quarter sheet of paper (dimensions: 4.25 
5.50 inches) and a pen were provided. To provide stringent
tests of our effects, in all of our studies we elected to use
products that past research has shown are most likely to be
trashed (e.g., small pieces of paper, small cups, dented
cups; Trudel and Argo 2013). In other words, given that
the disposal default for a small piece of paper is to be
trashed, any increases in recycling when the product is
identity linked provides strong support for our prediction.
Under the guise of testing a brand of pen, participants
were asked to evaluate the pen’s “writing ability and
smoothness.” Participants were randomly assigned to either
a self-identity–linked condition or self-identity–not linked
condition. Those in the identity-linked condition wrote
their first name only (rather than their full name, to in-
crease anonymity) five times on the piece of paper
provided while those in the identity–not linked condition
wrote the name “Avery” five times. The name Avery was
chosen because it is gender neutral, and no participants
who had registered for the sessions had that name. Once
TRUDEL, ARGO, AND MENG 249

participants finished writing on the paper, to be consistent
with our cover story they evaluated the pen via an online
questionnaire (not used in the analysis). Participants then
completed various demographic questions. We assessed
the effects of age and gender in all of the studies, but the
results did not produce any significant effects. They are
therefore not discussed further.
After completing the study (which was the last in a ser-
ies of unrelated studies), participants were asked to “dis-
pose of your paper on the way out.” The wording of this
instruction was chosen to avoid saying “trash” or “re-
cycle,” which could influence participants’ behavior. A
trash can and recycle bin were placed outside the closed
door of the laboratory. Both bins were identical with white
flip-top lids to prevent participants from seeing what was
inside. The only difference between the two bins is that
one was labeled “Recycle” and had the universal recycling
symbol and the other was labeled “Trash.” Participants
exited the laboratory one at a time, to control for any social
influence on their disposal decision. The dependent vari-
able was whether participants placed the paper in the trash
can or recycle bin.
Results and Discussion
A logistic regression tested whether there were different
recycling rates across the self-identity–link conditions.
The dependent variable was coded as Recycle ¼1 and
Trash ¼0 and the independent variable was coded as
Identity not linked ¼1 and Identity linked ¼1. The ana-
lysis revealed a significant main effect of self-identity-link,
where those who wrote their own first name were signifi-
cantly more likely to recycle the paper (36%) compared to
those who wrote the name “Avery” (23%, b¼0.320, v
2
(1)
¼4.248, p<.05, odds ratio [OR] ¼1.90). Study 1 pro-
vides initial support for our prediction that identity-linked
products are more likely to be recycled than products that
are not linked to consumers’ identities. In the next study,
we test the effect of identity on disposal behavior with a
more practically relevant manipulation that allows us to
demonstrate the ecological validity of our results.
STUDY 2
Study 2 was motivated by a recent trip to a coffee shop
where the barista misspelled the first author’s name on his
cup. Linking patron names with orders by writing names
on coffee cups has become standard procedure at many
coffee shops across the United States and provides a good
opportunity to investigate the practical implications of
doing so on disposal behavior. While prior research has
shown that people strongly identify with their names
(Stru¨mpfer 1978), study 2 tests the role of identity more
directly by showing that people are only more likely to re-
cycle a product that is “correctly” linked to their identity.
When a product–identity link is similar but not exact, con-
sumers should dispose of the product in the same way that
they would dispose of a neutral (nonlinked) product. In
contrast, writing a consumer’s name correctly on a cup
links the product to the consumer’s identity and should
make it more likely to be recycled.
Method
Participants and Design. A total of 164 undergraduates
(91 females, M
age
¼19.65) at a large northeastern US uni-
versity participated in the laboratory study. The experiment
used a single factorial between-subjects design, resulting in
three conditions (self-identity: linked vs. not linked vs.
control).
Procedure. Upon arrival, participants were asked their
names so they could be ostensibly checked in for their ses-
sion. Next, the research assistant provided them with a
small paper cup (3 fl. oz.) under the guise of participating
in a sampling study. Participants were told that in the sam-
pling study we were interested in whether consumers could
differentiate between tap water and bottled spring water.
Depending on the condition to which they were assigned,
participants were either given a blank cup (control condi-
tion), a cup with their name on it written by the research
assistant and spelled correctly (identity linked), or a cup
with their name on it written by the research assistant and
purposely spelled incorrectly (identity not linked). In the
latter condition, the research assistant was instructed to
change the name enough so that it could still be interpreted
but was obviously incorrect (e.g., Ashley might become
Ashli). After being seated at individual computer stations
and having the first water sample poured into their cups
from an unmarked bottle, participants were told to taste the
water sample and complete some online questions related
to the water. The research assistant then poured the second
water sample into their cup from a second identical un-
marked bottle and participants repeated the same process.
After tasting both of the water samples, answering the on-
line evaluation questions comparing the two different
water samples, and completing three other unrelated stud-
ies, participants were instructed to “dispose of your cup on
the way out.” The same trash can and recycle bin used in
our previous study were placed outside the closed door of
the laboratory. Participants exited one at a time. The de-
pendent variable was whether the cup was disposed of in
the trash or the recycle bin.
Results and Discussion
Five participants left with their cup in hand, leaving us
with 159 data points. We assume that these participants ei-
ther did not see the bins or opted to keep the cups. A logis-
tic regression analysis tested whether there were different
recycling rates across the cup conditions. Two dummy
250 JOURNAL OF CONSUMER RESEARCH

codes were created for the identity-linked
condition (Correctly spelled ¼2, Incorrectly spelled ¼1,
Control ¼1) and the identity-not linked condition
(Correctly spelled ¼1, Incorrectly spelled ¼2, Control
¼1) to allow comparison with the control condition. The
dependent variable was coded as Trash ¼0 and Recycle ¼
1. The analysis revealed a significant main effect of iden-
tity link, where those who were given a cup with their
name spelled correctly on it were significantly more likely
to recycle the cup (48%) compared to the control condition
(26% ; b¼.33, v
2
¼5.34, p<.05, OR ¼1.39).
Participants given a cup with their name spelled incor-
rectly, that is, those in the identity-not linked condition,
were just as likely to recycle their cup (24%) as those in
the control condition (26%; b¼.05, v
2
¼.09, p>.75).
By including an incorrectly linked comparison group,
the results from study 2 extend those of our previous study
to show that when a product–identity link is similar but not
exact, consumers dispose of the product in the same way
that they would dispose of a nonlinked product. However,
linking a product by correctly writing a consumer’s name
on a cup increases recycling behavior significantly.
It is possible to argue that in addition to linking the prod-
uct to a participant’s self-identity, writing names on the
products in study 1 and 2 may decrease anonymity and in-
crease demand effects. In anticipation of anonymity con-
cerns we only used first names on the paper and cups
because when participants sign up and register for the
study, unique identifiers are generated using their student
IDs. Participant anonymity was ensured. Nonetheless, in
subsequent studies we move away from the focus on names
to more broad levels of identity to further alleviate ano-
nymity concerns. In terms of demand effects, we attempted
to minimize the issue in several ways. First, in all of our
behavioral studies, participants are thanked for their par-
ticipation, dismissed, and told to dispose of their study ma-
terials on the way out. The wording was carefully crafted
to avoid any demand effects (we do not mention “trash” or
“recycle”) and to avoid revealing the true motives of our
studies. Second, each of our studies had a cover story that
had nothing to do with disposal. Third, participants left the
study one at a time, and the bins were placed outside the
lab so that no one could observe their disposal behavior.
Finally, in subsequent studies we included interactions that
should further alleviate concerns related to demand effects.
STUDY 3
In our third study we test for the moderating role of
strength of identity connection and do so by testing our
self-identity effects in another domain with broad practical
implications: self-brand connection. It is well accepted that
consumers form strong connections with their brands, and
brands can become an important component of their
identities (Belk 1988;Escalas and Bettman 2005;Fournier
1998;Thomson, MacInnis, and Park 2005). The decision
of how to dispose of identity-linked products is expected to
be moderated by the strength of connection between a con-
sumer’s identity and a product, in this case the strength of
self-brand connection. We expect that the strength of the
connection between a consumer’s identity and a brand will
lead to different representations of identity threat when dis-
posing of products. In particular, focusing on cola brands
we predict that consumers with a high self-brand connec-
tion to Coke (i.e., strong self-identity-link) will be more
likely to recycle a can of Coke than a can of Pepsi. We
have no predictions for those who are low in self-brand
connection.
Method
Participants and Design. A total of 200 participants
were recruited from Amazon’s Mechanical Turk (MTurk).
Four participants did not complete the entire instrument,
and therefore 196 respondents (65 females; M
age
¼30.56)
participated in a 2 (brand: Coke vs. Pepsi) by self-brand
connection to Coke between-subjects design.
Procedure. After reading an introductory page, partici-
pants read the following: “People are generally Coke or
Pepsi drinkers. It is pretty rare that people do not have a
preference. Which do you prefer?” Participants
then completed a five-item self-brand connection scale
(1 ¼Strongly disagree; 7 ¼Strongly agree) adapted from
Escalas and Bettman (2005) as a measure of the extent to
which they associate Coke with their identity. The five
items were (1) The Coke brand reflects who I am, (2) I can
identify with the Coke brand, (3) I feel a personal
connection to Coke, (4) The Coke brand suits me well, and
(5) I consider Coke to be “me.” The five items were com-
bined to form a reliable measure of self-brand connection
(a¼.96). Participants were then randomly assigned to one
of two brand conditions. Those in the Coke (Pepsi) condi-
tion were shown an image of a dented Coke (Pepsi) can
and read, “Imagine that this empty, dented can has been
left behind in the lunchroom at your work.” Participants
were then asked to respond to two randomly presented dis-
posal intentions items assessed on 7 point scales (1 ¼Very
unlikely; 7 ¼Very likely): “How likely are you to throw
this empty COKE (PEPSI) can in the garbage?” and “How
likely are you to throw this empty COKE (PEPSI) can in
the recycle bin?” The likelihood to trash item was reverse
coded, and the two items were combined to form a
reliable recycle index that served as our dependent variable
(a¼.71).
Results and Discussion
A regression with brand condition, mean centered self-
brand connection to Coke, and their interaction term as
TRUDEL, ARGO, AND MENG 251

predictors and the recycle index as the dependent variable
revealed a marginally significant effect for self-brand con-
nection (b¼.11, t¼1.87, p¼.06). The interaction be-
tween brand condition and self-brand connection was also
significant (b¼.15, t¼2.60, p<.01; figure 1). This inter-
action was analyzed using the Johnson-Neyman technique
to show the ranges of self-brand connection where the ef-
fect of brand condition (i.e., the difference between recycle
intentions for Coke vs. Pepsi) is significant and where it is
not (Hayes and Matthes 2009;Johnson and Neyman 1936).
The mean self-brand connection value was 3.52 (standard
deviation [SD] ¼1.69). The Johnson-Neyman point where
the probability of recycling Coke versus Pepsi is different
occurs at self-brand connection values of 2.46 (t¼1.98,
p¼.05) and at self-brand connection values of 6.71 (t¼
1.98, p¼.05). Stated differently, participants’ intentions to
recycle the Pepsi can were significantly less than intentions
to recycle the Coke can when self-brand connection to
Coke was greater than 6.71, or 1.89 SDs above the mean.
However, participants’ intentions to recycle the Pepsi can
was significantly greater than intentions to recycle the
Coke can when their self-brand connection to Coke was
less than 2.46, or .63 SDs below the mean. In sum, partici-
pants’ self-connection with the brand Coke significantly
and differentially impacted their intentions to recycle an
empty Coke versus an empty Pepsi can.
The results of study 3 show that consumers with a high
self-brand connection to Coke are more likely to recycle a
can of Coke than a can of Pepsi. Interestingly, the pattern
flipped for those with a low self-brand connection to Coke;
participants were more likely to recycle a can of Pepsi than
a can of Coke. Intuitively, those cola drinkers with a low
self-brand connection to Coke may have had high connec-
tions with other brands including Pepsi, or they could even
be considered “anti-Coke.” Importantly, this study pro-
vides initial support for our process by demonstrating the
effect of strength of connection on disposal. In the next
study, we set out to obtain more direct process evidence
and explore identity at the social level.
As we have done in the first three studies, research on
identity and products are often conceptualized and investi-
gated at the level of the individual (Ahuvia 2005,Belk
1988,Kettle and H€
aubl 2011). However, identity is often
derived by the groups to which people belong or aspire to
belong. The importance of group membership to one’s
identity is captured conceptually by one’s social identity
(Tajfel 1981,1982;Tajfel and Turner 1979;Turner 1987).
Social identity is defined as “the part of an individual’s
self-concept which derives from his knowledge of his
membership in a social group together with the value and
emotional significance attached to that membership”
(Tajfel 1981, 255). In the studies that follow, we focus on
social or collective identity (social identities are also com-
monly referred to as collective identities; Luhtanen and
Crocker 1992) to demonstrate that our effects are general-
izable beyond individual-level identities. This is important
given that many consumer product decisions are driven by
social identities and how they view the social groups to
which they belong (Berger and Heath 2007;Chan et al.
2012;White and Argo 2009;White and Dahl 2007).
STUDY 4
The goal of this study is to extend the initial findings
and show that, when no longer needed, everyday products
are also more likely to be recycled as compared to trashed
if they are linked to a consumer’s social identity.
Specifically, we examine the influence of national identity,
which is often a central part of one’s self (Druckman
1994). Our main prediction is that American respondents
will be more likely to recycle a product that contains an
image of the US flag as compared to the image of a flag
from another country that does not represent their national
identity. However, we also believe that this main effect
will be moderated by the strength of connection between a
consumer’s identity and the product.
We investigate social identity and measure participants’
strength of US identification to replicate the strength of
connection results from study 3 and to provide additional
support for our process explanation. We expect those who
identify most strongly with being American (i.e., being
American is an important part of their identity) to form
stronger connections to American branded/imprinted prod-
ucts. To confirm this we ran a pretest with 60 participants
(22 women, M
age
¼34.05) recruited from Amazon’s
MTurk. Participants first responded to demographic ques-
tions related to age, gender, and education, and then re-
sponded to eight US identification statements that were
FIGURE 1
THE INTERACTIVE IMPACT OF IDENTITY AND SELF-BRAND
CONNECTION ON INTENTIONS TO RECYCLE
1
2
3
4
5
6
7
12 34 5 6 7
Likelihood of Recycling
Coke Self-Brand Connection
Coke Pepsi
NOTE.—Johnson-Neyman region of significance is less than 2.46 and greater
than 6.71.
252 JOURNAL OF CONSUMER RESEARCH

averaged to create the strength of US identification meas-
ure (adapted from White and Dahl 2007). Next, all partici-
pants were asked to indicate how connected they felt
(1 ¼Not at all; 7 ¼Extremely) to nine different products
(three T-shirts, three cups, and three figurines). Each prod-
uct was either identity linked (US branded; e.g., a T-shirt
with an American flag on it), identity not linked (UK
branded; e.g., a T-shirt with the Union Jack on it), or con-
trol (no national identity-related aspects). The reported
connections for the three US-branded products were aver-
aged to create the US product identity-link index. A regres-
sion analysis revealed a significant effect of US identity
strength on US product–identity link (b¼.88, t¼9.55,
p<.001). That is, the stronger an individual’s US identity,
the stronger their connections with US-branded products.
The same procedure was performed for the UK-branded
products (b¼.14, t¼1.49, p>.14) and the control prod-
ucts (b¼.14, t¼1.39, p>.17); however, no significant
effects were found. These results support our contention
that the strength of connection with a product is highest
when it reflects an important part of a consumer’s identity.
In addition to testing our predictions, we also had two
other objectives for the study. First, it is possible that con-
sumers are more inclined to recycle products that they like
(i.e., about which they have positive evaluations). To test
this alternative explanation, we include product evaluation
items in this study. Second, in the first three studies partici-
pants were always presented with a forced choice for dis-
posal between recycling and trashing. This was intentional
given that our focal question is how people dispose of
everyday products that are linked versus not linked to their
identities, and trashing and recycling products are the two
most common methods of disposal. Thus in our context we
believe that this is realistic. However, in study 4 we re-
move the forced-choice measure for two reasons: (1) pre-
senting a forced choice does not provide participants with
the option to dispose of identity-linked products in an alter-
native manner and perhaps identity-linked products are dis-
posed of in a unique fashion that we had not considered,
and (2) removing a forced choice allows us to further rule
out another alternative account related to social norms (i.e.,
that disposing of an American flag goes against social
norms). In particular, a social norm account would predict
that the majority of participants would choose to neither
trash nor recycle a piece of paper depicting an American
flag.
Method
Participants and Design. A total of 240 American con-
sumers were recruited from Amazon’s MTurk website to
participate in this experiment (92 females, M
age
¼32.18).
The design was a single factorial with three between-sub-
jects conditions (national identity: linked vs. not linked vs.
control) and a measured individual difference to capture
the strength of their US identity.
Procedure. After giving consent to participate, partici-
pants responded to eight US identification statements
adapted from White and Dahl (2007) to capture how cen-
tral “being American” was to each individual’s identity.
These are the same items used in the pretest (all items are
included in online appendix A). For instance, one item
asked, “How important is being American to how you
view yourself?” (1 ¼Not at all important; 7 ¼Extremely
important). Another item asked, “How strong are the ties
you feel to your American identity?” (1 ¼Not at all strong;
7¼Extremely strong). The eight items were averaged to
create our strength of US identification measure (a¼.98).
Participants were then randomly assigned to one of the
experimental conditions and read the following scenario:
“Imagine that you had been given some plastic cups. On a
hot summer day, you decided to pour yourself a cold drink
in a plastic cup. After finishing the drink you decide its
time to go and leave.” Participants in the national identity-
linked condition were shown a picture of a plastic cup with
the US flag printed on it, those in the national identity not
linked condition were shown a picture of a plastic cup with
the Union Jack (flag of the United Kingdom) printed on it,
and participants in the control condition were shown a pic-
ture of a plain white plastic cup. Participants were then
asked to choose between the following three options: (1) I
would throw this cup in the garbage on the way out, (2) I
would throw this cup in the recycling on the way out, and
(3) I would do neither. Those who answered that they
would do neither were asked to “please write what you
would do with the cup” and were provided with a text entry
space. Finally, participants responded to three product
evaluation items. (1) How much do you like the look of
this cup? (1 ¼Not at all; 5 ¼Very much) (2) How much
do you like the cup itself? (1 ¼Not at all; 5 ¼Very much)
(3) How would you rate the quality of this cup (1 ¼Not at
all good; 5 ¼Very good).
Results and Discussion
Disposal Decision. Disposal results are summarized in
table 1 including the qualitative results. The results show
that participants are significantly more likely to dispose of
the cup in either the recycling or trash as opposed to doing
neither (Recycle vs Neither; v
2
(1) ¼134.33, p<.001 and
Trash vs Neither; v
2
(1) ¼43.13, p<.001). The likelihood
of doing neither was not significantly different across iden-
tity conditions (all comparisons v
2
<1; not significant
[NS]). As a result, we focus our analyses on the probability
of recycling versus trashing.
We expected that participants in the national identity-
linked condition would be significantly more likely to re-
cycle the plastic cup than those in the national identity not
TRUDEL, ARGO, AND MENG 253

linked condition and control conditions. We ran a logistic
regression to test our prediction, with the dependent vari-
able being disposal (Recycle ¼1; Trash ¼0). Two dummy
variables were created to capture our three conditions and
were used as independent variables; US dummy (Coded
US ¼2, UK ¼1, Control ¼1) and UK dummy (coded
US ¼1, UK ¼2, control ¼1). The US dummy proved
to be a significant predictor of recycling behavior (b¼.42,
v
2
¼10.65, p<.001, OR ¼1.53) while the UK dummy
was not significant (b¼.11, v
2
¼.95, p>.30). Planned
comparisons revealed that participants were more likely to
recycle the national identity-linked cup (US cup ¼80%)
than the national identity not linked (UK cup ¼62.5%;
b¼.41, v
2
¼5.0, p<.05, OR ¼1.51) and the control cup
(plain white cup ¼60%; b¼.60, v
2
¼10.53, p<.001,
OR ¼1.83; figure 2). Comparisons in disposal behavior
between the identity not linked (UK cup) and control
conditions proved not to be statistically different (b¼.2,
v
2
¼1.21, p>.25).
Strength of Connection. Next, we wanted to probe the
possible interaction between the national identity condi-
tions (linked, not linked, and control) and strength of US
identity. We ran a second logistic regression with the same
dependent variable and dummy variables. This time US
identity strength was included as an independent variable
and also as an interaction with each dummy variable. The
results revealed significant main effect of strength of US
identity (b¼.23, v
2
¼5.20, p<.05, OR ¼1.25) and a
significant US dummy by strength of US identity inter-
action (b¼.20, v
2
¼5.57, p<.05, OR ¼1.22). All other
effects failed to reach significance. The interaction is
graphed in figure 2 using the Johnson-Neyman technique
to explore the ranges of where the strength of US identity
on disposal is significant and where it is not. The UK
dummy and its interaction with strength of US identity
were included as covariates in Hayes and Matthes (2009)
MODPROBE procedure for probing interactions. The
Johnson-Neyman point where the probability of recycling
is significantly different between the identity-linked condi-
tion and the control condition is when the strength of US
identity value is 2.87 (Z¼1.96, p¼.05) or .90 SDs below
the mean of 4.43.This indicates that the probability of recy-
cling the identity-linked plastic cup is significantly greater
than the probability of recycling the blank plastic cup
when participants’ strength of US identity is greater
than 2.87.
Evaluations. The three evaluation items were averaged
to form a reliable index (a¼.86). Analysis of variance re-
vealed significant differences across conditions (F(2, 237)
¼28.75, p<001). Follow-up tests showed that partici-
pants evaluated the plain white control cup (M¼2.28,
SD ¼.81) less favorably than the US cup (M¼3.11,
TABLE 1
RESULTS AND WRITTEN RESPONSES FROM STUDY 4
Condition Trash Recycle Neither Comment if answered neither
Identity linked [US cup] 15%
(12/80)
80%
(64/80)
5%
(4/80)
If salvageable, I would try to reuse it
I would like to keep it
I would just leave it
Nothing
Keep it and reuse it until it’s not functional anymore
Identity not linked [UK cup] 30%
(24/80)
62.5%
(50/80)
7.5%
(6/80)
I wouldn’t buy it and I would probably just put it in a closet or something
I would rinse out the cup and put it away in order to use it again
I would drink out of it
I would keep it
I would keep it just in case I need to use it again
I would keep the cup
Control [white cup] 40%
(32/80)
60%
(48/80)
0%
FIGURE 2
INTERACTIVE INFLUENCE OF IDENTITY-LINKED PRODUCTS
AND STRENGTH OF IDENTITY ON DISPOSAL CHOICE
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1234567
Probability of Recycling
Strength of American Identity
USA Control
NOTE.—Johnson-Neyman region of significance when strength of identity is
greater than 2.88.
254 JOURNAL OF CONSUMER RESEARCH

SD ¼.94, p<.001) and less favorably than the UK cup
(M¼3.24, SD ¼.84, p<.001). The US cup and the UK
cup were evaluated similarly (p¼.36). Including the
evaluation index in the previous logistic regression ana-
lyses as a covariate does not affect the results. A logistic
regression on disposal (Recycle ¼1; Trash ¼0) with the
evaluation index as the lone independent variable was also
not significant (b¼.1, v
2
¼.41, p>.50). An additional
logistic regression on disposal with the evaluation index,
strength of US identity, and the evaluation by strength of
US identity interaction as independent variables did not re-
veal any significant effects (all v
2
<.5). As a result, we
can confidently claim that product liking or evaluations
cannot account for our disposal results.
The results of study 4 replicate the findings related to
identity strength from study 3 whereby participants who
strongly identify as being “American” are more likely to
recycle US-branded products, compared to those who do
not consider being American as an important part of their
identity. Further, even when participants’ disposal choices
were not limited to trashing or recycling, participants were
still significantly more likely to recycle the cup containing
the US flag compared to the cups that were not linked to
their identity. Regardless of how they identify with being
American, the norm regarding how paper or a cup with an
American flag printed on it should be disposed of is the
same for all participants. Finally, participants’ evaluations
of a product does not influence whether it will be recycled.
In sum, the differential behaviors demonstrated here cannot
be explained by social norms or product evaluations.
STUDY 5
In the previous two studies we provided process sup-
port by demonstrating that strength of connection moder-
ates the effect of product–identity links on disposal. We
have shown that as an individual’s strength of connection
with a product increases, they will be less likely to trash
the identity-linked (vs. non–identity-linked) product. Our
key proposition is that as the strength of connection be-
tween an individual and a product increases, so too does
the potential identity threat that would result from trashing
that product. Moreover, when the strength of connection
decreases, the identity threat will similarly decrease and, as
we observed in study 4, there are no differences in disposal
behavior as a function of the presence/absence of an iden-
tity link. Study 5 investigates the role of negative emotions
that are evoked during identity threat situations to provide
additional process support for our effects. According to
prior research, identity threats can often result in an in-
crease in negative emotions (Cheryan and Monin, 2005;
Guendelman, Cheryan, and Monin 2011;Packard and
Wooten 2013). Based on this literature, we expect that con-
sumers with a stronger connection to a particular product
will experience greater negative emotions as a result of a
potential identity threat from trashing that product, which
will ultimately impact their disposal decisions. Stated dif-
ferently, negative emotions should mediate the effect of
strength of connection on disposal behavior only when
there exists a product identity link. To test this possibility,
study 5 examines how anticipated identity threat from
trashing an identity-linked products leads to negative emo-
tions that influence consumers’ disposal decisions.
Method
Participants and Design. Two hundred American con-
sumers were recruited from Amazon’s MTurk website to
participate in this experiment (88 females, M
age
¼33.85).
The design was a single factorial with two between-sub-
jects conditions (national identity: linked vs. control) and
an individual difference measure to capture the strength of
connection to their US identity (adapted from White and
Dahl 2007).
Procedure. After giving consent to participate, partici-
pants responded to the same eight strength of US identifi-
cation statements used previously (adapted from White and
Dahl 2007) to capture how central “being American” was
to each individual’s identity. Participants were then ran-
domly assigned to one of the experimental conditions and
read the following scenario: “Imagine that you had been
given some plastic cups. On a hot summer day, you
decided to pour yourself a cold drink in a plastic cup.”
Participants in the national identity–linked condition were
shown a picture of a plastic cup with a US flag printed on
it, and participants in the control condition were shown a
picture of a plain white plastic cup. All participants then
read, “After finishing the drink you dispose of the cup in
the GARBAGE.” Therefore all participants imagined dis-
posing of the cup in the trash. If our theorizing is correct,
disposing of the product in the garbage should present an
identity threat for those participants in the identity-linked
conditions.
Next participants were asked how much they agreed
with three statements (1 ¼Strongly disagree; 7 ¼Strongly
agree): (1) I feel ashamed for throwing the cup in the gar-
bage. (2) I feel guilty for throwing this cup in the garbage.
(3) I regret throwing this cup in the garbage. The objective
of these items was to capture negative emotions related to
“trashing” the recyclable cup and in particular to see if a
potential identity threat evoked negative emotions that led
to decisions to recycle. The items were combined and aver-
aged to form a reliable negative emotions index (a¼.98).
Participants were then asked if they would like to change
their disposal decision. They were specifically asked,
“How likely are you to reach into the garbage can, pick up
the USA (white) cup that you had previously disposed of in
the garbage, and dispose of it in the recycling bin nearby?”
TRUDEL, ARGO, AND MENG 255

(1 ¼Very unlikely; 7 ¼Very likely) that was used as the
dependent variable. The experiment concluded with par-
ticipants answering demographic questions.
Results and Discussion
Disposal Decision. A regression analysis with identity
condition (Identity-linked ¼1 and Control ¼1), mean
centered strength of US identification, and their interaction
term on disposal behavior revealed significant main effects
of identity condition (b¼.43, t¼3.00, p<.01) and US
identification (b¼.28, t¼2.97, p<.01) and a significant
two-way interaction (b¼.23, t¼2.43, p<.05). The inter-
action was further analyzed using the Johnson-Neyman
technique (Hayes and Matthes 2009;Johnson and Neyman
1936) to explore the ranges where the influence of strength
of US identification on disposal is and is not significant.
The Johnson-Neyman point where the probability of recy-
cling is significantly different between the identity-linked
condition and the control condition is when the strength of
US identification value is 4.26 (t¼1.97, p¼.05) or .37
SDs below the mean of 4.82. Stated differently, the likeli-
hood of reaching into the trash can to retrieve a previously
trashed identity-linked plastic cup to redispose of it in the
recycling, in comparison to a previously trashed white
plastic cup, is significantly greater when participants’
strength of US identification is greater than 4.26.
Mediating Role of Negative Emotions. We predicted
that negative emotions should mediate the effect of
strength of connection on disposal behavior only when
there exists a product identity link. We conducted a moder-
ated mediation analysis using the PROCESS macro for
SPSS (Model 7, Hayes 2013) with mean-centered strength
of US identification as the independent variable, identity
condition (Identity linked ¼1 and Control ¼1) as the
moderator and negative emotions as the mediator. The re-
sults indicated that negative emotions were predicted by
strength of US identification (b¼.17, t¼2.16, p<
.05) and by the identity link by US identification inter-
action (b¼.21, t¼2.57, p¼.01) in the mediator
model. The main effect of identity condition was not sig-
nificant (t¼1.00, p>.30). Using bias-corrected bootstrap-
ping (n ¼1000, Hayes 2013) to generate 95% confidence
intervals (CIs) , the conditional indirect effect of US identi-
fication on disposal behavior through emotions was posi-
tive and significant in the identity-linked condition
(indirect effect ¼.17; 95% CI excluding zero ¼.07–.31)
but not in the control condition (95% CI, .13 to .08). The
direct effect of US identification is also positive (.22) and
significant (p <.05; 95% CI, .04–.40) and therefore it is
complementary mediation (Zhao, Lynch, and Chen 2010).
The results from study 5 provide further support that
consumers are more likely to recycle identity-linked prod-
ucts. Participants were more likely to reach into the trash
to retrieve an identity-linked product as compared to a non-
linked product and put it in the recycling. Furthermore,
those who most strongly identified with being American
(as compared to those who weakly identified with being
American) were even more likely to recycle a cup with a
US flag imprinted on it. Finally, we were able to gain add-
itional process support by demonstrating that negative
emotions mediate the effect of strength of connection on
disposal behavior only when there exists a product identity
link. When consumers with strong connections to products
anticipate an identity threat from trashing, they feel
increased negative emotions that then lead to their decision
to recycle.
STUDY 6
The goal of study 6 is to build on these previous studies
by providing more direct evidence that trashing an item
linked to one’s identity indeed threatens that identity.
Identity threat is defined as occurring when favorable
views of oneself are put into question such that it nega-
tively influences one’s self-conception (Baumeister, Smart,
and Boden 1996;Campbell and Sedikides 1999;Steele
1988). We have theorized that throwing an identity-linked
product in the trash presents an identity threat because it is
symbolically similar to throwing a piece of yourself in the
trash. Therefore, if throwing an identity-linked product in
the trash is indeed an identity threat, then we should expect
that doing so will lower a consumer’s sense of self-concept
(i.e., esteem) as compared to throwing an identity-linked
product in the recycling (Campbell and Sedikides 1999;
Park and Maner 2009;Steele 1988). As in the previous
study, study 6 again tests consumers’ national identity.
Given that nationality, the focal identity we test in this
study, is an aspect of one’s social identity, consistent with
prior work, we explore esteem at the collective level as the
focal dependent variable (Escalas and Bettman 2005;
Luhtanen and Crocker 1992;White and Argo 2009).
Method
Participants and Design. Four hundred Americans
were recruited from Amazon’s MTurk website to partici-
pate in this experiment. Three participants did not com-
plete the experiment, and six others were removed because
they were not American citizens, leaving 391 participants
(152 females; M
age
¼31.76). Participants were randomly
assigned to one of four conditions in a 2(national identity:
linked vs. control) 2(disposal: trash vs. recycle) be-
tween-subjects design. The focal dependent variable was a
state measure of collective self-esteem.
Procedure. Participants were introduced to the experi-
ment where they read a vignette that asked them to imagine
that “On a hot summer day, you decided to pour yourself a
256 JOURNAL OF CONSUMER RESEARCH

cold drink in a plastic cup.” In the national identity–linked
conditions, the plastic cup had a US flag design on it; in
the control group the plastic cup was plain white. To ma-
nipulate disposal, participants were randomly assigned to
read one of the two disposal conditions (trash vs. recy-
cling). That is, they read, “After finishing the drink you
crumple/dent the plastic cup before disposing of it in the
trash (recycling).” The cup was described as being
crumpled/dented to provide a stringent test of our predic-
tion because past research has found that distorting the
form of a product decreases the usefulness and increases
consumers’ likelihood of trashing the product (Trudel and
Argo 2013). Participants then completed the shortened,
adapted state collective self-esteem scale that served as our
dependent variable (Luhtanen and Crocker 1992). The ori-
ginal state collective self-esteem scale consists of 16 items
that load onto four different factors (i.e., membership, pri-
vate, public, and identity). Since the membership and iden-
tity factors accurately capture the elements of group
identity that we are interested in, we included only those
eight items in our experiment. The collective state self-es-
teem scale was introduced in the following manner: “We
are all members of different social groups. Please respond
to the following statements based on how you
CURRENTLY feel when you think about your
NATIONALITY and your membership in this social group
as an AMERICAN.” For example, statements included
“Right now this social group is an important reflection of
who I am” and “Right now, my group membership has
very little to do with how I feel about myself” (Strongly
disagree ¼1; Strongly agree ¼5). To finish the study, par-
ticipants indicated their nationality.
Results and Discussion
A 2(national identity) 2(disposal) analysis of variance
with the collective state self-esteem index (a¼.86) as the
dependent variable revealed a significant main effect of
disposal (F(1, 387) ¼6.22, p<.05, partial g
2
¼.016)
and the predicted group identity by disposal interaction
(F(1, 387) ¼11.21, p<.001, partial g
2
¼.028).
Consistent with our hypothesis, collective state self-esteem
was significantly lower when the identity-linked plastic
cup (US) was trashed (M¼3.10, SD ¼.79) as compared
to recycled (M¼3.51, SD ¼.66, F(1,387) ¼17.79, p<
.001, partial g
2
¼.043 ; figure 3). Collective self-esteem
did not differ in the control condition (i.e., plain white plas-
tic cup) as a function of the form of disposal (recycled: M
¼3.24, SD ¼.67, trashed: M¼3.30, SD ¼.64, F(1,387)
¼.36, p>.50). Planned comparisons also showed that col-
lective state self-esteem was higher when participants
recycled an identity-linked plastic cup (M¼3.51, SD ¼
.66) as compared to the plain white plastic cup (M¼3.24,
SD ¼.67; F(1,387) ¼7.24, p<.01, partial g
2
¼.018).
Finally, collective state self-esteem was lower when
participants trashed an identity-linked plastic cup (M¼
3.10, SD ¼.79) as compared to the plain white plastic cup
(M¼3.30, SD ¼.64; F(1,387) ¼4.18, p<.05, partial g
2
¼.011).
The results of study 6 provide direct process support for
our theoretical account. Collective state self-esteem is dam-
aged when social identity–linked products are disposed of in
the trash. In particular, participants had significantly lower
collective state self-esteem after they imagined disposing of a
national identity–linked cup in the trash, which was akin to
trashing an important part of themselves, in comparison to
participantswhoimagineddisposingofthesamecupinthe
recycling. The identity threat was also greater in the identity-
linked trash condition than the two control conditions.
Participants who trashed versus recycled a plain white cup
did not experience an identity threat and had equivalent levels
of collective state self-esteem. Since consumers are motivated
to maintain a positive self-view and will act to protect the self
(Steele 1988), the results of study 6 provide process evidence
for our proposition that throwing an identity-linked product in
the trash is an identity threat as there was a decrease in col-
lective state self-esteem.
STUDY 7
Our focus in this research up until now has been on posi-
tively valenced product identity links, that is, a product is
linked to a positive component of a consumer’s identity for
which there are positive associations. As we have observed
in these situations, the decision to throw the identity-linked
product in the trash threatens that identity and consumers
FIGURE 3
THE EFFECT OF DISPOSAL ON COLLECTIVE STATE
SELF-ESTEEM
2.7
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
Trash Recycle
Collective Self-Esteem
Disposal Condition
Control USA
NOTE.—Error bars represent 61 SD.
TRUDEL, ARGO, AND MENG 257

are more likely to recycle (vs. trash) because it is the least
threatening means to dispose of a positive identity-linked
product. But not all products form positive-identity links.
Products can also form links to a component of one’s iden-
tity for which there are negative associations. In this study
we examine the impact of negatively valenced product–
identity links on disposal decisions to gain further process
evidence. More specifically, in this study we make the op-
posite prediction for negatively valenced product–identity
links as we do for positively valenced product–identity
links. When a product is linked to a component of a con-
sumer’s identity for which there are negative associations,
trashing the product does not create a self-threat. We ex-
pect that trashing a product with a negative identity-link
may serve as a symbolic act of dissociation from that iden-
tity, and therefore, we predict that consumers will be more
likely to dispose of a negative identity-linked product in
the trash (vs. the recycling).
In this experiment we manipulate the valance of the
identity link by having students read positive versus nega-
tive news stories about their university because university
affiliation is a significant component of undergraduates’
identities (e.g., Ferraro, Bettman, and Chartrand 2009;
White, Argo, and Sengupta 2012). We predict that those
who read shameful (prideful) information about their uni-
versity will subsequently experience a more negative (posi-
tive) identity link that will increase the likelihood that the
product is trashed (recycled). Further, in all of our prior be-
havioral experiments, we told participants to dispose of the
products on their way out of the laboratory. Participants
were therefore left with a decision that was either “recycle”
or “trash.” As a more stringent test of our effect, similar to
study 4, we give participants the additional option to keep
the product. Based on the results of study 4 we expect that
participants would be equally likely to keep the product
across conditions.
Method
Participants and Design. A total of 118 undergradu-
ates (63 females; M
age
¼19.63) at a large northeastern US
university participated in a laboratory-based experiment.
The experiment used a 2(product: linked vs. not linked) 
2(social identity valence: positive vs. negative) between-
subjects design.
Procedure. Before beginning the session, each partici-
pant was given a piece of paper measuring 1.5 5 inches.
The paper for those in the product-linked condition con-
tained the logo of the participants’ current university,
whereas those in the identity not linked condition were pro-
vided with a piece of blank white paper. Each piece of
paper contained a 3 to 4 digit code, and participants were
instructed to enter this number into their computer when
prompted in order to begin their session. This number was
used to track their disposal behavior and match it with our
conditions.
Under the cover story of assessing online articles, par-
ticipants were randomly assigned to read either a positive
article describing a recent successful graduate of their uni-
versity (positively valenced social identity condition) or a
negative article describing recent graduates from their uni-
versity who were under criminal investigation (negatively
valenced social identity condition). Consistent with the
cover story, participants were asked questions related to
online articles (e.g., “How often do you read news articles
online?”). After reading the article, participants were asked
to indicate how they felt about being a member of the uni-
versity community (1 ¼Ashamed; 7 ¼Proud). Responses
to this item were included as a manipulation check for so-
cial identity valence.
At the conclusion of the session, participants were in-
structed to take the paper with them out of the lab without
mention of disposal. This eliminated the forced choice be-
tween recycling and trashing. Similar to previous studies, a
trash can and recycle bin were placed outside the closed
door of the laboratory. The dependent variable was
whether participants placed the paper in the trash can, re-
cycle bin, or kept it.
Results and Discussion
Manipulation Check. Participants in the positive social
identity valence condition reported significantly more
pride in being a member of the community (M¼5.86,
SD ¼1.47) compared to those in the negative social iden-
tity valence condition (M¼4.64, SD ¼1.91, t¼3.89, p<
.001). Thus we were successful in manipulating the social
identity valence.
Disposal Decision. We did not predict any differences
in disposal behavior in our control conditions (the two
product not linked conditions where participants were
given a blank piece of paper). As expected, there were no
differences in rates of trashing, recycling, or keeping the
not-linked paper across social identity valence conditions
(all comparisons X
2
<1.5, NS). As a result we combined
the two product not linked conditions to make a single
hanging control condition, leaving us with three conditions
that we will refer to as the control condition, the positively
valenced social identity condition, and the negatively
valenced social identity condition.
We used a multinomial logistic regression to explore the
impact of social identity valence on the disposal of a linked
product compared to the control condition. To do so we
created two contrast-coded dummy variables. The first
dummy variable (positively valenced social identity
dummy) was coded as follows: Positive ¼2, Negative ¼
1, Control ¼1. The second dummy variable (con-
trol dummy) was coded as follows: Positive ¼1,
258 JOURNAL OF CONSUMER RESEARCH

Negative ¼1, Control ¼2. This coding ensured that the
negatively valenced social identity condition was the com-
parison group. A significant main effect of the positive so-
cial identity valence dummy emerged (X
2
(2) ¼12.57, p<
.01) while the control dummy was not significant (X
2
(2) ¼
2.76, p>.25). The decomposed results are presented in
figure 4.
Planned comparisons between positive and negative so-
cial identity valence conditions revealed that participants
in the positive condition (46%) were significantly more
likely to recycle compared to those in the negative condi-
tion (15%; X
2
(1) ¼7.39, p<.01, OR ¼2.38). Participants
in the positively valenced social identity condition were
also significantly less likely to trash the university logo im-
printed paper than participants in the negatively valenced
social identity condition (28% vs. 73%; X
2
(1) ¼10.73, p<
.001, OR ¼.38). There was no difference in the number of
participants who chose to keep the paper across positive
and negative social identity valence conditions (25% vs.
13%; X
2
(1) ¼1.31, p¼.25).
In comparison to the participants in the control condi-
tion, participants in the positively valenced social identity
condition were less likely to trash (28% vs. 57%; X
2
(1) ¼
3.73, p¼.05, OR ¼.64) and directionally more likely to
recycle (46% vs. 28%; X
2
(1) ¼2.18, p¼.14) the paper.
There were no differences between the control condition
and the negatively valenced social identity condition (all
comparison X
2
<1, p’s >.30).
Finally, participants were equally likely to keep the
paper across all of the conditions (positive: 25%; negative:
13%; control: 17%; all comparison X
2
<1.31, p’s>.25),
which was in line with our prediction.
The results of study 7 reveal that if a consumer is pro-
vided with negative information regarding an aspect of
their social identity, in this case university affiliation, this
can lead to a negative social identity link that directly re-
sults in an increase in trashing rates. This study extends our
previous results and provide a more complete demonstra-
tion of how identity-linked products are treated differently
at disposal. Study 7 clearly demonstrates that this phenom-
enon does not just depend on the presence or strength of an
identity link but also on the valence of the component of
the identity to which the product is linked.
GENERAL DISCUSSION
We have known for some time that consumers’ identities
influence purchasing decisions and that meaningful posses-
sions are treated differently during the divestment process,
but until now, research had not examined whether con-
sumers’ identities can directly influence their decision to
trash or recycle an everyday product. We hypothesized that
when an everyday product (e.g., paper, cups) is linked to a
consumer’s identity, it is more likely to be recycled.
Across seven studies (plus three additional studies that we
report in online appendix B), we showed that the presence,
strength, and valence of an identity causes consumers to
treat functionally similar everyday products differently
during disposal. Further, we find that consumers are more
likely to recycle a product linked to the self because trash-
ing such a product creates an identity threat.
In study 1 we found initial support for our prediction,
where participants recycled paper with their names written
on it more so than paper with a different name written on
it. Study 2 replicated this finding while also demonstrating
that even when an unknown third party writes a con-
sumer’s name on an item, as is often experienced in coffee
shops, this is enough to activate the identity link, provided
their name is spelled correctly. In study 3, we extended our
effects to self-brand connections and found that partici-
pants who highly identified with Coca-Cola were more
likely to recycle a Coke can compared to a Pepsi can. In
contrast, those with a low self-brand connection to Coke
were more likely to recycle the Pepsi can.
In study 4, we sought to extend the scope of our research
to include a different type of identity: social identity. We
also introduced an alternative to trashing or recycling
where participants could indicate how they would dispose
of a product. Even with this freedom of choice, US-born
participants were still more likely to recycle a plastic cup
depicting the US flag compared to a plastic cup depicting
the UK flag or a blank plastic cup. This study also ruled
out familiarity, evaluations, and social norms as alternative
explanations. In study 5, US-born participants were more
likely to recycle plastic cups with a US flag on it compared
to a blank cup. This study revealed that the stronger the
identity-link, the stronger the negative emotions experi-
enced when trashing an identity-linked product. In study 6
FIGURE 4
THE EFFECT OF POSITIVE AND NEGATIVE SOCIAL IDENTITY
VALENCE ON DISPOSAL BEHAVIOR
57%
28%
73%
27%
46%
13%
17%
25%
13%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Control Positive Negative
Percentage Recycled
Trashed Recycled Kept
TRUDEL, ARGO, AND MENG 259

we provided support for the idea that trashing an identity-
linked product creates an identity threat. The results
showed that US-born participants reported lower state col-
lective self-esteem when they imagined trashing a plastic
cup with the US flag on it, compared to when they imag-
ined recycling it. Finally, in study 7, we revealed that the
valence of the identity-link has a differential effect on dis-
posal decisions, even with a no forced disposal choice.
That is, students who read a negative article about their
university were more likely to trash and less likely to re-
cycle paper depicting their university’s logo, compared to
those who read a positive article about their university.
Overall, we believe this research makes important con-
tributions from both a theoretical and practical perspective.
From a theoretical perspective, this research explores the
impact of intangible characteristics related to the product
itself—the extent to which it has an identity that is import-
ant to consumers and thus links the product to the individ-
ual. We have shown that identity links are not exclusive to
meaningful (i.e., special) possessions (Belk 1988) but can
also be formed with everyday products, even when they
are no longer needed or wanted. Like meaningful posses-
sions that are generally disposed of in purposeful ways so
as to preserve their meaningfulness and the identity with
which they are imbued (Belk 1988;Belk, Sherry, and
Wallendorf 1988;Belk, Wallendorf, and Sherry 1989;
Burris and Rempel 2004;Ferraro et al. 2011), everyday
products are often intrinsically linked to consumers’ iden-
tity and also get treated differentially at disposal. Unlike
the meaningful possessions studied by Belk and col-
leagues, the disposal options for everyday products typic-
ally involve a decision between either throwing the product
in the trash or in the recycling bin. Nonetheless, the self-
and social-identity imbued in products substantively influ-
ences consumers’ disposal decisions.
Research on disposal decision making has been largely
limited to investigating the impact of factors related to the
promotional message (Lord 1994;Smith and Petty 1996;
White et al. 2011), the consumer’s personality (McCarty
and Shrum 2001), and tangible product characteristics
(Trudel and Argo 2013) on recycling behaviors.
Importantly our results suggest that placing these identity-
linked products in the trash threatens that identity, a situ-
ation consumers are motivated to avoid. When examining
the relationship between identity and behavior, previous re-
search has focused on consumers’ memberships within
groups and how they tend to behave in a manner consistent
with others. For example, individuals may purchase or-
ganic vegetables when they identify themselves as a “green
consumer” or purchase locally manufactured goods due to
their meaningful association with the local community
(Lantz and Loeb 1998;Sparks and Shepard 1992).
Generally, the link between identity and recycling behavior
has often looked at group-level conformity and individual
attitudes, rather than product-level influencers. For
example, Terry, Hogg, and White (1999) recognized that
when an individual strongly identifies with a particular ref-
erence group, such as friends, family, or neighbors
(Hanson 1980), they are more motivated to recycle if recy-
cling is viewed as an important aspect of that reference
group.
From a practical perspective, we believe it is important
to enhance our understanding of the ways through which
we may encourage sustainable consumer behaviors if we
are to ensure a healthy Earth, not only for the present gen-
eration but for those who will follow. Our research adds to
the growing discourse on environmental issues by present-
ing a different perspective on what influences consumers’
recycling decisions. Specifically, our findings reveal that
by creating an identity link or making an existing identity
link salient that consumers may be less likely to trash prod-
ucts that should go into the recycling bin. There are many
examples of firms that already link products to our identi-
ties but may not be aware of the disposal consequences.
For instance, Coca-Cola’s “Share a Coke” campaign
(http://www.coca-colacompany.com/coca-cola-unbottled/
is-your-name-on-a-coke-bottle-find-out-here/#TCCC),
where consumers find their names on bottles of Coke, is
likely to increase recycle rates for those who drink from a
bottle with their name on it, however, drinking from a bot-
tle with a different name may lead to increased trashing if
consumers have a negative association with that name. As
another example, Starbucks and other coffee shops may be
able to increase recycling rates of their cups by instructing
their baristas to ensure that they spell their customers
names correctly.
Firms often appeal to consumers’ identities to encourage
acquisition of their products;, therefore if the manufacturer
includes an aspect of this identity on the packaging or the
product itself, this could potentially increase the likelihood
that the product will subsequently be recycled. For ex-
ample, Budweiser is marketed as “The Great American
Lager,” often using US-centric imagery in advertisements
to appeal to American consumers. Periodically, Budweiser
releases limited edition bottles and packaging containing
the American flag or the Statue of Liberty, for example,
with both providing very strong links with consumers’ na-
tional identity. Our research would suggest that these lim-
ited edition bottles and packaging would be more likely
placed in the recycle bin due to the stronger product–iden-
tity link. Finally, our research also points to an additional
benefit of a strong brand. In this work we demonstrate how
consumers’ connections to brands increases the likelihood
that the products will be recycled rather than trashed.
Limitations and Future Research
The effects of social norms, social influence, and de-
mand are always a concern when conducting behavioral re-
search in areas that might have a socially desirable way to
260 JOURNAL OF CONSUMER RESEARCH

respond. Through the careful design of our experiments we
feel that we have mitigated these issues, but nonetheless
we discuss some of the potential limitations and issues of
our work and in studying disposal behavior more generally.
While it certainly would violate social norms to throw an
actual American flag in the trash, it is not clear that there
are such established social norms about how to dispose of
pieces of paper or plastic cups imprinted with American
flags or paper adorned with a university logo. Nonetheless,
we believe that studies 4, 5, and 7 help to rule out this ex-
planation. In particular, we found that those individuals
low in American identification (being American is not an
important part of their identity) are less likely to recycle
the identity-linked product than those high in American
identification (being American is an important part of their
identity). Regardless of how consumers identify with being
American, the norm regarding how a cup (or paper, etc.)
with an American flag printed on it should be disposed of
is the same for all Americans. Thus we believe that the
interactions found rule out the social norms explanation.
Similarly, we manipulated university pride in study 7 to
show differential disposal behavior, but manipulating uni-
versity pride should also not affect social norms regarding
the disposal of paper imprinted with a university logo. It
could be argued that the social norm is to recycle all recyc-
lable materials; however, we know that a large proportion
of what is recyclable ends up in landfills. In all our studies
we show differential behaviors that cannot be explained by
social norms alone.
We also believe that our results cannot be explained by
social influence or lack of perceived anonymity for several
reasons. First, in all our studies, the trash and recycle bins
had flip lids to conceal what had been previously deposited
into the bins. Second, participants left one at a time and the
recycle and trash bins were in the hallway outside the lab
as participants left the room. This meant participants’ dis-
posal behavior was done outside of any direct observation.
Third, only first names were ever used on paper and cups
that were disposed of, which allowed us to maximize ano-
nymity because disposed of products could not be linked
back to our participants. Last, the MTurk studies were
completely anonymous.
Finally, we also believe our results cannot be explained
by demand. In the present research, participants were al-
ways taking part in an “unrelated study” and asked to clean
up their computer terminals and “dispose” of any study
materials on the way out. We feel that if participants had
shown any demand effects, the actual recycling rates would
have been much higher. While we did not specifically in-
clude suspicion probes in this work, we did include suspi-
cion probes in prior work that we have done in this domain
using the exact same experimental paradigm and no par-
ticipants identified disposal behavior as the purpose of the
experiments (Trudel and Argo 2013). Finally, to further
guard against demand, all of our experiments were between
subjects, and we found interaction effects in study 4, 5, 6,
and 7 (which further rule out any demand effects).
Given that our research is only a first step in understand-
ing the impact of intangible product properties on disposal
behaviors, there are many avenues for future research. For
example, in the present research the focal social identities
were made salient using images. Future research could
examine whether consumer’s product–identity links can be
activated through the use of text such as “Made in the
USA” or even through the use of certain color combin-
ations such as red, white, and blue (i.e., the American na-
tional colors). As a second example, across all of our
studies we used relatively subtle links between identity and
disposal such that consumers had to draw conclusions as to
the implications of their disposal decision. Future research
could study whether the use of explicit links might further
enhance our effects. In other words, if Coke drinkers read
either an advertising appeal or saw a message on the can
that makes the link between Coke identity and recycling
more salient (e.g., “You bought Coca-Cola because you are
a Coca-Cola drinker. Now be a Coca-Cola recycler”) might
this further increase the recycling rates of these products?
Relatedly, consumers have been shown to differ in envir-
onmental concern and recycling attitudes and this impacts
behaviors (Schultz and Oskamp 1996). While we did not
include measures for these individual differences, future
research could explore the interplay between environmen-
tal attitudes and identity-product links to predict recycling
behaviors. Stated differently, are identity links more,
equally, or less influential than attitudes in predicting how
an everyday product will be disposed? While we focus on
the impact of membership identities (i.e., identities we cur-
rently belong to) on disposal decisions, future research
could examine the impact of both aspirational groups (i.e.,
group identities we would like to have) and dissociative
groups (i.e., group identities we do not want to have) on
recycling behaviors.
Finally, although we differentiate between everyday
products and the previously researched “meaningful” (i.e.,
special; Belk 1988) possessions, if a product is linked to a
consumer’s identity, then by definition this product would
be imbued with more meaning. However, this meaning
may change over time as personal growth and major role
transitions shift peoples’ self-concepts (Schouten 1991),
which may also influence the relationship consumers have
with their products. A child’s teddy bear may represent
comfort and security and a tight link to a child’s identity,
but as that child ages they may see it as immature when
they are adolescents but nostalgic as adults, even though
there is still arguably an identity link continually present.
Therefore in a more general sense, future research could
examine the influence of this shifting self and how differ-
ent meanings given to possessions over time can influence
how they are disposed. This would ascertain whether the
TRUDEL, ARGO, AND MENG 261

influence of identity links is embedded within the larger
construct of meaningfulness.
In sum, we have taken an initial step toward understand-
ing how the presence of an identity link with an everyday
product can increase the frequency with which the product
is recycled. These findings can subsequently be used to in-
form future product and packaging design as well as mar-
keting communication development with the aim to
increase recycling rates by linking the product to the con-
sumer’s identity.
DATA COLLECTION INFORMATION
The third author collected the data for study 1 in the au-
tumn of 2014 at Boston University’s Behavioral Research
Lab. Data for study 2 (spring 2015), and study 7 (autumn
2015) were also collected by the third author at Boston
University’s Behavioral Lab. All data collection for stud-
ies1, 2, and 7 additionally employed research assistants
supervised by the third author. The first and third author
jointly analyzed the data for these studies. The first author
collected and analyzed the data for study 3 in the autumn of
2014, study 4 in the autumn of 2015, study 5 in the autumn
of 2015, and study 6 in summer 2014. Data for Study 3, 4,
5, and 6 were collected through Amazon’s Mechanical
Turk website. The manuscript was jointly authored.
REFERENCES
Aaker, Jennifer, Susan Fournier, and S. Adam Brasel (2004),
“When Good Brands Do Bad,” Journal of Consumer
Research, 31 (1), 1-16.
Ahuvia, Aaron C. (2005), “Beyond the Extended Self: Loved
Objects and Consumers’ Identity Narratives,” Journal of
Consumer Research, 32 (1), 171-84.
Argo, Jennifer J., Darren W. Dahl, and Rajesh V. Manchanda
(2005), “The Influence of Mere Social Presence in a Retail
Context,” Journal of Consumer Research, 32 (2), 207-12.
Baumeister, Roy F., Laura Smart, and Joseph M. Boden (1996),
“Relation of Threatened Egotism to Violence and
Aggression: The Dark Side of High Self-Esteem,”
Psychological Review, 103 (1), 5-33.
Belk, Russell W. (1988), “Possessions of the Extended Self,”
Journal of Consumer Research, 15 (2), 139-68.
Belk, Russell W., John F. Sherry Jr., and Melanie Wallendorf
(1988), “A Naturalistic Inquiry into Buyer and Seller
Behavior at a Swap Meet,” Journal of Consumer Research,
14 (4), 449-70.
Belk, Russell W., Melanie Wallendorf, and John F. Sherry Jr.
(1989), “The Sacred and the Profane in Consumer Behavior:
Theodicy on the Odyssey,” Journal of Consumer Research,
16 (1), 1-38.
Berger, Jonah and Chip Heath (2007), “Where Consumers
Diverge from Others: Identity Signaling and Product
Domains,” Journal of Consumer Research, 34 (2), 121-34.
Bolton, Lisa E. and Americus Reed (2004), “Sticky Priors: The
Perseverance of Identity Effects on Judgment,” Journal of
Marketing Research, 41 (4), 397-410.
Bostrom, Ann, M., Granger Morgan, Baruch Fischhoff, and
Daniel Read (1994), “What Do People Know About Global
Climate Change? 1. Mental Models,” Risk Analysis, 14 (6),
959-70.
Brendl, C. Miguel, Amitava Chattopadhyay, Brett W. Pelham, and
Mauricio Carvallo (2005), “Name Letter Branding: Valence
Transfers When Product Specific Needs Are Active,” Journal
of Consumer Research, 32 (3), 405-15.
Brewer, Marilynn B. (1991), “The Social Self: On Being the Same
and Different at the Same Time,” Personality and Social
Psychology Bulletin, 17 (5), 475-82.
Brough, Aaron R. and Mathew S. Isaac (2012), “Finding a Home
for Products We Love: How Buyer Usage Intent Affects the
Pricing of Used Goods,” Journal of Marketing, 76 (4), 78-91.
Burris, Christopher T. and John K. Rempel (2004), “‘It’s the End
of the World as We Know It’: Threat and the Spatial-
Symbolic Self,” Journal of Personality and Social
Psychology, 86 (1), 19-42.
Campbell, W. Keith and Constantine Sedikides (1999), “Self-
Threat Magnifies the Self-Serving Bias: A Meta-Analytic
Integration,” Review of General Psychology, 3 (1), 23-43.
Catlin, Jesse R. and Yitong Wang (2012), “Recycling Gone Bad:
When the Option to Recycle Increases Resource
Consumption.” Journal of Consumer Psychology, 23 (1),
122-27.
Chan, Cindy, Jonah Berger, and Leaf Van Boven (2012),
“Identifiable But Not Identical: Combining Social Identity
and Uniqueness Motives in Choice,” Journal of Consumer
Research, 39, 561-73.
Cheryan, Sapna and Benoıˆt Monin (2005), “Where Are You
Really From?: Asian Americans and Identity Denial,”
Journal of Personality and Social Psychology 89 (5), 717-30.
Deaux, Kay (1991), “Social Identities: Thoughts on Structure and
Change,” in The Relational Self: Theoretical Convergences
in Psychoanalysis and Social Psychology, ed. Rebecca
Coleman Curtis, New York: Guilford Press, 77-93.
Deshpande
´, Rohit and Douglas M. Stayman (1994), “A Tale of
Two Cities: Distinctiveness Theory and Advertising
Effectiveness,” Journal of Marketing Research, 31 (1),
57-64.
Druckman, Daniel (1994), “Nationalism, Patriotism, and Group
Loyalty: A Social Psychological Perspective,” Mershon
International Studies Review, 38: 43-68.
Ebreo, Angela and Joanne Vining (2001), “How Similar Are
Recycling and Waste Reduction? Future Orientation and
Reasons for Reducing Waste as Predictors of Self-Reported
Behavior,” Environment and Behavior, 33 (3), 424-48.
Environmental Protection Agency (EPA) (2012), “Municipal
Solid Waste Generation (MSW) in the United States: Facts
and Figures for 2010.” http://www.epa.gov/waste/nonhaz/mu
nicipal/pubs/2012_msw_fs.pdf.
Escalas, Jennifer Edson and James R. Bettman (2005), “Self-
Construal, Reference Groups, and Brand Meaning,” Journal
of Consumer Research, 32 (3), 378-89.
Ethier, Kathleen A. and Kay Deaux (1994), “Negotiating Social
Identity When Contexts Change: Maintaining Identification
and Responding to Threat,” Journal of Personality and
Social Psychology, 67 (2), 243-51.
Ewing, Gordon (2001), “Altruistic, Egoistic, and Normative
Effects on Curbside Recycling,” Environment and Behavior,
33 (6), 733-64.
Ferraro, Rosellina, James R. Bettman, and Tanya L. Chartrand
(2009), “The Power of Strangers: The Effect of Incidental
262 JOURNAL OF CONSUMER RESEARCH

Consumer Brand Encounters on Brand Choice,” Journal of
Consumer Research, 35 (5), 729-41.
Ferraro, Rosellina, Jennifer E. Escalas, and James R. Bettman
(2011), “Our Possessions, Our Selves: Domains of Self-
Worth and the Possession-Self Link,” Journal of Consumer
Psychology, 21 (2), 169-77.
Forehand, Mark R. and Rohit Deshpande´ (2001), “What We See
Makes Us Who We Are: Priming Ethnic Self-Awareness and
Advertising Response,” Journal of Marketing Research,38
(3), 336-48.
Fournier, Susan (1998), “Consumers and their Brands:
Developing Relationship Theory in Consumer Research,”
Journal of Consumer Research, 24 (4), 343-53.
Fullerton, Don and Thomas C. Kinnaman (1996), “Household
Responses to Pricing Garbage by the Bag,” American
Economic Review, 86 (4): 971-84.
Gallup (2015), “Environment Survey,” http://www.gallup.com/
poll/1615/environment.aspx#.
Goldstein, Noah J., Robert B. Cialdini, and Vladas Griskevicius
(2008), “A Room with a Viewpoint: Using Social Norms to
Motivate Environmental Conservation in Hotels,” Journal of
Consumer Research, 35 (3), 472-82.
Guendelman, Maya D., Sapna Cheryan, and Benoıˆt Monin (2011),
“Fitting in But Getting Fat: Identity Threat and Dietary
Choices Among US Immigrant Groups,” Psychological
Science, 22 (7), 959-67.
Halvorsen, Bente (2008), “Effects of Norms and Opportunity Cost
of Time on Household Recycling,” Land Economics, 84 (3),
501-16.
Hanson, James W. (1980), “A Proposed Paradigm for Consumer
Product Disposition Processes,” Journal of Consumer
Affairs, 14 (1), 49-67.
Hayes, Andrew F. (2013), Introduction to Mediation, Moderation,
and Conditional Process Analysis: A Regression-Based
Approach, New York: Guilford Press.
Hayes, Andrew F. and Jo¨ rg Matthes (2009), “Computational
Procedures for Probing Interactions in OLS and Logistic
Regression: SPSS and SAS Implementations,” Behavior
Research Methods, 41 (3), 924-36.
Hoornweg, Daniel and Perinaz Bhada-Tat (2012), “What a Waste:
A Global Review of Solid Waste Management,” http://www.
openknowledge.worldbank.org/handle/10986/17388.
Jenkins, Robin R., Salvador A. Martinez, Karen Palmer, and
Michael J. Podolsky (2003), “The Determinants of
Household Recycling: A Material-Specific Analysis of
Recycling Program Features and Unit Pricing,” Journal of
Environmental Economics and Management, 45 (2),
294-318.
Johnson, Palmer O. and Jerzy Neyman (1936), “Tests of Certain
Linear Hypotheses and Their Application to Some
Educational Problems,” Statistical Research Memoirs,1,
57-93.
Jones, John T., Brett W. Pelham, Matthew C. Mirenberg, and John
J. Hetts (2002), “Name Letter Preferences Are Not Merely
Mere Exposure: Implicit Egotism as Self-Regulation,”
Journal of Experimental Social Psychology, 38 (2), 170-77.
Kettle, Keri L. and Gerald H€
aubl (2011), “The Signature Effect:
Signing Influences Consumption-Related Behavior by
Priming Self-Identity,” Journal of Consumer Research,38
(3), 474-89.
Kidwell, Blair, Adam Farmer, and David M. Hardesty (2013),
“Getting Liberals and Conservatives to Go Green: Political
Ideology and Congruent Appeals,” Journal of Consumer
Research, 40 (2), 350-67.
Lantz, Garold and Sandra Loeb (1998), “An Examination of the
Community Identity and Purchase Preferences Using the
Social Identity Approach,” in NA - Advances in Consumer
Research, ed. Joseph W. Alba and J. Wesley Hutchinson,
Provo, UT: Association for Consumer Research, 486-91.
Laverie, Debra A. and Dennis B. Arnett (2000), “Factors
Affecting Fan Attendance: The Influence of Identity Salience
and Satisfaction,” Journal of Leisure Research, 32 (2),
225-46.
Lee, Kyoungmi, Hakkyun Kim, and Kathleen D. Vohs (2011),
“Stereotype Threat in the Marketplace: Consumer Anxiety
and Purchase Intentions.” Journal of Consumer Research,38
(2), 343-357.
Lord, Kenneth R. (1994), “Motivating Recycling Behavior: A
Quasi-Experimental Investigation of Message and Source
Strategies,” Psychology & Marketing, 11 (4), 341-58.
Luhtanen, Riia and Jennifer Crocker (1992), “A Collective Self-
Esteem Scale: Self-Evaluation of One’s Social Identity,”
Personality and Social Psychology Bulletin, 18 (3), 302-18.
Mannetti, Lucia, Antonio Pierro, and Stefano Livi (2004),
“Recycling: Planned and Self-Expressive Behavior,” Journal
of Environmental Psychology, 24 (2), 227-36.
McCarty, John A. and L. J. Shrum (2001), “The Influence of
Individualism, Collectivism, and Locus of Control on
Environmental Beliefs and Behavior,” Journal of Public
Policy & Marketing, 20 (1), 93-104.
Packard, Grant M. and David B. Wooten (2013), “Compensatory
Knowledge Signaling in Consumer Word-of-Mouth.”
Journal of Consumer Psychology, 23 (4), 434-50.
Paharia, Neeru, Anat Keinan, Jill Avery, and Juliet B. Schor
(2011), “The Underdog Effect: The Marketing of
Disadvantage and Determination Through Brand
Biography,” Journal of Consumer Research, 37 (5), 775-90.
Park, C. Whan, Deborah J. MacInnis, Joseph Priester, Andreas B.
Eisingerich, and Dawn Iacobucci (2010), “Brand Attachment
and Brand Attitude Strength: Conceptual and Empirical
Differentiation of Two Critical Brand Equity Drivers,”
Journal of Marketing, 74 (6), 1-17.
Park, Lora E. and Jon K. Maner (2009), “Does Self-Threat
Promote Social Connection? The Role of Self-Esteem and
Contingencies of Self-Worth,” Journal of Personality and
Social Psychology, 96 (1), 203-17.
Pelham, Brett W., Matthew C. Mirenberg, and John T. Jones
(2002), “Why Susie Sells Seashells by the Seashore: Implicit
Egotism and Major Life Decisions,” Journal of Personality
and Social Psychology, 82 (4), 469-87.
Price, Linda L., Eric J. Arnould, and Carolyn Folkman Curasi
(2000), “Older Consumers’ Disposition of Special
Possessions,” Journal of Consumer Research, 27 (2),
179-201.
Reed, Americus II (2004), “Activating the Self-Importance of
Consumer Selves: Exploring Identity Salience Effects on
Judgments,” Journal of Consumer Research, 31 (2), 286-95.
Reschovsky, James D. and Sarah E. Stone (1994), “Market
Incentives to Encourage Household Waste Recycling: Paying
for What You Throw Away,” Journal of Policy Analysis and
Management, 13 (1), 120-39.
Reynolds, Travis William, Ann Bostrom, Daniel Read, and M.
Granger Morgan (2010), “Now What Do People Know
About Global Climate Change? Survey Studies of Educated
Laypeople,” Risk Analysis, 30 (10), 1520-38.
TRUDEL, ARGO, AND MENG 263

Schouten, John W. (1991), “Selves in Transition: Symbolic
Consumption in Personal Rites of Passage and Identity
Reconstruction.” Journal of Consumer Research, 17 (4),
412-25.
Schultz, P. Wesley and Stuart Oskamp (1996), “Effort as a
Moderator of the Attitude-Behavior Relationship: General
Environmental Concern and Recycling,” Social Psychology
Quarterly, 59 (4), 375-83.
Smith, Stephen M. and Richard E. Petty (1996), “Message
Framing and Persuasion: A Message Processing Analysis,”
Personality and Social Psychology Bulletin, 22 (3), 257-68.
Sparks, Paul and Richard Shepherd (1992), “Self-Identity and the
Theory of Planned Behavior: Assessing the Role of
Identification with ‘Green Consumerism,’” Social
Psychology Quarterly, 55 (4), 388-99.
Spencer, Steven J., Claude M. Steele, and Diane M. Quinn (1999),
“Stereotype Threat and Women’s Math
Performance,” Journal of Experimental Social
Psychology, 35 (1), 4-28.
Sprott, David, Sandor Czellar, and Eric Spangenberg (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, Claude M. (1988), “The Psychology of Self-Affirmation:
Sustaining the Integrity of the Self,” in Advances in
Experimental Social Psychology, ed. Leonard Berkowitz,
San Diego, CA: Academic Press, 261-302.
Steele, Claude M. and Joshua Aronson (1995), “Stereotype Threat
and the Intellectual Test Performance of African
Americans,” Journal of Personality and Social Psychology,
69 (5), 797-811.
Stru¨ mpfer, D. J. W. (1978), “Relationship Between Attitudes
Toward One’s Names and Self-Esteem,” Psychological
Reports, 43 (3), 699-702.
Tajfel, Henri (1981), Human Groups and Social Categories:
Studies in Social Psychology, Cambridge: Cambridge
University Press
——— (1982), “Social Psychology of Intergroup Relations,”
Annual Review of Psychology, 33 (1), 1-39.
Tajfel, Henri and John C. Turner (1979), “An Integrative Theory
of Intergroup Conflict,” in The Social Psychology of
Intergroup Relations, ed. William G. Austin and Stephen
Worchel, Pacific Grove, CA: Brooks/Cole, 33-48.
Taylor, Shirley and Peter A. Todd (1995), “An Integrated Model
of Waste Management Behavior—a Test of Household
Recycling and Composting Intentions,” Environment and
Behavior, 27 (5), 603-30.
Terry, Deborah J., Michael A. Hogg, and Katherine M. White
(1999), “The Theory of Planned Behaviour: Self-Identity,
Social Identity and Group Norms,” British Journal of Social
Psychology, 38 (3), 225-44.
Thomson, Matthew, Deborah J. MacInnis, and C. Whan Park
(2005), “The Ties That Bind: Measuring the Strength of
Consumers’ Emotional Attachments to Brands,” Journal of
Consumer Psychology, 15 (1), 77-91.
Trudel, Remi and Jennifer J. Argo (2013), “The Effect of Product
Size and Form Distortion on Consumer Recycling Behavior,”
Journal of Consumer Research, 40 (4), 632-43.
Trudel, Remi, Jennifer J. Argo, and Matthew D. Meng (2015),
“Trash or Recycle? How Product Distortion Leads to
Categorization Error During Disposal,” Environment and
Behavior: 0013916515577635.
Turner, John C. (1987), “The Analysis of Social Influence,” in
Rediscovering the Social Group: A Self-Categorization
Theory, ed. John C. Turner, Michael A. Hogg, Penelope J.
Oakes, Stephen D. Reicher, and Margaret S. Wetherell,
Milton Keynes, UK: Open University Press, 68-88.
Vining, Joanne and Angela Ebreo (1992), “Predicting Recycling
Behavior from Global and Specific Environmental Attitudes
and Changes in Recycling Opportunities,” Journal of Applied
Social Psychology, 22 (20), 1580-1607.
White, Katherine and Jennifer J. Argo (2009), “Social Identity
Threat and Consumer Preferences,” Journal of Consumer
Psychology, 19 (3), 313-25.
White, Katherine M., Jennifer J. Argo, and Jaideep Sengupta
(2012),”Dissociative Versus Associative Responses to Social
Identity Threat: The Role of Consumer Self-Construal,”
Journal of Consumer Research, 39 (4), 704-19.
White, Katherine M. and Darren W. Dahl (2007), “Are All Out-
Groups Created Equal? Consumer Identity and Dissociative
Influence,” Journal of Consumer Research, 34 (4), 525-36.
White, Katherine M., Rhiannon MacDonnell, and Darren W. Dahl
(2011), “It’s the Mind-Set That Matters: The Role of
Construal Level and Message Framing in Influencing
Consumer Efficacy and Conservation Behaviors,” Journal of
Marketing Research, 48 (3), 471-85.
Winterich, Karen Page, Vikas Mittal, and William T. Ross (2009),
“Donation Behavior Toward In-Groups and Out-Groups: The
Role of Gender and Moral Identity,” Journal of Consumer
Research, 36 (2), 199-214.
Zhao, Xinshu, John G. Lynch Jr., and Qimei Chen (2010),
“Reconsidering Baron and Kenny: Myths and Truths about
Mediation Analysis,” Journal of Consumer Research, 37 (2),
197-206.
264 JOURNAL OF CONSUMER RESEARCH