Do People Embrace Praise Even When They Feel Unworthy? A Review of Critical Tests of Self-Enhancement Versus Self-Verification

Abstract

Some contemporary theorists contend that the desire for self-enhancement is prepotent and more powerful than rival motives such as self-verification. If so, then even people with negative self-views will embrace positive evaluations. The authors tested this proposition by conducting a meta-analytic review of the relevant literature. The data provided ample evidence of self-enhancement strivings but little evidence of its prepotency. Instead, the evidence suggested that both motives are influential but control different response classes. In addition, other motives may sometimes come into play. For example, when rejection risk is high, people seem to abandon self-verification strivings, apparently in an effort to gratify their desire for communion. However, when rejection risk is low, as is the case in many secure marital relationships, people prefer self-verifying evaluations. The authors conclude that future researchers should broaden the bandwidth of their explanatory frameworks to include motives other than self-enhancement.

Full text

http://psr.sagepub.com/
Review
Personality and Social Psychology
http://psr.sagepub.com/content/14/3/263
The online version of this article can be found at:
DOI: 10.1177/1088868310365876
2010 14: 263 originally published online 30 April 2010Pers Soc Psychol Rev
Tracy Kwang and William B. Swann, Jr
Self-Enhancement Versus Self-Verification
Do People Embrace Praise Even When They Feel Unworthy? A Review of Critical Tests of
Published by:
http://www.sagepublications.com
On behalf of:
Society for Personality and Social Psychology
can be found at:Personality and Social Psychology ReviewAdditional services and information for
http://psr.sagepub.com/cgi/alertsEmail Alerts:
http://psr.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://psr.sagepub.com/content/14/3/263.refs.htmlCitations:
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Personality and Social Psychology Review
14(3) 263 –280
© 2010 by the Society for Personality
and Social Psychology, Inc.
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1088868310365876
http://pspr.sagepub.com
Do People Embrace Praise Even
When They Feel Unworthy? A Review
of Critical Tests of Self-Enhancement
Versus Self-Verification
Tracy Kwang1 and William B. Swann, Jr.1
Abstract
Some contemporary theorists contend that the desire for self-enhancement is prepotent and more powerful than rival
motives such as self-verification. If so, then even people with negative self-views will embrace positive evaluations. The authors
tested this proposition by conducting a meta-analytic review of the relevant literature. The data provided ample evidence
of self-enhancement strivings but little evidence of its prepotency. Instead, the evidence suggested that both motives are
influential but control different response classes. In addition, other motives may sometimes come into play. For example,
when rejection risk is high, people seem to abandon self-verification strivings, apparently in an effort to gratify their desire
for communion. However, when rejection risk is low, as is the case in many secure marital relationships, people prefer self-
verifying evaluations. The authors conclude that future researchers should broaden the bandwidth of their explanatory
frameworks to include motives other than self-enhancement.
Keywords
self-identity, motivation, goals, self-presentation, attribution, individual differences, interpersonal processes, self-verification,
self-enhancement
People love to be admired and praised. Indeed, the notion
that people prefer positive, “self-enhancing” evaluations is
among the most prominent motivational assumptions in
Western Psychology (e.g., Jones, 1973; Leary, 2007). Never-
theless, people’s allegedly voracious appetite for adulation
may be qualified by a competing desire for “self-verifying”
evaluations—that is, evaluations that confirm their enduring
self-views (e.g., Lecky, 1945; Swann, 1983). In this article,
we use meta-analytic techniques to evaluate the relative
strength of these competing motives for self-enhancement
and self-verification. We begin with a discussion of the older
of the two theories, self-enhancement.
Self-Enhancement Theory
The seeds of self-enhancement theory were sewn more than
seven decades ago when Gordon Allport (1937) asserted that
there exists a vital human need to view oneself positively. In
this tradition, modern self-enhancement theorists assert that
people desire to increase the positivity—or reduce the
negativity—of their self-views (see Leary, 2007, for a review).
The proposal that there exists a pervasive desire for positiv-
ity has inspired dozens of studies. At this juncture, people are
thought to engage in a host of self-serving biases that pre-
sumably enable them to maintain positive conceptions of
themselves. One of the most prevalent of such biases is the
tendency to attribute positive outcomes to the self and nega-
tive outcomes to external circumstances (e.g., Blaine &
Crocker, 1993; Fitch, 1970). Similarly, people routinely claim
to be better off than the average person, by, for example, pre-
dicting better futures for themselves than for the average
person (Taylor & Brown, 1988) and even asserting that they
are less susceptible to bias than are others (Pronin, Gilovich
& Ross, 2004). When people receive feedback, they selec-
tively attend to information that preserves their self-esteem
(Ditto & Lopez, 1993) and report feeling better after receiv-
ing positive as compared to negative feedback (e.g., Korman,
1968; Swann, Griffin, Predmore, & Gaines, 1987). Finally,
self-enhancement biases are not “all in the head,” in that they
are believed to motivate people to strategically present them-
selves in a flattering light (Baumeister, 1982).
In addition to inspiring dozens of empirical investiga-
tions, the notion that people are fundamentally motivated to
acquire positive evaluations has developed into one of social
1University of Texas at Austin
Corresponding Author:
William B. Swann, Jr., University of Texas at Austin, Department of
Psychology, Austin, TX 78712
Email: swann@mail.utexas.edu
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

264 Personality and Social Psychology Review 14(3)
psychology’s most influential theoretical assumptions (e.g.,
Hoyle, Kernis, Leary, & Baldwin, 1999; Leary, 2007). In fact,
despite some early literature reviews indicating that self-
enhancement strivings influenced affective but not cognitive
reactions (e.g., Shrauger, 1975), over the past few decades it
has arguably become one of social psychology’s most widely
accepted theory. Today self-enhancement themes can be
found in most of the field’s most influential theories,
including terror management (Greenberg, Pyszczynski, &
Solomon, 1986), self-evaluation maintenance (Tesser, 1988),
positive illusions (Murray, Holmes, & Griffin, 1996; Taylor
& Brown, 1988), self-affirmation (Steele, 1988), and contin-
gencies of self-worth (e.g., Crocker & Wolfe, 2001).
Although the assumption that people value and prefer pos-
itive evaluations is the core proposition underlying most
variations of self-enhancement theory, some contemporary
proponents of the theory have taken the argument further.
Impressed by wide-ranging evidence of self-enhancing biases
in human information processing and behavior, several
authors have asserted that the desire for self-enhancement
overrides the desire for accurate self-knowledge (Copleston,
1957; Jahoda, 1958; Rogers, 1951; for a review, see Colvin &
Griffo, 2007). The authors of one landmark article, for exam-
ple, defined self-enhancement as a tendency to entertain
unrealistically positive self-evaluations that reflect a “general,
enduring pattern of error” (Taylor & Brown, 1988, p. 194).
Others have recently added that the self-enhancement motive
is both prepotent and universal, a “cornerstone” of psycho-
logical activity (Sedikides & Gregg, 2008).
To be sure, some authors have dissented from the notion
that self-enhancement strivings are completely unbridled
(e.g., Baumeister, 1989), including even some of the original
advocates of the notion that they are pervasive (e.g., Taylor &
Gollwitzer, 1995). Yet for the most part researchers interested
in self-enhancement have sought to collect additional support
for the motive with little attention to countervailing motiva-
tional forces. This focus, in combination with recent claims
for the prepotency of the self-enhancement motive, implies
that there exists a fundamental imbalance in human priorities
rather than a delicately balanced system of motivational
checks and balances (Sedikides & Gregg, 2008). In particu-
lar, it would appear that the desire for positivity is so powerful
that it overrides competing accuracy-related motives such as
self-verification (Swann, 1983) and self-assessment (Trope,
1983). To test this provocative claim, we conducted a meta-
analysis of research designed to compare the relative strength
of self-enhancement strivings with one of its historic com-
petitors, self-verification.1 To set the stage for this analysis,
we briefly characterize this competing motive.
The Desire for Self-Verification
Self-verification theory (e.g., Swann, 1983) assumes that
people have a powerful desire to confirm and stabilize their
firmly held self-views. This desire for stable self-views can
be understood by considering how and why people develop
self-views in the first place. Theorists have long assumed
that people form their self-views by observing how others
treat them (e.g., Cooley, 1902; Mead, 1934). As they acquire
more and more evidence to support their self-views, people
become increasingly certain of these views. Once firmly
held, self-views enable people to make predictions about
their worlds, guide behavior, and maintain a sense of conti-
nuity, place, and coherence. In this way, stable self-views not
only serve the pragmatic function of guiding behavior but
also serve the epistemic function of affirming people’s sense
that things are as they should be. Indeed, firmly held self-
views develop into the centerpiece of their knowledge
systems. As such, it is not surprising that by mid-childhood
children begin to display a preference for evaluations that
confirm and stabilize their self-views (e.g., Cassidy, Ziv,
Mehta, & Feeney, 2003).
Self-verification theory’s most provocative prediction is
that people should prefer self-confirming evaluations even
if the self-view in question is negative. For example, contrary
to self-enhancement theory, self-verification theory predicts
that those who see themselves as disorganized or unintelli-
gent should prefer evidence that others also perceive them as
such. Support for the theory has come from studies that have
examined the relationship of people’s enduring self-views to
their choice of feedback and interaction partners, attention,
overt behavior, recall, and relationship quality (for a review,
see Swann, in press).
Yet the inherently social nature of the self-verification
process points to the existence of at least one critically
important boundary condition of the effect. That is, if people
are to receive a steady supply of self-verifying feedback,
they must maintain the “vehicles” for the delivery of such
feedback—their ongoing relationships. Simply put, no rela-
tionship, no self-verification. This means that when people
want the relationship to survive, feedback may be eschewed
not only when it threatens the desire for self-verification,
but also when it threatens the future of the relationship.
Hence, overly positive evaluators will be avoided because
they might eventually be disappointed and leave; overly
negative evaluators will be avoided because their negativity
calls the very existence of the relationship into question.
Wariness of negative evaluators may be magnified insofar
as the relationship is provisional or uncommitted, for termi-
nating such relationships is far easier than ending
relationships that involve significant long-term commit-
ment. The general principle, then, is that people will seek
self-verification only insofar as doing so does not put them
at risk of being abandoned, for abandonment would frus-
trate their communion motive (e.g., Baumeister & Leary,
1995; Heine, Proulx, & Vohs, 2006; Wiggins & Broughton,
1985) and sever their supply of self-verification (see
also Hardin & Higgins’s, 1996, discussion of people’s
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 265
unwillingness to embrace epistemic truth if it undermines
the relationship aspect of shared realities).
This reasoning suggests that as long as self-verifying nega-
tive feedback does not portend rejection and relationship
termination, people with negative self-views will prefer nega-
tive evaluations. Similarly, people with positive self-views
will display a corresponding preference for positive evalua-
tions. In contrast, self-enhancement theory predicts that all
people will prefer positive evaluations, regardless of the posi-
tivity of their self-views. At this juncture, these competing
predictions have been tested in a large number of studies (for
an early review, see Shrauger, 1975). Despite this, some of the
most prominent reviewers of the literature (e.g., Baumeister,
1998; Leary, 2007) have referenced a single article by
Sedikides (1993) as offering definitive evidence for the pre-
potency of the self-enhancement motive. To burden a single
article with the resolution of such a complex controversy is
hazardous in itself, but the choice of this particular article is
especially problematic. Witness that the author himself
acknowledged that five of his six studies are irrelevant to self-
verification theory because he failed to measure chronic
self-views and “an adequate testing of [the self-verification]
perspective would require that subjects’ preexisting (both
positive and negative) self-conceptions be at stake during the
self-evaluation process” (p. 329). In the only study that did
include a measure of self-views, the self-enhancement effect
(r = .50) was no larger than the self-verification effect (r =
.46), rendering the study absolutely inconclusive with respect
to the prepotency of self-enhancement.
Of course, even if the Sedikides (1993) article did inform
the debate between advocates of self-enhancement and self-
verification theory, there is a larger point here: In light of the
existence of numerous relevant studies, the most appropriate
means of testing the relative merits of self-enhancement
versus self-verification approaches is to review all available
studies that meet the design criteria specified by the two the-
ories. In the next section, we report such a meta-analysis.
Evidence that self-enhancement strivings are significantly
stronger than self-verification strivings will buttress recent
claims that self-enhancement is the prepotent social psycho-
logical motive (e.g., Sedikides & Gregg, 2008). Evidence
that self-verification strivings are equal to, or stronger than,
self-enhancement strivings will point to the existence of a
more balanced and variegated motive system.
Critical Tests of Self-Enhancement
Versus Self-Verification
Both self-enhancement and self-verification theories make
similar predictions for people with positive self-views. That
is, both theories predict that people with positive self-views
will embrace positive evaluations because, for such indi-
viduals, positive evaluations are both self-enhancing and
self-verifying. The two theories make competing predictions
for people with negative self-views, however. Self-enhancement
theory predicts that people with negative self-views will
prefer positive over negative evaluations. Statistically, self-
enhancement will be reflected by a main effect of the
evaluation factor in an analysis of variance (ANOVA) or
regression. The effect size of the main effect will be reported
regardless of whether there was an interaction effect. In con-
trast, self-verification theory assumes that the match between
the evaluation and the self-view is crucial (Swann, Chang-
Schneider, & McClarty, 2007). For this reason, people with
positive self-views should prefer positive over negative eval-
uations and people with negative self-views should prefer
negative over positive evaluations. Statistically, this will be
reflected in an interaction between self-view and evaluation
in an ANOVA or regression.2
The most straightforward form of support for self-
verification theory would be for people with positive
self-views to prefer positive evaluations and people with
negative self-views to prefer negative evaluations. Neverthe-
less, because self-verification theory holds that it is the match
between the self-view and evaluation that is crucial, a given
evaluation could be non-matching even though it has the
same valence as the self-view. Support for this possibility
comes from a study in which people with positive self-views
were less intimate with their spouses not only when the
spouses’ appraisals were negative but also when their apprais-
als were extremely positive (Swann, De La Ronde, & Hixon,
1994). For this reason, from the perspective of self-verifica-
tion theory, the crucial issue is whether there is an interaction
between self-views and evaluations, such that the preference
for positivity (or aversion to negativity) is stronger among
people with positive as compared to negative self-views.
Search Procedure and Inclusion Criteria. We searched the
references sections of published review articles on self-
enhancement and self-verification (e.g., Blaine & Crocker,
1993; Shrauger, 1975; Swann, 1990) for critical tests of the
two self-motives. In addition, we searched for relevant arti-
cles using Academic Search Premier, Medline, PsycINFO,
PsycARTICLES, and Sociological Collection using key-
words such as self-verification, self-enhancement, self-esteem,
feedback, attribution, cognitions, reactions, behavior, emo-
tion, and affect. Finally, we e-mailed appropriate listservs and
contacted prominent researchers to request relevant articles
that we might have been missed.
The first author read the abstracts of all of the articles. If
the abstract was at all promising, she read the article itself to
determine if the authors provided the information required to
compute effect sizes for self-enhancement and self-verification.
This led to the deletion of two types of studies (specific
citations are provided below). Some articles were deleted
because the authors discussed self-verification but failed to
measure participants’ self-views—a requirement for com-
puting self-verification effects. Other studies were deleted
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

266 Personality and Social Psychology Review 14(3)
because the authors did not provide sufficient information to
calculate effect sizes for self-enhancement and self-verification
in their result sections and our efforts to obtain this informa-
tion were fruitless.
All of the studies that provided the information required to
compute effect sizes were listed in our tables and included in
an initial meta-analysis. Results were standardized to Pear-
son’s r effect size and corrected for sampling error, according
to the meta-analysis methods of Hunter and Schmidt (1990)
and Hunter, Schmidt, and Jackson (1982) and as outlined by
Lyons (2003) as well as Rosenthal and Rosnow (2008).
After conducting the initial meta-analysis, we excluded
some studies from the final meta-analysis because character-
istics of the design or methodology compromised their
capacity to provide a fair test of the two theories. For exam-
ple, some researchers included only participants with negative
self-views, rendering the design incomparable to those in
which researchers recruited participants with both positive
and negative self-views. Other researchers manipulated feed-
back that was disjunctive with the self-view, a feature that
made it impossible to say whether the feedback was verifying
or non-verifying. For instance, some researchers assessed the
relationship between global self-esteem and a specific self-
view (e.g., athletic ability). In such instances, the global and
specific self-views are mismatched, thus violating the speci-
ficity matching principle. Without a clear match between the
self-view and feedback, self-verification theory does not
make clear predictions (see Swann et al.’s, 2007, discussion
of the specificity matching principle). Still other researchers
manipulated or measured variables that were not comparable
to those examined in studies included in the meta-analysis.
For example, one investigator manipulated mood state rather
than giving participants positive versus negative feedback;
others measured the extent to which participants believed
feedback would foster self-improvement or predict future
success. Finally, two additional studies were excluded from
the final meta-analysis because of statistical irregularities.
Specifically, the researchers partialed out the effects of par-
ticipants’ performance expectations while assessing the
effects of their self-views, a procedure that almost surely
diminishes the contribution of self-views to the outcome.
To ensure that we were comparing apples to apples, we
organized our meta-analyses into sections based on the
dependent measures used by the researchers. To determine
straightaway if we replicated the results of the most expan-
sive previous review of this literature (Shrauger, 1975), we
began with cognitive and affective responses to feedback.
We followed with three new categories of dependent vari-
ables, namely, behavioral reactions, feedback seeking, and
relationship quality.
Cognitive Processes. These processes refer to the extent to
which people perceive the feedback they receive to be accu-
rate, diagnostic, attributable to themselves, and delivered by a
competent evaluator. Operationally, self-enhancement was
defined as a tendency for participants to devote more atten-
tion to positive evaluations and perceive them to be more
accurate, diagnostic, and so on than negative evaluations.
Self-enhancement was also defined as the tendency to attri-
bute positive events or outcomes to personal, stable, and
global qualities, whereas negative events or outcomes are
attributed to situational, unstable, and specific qualities. In
contrast, self-verification was defined by the tendency for
participants with positive self-views to devote more attention
and impute more accuracy to positive feedback and evalua-
tors whereas people with negative self-views displayed the
opposite pattern. Self-verification was also defined by the
tendency for those with positive self-views to attribute posi-
tive events as internal, stable, and global and negative events
as external, instable, and specific, whereas those with nega-
tive self-views displayed the opposite pattern.
Twenty-six studies were not included in the tables because
there were insufficient data to calculate r (e.g., Bell & Arthur,
2008; Crary, 1966; Sedikides & Green, 2004). This left forty-
five studies in Table 1. Seven studies did not meet our further
inclusion criteria. Anseel and Lievens (2006) measured utility
of feedback for improving the self rather than the accuracy or
diagnosticity of the feedback with respect to the actual self.
Similarly, McFarlin and Blascovich (1981) violated the spec-
ificity matching principle in that the measure of self-view
was social skill and the outcome measure was performance
on a spatial task. Also, Bellavia and Murray (2003) manipu-
lated the mood state of the participants rather than giving
them valenced feedback. Wood, Heimpel, Newby-Clark, and
Ross (2005) was eliminated because participants were not
given negative feedback. In two studies (Moreland & Sweeney,
1984; Shrauger & Osberg, 1980), the authors partialed out the
effects of performance expectations from self-views. Because
self-views presumably exert their influence by shaping
expectations, covarying out expectations may neutralize the
effects of self-views, a possibility that was supported by the
fact that expectations significantly influenced responses.
Finally, we eliminated Study 2 of Rudich and Vallacher
(1999) because the authors confounded negative feedback
with rejection, which disqualifies it as a test of self-verification
because rejection cuts off one’s supply of self-verification.
When the foregoing studies were eliminated, thirty-eight
studies remained.3 We further categorized the studies by spe-
cific dependent variables: accuracy (i.e., how accurate or
valid is the feedback), attention (i.e., how much time is spent
scrutinizing the feedback), attribution (i.e., are successes or
failures because of internal or external causes), over-claiming
bias (i.e., claiming more positive evaluations than objective
measures suggest), predictive ability (i.e., does the feedback
inform future results), and recall accuracy (i.e., remembering
the results as being more positive or negative than they were).
Self-enhancement and cognitive processes. As shown in
Table 1, the average effect size for the self-enhancement
effect in the entire sample was r = .19. Broken down by spe-
cific dependent variables, the average effect size for accuracy
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 267
Table 1. Accumulated Effect Sizes, Corrected for Sampling Error for Cognitive Processes
Enhancement Verification
Study N Dependent variable effect size effect size
Anseel and Lievens 2006a† 389 Accuracy .16 -.08
Bosson and Swann 1999 (Study 1)b 74 Accuracy .30 .33
Campbell, Lackenbauer, and Muise 2006c 103 Accuracy .00 .28
De La Ronde and Swann, 1998c 61 Accuracy .43 .35
Giesler, Josephs, and Swann, 1996cd 73 Accuracy .23 .76
Moreland and Sweeney, 1984e† 166 Accuracy .31 -.06
Quinlivan and Leary 2005 81 Accuracy .26 .30
Robinson and Smith-Lovin 1992 (Study 1)c 75 Accuracy .26 .41
Rudich and Vallacher 1999 (Study 1) 47 Accuracy .32 .22
Rudich and Vallacher 1999 (Study 2)† 50 Accuracy .34 .42
Shrauger and Kelly 1988 (Study 1)bcf 31 Accuracy .14 .43
Shrauger and Kelly 1988 (Study 2)bc 39 Accuracy .56 .37
Shrauger and Lund 1975c 48 Accuracy .40 .33
Shrauger and Rosenberg 1970 36 Accuracy .29 .48
Stake 1982 (Study 1) 236 Accuracy .44 .11
Swann, Griffin, Predmore, and Gaines 1987c 98 Accuracy .30 .28
Swann and Read 1981a (Study 3) 74 Accuracy .00 .84
Woo and Mix 1997c 72 Accuracy .59 .10
Wood, Heimpel, Newby-Clark, and Ross 2005 (Study 2)cg† 79 Accuracy .11 .20
Swann and Read 1981b (Study 1) 64 Attention .00 .28
Bellavia and Murray 2003† 81 Attribution .35 .35
Brown, Cai, Oakes, and Deng 2009 (Study 1)d 91 Attribution .32 .12
Burke 1978 90 Attribution .17 .29
Campbell, Chew, and Scratchley 1991 (Study 1) 67 Attribution .36 .26
Chandler, Lee, and Pengilly 1997 254 Attribution .29 .45
Feather 1969 167 Attribution .00 .00
Feather and Simon 1971h 85 Attribution .19 .28
Feather and Simon 1973h 265 Attribution .10 .11
Fielstein et al. 1985 201 Attribution .00 .30
Fitch 1970 135 Attribution .25 .22
Gilmor and Minton 1974h 80 Attribution .22 .75
Girodo, Dotzenroth, and Stein 1981 78 Attribution .01 .35
Jussim, Yen, and Aiello 1995 172 Attribution .07 .31
McMahan 1973h 336 Attribution .24 .19
Piers 1977 297 Attribution .00 .30
Raps, Peterson, Reinhard, Abramson, and Seligman 1982 106 Attribution .17 .24
Rizley 1978 (Study 1) 38 Attribution .27 .23
Rizley 1978 (Study 2) 38 Attribution .20 .26
Shrauger and Osberg 1980† 60 Attribution .27 .00
Stroebe, Eagly, and Stroebe 1977 56 Attribution .00 .42
Tennen, Herzberger, and Nelson 1987 (Study 1) 55 Attribution .17 .48
Tennen et al. 1987 (Study 2) 23 Attribution .34 .70
Kwan, John, Kenny, Bond, and Robins 2004 128 Over-claiming bias -.11 .50
McFarlin and Blascovich 1981† 64 Predictive ability .00 .41
Story 1998 (Study 1) 22 Recall .45 .50
Total 4,863 Sample-weighted mean r .19 .25
Total of studies that do not partial out expectation 3,974 Sample-weighted mean r .18 .30
aThe necessary standardized data were not published in the article, but Anseel kindly provided them.
bF statistics were calculated from cell means and standard deviations according to the formulas outlined in Cohen (2002).
cThe dependent variable in the actual study was labeled differently in the original article. However, on closer examination, the question items did not differ
from items measuring accuracy.
dAnalyzed American sample only.
eWe averaged the effect sizes for accuracy and attribution. The accuracy effect size for self-enhancement is r = .22 and for self-verification is r = –.09. The
attribution effect size for self-enhancement is r = .39 and for self-verification is r = –.03. We separated these effect sizes when we analyzed the specific
dependent variables.
fWe did not include effect sizes for recall because almost all participants correctly recognized all the feedback they received.
gThe necessary data were not published in the article, but Wood kindly provided them.
hMeasured confidence as self-view.
†Study does not fit inclusion criteria.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

268 Personality and Social Psychology Review 14(3)
was r = .27, attention was r = .00, attribution was r = .15,
over-claiming bias was r = –.11, predictive ability was r =
.00, and recall accuracy was r = .45.
When we eliminated the studies that did not fit the inclu-
sion criteria, the average effect size for the self-enhancement
effect was r = .18. Broken down by specific dependent vari-
ables, the average effect size for accuracy was r = .31,
attention was r = .00, attribution was r = .14, over-claiming
bias was r = –.11, and recall was r = .45.
Self-verification and cognitive processes. The average effect
size for the self-verification effect in our preliminary sample
was r = .25. Broken down by specific dependent variables, the
average effect size for accuracy was r = .20, attention was r =
.28, attribution was r = .27, over-claiming bias was r = .50,
predictive ability was r = .41, and recall accuracy was r = .50.
When we eliminated the studies that did not fit the inclu-
sion criteria, the average effect size for the self-verification
effect was r = .30. Broken down by specific dependent vari-
ables, the average effect size for accuracy was r = .33,
attention was r = .28, attribution was r = .27, over-claiming
bias was r = .50, and recall was r = .50.
In sum, there were significant enhancement and verifica-
tion effects on cognitive processes. To determine the relative
strength of the self-enhancement versus self-verification
effects, we found the mean difference between the effect
sizes (M = –.067, SD = .27) based on a random effects model
and calculated the confidence interval for the weighted mean
differences. Overall, self-verification effects were greater
than self-enhancement effects, t(4884) = –17.42, p < .001,
CI.95 = –.0735, –.0587, rs = .30 versus .18, respectively.
When broken down by specific dependent variables, self-
enhancement effects were greater than self-verification
effects for accuracy, t(1831) = 8.945, p < .001, CI.95 = .0498,
.0778, whereas self-verification effects were greater than
self-enhancement effects for attribution, t(1774) = –33.89,
p < .001, CI.95 = –.1206, –.1074. There were not enough
studies to test mean differences for the other dependent vari-
ables. Therefore, consistent with Shrauger (1975), overall,
the self-verification effects were stronger than the self-
enhancement effects for studies of cognitive processes.
Affective Responses. These responses refer to emotional and
affective responses to feedback, such as hostility, anxiety, dys-
phoria, liking, and positive and negative mood. Operationally,
self-enhancement was defined as the tendency for participants
to display or report more positive affect in response to positive
as compared to negative feedback. Self-verification was
defined as the tendency for people with positive self-views to
experience more positive or less negative affect when their
partners view them positively and people with negative self-
views to experience more positive or less negative affect when
their partners view them negatively.
Not included in the tables were 4 studies because of insuf-
ficient data to calculate r (e.g., Stets & Asencio, 2008; Swann,
Wenzlaff, Krull, & Pelham, 1992), leaving 28 studies listed in
Table 2. Of the studies included in the tables, Study 4 of
Sprecher and Hatfield (1982) was not excluded from the final
meta-analysis because the inclusion of four feedback condi-
tions (totally favorable, ambiguous favorable, rejecting, and
totally negative) rendered it incomparable to the other studies
in our sample. We also eliminated Studies 1 and 2 of Wood et
al. (2005) because participants received positive but not nega-
tive feedback. This left the 25 studies in the final sample pool.
We further categorized the studies by specific dependent vari-
ables: affect (i.e., both positive and negative emotional
reactions), attraction (i.e., how attracted are you to the evalu-
ator), negative affect only, positive affect only, satisfaction
(i.e., how satisfied are you with the feedback), and task liking
(i.e., how enjoyable did you find the task).
Self-enhancement and affective responses. The average
effect size for the self-enhancement effect in our preliminary
sample was r = .26. Broken down by specific dependent
variables, the average effect size for general affect was r =
.31, attraction was r = .26, negative affect only was r = .12,
positive affect only was r = .39, satisfaction was r = .62, and
task liking was r = .16.
When we eliminated the studies that did not fit further
inclusion criteria, the average effect size for the self-
enhancement effect was r = .29. The average effect size for
attraction rose to r = .34. The average effect sizes for all
other dependent variables stayed the same.
Self-verification and affective responses. Inspection of Table 2
reveals that the average effect size for the self-verification
effect in our preliminary sample was r = .13. Broken down by
specific dependent variables, the average effect size for
general affect was r = .05, attraction was r = .28, negative
affect only was r = .10, positive affect only was r = –.03,
satisfaction was r = .03, and task liking was r = .09.
When we eliminated the studies that did not fit further
inclusion criteria, the average effect size for the self-verification
effect was r = .13. The average effect size for attraction rose
to r = .35 and negative affect dropped only to r = .09. The
average effect sizes for all other dependent variables stayed
the same.
In sum, although there were some self-verification effects
for measures of affect, the self-enhancement effects seemed
considerably stronger. To determine the relative strength of
the self-enhancement versus self-verification effects, we
found the mean difference between the effect sizes for
enhancement and verification effects (M = .152, SD = .25)
based on a random effects model and calculated the confi-
dence interval for the weighted mean differences. Overall,
self-enhancement effects were greater than self-verification
effects, t(4111) = 39.12, p < .001, CI.95 = .1443, .1595, rs =
.29 versus .13, respectively. When broken down by specific
dependent variables, self-enhancement effects were greater
than self-verification effects for general affect, t(931) = 30.98,
p < .001, CI.95 = .2481, .2817; negative affect, t(1557) = 19.59,
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 269
CI.95 = .0967, .1182, and task liking, t(373) = 27.62, CI.95 =
.0625, .0720. Self-verification effects were greater than
self-enhancement effects for attraction, t(1336) = –4.43, CI.95
= –.0244, –.0094. Therefore, consistent with Shrauger (1975),
overall, the effect sizes for self-enhancement tended to exceed
those for self-verification for studies of affective responses.
Interpersonal Behaviors. This category refers to behaviors or
speech designed to bring interaction partners to see oneself in
a self-enhancing or self-verifying manner. Operationally, self-
enhancement strivings would be evidenced by a tendency for
participants to embrace positive feedback but eschew nega-
tive feedback. Self-verification effects would be reflected in a
tendency for participants’ positive self-views to embrace
positive feedback but to shun negative feedback; participants
with negative self-views display the opposite pattern.
We eliminated four studies because of insufficient data for
calculating r (e.g., Swann & Ely, 1984; Swann, Milton, &
Polzer, 2000; Swann & Read, 1981b; Swann, Stein-Seroussi,
& McNulty, 1992), leaving six studies, listed in Table 3. Tes-
sler and Schwartz (1972) manipulated locus of control instead
of positive or negative feedback, so we eliminated this study
from our final analysis. Finally, four studies (Baumeister &
Tice, 1985; McFarlin, Baumeister, & Blascovich, 1984; Stud-
ies 1 and 2 of Shrauger & Sorman, 1977) measured persistence
as the outcome variable. Persistence is an ambiguous outcome
as it can be explained by an increase in intrinsic motivation
(i.e., enjoying feedback and reveling in it) or it can be
Table 2. Accumulated Effect Sizes, Corrected for Sampling Error for Affective Reactions
Dependent Enhancement Verification
Study N variable effect size effect size
Gibbons and McCoy 1991a 111 Affect .28 .00
Jussim, Yen, and Aiello 1995 172 Affect .33 .05
Moreland and Sweeney 1984b 166 Affect .63 -.06
Quinlivan and Leary 2005 81 Affect .28 .00
Shrauger and Lund 1975 48 Affect .00 .00
Stets 2005 282 Affect .14 .17
Woo and Mix 1997 72 Affect .54 .00
Katz and Beach 2000 (Study 1) 143 Attraction .44 .39
Katz and Beach 2000 (Study 2) 198 Attraction .60 .87
Morling and Epstein 1997 (Study 1) 245 Attraction .34 .36
Sprecher and Hatfield 1982 (Study 1) 37 Attraction .56 .02
Sprecher and Hatfield 1982 (Study 2) 182 Attraction .17 .13
Sprecher and Hatfield 1982 (Study 3) 200 Attraction .12 .08
Sprecher and Hatfield 1982 (Study 4)† 332 Attraction .02 .04
Robinson and Smith-Lovin 1992 (Study 1) 75 Negative affect only .35 .00
Burke and Harrod 2005 286 Negative affect only .02 .06
Cast and Burke 2002 401 Negative affect only .05 .07
Ralph and Mineka 1998 160 Negative affect only .03 .19
Shrauger and Sorman 1977c 53 Negative affect only .54 .37
Swann, Griffin, Predmore and Gaines 1987 98 Negative affect only .38 .00
Wood, Heimpel, Newby-Clark, and Ross 2005 (Study 1)d† 67 Negative affect only .20 .18
Wood et al. 2005 (Study 2)d† 79 Negative affect only .06 .11
Campbell, Lackenbauer, and Muise 2006 103 Positive affect only .23 -.10
Stake 1982 (Study 1) 236 Positive affect only .46 .00
Anseel and Lievens 2006 389 Satisfaction .58 .07
Korman 1968 (Study 1) 71 Task liking .30 .16
Korman 1968 (Study 2) 129 Task liking .19 .18
Korman 1968 (Study 3) 174 Task liking .08 .00
Total 4,590 Sample weighted mean r .26 .13
Total of studies that fit inclusion criteria 4,112 Sample weighted mean r .29 .13
aWe averaged the effect sizes for Studies 1 and 2 because the data came from the same participant sample.
bWe averaged the effect sizes for affect and satisfaction. The affect effect size for self-enhancement is r = .57 and for self-verification is r = –.06. The
satisfaction effect size for self-enhancement is r = .69 and for self-verification is r = –.06. We separated these effect sizes when we analyzed the specific
dependent variables.
cThe dependent variables measured by Shrauger and Sorman are anxiety and satisfaction. We did not average the effect sizes across these two variables
because they are very different constructs. Instead, we chose to report effect sizes for anxiety only because it is the more common variable measured
in other studies and labeled it as “negative affect only.” The satisfaction effect size for self-enhancement is r = .75 and for self-verification is r = .00. We
separated these effect sizes when we analyzed the specific dependent variables.
†Study does not fit inclusion criteria.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

270 Personality and Social Psychology Review 14(3)
evidence of motivation to disprove the feedback. Because it
is unclear what task persistence signifies, we dropped all the
studies that measure task persistence from the final analysis.
This left only one study in the final sample pool. The study
measured feedback resistance (i.e., amount to which the par-
ticipant verbally questioned or refuted feedback).
Self-enhancement and interpersonal behaviors. As displayed
in Table 3, the average effect size for the self-enhancement
effect for our preliminary sample was r = .15. Broken down
by specific behaviors, the average effect size for persistence
was r = .12, help-seeking was r = .39, and feedback resis-
tance was r = .00.
Table 3 also indicates that only one study was included in
the final analysis. The effect size for the self-enhancement
effect was r = .00.
Self-verification and interpersonal behaviors. In our prelimi-
nary sample, the average effect size for the self-verification
effect was r = .18. Broken down by specific behaviors, the
average effect size for persistence was r = .18, help-seeking
was r = .00, and feedback resistance was r = .47.
Table 3 also indicates that only one study was included in
the final analysis. The effect size for the self-verification
effect was r = .47.
In sum, inspection of Table 3 reveals that the only study
that was included in the final analysis supported self-
verification (r = .47). Nevertheless, a single study does not
provide a sound basis for drawing conclusions regarding the
interpersonal behavior category.
Feedback Seeking. Feedback seeking refers to a tendency to
choose evaluations, or evaluators, who are inclined to offer
particular types of feedback. The conceptual predictor
variables were self-enhancement and self-verification. Oper-
ationally, self-enhancement was defined as a main effect for
feedback valence. Self-verification was defined as either the
interaction between feedback valence and self-views or the
(statistically equivalent) main effect of the congruence of
the feedback (with congruence referring to the degree of
match between the feedback valence and self-views). In
either instance, significant self-verification effects reflected
a tendency for participants with positive self-views to be
more inclined to prefer positive feedback (evaluators) than
people with negative self-views.
Omitted from Table 4 are 10 studies (e.g., Neiss,
Sedikides, Shahinfar, & Kupersmidt, 2006; Sedikides, 1993,
Experiments 1-3, 5-6) because the researchers did not mea-
sure self-views, leaving 22 studies. Studies 1 and 2 of Chen,
Chen, and Shaw (2004) were then excluded because they
measured only negative self-views (i.e., socially unskilled).
Study 2 of Rudich and Vallacher (1999) was deleted because
they confounded negative evaluation with rejection. This left
19 studies in the final sample pool. We further categorized
the studies by specific dependent variables: evaluator choice
(i.e., which evaluator do you want to have future interactions
with), evaluator preference (i.e., how much do you wish to
interact with the evaluator), feedback choice (i.e., which
feedback do you wish to see), and feedback preference (i.e.,
how much do you prefer to read each feedback).
Self-enhancement and feedback seeking. Table 4 reveals that
the average effect size for the self-enhancement effect in our
preliminary sample was r = .19. Broken down by specific
dependent variables, the average effect size for evaluator
choice was r = .17, evaluator preference was r = .00, feedback
choice was r = .11, and feedback preference was r = .41.
When we eliminated studies that did not meet the inclu-
sion criteria, the average effect size for the self-enhancement
effect was r = .22, as can be seen in Table 4. Broken down by
specific dependent variables, the average effect size for eval-
uator choice was at r = .13, feedback choice was r = .18, and
feedback preference stayed at r = .41.
Self-verification and feedback seeking. The average effect
size for the self-verification effect in our preliminary sample
was r = .25. Broken down by specific dependent variables,
the average effect size for evaluator choice was r = .10, eval-
uator preference was r = .38, feedback choice was r = .32,
and feedback preference was r = .27.
When we eliminated studies that did not fit the inclusion
criteria, the average effect size for the self-verification effect
Table 3. Accumulated Effect Sizes, Corrected for Sampling Error for Interpersonal Behaviors
Dependent Enhancement Verification
Study N variable effect size effect size
Baumeister and Tice 1985 (Study 1)† 61 Persistence .00 .41
McFarlin, Baumeister, and Blascovich 1984 (Study 1)† 93 Persistence .00 .00
Shrauger and Sorman 1977 (Study 1)† 53 Persistence .00 .36
Shrauger and Sorman 1977 (Study 2)a† 84 Persistence .42 .10
Tessler and Schwartz 1972† 48 Help seeking .39 .00
Swann and Hill 1982 24 Feedback resistance .00 .47
Total 363 Sample weighted mean r .15 .18
Total of studies that fit inclusion criteria 24 Sample weighted mean r .00 .47
aWe averaged the effect sizes for global and specific self-views. The global self-view effect size for self-enhancement was r = .41, and for self-verification was
r = .20. The specific self-view effect size for self-enhancement was r = .42 and for self-verification was r = .00.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 271
stayed at r = .25. Broken down by specific dependent vari-
ables, the average effect size for evaluator choice was r =
.17, feedback choice stayed at r = .32, and feedback prefer-
ence stayed at r = .27.
In sum, there were significant enhancement and verifica-
tion effects on feedback seeking. To determine the relative
strength of the self-enhancement versus self-verification
effects, we found the mean difference between the effect
sizes (M = –.028, SD = .25) based on a random effects model
and calculated the confidence interval for the weighted
means differences. Overall, self-verification effects were
greater than self-enhancement effects, t(1623) = –4.46, p <
.001, CI.95 = –.0401, –.0156, rs = .25 versus .22, respectively.
When broken down by specific dependent variables, self-
verification effects were greater than self-enhancement
effects for evaluator choice, t(580) = –9.42, p < .001,
CI.95 = –.0424, –.0278, and feedback choice, t(490) = –25.20,
CI.95 = –.23 11 , –.1976, whereas self-enhancement effects
were greater than self-verification effects for feedback
preference, t(551) = 11.76, p < .001, CI.95 = .1214, .1701.
Therefore, the tendency for self-verification to override
self-enhancement overall was driven by the relative
strength of self-verification on indices of actual choice; self-
enhancement strivings were stronger when preference
ratings were examined.
The results of our analysis of feedback seeking have impli-
cations for the following section on relationship quality. In
particular, although Rudich and Vallacher’s (1999) second
study was eliminated from the final meta-analysis because
they confounded negative evaluations with rejection, their
findings are nevertheless revealing and important. As shown
in Table 4, when participants perceived that the negative eval-
uator was apt to reject them, their self-verification strivings
were completely overridden by their desire for self-enhance-
ment. This evidence that people with negative self-views are
unusually wary of rejection is also supported by evidence that
low self-esteem persons are hesitant to enter novel social situ-
ations unless acceptance is virtually guaranteed (Anthony,
Holmes, & Wood, 2007, Study 4; Anthony, Wood, & Holmes,
2007). Moreover, when their feelings are hurt, people with
Table 4. Accumulated Effect Sizes, Corrected for Sampling Error for Feedback Seeking
Dependent Enhancement Verification
Study N variable effect size effect size
Robinson and Smith-Lovin 1992 (Study 2) 78 Evaluator choice .38 .50
Rudich, Sedikides, and Gregg 2007 (Study 1) 148 Evaluator choice .00 .00
Rudich and Vallacher 1999 (Study 1) 47 Evaluator choice .06 .36
Rudich and Vallacher 1999 (Study 2)† 50 Evaluator choice .60 –.62
Swann and Pelham 2002 227 Evaluator choice .14 .13
Swann, Stein-Seroussi, and Giesler 1992 (Study 1) 81 Evaluator choice .14 .13
Chen, Chen, and Shaw 2004 (Study 1)† 107 Evaluator preference .00 .35
Chen et al. 2004 (Study 2)† 211 Evaluator preference .00 .40
Bosson and Swann 1999 (Study 1) 74 Feedback choice .05 .33
Giesler, Josephs, and Swann 1996a 73 Feedback choice –.21 .36
Sedikides 1993 (Study 4) 120 Feedback choice .50 .46
Swann and Read, 1981a (Study 1)b 79 Feedback choice .00 .24
Swann and Read, 1981a (Study 2) 120 Feedback choice .00 .20
Swann, Wenzlaff, and Tafarodi 1992 (Study 1) 25 Feedback choice .17 .36
Joiner 1995 100 Feedback preference .48 .29
Kraus and Chen 2009c 104 Feedback preference .32 .23
Petit and Joiner 2001 101 Feedback preference .45 .31
Silvera and Neilands 2004 89 Feedback preference .36 .22
Silvera and Seger 2004 76 Feedback preference .90 .18
Swann, Hixon, Stein-Seroussi, and Gilbert 1990 (Study 3) 51 Feedback preference .00 .30
Swann, Pelham, and Krull 1989 (Study 1) 21 Feedback preference .00 .45
Swann, Pelham, and Krull 1989 (Study 2)d 10 Feedback preference .00 .45
Total 1,992 Sample weighted mean r .19 .25
Total of studies that fit inclusion criteria 1,624 Sample weighted mean r .22 .25
aBecause the feedback choice response between low self-esteem group and depressed group did not achieve statistical significance (z = 1.5, p < .07), we
collapsed the two groups.
bWe reported the average effect size for both emotionality and assertiveness. The effect sizes for emotionality were r = .00 for enhancement and r = .23
for verification. The effect sizes for assertiveness were r = .00 for enhancement and r = .26 for verification.
cWe averaged the effects sizes across both the significant other and the acquaintance groups. Insufficient data were available for the low-certainty condi-
tion group so only the results for the high-certainty condition were reported.
dThe necessary data were not provided in the published study but were provided by Swann.
†Study does not fit inclusion criteria.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

272 Personality and Social Psychology Review 14(3)
low self-esteem respond to hurt feelings with avoidance (e.g.,
Murray, 2005) and limit their risk of rejection by romantic
partners by emotionally distancing themselves from their
partner (Murray, Rose, Bellavia, Holmes, & Kusche, 2002).
Together, the foregoing findings suggest that when we
examine the impact of enhancement and verification on rela-
tionship quality, it will be important to distinguish the extent
to which the threat of rejection is low versus high. Most
important, the tendency for people with negative self-views
to seek self-verification should manifest itself when the risk
of rejection is low but not when the risk of rejection is high.
This insight guided our approach to the meta-analysis of
studies of relationship quality.
Relationship Quality. Relationship quality refers to feelings
about the relationship such as intimacy, satisfaction, and
thoughts about, or actual rates of, separation and divorce.
Operationally, self-enhancement was defined as a tendency
for relationship quality to be higher insofar as the partner
evaluation was positive. Self-verification effects reflected a
tendency for participants to report superior relationship qual-
ity when the partner saw participants as they saw themselves.
Specifically, self-verification occurred insofar as people with
positive self-views reported better relationship quality when
their partner viewed them positively, and people with nega-
tive self-views reported better relationship quality when
their partner viewed them negatively.
Four studies were not included in the tables because of
insufficient data to calculate r (i.e., Burke & Stets, 1999;
Carnelley, Ruscher, & Shaw, 1999; Ritts & Stein, 1995;
Schafer, Wickrama, & Keith, 1996). Of the studies that were
included in the tables and initial meta-analysis, three studies
were eliminated from the final meta-analysis because they did
not fit into our inclusion criteria. Katz, Beach, and Anderson
(1996) measured perceived verification rather than actual
verification; this introduces the possibility that their
responses were influenced by projection biases or theories
regarding socially appropriate responding. Katz, Arias, and
Beach (2000) appear to have violated the specificity match-
ing principle (e.g., Swann et al., 2007), which is important
because one would not expect self-verification effects if
there was a mismatch between the self-view and evaluative
feedback. Specifically, Katz and colleagues matched partici-
pants’ global self-esteem with partners’ psychological and
physical abuse, a problem because having low self-esteem
does not mean that one expects or desires to be physically
abused. Finally, Cast and Burke (2002) measured verifying
role identities, which is how much spouses agreed on mean-
ings and expectations on spousal roles rather than actual
self-views. Because this operationalization of self-views was
not consistent with how self-views were measured in the
other studies, we dropped this study from the final analysis.
All of the remaining studies were included in the final
meta-analysis. In line with the foregoing discussion of
evidence that people with negative self-views are adverse
to rejection (e.g., Anthony et al., 2007; Anthony et al.,
2007; Murray, 2005; Murray et al., 2002; Rudich & Val-
lacher, 1999), we divided the studies into two categories,
one high and one low in rejection risk. Studies of dating
relationships were placed in the high-rejection-risk cate-
gory because such relationships are essentially extended
qualifying exams in which commitment is generally lack-
ing. For this reason, negative evaluations often serve as an
ominous sign that rejection may be imminent. Although
rejection risk is generally much lower in marital relation-
ships (Swann et al., 1994), the stability of such relationships
can be threatened if one person decides that the partner is
unsuitable. In particular, if spouses verify negative quali-
ties that are high in relationship relevance (e.g., affectionate,
thoughtful, warm), they convey that their partner is unsatis-
factory and thus rejectable (“You are unaffectionate,
thoughtless, cold, etc.”). We accordingly placed marital
studies that focused on qualities that were high in relation-
ship relevance (i.e., the interpersonal qualities scale
developed by Murray et al., 1996) in a separate, “high-
rejection-risk” category; all other marital studies were
placed in the low-rejection-risk category.
Finally, studies that measured global traits such as global
worth and competency were placed in the high-rejection-risk
category. If one does not believe one’s partner is positive on
a global level (i.e., unworthy), the evaluation is overly nega-
tive and calls into question why the partner even wants to be
in relationship with the target. In contrast, evaluations about
specific traits (i.e., intelligence or athletic ability) have less
impact on the rejectability of the partner. For example, low
ratings on intelligence could be offset by high ratings on
attractiveness (see also Neff & Karney, 2002).
As can be seen in Tables 5 and 6, the foregoing procedure
left 10 studies in the high-rejection-risk category and 5 stud-
ies in the low-rejection-risk category in the final sample
pool.4 We further categorized the studies by specific depen-
dent variables: commitment (i.e., how committed are you to
your partner), intimacy (i.e., how intimate is your relation-
ship), satisfaction (i.e., how satisfied are you with your
relationship), and separation or divorce (i.e., did the couple
stay together or separate or divorce).
Finally, note that two studies by Neff and Karney (2005)
measured both global and specific self-views. In keeping
with the reasoning outlined above, findings derived from
measures of global self-views belong in the high-rejection-
risk category and are displayed in Table 5; findings derived
from specific self-views belong in the low-rejection-risk cat-
egory and are displayed in Table 6.
Self-enhancement and relationship quality in high-rejection-
risk studies. As shown in Table 5, the average effect size for
the self-enhancement effect in our preliminary sample was
r = .32. Broken down by specific relationship quality, the
average effect size for dating commitment was r = .55, dating
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 273
intimacy was r = .27, dating satisfaction was r = .34, marital
satisfaction was r = 38, and divorce was r = .05.
When we eliminated studies that did not fit the inclusion
criteria, the average effect size for the self-enhancement
effect was r = .31. The effect size for dating intimacy dropped
to r = .24. All other effect sizes stayed the same.
Self-verification and relationship quality in high-rejection-risk
studies. The average effect size for the self-verification effect
for our preliminary sample was r = .08. Broken down by
specific relationship quality, the average effect size for dating
commitment was r = .88, dating intimacy was r = .08, dating
satisfaction was r = –.12, marital satisfaction was r = –.06,
and divorce was r = –.03.
When we eliminated studies that did not fit the inclusion
criteria, the average effect size for the self-verification effect
stayed at r = .08, as can be seen in Table 5. The effect size for
dating intimacy dropped to r = .07. All other effect sizes
stayed the same.
To determine the relative strength of the self-enhancement
versus self-verification effects in the high-rejection-risk stud-
ies, we found the mean difference between the effect sizes
(M = .234, SD = .27) based on a random effects model and calcu-
lated the confidence interval for the weighted means
differences. Overall, self-enhancement effects were greater
than self-enhancement effects, t(1528) = 33.69, p < .001,
CI.95 = .2206, .2479, rs = .31 versus .08, respectively. When
broken down by specific dependent variables, self-enhance-
ment effects were greater than self-verification effects for
dating intimacy, t(281) = 97.53, p < .001, CI.95 = .1647, .1715;
dating satisfaction, t(316) = 282.31, p < .001, CI.95 = .4597,
.4661; marital satisfaction, t(480) = 88.60, p < .001,
CI.95 = .4271, .4465; and divorce, t(250) = 64.84, p < .001,
Table 6. Accumulated Effect Sizes, Corrected for Sampling Error for Relationship Quality in Low-Rejection-Risk Studies
Dependent Enhancement Verification
Study N variable effect size effect size
De La Ronde and Swann 1998 160 Marital intimacy .08 .24
Swann, De La Ronde, and Hixon 1994 165 Marital intimacy .00 .30
Katz, Beach, and Anderson 1996† 265 Marital satisfaction .65 .17
Cast and Burke 2002a† 460 Separation or divorce .05 .12
Burke and Harrod 2005a 286 Separation or divorce .05 .12
Neff and Karney 2005 (Study 1)b 82 Divorce .07 .14
Neff and Karney 2005 (Study 2)b 169 Divorce .04 .17
Total 1,587 Sample weighted mean r .15 .17
Total of studies that fit inclusion criteria 862 Sample weighted mean r .05 .19
aThe necessary c2 statistic was not reported in the article but was kindly provided to us by Burke.
bEffect sizes for specific self-views only.
†Study does not fit inclusion criteria.
Table 5. Accumulated Effect Sizes, Corrected for Sampling Error for Relationship Quality in High-Rejection-Risk Studies
Dependent Enhancement Verification
Study N variable effect size effect size
Katz and Beach 2000 (Study 2) 198 Dating commitment .55 .88
Campbell, Lackenbauer, and Muise 2006 103 Dating intimacy .33 .20
Katz, Arias, and Beach 2000a† 82 Dating intimacy .37 .11
Swann, De La Ronde, and Hixon 1994 179 Dating intimacy .19 .00
Murray, Holmes, and Griffin 1996 196 Dating satisfaction .25 –.19
Murray, Holmes, and Griffin 2000 121 Dating satisfaction .49 –.01
Murray et al. 1996 178 Marital satisfaction .34 .02
Murray et al. 2000 105 Marital satisfaction .60 –.01
Sacco and Phares 2001b 198 Marital satisfaction .30 –.15
Neff and Karney 2005 (Study 1)c 82 Divorce .08 .03
Neff and Karney 2005 (Study 2)c 169 Divorce .03 –.06
Total 1,611 Sample weighted mean r .32 .08
Total of studies that fit inclusion criteria 1,529 Sample weighted mean r .31 .08
aWe averaged the effect sizes for both intimacy and stability outcomes across Time 1 and Time 2.
bWe averaged the effect sizes for global self-views of depression and self-esteem because of similar patterns in results.
cEffect sizes for global self-views only.
†Study does not fit inclusion criteria.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

274 Personality and Social Psychology Review 14(3)
CI.95 = .0746, .0793. In short, when rejection risk was high,
self-enhancement strivings prevailed over self-verification
strivings.
Self-enhancement and relationship quality in low-rejection-risk
studies. As shown in Table 6, the average effect size for the
self-enhancement effect in our preliminary sample was r =
.15. Broken down by specific relationship quality, the average
effect size for marital intimacy was r = .04, marital satisfac-
tion was r = .65, and separation or divorce was r = .05.
When we eliminated studies that did not fit the inclusion
criteria, the average effect size for the self-enhancement
effect was r = .05. Broken down by specific relationship
quality, the effect size for marital intimacy was r = .04, and
separation or divorce was r = .05.
Self-verification and relationship quality in low-rejection-risk
studies. As shown in Table 6, the average effect size for the
self-verification effect for our preliminary sample was r =
.17. Broken down by specific relationship quality, the average
effect size for marital intimacy was r = .27, marital satisfac-
tion was r = .17, and separation or divorce was r = .13.
When we eliminated studies that did not fit the inclusion
criteria, the average effect size for the self-verification effect
was r = .19. Broken down by specific relationship quality,
the average effect size for marital intimacy stayed at r = .27,
and separation or divorce was r = .14.
To determine the relative strength of the self-enhancement
versus self-verification effects in the low-rejection-risk stud-
ies, we found the mean difference between the effect sizes
(M = –.143, SD = .08) based on a random effects model and
calculated the confidence interval for the weighted means dif-
ferences. Overall, self-verification effects were greater than
self-enhancement effects t(861) = –49.69, p < .001, CI.95 =
–.1481, –.1369, rs = .19 versus .05, respectively. When
broken down by specific dependent variables, self-verification
effects were greater than self-enhancement effects for marital
intimacy, t(324) = –59.43, p < .001, CI.95 = –.2387, –.2234,
and separation or divorce, t(536) = –73.85, p < .001, CI.95 =
–.0912, –.0865. In summary, just as the effect sizes for self-
enhancement tended to be higher in the high-rejection-risk
studies, the effect sizes for self-verification tended to be
higher in the low-rejection-risk studies.
General Discussion
The results of our meta-analysis confirm earlier evidence that
self-enhancement strivings influence affective responses but
self-verification strivings shape cognitive reactions (e.g.,
Shrauger, 1975; Swann et al., 1987). In addition, our findings
extend earlier reviews by showing that self-verification
strivings were significantly stronger predictors of feedback
seeking and—as long as rejection risk was low—relationship
quality. In studies of relationship quality in which rejection
risk was high (i.e., when a negative appraisal might signal
disinterest in maintaining the relationships), self-enhancement
strivings trumped self-verification. On the balance, the evi-
dence suggests that both motives are potent but that they
express themselves differently depending on the response
class under scrutiny as well as the relevance of other moti-
vational forces. The communion motive seemed particularly
important, as it seems likely that people abandoned their self-
verification strivings when rejection risk was high in an effort
to gratify their desire for communion.5
If one looks beyond a simple tally of wins and losses,
it appears that somewhat different processes may have
mediated responses to different dependent measures. In
part, this may reflect the fact that studies associated with
some response classes were conducted in the laboratory
whereas studies associated with other response classes were
conducted in naturally occurring settings. Research on feed-
back seeking versus relationship quality is a case in point.
Recall that in studies of feedback seeking, the size of the
self-verification effects equaled or exceeded self-enhance-
ment effects, but in studies of relationship quality,
self-verification effects prevailed in studies in which rejec-
tion risk was low but self-enhancement prevailed in studies
in which rejection risk was high. It may be that, in both
types of studies, rejection avoidance was a crucial mediat-
ing mechanism but that it manifested itself differently
because of unique properties of the laboratory versus field
settings. In the laboratory settings in which most studies of
feedback seeking were conducted, the interaction partner
was often a stranger in whom the participant had very little
investment. The overriding goal of participants in such
experiments was likely to complete the study with a mini-
mum of effort or duress. Participants were therefore
unconcerned with being rejected unless the experimenter
specifically alerted them to this possibility, as did Rudich
and Vallacher (1999, Study 2). Seeking verifying, negative
feedback in the typical feedback seeking study, then, is
something of a mixed bag: Although it is valued because it
seems accurate, it is unpleasant to receive (Swann et al.,
1987; Wood et al., 2005). It is thus not surprising that
there was evidence for both self-verification and self-
enhancement strivings on the measures of feedback seeking,
although the overall pattern favored self-verification.
In contrast, in ongoing relationships, evaluations from the
partner can serve as a bellwether of the relationship. In dating
relationships and other designs in which rejection risk is high,
negative evaluations represent a clear signal that all is not
well. Such evaluations are therefore unwelcome, even if they
contain a kernel of truth. In contrast, things are different in
marital relationships and other designs in which rejection risk
is low because of high levels of commitment. Here, negative
evaluations are not necessarily worrisome, as long as they do
not focus on qualities that are crucial to the survival of the
relationship, such as being loving or caring. By the same
token, overly positive evaluations can be troubling, as they
signal that the partner may expect more than the target of the
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 275
evaluations feels capable of delivering. In addition to low
levels of rejection risk, in marital relationships, negative eval-
uations can be welcome as they signal mutual understanding
of limitations and confirm an expectation that spouses should
know one another. Consistent with this reasoning, Campbell,
Lackenbauer, and Muise (2006) found that as relationship
length increased, couples came to desire self-verifying feed-
back over self-enhancing feedback, even when self-views
were negative. Of course, additional moderators may be at
play (e.g., Neff & Karney, 2002), and we urge future research-
ers to continue the search for such variables.
Conclusion
At the very least, the findings from the meta-analysis reported
here refute the contention that the desire for self-enhancement
routinely overrides the desire for self-verification (Sedikides
& Gregg, 2008). More interestingly, the robustness of self-
verification strivings in our results throws into question the
proper interpretation of dozens of studies that are widely
assumed to represent evidence for self-enhancement theory:
studies that were not included in our meta-analyses because
the investigators failed to include measures of self-views.
The failure of researchers to measure self-views in such
putative studies of self-enhancement introduces a glaring
interpretative ambiguity. That is, because most people in
most societies have positive self-views (Diener & Diener,
1995), in unselected samples roughly 70% will possess
positive self-views. This means that evidence of “self-
enhancement” in such samples may reflect self-verification
strivings of the majority of people who happen to have posi-
tive self-views (see also Kwan, John, Kenny, Bond, &
Robins, 2004; Kwan, John, Robins, & Kuang, 2008). Stated
differently, the dozens of “self-enhancement” studies in
which researchers failed to incorporate measures of self-
views may not support the theory after all. Among many
others, this includes studies that some (Heine & Hamamura,
2007; Heine, Kitayama, & Hamamura, 2007; Mezulis,
Abramson, Hyde, & Hanklin, 2004; Sedikides, Gaertner, &
Vevea, 2005) have taken as support for the cross-cultural gen-
erality of the self-enhancement motive.
Of course, advocates of self-enhancement theory could
respond to our evidence that neither self-verification nor
self-enhancement strivings dominated the responses of
participants by singling out the one arena in which self-
enhancement strivings were clearly strongest: affective
responses. Furthermore, they might add, people’s feelings
are the most telling indicator of what they really want. We
question this argument on two grounds. First of all, we see
no compelling reason to believe that affective responses are
any more diagnostic of what people want than any of the
other responses we examined. Second, although the affective
responses certainly did conform to the predictions of self-
enhancement theory, we believe that other motives could
also explain these responses. At a general level, we suspect
that people feel good about positive evaluations because,
for most people most of the time, positive evaluations are
associated with positive outcomes. As such, it is not so much
that people with negative self-views are smitten with the
positive evaluations per se, it is that they associate such eval-
uations with increased feelings of agency (e.g., getting a
good job, pay raise, or respect) or communion (e.g., social
acceptance, a friend or relationship partner). As such, peo-
ple’s affective responses to feedback could be driven by
people’s need for agency (competence and autonomy; Ryff,
1989) or communion (belongingness and interpersonal con-
nectedness; Baumeister & Leary, 1995; Heine et al., 2006;
Wiggins & Broughton, 1985). More work is needed to test
this hypothesis.
But if self-enhancement theory is not the master motive,
then what is? Clearly, it is not self-verification, as our find-
ings suggested that self-verification strivings were poor
predictors of affective reactions and relationship quality when
rejection risk was high. Could communion be the master
motive? Here again, we think not. For example, although the
desire for communion almost surely places some constraints
on the manner in which people pursue their self-verification
strivings, we suspect that the opposite is sometimes the case.
Consider, for example, evidence that people divorce partners
who see them too positively or too negatively (e.g., Burke &
Harrod, 2005; Cast & Burke, 2002; Neff & Karney, 2005).
Insofar as such evidence reflects a tendency for self-verification
strivings to override people’s desire for communion, it would
appear that the self-verification motive does not always sub-
serve a superordinate communion motive.
The upshot is this: Human social behavior is far too com-
plex and nuanced to be readily explained by any single
motive or even a hierarchy in which a single motive is the
“master” of the others. As such, it would behoove workers to
suspend their competitive quest to establish that their chosen
motive is “biggest.” In the place of this quest, we suggest
that researchers acknowledge the existence of multiple
motives and strive to develop a deeper understanding of the
delicate interplay between them.
Acknowledgments
We thank Michael Buhrmester, Matthew Brooks, Christine Chang-
Schneider, Serena Chen, Steven J. Heine, Sade Jones, Robin R.
Vallacher, and Morgan Ward for their feedback on earlier versions of
this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interests with respect
to the authorship and/or publication of this article.
Financial Disclosure/Funding
The authors received no financial support for the research and/or
authorship of this article.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

276 Personality and Social Psychology Review 14(3)
Notes
1. In principle, we could also have compared the predictive
power of self-enhancement to another major competitor, self-
assessment (Trope, 1983). Examination of the extant research
literature, however, revealed that there were far more studies
on self-verification than self-assessment, leading us to focus on
self-verification.
2. Although a case can be made for ignoring main effects when
they are qualified by a significant interaction, our theoretical
interest in the relative strength of self-enhancement and self-
verification strivings overrode such considerations.
3. We did not include Study 2 of Story (1998) because the findings
were so conflicting that any summary statistic would have been
misleading. Three dependent variables in the study fell under
the category of cognitive processes: recall, attention, and accu-
racy. The verification effect sizes for those variables were r =
.33, r = –.34, and r = .24, respectively; the enhancement effect
sizes were r = .11, r = .00, and r = .68, respectively. As such,
summary statistics would have masked these discrepancies.
4. In Footnote 15 Murray, Holmes, and Griffin (2000) report that
they found evidence of self-enhancement but not self-verification
in their married participants when the Self-Attributes Ques-
tionnaire was used as the index of self-views and appraisals.
Nevertheless, Murray (2005) subsequently acknowledged that a
substantial number of the “married” participants were actually
cohabitating. This is important because (a) there is compelling
evidence that cohabiting couples resemble dating couples rather
than married couples (e.g., Manning & Smock, 2005), (b) dating
couple do not display self-verification strivings (e.g., Swann
et al., 1994), and (c) when cohabitating couples were eliminated
from Murray et al.’s sample, a self-verification effect (p < .07)
emerged among male participants (Murray, 2005). The ambigu-
ity in the population of married couples is further reason to not
include the married sample of Murray et al. in the meta-analysis.
5. We should note that this effect obtained for “behavioral” mea-
sures such as intimacy and divorce but not marital satisfaction.
Caution should be exercised in interpreting this finding, how-
ever, as in this category there was only one study in which mari-
tal satisfaction was reported.
References
References marked with an asterisk indicate studies included in the
meta-analysis.
Allport, G. W. (1937). Personality: A psychological interpretation.
New York, NY: Holt.
*Anseel, F., & Lievens, F. (2006). Certainty as a moderator for
feedback reactions? A test of the strength of the self-verification
motive. Journal of Occupational Psychology, 79, 533-551.
Anthony, D. B., Holmes, J. G., & Wood, J. V. (2007). Social
acceptance and self-esteem: Tuning the sociometer to inter-
personal value. Journal of Personality and Social Psychology,
92, 1024-1039.
Anthony, D. B., Wood, J. V., & Holmes, J. G. (2007). Testing soci-
ometer theory: Self-esteem and the importance of acceptance
for social decision-making. Journal of Experimental Social
Psychology, 43, 425-432.
Baumeister, R. F. (1982). A self-presentational view of social phe-
nomena. Psychological Bulletin, 91, 3-26.
Baumeister, R. F. (1989). The optimal margin of illusion. Journal of
Social and Clinical Psychology, 8, 176-189.
Baumeister, R. F. (1998). The self. In D. T. Gilbert, S. T. Fiske,
& G. Lindzey (Eds.), The handbook of social psychology
(pp. 680-740). New York, NY: McGraw-Hill.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong:
Desire for interpersonal attachments as a fundamental human
motivation. Psychological Bulletin, 117, 497-529.
*Baumeister, R. F., & Tice, D. M. (1985). Self-esteem and responses
to success and failure: Subsequent performance and intrinsic
motivation. Journal of Personality, 53, 450-467.
Bell, S. T., & Arthur, W., Jr. (2008). Feedback acceptance in devel-
opmental assessment centers: The role of feedback message,
participant personality, and affective response to the feedback
session. Journal of Organizational Behavior, 29, 681-703.
*Bellavia, G., & Murray, S. (2003). Did I do that? Self-esteem-
related differences in reactions to romantic partners’ moods.
Personal Relationships, 10, 77-95.
Blaine, B., & Crocker, J. (1993). Self-esteem and self-serving
biases in reactions to positive and negative events: An integra-
tive review. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of
low self-regard (pp. 55-85). Hillsdale, NJ: Erlbaum.
*Bosson, J., & Swann, W. B., Jr. (1999). Self-liking, self-competence,
and the quest for self-verification. Personality and Social Psy-
chology Bulletin, 25, 1230-1241.
*Brown, J. D., Cai, H., Oakes, M. A., & Deng, C. (2009). Cultural
similarities in self-esteem functioning: East is east and west is
west, but sometimes the twain do meet. Journal of Cross-
Cultural Psychology, 40, 140-157.
*Burke, J. P. (1978). On causal attribution: The interactive relation-
ship between self-esteem and task performance. Social Behav-
ior and Personality, 6, 211-221.
*Burke, P. J., & Harrod, M. M. (2005). Too much of a good thing?
Social Psychology Quarterly, 68, 359-374.
Burke, P. J., & Stets, J. E. (1999). Trust and commitment
through self-verification. Social Psychology Quarterly, 62,
347-366.
*Campbell, J. D., Chew, B., & Scratchley, L. S. (1991). Cogni-
tive and emotional reactions to daily events: The effects of
self-esteem and self-complexity. Journal of Personality, 59,
473-505.
*Campbell, L., Lackenbauer, S. D., & Muise, A. (2006). When is
being known or adored by romantic partners most beneficial?
Self-perceptions, relationship length, and responses to partner’s
verifying and enhancing appraisals. Personality and Social Psy-
chology Bulletin, 21, 1283-1294.
Carnelley, K. B., Ruscher, J. B., & Shaw, S. K. (1999). Meta-
accuracy about potential relationship partners’ models of others.
Personal Relationships, 6, 95-109.
Cassidy, J., Ziv, Y., Mehta, T. G., & Feeney, B. C. (2003). Feed-
back seeking in children and adolescents: Associations with
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 277
self-perceptions, attachment representations, and depression.
Child Development, 74(2), 612-628.
*Cast, A. D., & Burke, P. J. (2002). A theory of self-esteem. Social
Forces, 83(3), 1041-1068.
*Chandler, T. A., Lee, M. S., & Pengilly, J. W. (1997). Self-esteem
and causal attributions. Genetic, Social, and General Psychol-
ogy Monographs, 123, 479-491.
*Chen, S., Chen, K. Y., & Shaw, L. (2004). Self-verification motives
at the collective level of self-definition. Journal of Personality
and Social Psychology, 86, 77-94.
Cohen, B. H. (2002). Calculating a factorial ANOVA from means
and standard deviations. Understanding Statistics, 1(3), 191-203.
Colvin, C. R., & Griffo, R. (2007). The psychological costs of
self-enhancement. In E. C. Chang (Ed.), Self-criticism and
self-enhancement: Theory, research, and clinical implica-
tions (pp. 123-140). Washington, DC: American Psychologi-
cal Association.
Cooley, C. H. (1902). Human nature and the social order. New York,
NY: Scribner’s.
Copleston, F. (1957). A history of philosophy (Vol. 1). Westminster,
MD: Newman Press.
Crary, W. G. (1966). Reactions to incongruent self-experiences.
Journal of Consulting Psychology, 30, 246-252.
Crocker, J., & Wolfe, C. T. (2001). Contingencies of self-worth.
Psychological Review, 108, 593-623.
*De La Ronde, C., & Swann, W. B., Jr. (1998). Partner verification:
Restoring shattered images of our intimates. Journal of Person-
ality and Social Psychology, 75, 374-382.
Diener, E., & Diener, M. (1995). Cross-cultural correlates of life
satisfaction. Journal of Personality and Social Psychology, 68,
653-663.
Ditto, P. H., & Lopez, D. A. (1993). Motivated skepticism: Use
of differential decision criteria for preferred and nonpreferred
conclusions. Journal of Personality and Social Psychology, 63,
568-584.
*Feather, N. T. (1969). Attribution of responsibility and valence
of success and failure in relation to initial confidence and task
performance. Journal of Personality and Social Psychology, 13,
129-144.
*Feather, N. T., & Simon, J. G. (1971). Causal attributions for
success and failure in relation to expectations of success based
upon selective or manipulative control. Journal of Personality,
39, 527-541.
*Feather, N. T., & Simon, J. G. (1973). Fear of success and causal
attribution for outcome. Journal of Personality, 41, 525-542.
*Fielstein, E., Klein, M. S., Fischer, M., Hanan, C., Koburger, P.,
Schneider, M. J., & Leitenberg, H. (1985). Self-esteem and
causal attributions for success and failure in children. Cognitive
Therapy and Research, 9, 381-398.
*Fitch, G. (1970). Effects of self-esteem, perceived performance,
and choice on causal attributions. Journal of Personality and
Social Psychology, 16, 311-315.
*Gibbons, F. X., & McCoy, S. B. (1991). Self-esteem, similarity,
and reactions to active versus passive downward comparison.
Journal of Personality and Social Psychology, 60, 414-424.
*Giesler, R. B., Josephs, R. A., & Swann, W. B., Jr. (1996). Self-
verification in clinical depression. Journal of Abnormal Psychol-
ogy, 105, 358-368.
*Gilmor, T. M., & Minton, H. L. (1974). Internal versus external
attribution of task performance as a function of locus of control,
initial confidence, and success-failure outcome. Journal of Per-
sonality, 42, 159-174.
*Girodo, M., Dotzenroth, S. E., & Stein, S. J. (1981). Causal attri-
bution bias in shy males: Implications for self-esteem and self-
confidence. Cognitive Therapy and Research, 5, 325-338.
Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The causes
and consequences of a need for self-esteem: A terror manage-
ment theory. In R. F. Baumeister (Ed.), Public self and private
self (pp. 189-212). New York, NY: Springer-Verlag.
Hardin, C. D., & Higgins, E. T. (1996). Shared reality: How social
verification makes the subjective objective. In E. T. Higgins &
R. M. Sorrentino (Eds.), Handbook of motivation and cogni-
tion: The interpersonal context (Vol. 3, pp. 1-77). New York:
Guilford.
Heine, S. J., & Hamamura, T. (2007). In search of East Asian self-
enhancement. Personality and Social Psychology Review, 11, 1-24.
Heine, S. J., Kitayama, S., & Hamamura, T. (2007). The inclusion
of additional studies yields different conclusions: A reply to
Sedikides, Gaertner, & Vevea (2005), JPSP. Asian Journal of
Social Psychology, 10, 49-58.
Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999).
Is there a universal need for positive self-regard? Psychological
Review, 106, 766-794.
Heine, S. J., Proulx, T., & Vohs, K. D. (2006). Meaning mainte-
nance model: On the coherence of human motivations. Person-
ality and Social Psychology Review, 10, 88-110.
Hoyle, R. H., Kernis, M. H., Leary, M. R., & Baldwin, M. W. (1999).
Selfhood: Identity, esteem, regulation. Boulder, CO: Westview.
Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis:
Correcting error and bias in research findings. Newbury Park,
CA: Sage.
Hunter, J. E., Schmidt, F. L., & Jackson, G. B. (1982). Meta-analysis:
Accumulating research findings across studies. Beverly Hill, CA:
Sage.
Jahoda, M. (1958). Current concepts of positive mental health.
New York: Basic Books.
*Joiner, T. E., Jr. (1995). The price of soliciting and receiving negative
feedback: Self-verification theory as a vulnerability to depression
theory. Journal of Abnormal Psychology, 104, 364-372.
Jones, S. C. (1973). Self and interpersonal evaluation: Esteem theories
versus consistency theories. Psychological Bulletin, 79, 185-199.
*Jussim, L., Yen, H., & Aiello, J. R. (1995). Self-consistency, self-
enhancement, and accuracy in reactions to feedback. Journal of
Experimental Social Psychology, 31, 322-356.
*Katz, J., Arias, I., & Beach, S. R. H. (2000). Psychological abuse,
self-esteem, and women’s dating relationship outcomes. Psy-
chology of Women Quarterly, 24, 349-357.
*Katz, J., & Beach, S. R. H. (2000). Looking for love? Self-verification
and self-enhancement effects on initial romantic attraction. Per-
sonality and Social Psychology Bulletin, 26, 1526-1539.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

278 Personality and Social Psychology Review 14(3)
*Katz, J., Beach, S. R. H., & Anderson, P. (1996). Self-enhancement
versus self-verification: Does spousal support always help? Cog-
nitive Therapy and Research, 20, 345-360.
*Korman, A. K. (1968). Task success, task popularity, and self-
esteem as influences on task liking. Journal of Applied Psy-
chology, 52, 484-490.
*Kraus, M. W., & Chen, S. (2009). Striving to be known by sig-
nificant others: Automatic activation of self-verification goals
in relationship contexts. Journal of Personality and Social Psy-
chology, 97, 58-73.
*Kwan, V. S., John, O. P., Kenny, D. A., Bond, M. H., &
Robins, R. W. (2004). Reconceptualizing individual differences
in self-enhancement bias: An interpersonal approach. Psycho-
logical Review, 111, 94-111.
Kwan, V. S., John, O. P., Robins, R. W., & Kuang, L. L. (2008).
Conceptualizing and assessing self-enhancement bias: A com-
ponential approach. Journal of Personality and Social Psychol-
ogy, 94, 1062-1077.
Leary, M. R. (2007). Motivational and emotional aspects of the self.
Annual Review of Psychology, 58, 317-344.
Lecky, P. (1945). Self-consistency: A theory of personality. New York,
NY: Island Press.
Lyons, L. C. (2003). Meta-analysis: Methods of accumulating
results across research domains. Retrieved from http://www
.lyonsmorris.com/MetaA/index.htm
Manning, W. D., & Smock, P. J. (2005). Measuring and modeling
cohabitation: New perspectives from qualitative data. Journal of
Marriage and Family, 67, 989-1002.
*McFarlin, D. B., Baumeister, R. F., & Blascovich, J. (1984). On
knowing when to quit: Task failure, self-esteem, advice, and non-
productive persistence. Journal of Personality, 52, 138-155.
*McFarlin, D. B., & Blascovich, J. (1981). Effects of self-esteem
and performance feedback on future affective preferences and
cognitive expectations. Journal of Personality and Social Psy-
chology, 40, 521-531.
*McMahan, I. D. (1973). Relationships between causal attributions
and expectancy of success. Journal of Personality and Social
Psychology, 28, 108-114.
Mead, G. H. (1934). Mind, self and society. Chicago, IL: University
of Chicago Press.
Mezulis, A. H., Abramson, L. Y., Hyde, J. S., & Hanklin, B. L.
(2004). Is there a universal positivity bias in attributions? A
meta-analytic review of individual, development, and cultural
differences in self-serving attributional bias. Psychological
Bulletin, 130, 711-747.
*Moreland, R. L., & Sweeney, P. D. (1984). Self-expectancies and
reactions to evaluations of personal performance. Journal of
Personality, 52, 156-176.
*Morling, B., & Epstein, S. (1997). Compromises produced
by the dialectic between self-verification and self-enhance-
ment. Journal of Personality and Social Psychology, 73,
1268-1283.
Murray, S. L. (2005). Regulating the risks of closeness: A relation-
ship-specific sense of felt security. Current Directions in Psy-
chological Science, 14, 74-78.
*Murray, S. L., Holmes, J. G., & Griffin, D. W. (1996). The ben-
efits of positive illusions: Idealization and the construction of
satisfaction in close relationships. Journal of Personality and
Social Psychology, 70, 79-98.
*Murray, S. L., Holmes, J. G., & Griffin, D. W. (2000). Self-esteem
and the quest for felt security: How perceived regard regulates
attachment processes. Journal of Personality and Social Psy-
chology, 78, 478-498.
Murray, S. L., Rose, P., Bellavia, G. M., Holmes, J. G., &
Kusche, A. G. (2002). When rejection stings: How self-esteem
constrains relationship-enhancement processes. Journal of Per-
sonality and Social Psychology, 83, 556-573.
Neff, L. A., & Karney, B. R. (2002). Judgments of a relationship
partner: Specific accuracy but global enhancement. Journal of
Personality, 70, 1079-1112.
*Neff, L. A., & Karney, B. R. (2005). To know you is to love you:
The implications of global adoration and specific accuracy for
marital relationships. Journal of Personality and Social Psychol-
ogy, 88, 480-497.
Neiss, M. B., Sedikides, C., Shahinfar, A., & Kupersmidt, J. B.
(2006). Self-evaluation in a naturalistic context: The case of
juvenile offenders. British Journal of Social Psychology, 45,
499-518.
*Petit, J. W., & Joiner, T. E., Jr. (2001). Negative life events pre-
dict negative feedback seeking as a function of impact on self-
esteem. Cognitive Therapy and Research, 25, 733-741.
*Piers, E. V. (1977). Children’s self-esteem, level of esteem cer-
tainty, and responsibility for success and failure. Journal of
Genetic Psychology, 130, 295-304.
Pronin, E., Gilovich, T., & Ross, L. (2004). Objectivity in the eye of
the beholder: Divergent perceptions of bias in self versus others.
Psychological Review, 3, 781-799.
*Quinlivan, E., & Leary, M. R. (2005). Women’s perceptions
of their bodies: Discrepancies between self-appraisals and
reflected appraisals. Journal of Social and Clinical Psychology,
24, 1139-1163.
*Ralph, J. A., & Mineka, S. (1998). Attributional style and self-
esteem: The prediction of emotional distress following a mid-
term exam. Journal of Abnormal Psychology, 107, 203-215.
*Raps, C. S., Peterson, C., Reinhard, K. E., Abramson, L. Y., &
Seligman, M. E. P. (1982). Attributional style among depressed
patients. Journal of Abnormal Psychology, 91, 102-108.
Ritts, V., & Stein, J. R. (1995). Verification and commitment in
marital relationships: An exploration of self-verification theory in
community college students. Psychological Reports, 76, 383-386.
*Rizley, R. (1978). Depression and distortion in the attribution of
causality. Journal of Abnormal Psychology, 87, 32-48.
*Robinson, D. T., & Smith-Lovin, L. (1992). Selective interaction
as a strategy for identity maintenance: An affect control model.
Social Psychology Quarterly, 55, 12-28.
Rogers, C. (1951). Client-centered therapy: Its current practice,
implications and theory. London, England: Constable.
Rosenthal, R., & Rosnow, R. L. (2008). Essentials of behavioral
research: Methods and data analysis (3rd ed.). New York:
McGraw-Hill.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

Kwang and Swann 279
*Rudich, E. A., Sedikides, C., & Gregg, A. P. (2007). Self-
esteem moderates preference for accepting versus rejection
interaction partners. European Journal of Social Psychology,
37, 955-967.
*Rudich, E. A., & Vallacher, R. R. (1999). To belong or to self-
enhance? Motivational bases for choosing interaction partners.
Personality and Social Psychology Bulletin, 25, 1387-1404.
Ryff, C. D. (1989). Happiness is everything, or is it? Explorations
on the meaning of psychological well-being. Journal of Person-
ality and Social Psychology, 57, 1069-1081.
*Sacco, W. P., & Phares, V. (2001). Partner appraisal and marital
satisfaction: The role of self-esteem and depression. Journal of
Marriage and Family, 63, 504-513.
Schafer, R. B., Wickrama, K. A. S., & Keith, P. M. (1996). Self-
concept disconfirmation, psychological distress, and marital
happiness. Journal of Marriage and Family, 58, 167-177.
*Sedikides, C. (1993). Assessment, enhancement, and verification
determinants of the self-evaluation process. Journal of Person-
ality and Social Psychology, 65, 317-338.
Sedikides, C., Gaertner, L., & Vevea, J. L. (2005). Pancultural self-
enhancement reloaded: A meta-analytic reply to Heine (2005).
Journal of Personality and Social Psychology, 89, 539-551.
Sedikides, C., & Green, J. D. (2004). What I don’t recall can’t hurt
me: Information negativity versus information inconsistency
as determinants of memorial self-defense. Social Cognition,
22(1), 4-29.
Sedikides, C., & Gregg, A. P. (2008). Self-enhancement: Food for
thought. Perspectives on Social Psychology, 3(2), 102-116.
Shrauger, J. S. (1975). Responses to evaluation as a function of ini-
tial self-perceptions. Psychological Bulletin, 82, 581-596.
*Shrauger, J. S., & Kelly, R. J. (1988). Global self-evaluation and
changes in self-description as a function of information discrep-
ancy and favorability. Journal of Personality, 56, 709-728.
*Shrauger, J. S., & Lund, A. K. (1975). Self-evaluation and reactions
to evaluations from others. Journal of Personality, 43, 94-108.
*Shrauger, J. S., & Osberg, T. M. (1980). The relationship of time
investment and task outcome to causal attributions and self-
esteem. Journal of Personality, 48, 360-378.
*Shrauger, J. S., & Rosenberg, S. E. (1970). Self-esteem and the
effects of success and failure feedback on performance. Journal
of Personality, 38, 404-417.
*Shrauger, J. S., & Sorman, P. B. (1977). Self-evaluations, initial
success and failure, and improvement as determinants of per-
sistence. Journal of Consulting and Clinical Psychology, 45,
784-795.
*Silvera, D. H., & Neilands, T. (2004). Interpreting the uninterpre-
table: The effect of self-esteem on the interpretation of meaning-
less feedback. Scandinavian Journal of Psychology, 45, 61-66.
*Silvera, D. H., & Seger, C. R. (2004). Feeling good about our-
selves: Unrealistic self-evaluations and their relation to
self-esteem in the United States and Norway. Journal of Cross-
Cultural Psychology, 35, 571-585.
*Sprecher, S., & Hatfield, E. (1982). Self-esteem and romantic
attraction: Four experiments. Recherches de Psychologie Soci-
ale, 4, 61-81.
*Stake, J. E. (1982). Reactions to positive and negative feedback:
Enhancement and consistency effects. Social Behavior and Per-
sonality, 10, 151-156.
Steele, C. M. (1988). The psychology of self-affirmation: Sustain-
ing the integrity of the self. In L. Berkowitz (Ed.), Advances
in experimental social psychology (Vol. 21, pp. 261-302). San
Diego, CA: Academic Press.
*Stets, J. E. (2005). Examining emotions in identity theory. Social
Psychology Quarterly, 68, 39-56.
Stets, J. E., & Asencio, E. K. (2008). Consistency and enhance-
ment processes in understanding emotions. Social Forces,
86(3), 1055-1078.
*Story, A. L. (1998). Self-esteem and memory for favorable and
unfavorable personality feedback. Personality and Social Psy-
chology Bulletin, 24, 51-64.
*Stroebe, W., Eagly, A. H., & Stroebe, M. S. (1977). Friendly or
just polite? The effect of self-esteem on attributions. European
Journal of Social Psychology, 7, 265-274.
Swann, W. B., Jr. (1983). Self-verification: Bringing social reality
into harmony with the self. In J. Suls & A. G. Greenwald (Eds.),
Social psychological perspectives on the self (Vol. 2, pp. 33-66).
Hillsdale, NJ: Erlbaum.
Swann, W. B., Jr. (1990). To be adored or to be known: The inter-
play of self-enhancement and self-verification. In R. M. Sorrentino
& E. T. Higgins (Eds.), Foundations of social behavior (Vol. 2,
pp. 408-448). New York: Guilford.
Swann, W. B., Jr. (in press). Self-verification theory. In P. Van Lang,
A. Kruglanski, & E. T. Higgins (Eds.), Handbook of theories of
social psychology. Thousand Oaks, CA: Sage.
Swann, W. B., Jr., Chang-Schneider, C., & McClarty, K. L. (2007).
Do people’s self-views matter? Self-concept and self-esteem in
everyday life. American Psychologist, 62(2), 84-94.
*Swann, W. B., Jr., De La Ronde, C., & Hixon, J. G. (1994).
Authenticity and positivity strivings in marriage and courtship.
Journal of Personality and Social Psychology, 66, 857-869.
Swann, W. B., Jr., & Ely, R. J. (1984). A battle of wills: Self-
verification versus behavioral confirmation. Journal of Person-
ality and Social Psychology, 46, 1287-1302.
*Swann, W. B., Jr., Griffin, J. J., Predmore, S., & Gaines, B. (1987).
The cognitive-affective crossfire: When self-consistency con-
fronts self-enhancement. Journal of Personality and Social Psy-
chology, 52, 881-889.
*Swann, W. B., Jr., & Hill, C. A. (1982). When our identities are
mistaken: Reaffirming self-conceptions through social interac-
tion. Journal of Personality and Social Psychology, 43, 59-66.
*Swann, W. B., Jr., Hixon, J. G., Stein-Seroussi, A., & Gilbert, D. T.
(1990). The fleeting gleam of praise: Behavioral reactions to
self-relevant feedback. Journal of Personality and Social Psy-
chology, 59, 17-26.
Swann, W. B., Jr., Milton, L. P., & Polzer, J. T. (2000). Should we
create a niche or fall in line? Identity negotiation and small group
effectiveness. Journal of Personality and Social Psychology, 79,
238-250.
*Swann, W. B., Jr., & Pelham, B. W. (2002). Who wants out when
the going gets good? Psychological investment and preference
by William Swann on August 18, 2010psr.sagepub.comDownloaded from

280 Personality and Social Psychology Review 14(3)
for self-verifying college roommates. Journal of Self and Iden-
tity, 1, 219-233.
*Swann, W. B., Jr., Pelham, B. W., & Krull, D. S. (1989). Agreeable
fancy or disagreeable truth? Reconciling self-enhancement and
self-verification. Journal of Personality and Social Psychology,
57, 782-791.
*Swann, W. B., Jr., & Read, S. J. (1981a). Acquiring self-knowledge:
The search for feedback that fits. Journal of Personality and
Social Psychology, 41, 1119-1128.
*Swann, W. B., Jr., & Read, S. J. (1981b). Self-verification pro-
cesses: How we sustain our self-conceptions. Journal of Experi-
mental Social Psychology, 17, 351-372.
*Swann, W. B., Jr., Stein-Seroussi, A., & Giesler, B. (1992). Why
people self-verify. Journal of Personality and Social Psychol-
ogy, 62, 392-401.
Swann, W. B., Jr., Stein-Seroussi, A., & McNulty, S. E. (1992).
Outcasts in a white-life society: The enigmatic worlds of peo-
ple with negative self-conceptions. Journal of Personality and
Social Psychology, 62, 618-624.
Swann, W. B., Jr., Wenzlaff, R. M., Krull, D. S., & Pelham, B. W.
(1992). The allure of negative feedback: Self-verification striv-
ings among depressed persons. Journal of Abnormal Psychology,
101, 293-306.
*Swann, W. B., Jr., Wenzlaff, R. M., & Tafarodi, R. W. (1992).
Depression and the search for negative evaluations: More evi-
dence of the role of self-verification strivings. Journal of Abnor-
mal Psychology, 101, 314-317.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A
social psychological perspective on mental health. Psychologi-
cal Bulletin, 103, 193-210.
Taylor, S. E., & Gollwitzer, P. M. (1995). Effects of mindset on
positive illusions. Journal of Personality and Social Psychol-
ogy, 69, 213-226.
*Tennen, H., Herzberger, S., & Nelson, H. F. (1987). Depressive
attributional style: The role of self-esteem. Journal of Person-
ality, 55, 631-660.
Tesser, A. (1988). Toward a self-evaluation maintenance model of
social behavior. In L. Berkowitz (Ed.), Advances in experimen-
tal social psychology (Vol. 21, pp. 181-227). New York, NY:
Academic Press.
*Tessler, R. C., & Schwartz, S. H. (1972). Help seeking, self-
esteem, and achievement motivation: An attributional anal-
ysis. Journal of Personality and Social Psychology, 21,
318-326.
Trope, Y. (1983). Self-assessment in achievement behavior. In
J. M. Suls & A. G. Greenwald (Eds.), Psychological per-
spectives on the self (Vol. 2, pp. 93-121). Hillsdale, NJ:
Erlbaum.
Wiggins, J. S., & Broughton, R. (1985). The interpersonal circle:
A structural model for the integration of personality research.
In R. Hogan & W. H. Jones (Eds.), Perspectives in personality
(Vol. 1, pp. 1-47). Greenwich, CT: JAI.
*Woo, T. O., & Mix, P. (1997). Self-enhancing reactions to perfor-
mance feedback in an academic setting. Journal of Social Behav-
ior and Personality, 12, 481-500.
*Wood, J. V., Heimpel, S. A., Newby-Clark, I. R., & Ross, M.
(2005). Snatching defeat from the jaws of victory: Self-
esteem differences in the experience and anticipation of
success. Journal of Personality and Social Psychology, 89,
764-780.
by William Swann on August 18, 2010psr.sagepub.comDownloaded from