Self‐esteem and culture: Differences in cognitive self‐evaluations or affective self‐regard?

Abstract

Although people from East Asian countries consistently report lower self-esteem than do those from Western countries, the origins of this difference are unclear. We conducted two studies to illuminate this issue. Study 1 found that Chinese participants appraised themselves less positively than American participants on a cognitive measure of self-evaluations, but cultural differences were absent on a measure of affective self-regard. Moreover, cultural differences in global self-esteem were eliminated once cognitive self-evaluations were statistically controlled. Study 2 found that cultural differences in modesty underlie cultural differences in cognitive self-evaluations. These findings suggest that Chinese feel as positively toward themselves as Americans do, but are less inclined to evaluate themselves in an excessively positive manner.

Full text

Self-esteem and culture: Differences in cognitive self-evaluations
or affective self-regard?
Huajian Cai,1Jonathon D. Brown,2Ciping Deng3and Mark A. Oakes4
1Sun Yat-Sen University, Guangzhou, 3East China Normal University, Shanghai, China; 2University of Washington,
Seattle, Washington and 4Hamilton College, Clinton, New York, USA
Although people from East Asian countries consistently report lower self-esteem than do those from Western
countries, the origins of this difference are unclear. We conducted two studies to illuminate this issue. Study 1
found that Chinese participants appraised themselves less positively than American participants on a cognitive
measure of self-evaluations, but cultural differences were absent on a measure of affective self-regard. Moreover,
cultural differences in global self-esteem were eliminated once cognitive self-evaluations were statistically
controlled. Study 2 found that cultural differences in modesty underlie cultural differences in cognitive self-
evaluations. These findings suggest that Chinese feel as positively toward themselves as Americans do, but are
less inclined to evaluate themselves in an excessively positive manner.
Key words: competence, culture, self-esteem, self-love.
Introduction
People from East Asian countries score lower on self-report
measures of global self-esteem than do those from Western
countries (Schmitt & Allik, 2005).Although the effect itself
is well established, its interpretation remains unclear. It
could mean that East Asians like themselves less than do
Westerners, or it could also mean that East Asians like
themselves as much as Westerners, but are less inclined to
evaluate themselves in an excessively positive manner.
These possibilities highlight that global self-esteem
scales measure two related, but conceptually distinct,
aspects of self-worth: cognitively based self-evaluations
and affectively based feelings of self-regard (Tafarodi &
Swann, 1996). The former term refers to judgments about
one’s competencies, talents, and attributes (e.g. I am intel-
ligent or I am incompetent), whereas the latter refers to how
people feel about themselves (e.g. I am proud of myself or
I am ashamed of myself).1
In the present report, we use this distinction to clarify the
nature of cultural differences in self-esteem. Our research
was guided by three hypotheses. (a) First, we predicted that
cultural differences would be more apparent for cognitive
self-evaluations than for affective self-regard. (b) Second,
we predicted that cultural differences in cognitive self-
evaluations would mediate cultural differences in global
self-esteem. (c) Third, we predicted that norms of
modesty would underlie cultural differences in cognitive
self-evaluations.
With respect to the first of these hypotheses, cultural
differences are commonly found when it comes to cogni-
tively based self-evaluations. Across a range of attributes,
Americans, Canadians, and Western Europeans describe
themselves more positively than do East Asians. These
differences tend to be especially large when people evaluate
their agentic qualities (e.g. ‘How capable are you?’)
(Sedikides, Gaertner, & Toguchi, 2003; Sedikides, Gaert-
ner, & Vevea, 2005), but they also occur when people
evaluate their communal qualities (e.g. ‘How cooperative
are you?) (Heine, Kitayama, & Hamamura, 2007). Because
self-esteem scales tap these cognitively based judgments,
these differences could explain why Americans (and other
Westerners) score higher on self-esteem scales than do East
Asians.
Whether cultural differences emerge for items that assess
affective self-regard is less clear. Tafarodi and Swann
(1996) found that American college students scored higher
than Chinese college students on a measure of self-
competence, but the reverse was true for a measure of
self-liking (once self-competence had been statistically
removed). If we assume that self-liking is fundamentally
an affective construct, this finding suggests that EastAsians
do not feel worse about themselves than do Westerners.
Implicit measures of self-esteem also support this
hypothesis. Cultural differences are generally small or non-
existent when implicit attitudes toward the self are assessed
(Kitayama & Karasawa, 1997; Kobayashi & Greenwald,
2003; Yamaguchi et al., 2007). Because implicit attitudes
are thought to be more strongly influenced by affective
associations than cognitive evaluations (Gawronski &
Correspondence: Jonathon D. Brown, Department of Psychol-
ogy, Box 351525, University of Washington, Seattle, WA 98195-
1525, USA. Email: jdb@u.washington.edu
Portions of this research were presented at the 7th Annual Meeting
of the Society of Personality and Social Psychology, 26–28
January, 2006, Palm Springs, CA, USA.
Received 23 August 2006; accepted 10 January 2007.
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association
Asian Journal of Social Psychology (2007), 10, 162–170 DOI: 10.1111/j.1467-839X.2007.00222.x

Bodenhausen, 2006), the lack of cultural differences in
implicit self-esteem provides additional evidence that cul-
tural differences are small when we examine how people
feel about themselves.
Finally, there are theoretical reasons to question whether
cultures differ in affective self-regard. Self-love is pre-
sumed by many theorists to be a universal human motiva-
tion (Becker, 1968), and even researchers who emphasize
the importance of cross-cultural differences in self-
evaluations acknowledge that Asians like themselves every
bit as much as do Westerners (Heine, 2003; Heine et al.,
2007). To the extent that this is so, we should find small
cultural differences in affective self-regard.
China presents a particularly promising population for
investigating these issues. Partially owing to the single
child policy currently in effect, Chinese children receive a
great deal of parental attention and care (Wang & Ollen-
dick, 2001). Theoretically, this attention should foster a
secure attachment and instil feelings of self-love and affec-
tive self-regard (Bowlby, 1979). At the same time, the Con-
fucian tradition emphasizes modesty, deference, and self-
effacement. Early in life, Chinese children are discouraged
from bragging about themselves and are taught to avoid
self-aggrandizement, especially at the expense of others. In
concert, these childrearing traditions should shape the
expression of self-esteem, leading to high levels of self-love
but low levels of expressed self-competence (Tafarodi &
Swann, 1996).
Study 1
In an initial investigation, we asked American and Chinese
participants to complete three self-report measures that
assessed global self-esteem, cognitive self-evaluations, and
affective self-regard. We expected (a) to replicate the usual
tendency for East Asians to report lower self-esteem than
Westerners; (b) to find that cultural differences are greater
for cognitive self-evaluations than for affective self-regard;
and (c) to show that cultural differences in global self-
esteem are significantly reduced once cognitive self-
evaluations are statistically controlled.
Method
Participants
The American sample consisted of 36 undergraduates
attending the University of Washington (nine males). All
had identified themselves as being of European descent.
The Chinese sample consisted of 39 undergraduates
enrolled in introductory psychology courses at East China
Normal University (seven males). American students par-
ticipated in exchange for course credit, whereas Chinese
students were paid 10 Chinese yuan for their participation.
All participants were tested individually with question-
naires given via computer. All measures were presented in
random order.
Measures
The data used in this report were gathered as part of a larger
study of self-evaluations in America and China. All trans-
lations were performed by the first author and another indi-
vidual fluent in Chinese and English, with back-translations
conducted to ensure comparability.
Three questionnaires are of interest here. First, all par-
ticipants completed the Rosenberg Self-Esteem Scale
(Rosenberg, 1965). This scale is a well-validated and
widely used measure of global self-esteem. In the present
research, participants answered each of the 10 items using
a four-point Likert scale (0 =strongly disagree;
3=strongly agree). After reversing the scoring for five
negatively worded items, a total self-esteem score was
found by summing the 10 responses.
Two other self-report questionnaires were used to
supplement this measure. To assess affectively based feel-
ings of self-regard, participants were asked to indicate the
extent to which they generally feel four self-relevant emo-
tions (ashamed, humiliated, proud, pleased with myself).
Previous research has shown that these items are closely
related to, although not identical with, global self-esteem
(Brown & Dutton, 1995; Dutton & Brown, 1997; Brown &
Marshall, 2001; Bernichon, Cook, & Brown, 2003). Each
item was answered using a five-point Likert scale (1 =not
at all; 5 =very much) and, after reversing the scoring for
the negatively worded items, an affective self-regard scale
was formed by summing across the four items.
To assess cognitively based self-evaluations, participants
were asked to indicate how well eight attributes described
them. One-half of the items referred to positive qualities
(attractive, competent, intelligent, kind) and one-half
referred to negative qualities (unattractive, unfriendly, unin-
telligent, unkind). The participants indicated how well each
item described them using a seven-point Likert scale
(1 =not at all; 7 =very much). After reversing the scoring
for the negatively worded items, a cognitive self-evaluation
measure was formed by summing across the eight items.2
Results and discussion
Table 1 presents the reliabilities and intercorrelations
among the three measures, separately within each culture.
The reliabilities were generally acceptable, although the
measure of affective self-regard was less reliable in China
than in America. With respect to the correlations, the three
Self-esteem and culture 163
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association

measures tended to be highly correlated with one excep-
tion: In America, cognitive self-evaluations were unrelated
to affective self-regard. Finally, in both cultures, scores
on the Rosenberg Self-Esteem Scale were significantly
correlated with cognitive self-evaluations and affective
self-regard.
Rosenberg Self-Esteem Scale
Previous research has found that East Asians score lower on
self-report measures of self-esteem than do Americans,
Canadians, and Western Europeans (Schmitt & Allik,
2005). Replicating prior research, our Chinese participants
reported lower levels of global self-esteem (M=20.33)
than our European American participants (M=22.78),
t(73) =2.14, p<0.05, hp2=0.06.3
Cognitive self-evaluations and
affective self-regard
The present research was designed to distinguish between
cognitively based self-evaluations and affectively based
feelings of self-regard. To examine this issue, we submitted
the data to a 2 (Culture) ¥2 (Measure: Cognitive self-
evaluations vs Affective self-regard) mixed analysis of
variance (anova), with the latter factor treated as a
repeated measure. The anova revealed the predicted
Culture ¥Measure interaction, F1,73 =9.60, p<0.01,
hp2=0.12. As shown in Table 2, simple effects confirmed
that the Americans scored higher than the Chinese on the
cognitive self-evaluation measure, t(73) =2.60, p<0.01,
hp2=0.09, but the two groups scored comparably on the
affective measure of self-regard, t(73) =1.77, ns.4
Mediation
To this point, we have seen that Chinese score lower on the
Rosenberg Self-Esteem Scale than do European Americans,
and that cultural differences are greater for cognitive self-
evaluations than for affective self-regard. We have yet to
establish, however, that cultural differences in cognitive
self-evaluations mediate cultural differences in global self-
esteem.
We conducted a hierarchical regression analysis to
examine this issue. First, we used culture (dummy coded
0=America; 1 =China) and cognitive self-evaluations to
predict global self-esteem. Each predictor was centred
around its respective mean, and entered in Step 1. In Step 2,
we entered a cross-product term representing the interac-
tion between the two variables.
If cognitive self-evaluations mediate cultural differences
in self-esteem, we should find that the effect of culture on
self-esteem is reduced or eliminated once cognitive self-
evaluations are statistically controlled. The left-hand
column in Table 3 shows just such an effect. With
both predictors entered simultaneously, cognitive self-
evaluations predict self-esteem, but culture does not. A
Sobel’s test (Sobel, 1982) provided further evidence of
mediation, confirming that cognitive self-evaluations
significantly reduced cultural differences in global self-
esteem, Z=2.04, p<0.05. The lack of any Culture ¥
Cognitive self-evaluations interaction is also of interest, as
it suggests that the association between cognitive self-
evaluations, and global self-esteem does not vary between
the two cultures.
The right-hand side of Table 3 shows a comparable set of
analyses using affective self-regard as a predictor. Here,
both main effects are significant, indicating that cultural
differences in self-esteem are not due to cultural differences
Table 1 Reliabilities and intercorrelations among global self-esteem, cognitive self-evaluations, and affective self-
regard as a function of culture: Study 1
America China
Global
self-esteem
Cognitive
self-evaluations
Affective
self-regard
Global
self-esteem
Cognitive
self-evaluations
Affective
self-regard
Global self-esteem scale 0.89 0.54*** 0.40* 0.83 0.67*** 0.42**
Cognitive self-evaluations 0.74 0.01 0.82 0.54***
Affective self-regard 0.76 0.52
Within each country, reliabilities appear on the diagonal; correlations appear above the diagonal. *pⱕ0.05; **pⱕ0.01; ***pⱕ0.001.
Table 2 Cognitive self-evaluations and affective self-
regard as a function of culture: Study 1
Scale
Country
America China
Cognitive self-evaluations 5.90 (0.60) 5.54 (0.83)
Affective self-regard 3.68 (0.81) 3.93 (0.51)
Values in parentheses are standard deviations.
164 Huajian Cai et al.
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association

in affective self-regard. Finally, as before, the interaction is
not significant, indicating that the association between
affective self-regard and global self-esteem is the same in
China as in America.
Summary
The present study revealed two important findings. First,
although our Chinese participants appraised themselves
less favourably than our Americans on a cognitive measure
of self-evaluations, they did not appraise themselves less
favourably on an affective measure of self-regard. Second,
although cultural differences were found on a commonly
used measure of global self-esteem, these differences were
eliminated when cognitive self-evaluations were taken into
account. Collectively, these findings indicate that cultural
differences in self-regard are primarily cognitive in origin.
Admittedly, this interpretation rests on a null finding (i.e.
cultures didn’t differ on affective self-regard). Conceivably,
the lack of cultural differences on the affective self-regard
measure occurred not because Americans and Chinese feel
equally good about themselves (as we have argued), but
because the items don’t measure feelings of self-worth at
all. This interpretation is made more plausible by the
scale’s relatively low reliability.
Although we recognize these interpretations, we believe
other aspects of our data attest to the scale’s utility. First,
the face validity of the items is very high. Indeed, it would
be difficult to think of items that are more relevant to
affective self-regard than asking people to indicate how
proud of themselves, pleased with themselves, humiliated,
or ashamed of themselves they are. Second, these items
have been shown in previous research to be related to global
self-esteem (Brown & Dutton, 1995; Dutton & Brown,
1997; Brown & Marshall, 2001; Bernichon et al., 2003).
Finally, and most importantly, this was also true in the
present study. In China, as well as in America, self-esteem
and affective self-regard were significantly correlated (r=
0.42, p<0.01, and, r=0.40, p=0.01, for China and
America, respectively). It was not the case, then, that the
items failed to tap important components of self-regard.
Instead, it was simply the case that cultural differences in
affective self-regard were minimal.
Study 2
It is interesting to consider why East Asians are more
modest about their attributes and abilities than are Western-
ers. Two possibilities suggest themselves. First, they might
privately think they are just as competent as Americans,
but publicly refrain from saying so because of norms
of modesty (Kurman & Sriram, 2002; Kurman, 2003).
Alternatively, they might genuinely believe they lack
competence, perhaps because their culture encourages
people to acknowledge their weaknesses as a prelude to
self-improvement (Heine, Takata, & Lehman, 2000).
We conducted a second study to examine these issues.
In addition to completing measures of cognitive self-
evaluations and affective self-regard, we also had parti-
cipants complete a measure of modesty developed by
Whetstone, Okun, and Cialdini (1992) and used by Kurman
and colleagues (Kurman & Sriram, 2002; Kurman, 2003).
If cultural differences in self-competence are due to cultural
differences in modesty, these differences should be greatly
reduced or eliminated once cultural differences in modesty
are taken into account. Conversely, if cultural differences in
self-competence reflect true, private differences in per-
ceived competence, controlling for modesty ought not to
reduce cultural differences.
Method
Participants
The American sample comprised 64 undergraduates
attending the University of Washington (25 males), and the
Chinese sample comprised 68 undergraduates attending
East China Normal University (12 males). As in Study 1, all
Table 3 Hierarchical regression analyses predicting global self-esteem: Study 1
Cognitive self-evaluations Affective self-regard
bDR2bDR2
Step 1 0.39*** 0.22***
Culture -0.82 -3.23**
Self-regard 4.36*** 3.18**
Step 2
Culture ¥Self-regard -1.99 0.02 0.48 0.00
*pⱕ0.05; **pⱕ0.01; ***pⱕ0.001.
Self-esteem and culture 165
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association

of the students participated as part of a larger investigation
concerning self-evaluations in America and China, and
all were tested individually with questionnaires given via
computer.
Measures
Three measures were used in this investigation. As in Study
1, participants completed the cognitive measure of self-
evaluations and the affective measure of self-regard. Par-
ticipants also completed the nine-item Inclination Toward
Modesty subscale of the Modest Responding Scale (Whet-
sone et al., 1992).5Sample items include ‘I believe it is
impolite to talk excessively about one’s achievements, even
if they are outstanding’. and ‘It’s difficult for me to talk
about my strengths to others even when I know I possess
them’. Participants answered each item using a seven-point
Likert scale with appropriate endpoints. Finally, the three
measures were presented in counterbalanced order.
Results and discussion
Cognitive self-evaluations and
affective self-regard
After reversing the scoring for the negatively worded items
in each scale, we formed measures of cognitive self-
evaluations and affective self-regard. The internal reliabili-
ties of the two scales were similar to the values found
in Study 1. For the cognitive self-evaluation measure,
a=0.79 and a=0.81, for America and China, respectively.
For the affective self-regard measure, a=0.63 and
a=0.48, for America and China, respectively. Finally, in
contrast to the findings of Study 1, cognitive self-
evaluations and affective self-regard were significantly cor-
related in America (r=0.35, p<0.01) and in China (r=
0.57, p<0.001).
Next, we performed a 2 (Culture) ¥2 (Measure: Cogni-
tive self-evaluation vs affective self-regard) mixed anova,
with the latter factor treated as a repeated measure. Repli-
cating our earlier results, the anova revealed the predicted
Culture ¥Measure interaction, F1,130 =12.39, p=0.001,
hp2=0.09. Table 4 shows that, as in Study 1, theAmericans
scored higher than the Chinese on the self-evaluation
measure, t(130) =3.85, p<0.01, hp2=0.10, but the two
groups scored comparably on the affective measure of self-
regard, t(130) =1.13, ns.
Modesty
It is widely recognized that cultural norms of modesty are
stronger in East Asian countries than in North American
countries and the countries of Western Europe (Markus &
Kitayama, 1991). Accordingly, we predicted that our
Chinese participants would score higher on a measure of
modesty than would our American participants. This
proved to be the case. After summing participants’
responses to the nine-item modesty scale, we found that
our Chinese participants reported being more modest
(M=38.02) than our American participants (M=33.50),
t(130) =2.80, p<0.01, hp2=0.06.
Mediation
The preceding results set the stage for our primary analyses.
If cultural differences in cognitive self-evaluations depend
on cultural differences in modesty, we should find that the
effect of culture on cognitive self-evaluations is reduced or
eliminated once modesty is statistically controlled. To test
this hypothesis, we performed a hierarchical regression
analysis on cognitive self-evaluations, using mean centred
predictors of culture (dummy coded: 0 =America,
1=China) and modesty as predictors. The main effect of
modesty was highly significant (b=-0.03, p<0.001), but
the main effect of culture was not (p>0.10), suggesting
that cultural differences in modesty can explain cultural
differences in cognitive self-evaluations. A Sobel’s test
(Sobel, 1982) confirmed this interpretation, showing that
cultural differences in self-evaluations were reduced when
modesty was statistically controlled, Z=2.40, p<0.05.
Finally, the Culture ¥Modesty interaction did not even
approach significance (p>0.40), establishing that the rela-
tion between modesty and cognitive self-evaluations is
comparable across cultures.
Correlational analyses
Insofar as culture predicts cognitive self-evaluations and
modesty, but not affective self-regard, it would seem that
modesty is more strongly related to cognitive self-
evaluations than affective self-regard. After first determin-
ing that the correlations did not differ between the
American and Chinese samples, we tested this hypothe-
sis by collapsing across cultures and computing the
Table 4 Cognitive self-evaluations and affective self-
regard as a function of culture: Study 2
Scale
Country
America China
Cognitive self-evaluations 5.81 (0.65) 5.49 (0.73)
Affective self-regard 3.86 (0.63) 3.96 (0.47)
Values in parentheses are standard deviations.
166 Huajian Cai et al.
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association

correlations between modesty and the two forms of self-
regard. As predicted, modesty was more strongly correlated
with cognitive self-evaluations (r=-0.41) than with affec-
tive self-regard (r=-0.21), Z=2.26, p<0.025. Although
this finding might be due to the specific items on the scale
we used to measure modesty, it may also be that cultural
norms of modesty apply mostly to boasts about one’s com-
petencies and accomplishments, rather than to descriptions
about how one feels about oneself. Said differently, cultures
seem to differ more with regard to their tolerance for brag-
gadocio than their proscriptions against liking oneself.
It is also possible that affective self-regard is more stable
than cognitive self-evaluations. A great number of psy-
chologists of diverse theoretical stripe have argued that
affective self-regard forms early in life in response to tem-
peramental and relational factors (Erikson, 1963; Bowlby,
1979). In contrast, self-evaluations are thought to be more
variable and influenced by a host of contextual factors, such
as the social setting, recent events, priming effects, or frame
of reference. Modesty might constitute an additional con-
textual factor that exerts a greater influence on cognitive
self-evaluations than affective self-regard.
General discussion
Although previous research has found that people from
East Asian countries report lower levels of global self-
esteem than do those from Western countries (Schmitt &
Allik, 2005), the source of this difference has not been
clearly identified. In the present paper, we examined
whether the origins of this difference are primarily cogni-
tive or affective. Our findings point to cognitive factors.
Although our Chinese participants appraised themselves
less favourably than our Americans on a cognitive measure
of self-evaluation, they did not appraise themselves less
favourably on an affective measure of self-regard. More-
over, cultural differences in global self-esteem were elimi-
nated once cognitive self-evaluations were statistically
controlled. Taken together, these findings suggest that cul-
tural differences in global self-esteem reflect cognitive
rather than affective factors.
Study 2 examined why people from East Asian cultures
evaluate themselves less positively than do Americans. We
hypothesized that cultural differences in norms of modesty
underlie these differences. In accordance with our predic-
tions, our Chinese participants scored higher on a measure
of modesty than did our American participants, and cultural
differences in cognitive self-evaluations were greatly
reduced once modesty scores were statistically controlled.
Along with other research (Kurman & Sriram, 2002;
Kurman, 2003), these findings suggest that cultural norms
of modesty temper cognitive self-evaluations in China.
Of course, one could also conclude that cultural norms
promote self-aggrandizement in America, and that the
public self-evaluations of Americans are no less genuine
than are the public self-evaluations of Chinese. Unfortu-
nately, the modesty measure we used does not provide a
way to test this hypothesis. One could, however, experi-
mentally induce bragging and modesty in a laboratory
study, and then determine whether American or Chinese
participants are more affected by these manipulations.
Limitations
Before considering the implications of our findings, we
wish to call attention to some limitations. First, like all
cross-cultural research, subtle differences in translation and
context can influence the findings (Heine, Lehman, Peng, &
Greenholtz, 2002). Second, we assessed only explicit atti-
tudes that can be influenced by a lack of introspective
awareness or self-presentational biases. Although there is
reason to believe our findings would replicate when implicit
attitudes are assessed (a point we consider in more detail
below), we cannot be certain this is true.
Third, our sample sizes were rather small, and we studied
only college students. Whether our findings apply to older
adults or young adults is an unanswered question. We also
failed to study different ethnic groups within China, leaving
open the question of whether our findings apply to some
groups but not others (Kashima et al., 2004). The same is
true with our American sample. The category ‘European
American’ covers a lot of territory, and it may be that only
some people of European descent tout their competence.
For example, people from some of the Scandinavian coun-
tries (e.g. Sweden, Norway, or Finland) might be less self-
congratulatory (Silvera & Seger, 2004). We also have no
way of knowing whether our results apply to other East
Asian countries. This limitation may be particularly rel-
evant to Japan. Most cross-cultural research on self-
enhancement and self-esteem has compared Japan and
America (Heine, Lehman, Markus, & Kitayama, 1999).
Whether our findings would replicate in Japan is an inter-
esting topic for future research.
It is also important to note that our scales were quite
short, and our measure of cognitive self-evaluations
included very few communal traits. This issue is important
because some studies have found that cultural differences
are less pronounced for communal qualities than agentic
ones (Kurman, 2001; Brown & Kobayashi, 2002; Sedikides
et al., 2003). Although not all studies find such an effect
(Heine et al., 2007), it would be interesting to determine
whether our findings hold true with a measure that included
more communal traits.
Our results also appear to be inconsistent with findings
reported by Eid and Diener (2001). These investigators
found that Chinese college students felt it was less desirable
Self-esteem and culture 167
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association

and appropriate to experience pride than did American
college students. The Chinese students also reported feeling
proud less often and intensely than the Americans.
Although we can’t say for sure, we suspect that asking
participants ‘how appropriate is it to feel proud?’ might
have led them to construe this emotion in terms of public
conceit rather than a more personal self-love.
Finally, it’s important to note that some of the arguments
we have advanced in this paper have been offered else-
where. For example, Tafarodi and Swann (1996) have
argued that cultural differences are stronger for self-
competence than for self-liking, and Kurman and col-
leagues have argued that modesty can explain cultural
differences in self-evaluations (Kurman & Sriram, 2002;
Kurman, 2003). Our findings extend and integrate these
strands of research by showing that: (a) cultural differences
typically found using the Rosenberg Self-Esteem
Scale -which is the most common measure of global self-
esteem -are primarily due to cognitive self-evaluations
rather than affective self-regard; and (b) modesty is more
closely linked with cognitive self-evaluations than affective
self-regard in China and America.
Implications
We believe our findings carry some important implications.
First, they speak to the universal nature of self-love. Virtu-
ally all theories of human motivation accord a central role
to a general need for positive self-feelings (Maslow, 1943;
Rogers, 1951; Becker, 1968; Deci & Ryan, 1995). Support-
ing this contention, our Chinese participants reported liking
themselves every bit as much as our European American
participants. Along with other findings, these results
suggest that self-love is as strong in East Asia as in North
America and the countries of Western Europe (Brown,
2003; Heine, 2003; Sedikides et al., 2003).
We would not have reached this conclusion had we only
examined scores on the Rosenberg Self-Esteem Scale.
Because the Rosenberg scale assesses cognitive self-
competence and affective self-regard, cultural differences
are usually found when self-esteem is measured using this
scale (Schmitt & Allik, 2005). When we statistically con-
trolled for the effects of self-competence, however, leaving
only the affective component of self-regard, cultural differ-
ences on the Rosenberg scale disappeared. This finding
supports our claim that cultural differences in self-esteem
arise from cultural differences in self-evaluations, with
people from East Asian countries evaluating themselves
less positively than people from Western countries.
This facet of our data also calls attention to the need to
distinguish self-love from self-competence (Tafarodi &
Swann, 1995, 1996). By extension, it also highlights the
need to distinguish self-esteem from self-evaluations. Many
self-esteem scales include subscales that measure self-
evaluations in individual domains (Shavelson, Hubner, &
Stanton, 1976; Harter, 1990; Marsh, 1993). These scores
are not synonymous with having high self-esteem, as many
people who believe they are competent do not love them-
selves, and many people who love themselves do not boast
about their competencies (Brown & Marshall, 2006). Our
findings suggest that cross-cultural differences will be most
pronounced when self-evaluations are used as an index of
global self-esteem.
It was not the case, however, that the correlates of global
self-esteem varied from one culture to the next (Table 3).
Instead, both measures of self-regard predicted global self-
esteem to a comparable degree in China and in America.
Thus, even though cognitive self-evaluations are lower in
China than in America, they are not less predictive of global
self-esteem. This finding suggests that global self-esteem is
experienced similarly across dissimilar cultures (Tafarodi
& Swann, 1996; Kobayashi & Brown, 2003; Brown, Cai,
Oakes, & Deng, unpubl data, 2007).
In emphasizing their comparatively low self-evaluations,
we do not wish to leave the impression that our Chinese
participants were critical of themselves. In fact, their cog-
nitive self-evaluations fell well above the scale mid-point in
both of our studies. It was only in comparison with the
highly favourable self-views of Westerners that they
appeared self-deprecating (see also Falbo et al., 1997;
Brown & Kobayashi, 2002; Sedikides et al., 2003).
The present findings are also consistent with accumulat-
ing evidence from implicit measures of self-regard. Cul-
tural differences are rarely found when implicit measures of
self-regard are assessed (Kitayama & Karasawa, 1997;
Kobayashi & Greenwald, 2003; Yamaguchi et al., 2007).
This finding is readily interpretable if we conclude, as
others have argued, that implicit attitude measures princi-
pally tap the affective component of an attitude, and explicit
attitude measures reflect affective and cognitive compo-
nents (Gawronski & Bodenhausen, 2006). In terms of our
research, implicit measures of self-esteem may be said to
assess affective self-regard rather than cognitive self-
evaluations. Because cultures differ little in affective
self-regard, cultural differences in implicit measures of
self-esteem will generally be small or non-existent.
Our findings also bear on the cross-cultural correlates of
self-esteem. Diener and Diener (1995) found that self-
esteem was a better predictor of subjective well-being in
Western countries than in East Asian countries. These
researchers speculated that self-esteem may simply be less
important in East Asian countries, a conclusion shared by
other theorists (Markus & Kitayama, 1991; Heine et al.,
1999). Whether this conclusion applies to cognitive self-
evaluations or affective self-regard is an important topic for
future research.
Finally, we think it’s important to emphasize that a con-
sensus now exists on the nature of cultural differences in
168 Huajian Cai et al.
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association

self-esteem, with substantial agreement on the following
facts: (a) people the world over strive to feel good about
themselves; (b) cultures dictate what qualities people
should have in order to feel good about themselves; (c)
Asians score lower on self-esteem scales than do Western-
ers; (d) these differences arise because Asians evaluate their
characteristics less positively, not because they like them-
selves less than do Westerners; and (e) cultural differences
in self-evaluations depend, at least in part, on cultural dif-
ferences in norms of modesty (Brown, 2003; Heine, 2003;
Kurman, 2003; Heine et al., 2007). With these conclusions
now established, researchers are better positioned to
examine the precise ways in which cultures shape self-
evaluations and self-feelings.
End notes
1. Throughout this paper we are careful to distinguish among the
following three terms: global self-esteem (i.e. overall feelings
of self-acceptance and self-love); cognitive self-evaluations
(i.e. specific judgments about one’s abilities, attributes, or
traits); and affective self-regard (i.e. emotions that express how
proud or ashamed one is of oneself).
2. Preliminary analyses showed that scale valence did not interact
with any of the findings of interest in this article. Consequently,
we ignore this variable in all reported analyses.
3. In both studies reported in this paper, preliminary analyses
revealed no significant gender effects and the data are pre-
sented collapsed across this variable.
4. The two measures were assessed using two different scales in
order to minimize response biases. As a consequence, the
anova revealed a theoretically meaningless main effect of
measure, with scores on the cognitive measure being greater
than scores on the affective measure, F1,73 =372.43, p<0.001.
5. We used this subscale because of its face validity and because
Kurman and Sriram (2002) found it had the highest internal
reliability of the three subscales that comprise the overall
measure. The internal reliabilities in our sample were also high
(American, a=0.82; China, a=0.86).
References
Becker, E. (1968). The Structure of Evil. New York: George
Braziller.
Bernichon, T., Cook, K. E. & Brown, J. D. (2003). Seeking
self-evaluative feedback: The interactive role of global self-
esteem and specific self-views. Journal of Personality and
Social Psychology, 84, 194–204.
Bowlby, J. (1979). The Making and Breaking of Affectional
Bonds. London: Tavistock.
Brown, J. D. (2003). The self-enhancement motive in collectivis-
tic cultures: The rumors of my death have been greatly
exaggerated. Journal of Cross-Cultural Psychology, 34, 603–
605.
Brown, J. D. & Dutton, K. A. (1995). The thrill of victory, the
complexity of defeat: Self-esteem and people’s emotional reac-
tions to success and failure. Journal of Personality and Social
Psychology, 68, 712–722.
Brown, J. D. & Kobayashi, C. (2002). Self-enhancement in Japan
and America. Asian Journal of Social Psychology, 5, 145–
167.
Brown, J. D. & Marshall, M. A. (2001). Self-esteem and emotion:
Some thoughts about feelings. Personality and Social Psychol-
ogy Bulletin, 27, 575–584.
Brown, J. D. & Marshall, M. A. (2006). The three faces of self-
esteem. In: M. Kernis, ed. Self-Esteem: Issues and Answers, pp.
4–9. New York: Psychology Press.
Deci, E. L. & Ryan, R. M. (1995). Human autonomy: The basis
for true self-esteem. In: M. H. Kernis, ed. Efficacy, Agency, and
Self-Esteem, pp. 31–49. New York: Plenum Press.
Diener, E. & Diener, M. (1995). Cross-cultural correlates of life
satisfaction and self-esteem. Journal of Personality and Social
Psychology, 68, 653–663.
Dutton, K. A. & Brown, J. D. (1997). Global self-esteem and
specific self-views as determinants of people’s reactions to
success and failure. Journal of Personality and Social Psychol-
ogy, 73, 139–148.
Eid, M. & Diener, E. (2001). Norms for experiencing emotions in
different cultures: Inter- and intranational differences. Journal
of Personality and Social Psychology, 81, 869–885.
Erikson, E. H. (1963). Childhood and Society, 2nd edn. NewYork:
W. W. Norton.
Falbo, T., Poston, D. L. Jr, Triscari, R. S. & Zhang, X. (1997).
Self-enhancing illusions among Chinese schoolchildren.
Journal of Cross-Cultural Psychology, 28, 172–191.
Gawronski, B. & Bodenhausen, G. V. (2006). Associative and
propositional processes in evaluation: An integrative review of
implicit and explicit attitude change. Psychological Bulletin,
132, 692–731.
Harter, S. (1990). Causes, correlates, and the functional role of
global self-worth: A life-span perspective. In: R. J. Sternberg &
J. Kolligian, eds. Competence Considered, pp. 66–97. New
Haven, CT: Yale University Press.
Heine, S. J. (2003). Making sense of East Asian self-enhancement.
Journal of Cross-Cultural Psychology, 34, 596–602.
Heine, S. J., Kitayama, S. & Hamamura, T. (2007). Inclusion of
additional studies yields different conclusions: Comment on
Sedikides, Gaertner and Vevea (2005), Journal of Personality
and Social Psychology.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? Psy-
chological Review, 106, 766–794.
Heine, S. J., Lehman, D. R., Peng, K. & Greenholtz, J. (2002).
What’s wrong with cross-cultural comparisons of subjective
Likert scales?: The reference-group effect. Journal of Person-
ality and Social Psychology, 82, 903–918.
Heine, S. J., Takata, T. & Lehman, D. R. (2000). Beyond self-
presentation: Evidence for self-criticism among Japanese. Per-
sonality and Social Psychology Bulletin, 26, 71–78.
Kashima, Y., Kokubo, T., Kashima, E. S., Boxall, D., Yamgauchi,
S. & Macrae, K. (2004). Culture and self: Are there
Self-esteem and culture 169
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association

within-culture differences in self between metropolitan areas
and regional cities? Personality and Social Psychology Bulle-
tin, 30, 816–823.
Kitayama, S. & Karasawa, M. (1997). Implicit self-esteem in
Japan: Name letters and birthday numbers. Personality and
Social Psychology Bulletin, 23, 736–742.
Kobayashi, C. & Brown, J. D. (2003). Self-esteem and self-
enhancement in Japan and America. Journal of Cross-Cultural
Psychology, 34, 567–580.
Kobayashi, C. & Greenwald, A. G. (2003). Implicit-explicit dif-
ferences in self-enhancement for Americans and Japanese.
Journal of Cross-Cultural Psychology, 34, 522–541.
Kurman, J. (2001). Self-enhancement: Is it restricted to individu-
alistic cultures? Personality and Social Psychology Bulletin,
27, 1705–1716.
Kurman, J. (2003). Why is self-enhancement low in certain col-
lectivist cultures? An investigation of two competing explana-
tions. Journal of Cross-Cultural Psychology, 34, 496–510.
Kurman, J. & Sriram, N. (2002). Interrelationships among vertical
and horizontal collectivism, modesty, and self-enhancement.
Journal of Cross-Cultural Psychology, 33, 71–86.
Markus, H. R. & Kitayama, S. (1991). Culture and the self: Impli-
cations for cognition, emotion, and motivation. Psychological
Review, 98, 224–253.
Marsh, H. W. (1993). Academic self-concept: Theory, measure-
ment, and research. In: J. Suls, ed. Psychological Perspectives
on the Self, Vol. 4, pp. 59–98. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Maslow, A. (1943). A theory of human motivation. Psychological
Review, 50, 370–396.
Rogers, C. R. (1951). Client-Centered Therapy. New York:
Houghton Mifflin.
Rosenberg, M. (1965). Society and the Adolescent Self-Image.
Princeton, NJ: Princeton University Press.
Schmitt, D. P. & Allik, J. (2005). Simultaneous administration of
the Rosenberg Self-Esteem Scale in 53 nations: Exploring the
universal and culture-specific features of global self-esteem.
Journal of Personality and Social Psychology, 89, 623–
642.
Sedikides, C., Gaertner, L. & Toguchi, Y. (2003). Pancultural
self-enhancement. Journal of Personality and Social Psychol-
ogy, 84, 60–79.
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.
Shavelson, R. J., Hubner, J. J. & Stanton, G. C. (1976). Self-
concept: Validation of construct interpretations. Review of Edu-
cational Research, 46, 407–441.
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.
Sobel, M. E. (1982). Asymptotic confidence intervals for indirect
effects in structural equation models. In: S. Leinhart, ed. Socio-
logical Methodology, pp. 290–312. San Francisco, CA: Jossey-
Bass.
Tafarodi, R. W. & Swann, W. B. Jr (1995). Self-liking and
self-competence as dimensions of global self-esteem: Initial
validation of a measure. Journal of Personality Assessment, 65,
322–342.
Tafarodi, R. W. & Swann, W. B. Jr (1996). Individualism-
collectivism and global self-esteem: Evidence for a cultural
trade-off. Journal of Cross-Cultural Psychology, 27, 651–672.
Wang, Y. & Ollendick, T. H. (2001). A cross-cultural and
developmental analysis of self-esteem in Chinese and Western
children. Clinical Child and Family Psychology Review, 4,
253–271.
Whetsone, M. R., Okun, M. A. & Cialdini, R. B. (1992). The
modest responding scale. Paper presented at the convention of
the American Psychological Society, San Diego, CA.
Yamaguchi, S., Greenwald, A. G., Banaji, M. R., et al. (2007).
Apparent universality of positive implicit self-esteem. Psycho-
logical Science, 18, 498–500.
170 Huajian Cai et al.
© 2007 The Authors
© 2007 Blackwell Publishing Ltd with the Asian Association of Social Psychology and the Japanese Group Dynamics Association