Cogn Ther Res (2012) 36:1–14

DOI 10.1007/s10608-011-9376-9

ORIGINAL ARTICLE

A Test of the Usefulness of Perfectionism Theory Across Cultures:

Does Perfectionism in the US and Japan Predict Depressive
Symptoms Across Time?
Edward C. Chang • Rita Chang • Lawrence J. Sanna

Published online: 26 June 2011

Ó Springer Science+Business Media, LLC 2011

Abstract The present study examined cultural variations differentiate between perfectionistic standards stemming
on performance perfectionism (Chang in Cogn Therapy from within the individual and perfectionistic standards
Res 30:677–697, 2006; Cogn Therapy Res 33:334–344, stemming from an outside source. For example, Hewitt
2009) in 168 European American and 151 Japanese college and Flett’s (1991) model of perfectionism consists of
students. Results of between-groups analyses on perfor- three dimensions: self-oriented perfectionism, socially
mance perfectionism provided support for the general prescribed perfectionism, and other-oriented perfectionism.
notion of self-enhancement in the West and self-criticism Similarly, Frost et al.’s (1990) model consists of the fol-
in the East. Moreover, performance perfectionism was lowing dimensions: concerns over mistakes, doubts about
found to be associated with concurrent and prospective actions, personal standards, parental expectations, parental
(2 months later) depressive symptomatology in both cul- criticism, and organization. In studies of perfectionism,
tural groups. Results of regression analyses for both researchers have been able to identify a consistent link
European Americans and Japanese indicated that negative between perfectionism and maladjustment, though Shafran
self-oriented performance perfectionism predicted changes and Mansell (2001) have argued that self-oriented forms of
in depressive symptomatology at Time 2 even after con- perfectionism are more representative of the clinical con-
trolling for initial symptomatology at Time 1. Implications struction of perfectionism. Still, it is unclear which forms
of these findings for future research are discussed. of perfectionism are more deleterious to psychological
well-being. In student and clinical samples, socially pre-
Keywords Cultural differences
Perfectionism
scribed perfectionism has been linked to depressive
Depressive symptoms
Personality symptoms, but self-oriented perfectionism has also been
shown to be elevated in clinical samples and to interact
with stress in predicting depressive symptoms (Enns and
Introduction Cox 1999; Hewitt et al. 1996). Additionally, self-oriented
perfectionism (but not socially prescribed perfectionism)
In recent decades, a growing number of researchers have has been linked to low self-efficacy, which in turn may be
begun to focus their attention on the study of perfectionism, linked to depressive and anxious symptoms when indi-
a personality variable commonly marked by high standards viduals experience low levels of personal control (Hart
of performance (Burns 1980; Frost et al. 1990). There are et al. 1998). Finally, both self-oriented and socially pre-
several existing models of perfectionism, and most of them scribed perfectionism have been associated with suicidal
ideation (Hewitt et al. 1994).
Independent of the line of research on perfectionism, a
similar trend has also emerged in the literature involving
the study of outcome cognitions, thoughts involving posi-
E. C. Chang (&)
R. Chang
L. J. Sanna
tive and negative perceptions of past, present, and future
Department of Psychology, University of Michigan,
530 Church Street, Ann Arbor, MI 48109, USA outcomes (Bandura 1977; Scheier and Carver 1985; Snyder
e-mail: changec@umich.edu et al. 1991). Similar to findings on perfectionism, outcome

123
2 Cogn Ther Res (2012) 36:1–14

cognitions like optimism and pessimism have also been performance perfectionism and negative socially pre-
found to have reliable associations with adjustment (Chang scribed performance perfectionism). In contrast, for
2001; Hardin and Leong 2005; Snyder 2000), such that example, Hewitt and Flett’s (1991) original model pointed
individuals with high levels of optimism have higher levels to both self-oriented and socially prescribed perfectionism
of life satisfaction and positive affect and individuals with as potentially maladaptive.
high levels of pessimism have higher levels of depressive Positive self-oriented performance perfectionism is
symptoms and negative affect (e.g., Marshall et al. 1992). defined by high personal standards of performance that
Some researchers have suggested that optimism and pes- involve positive outcome cognitions for the individual
simism may influence psychological well-being by affect- (e.g., ‘‘My performance is always made better by the high
ing the way individuals cope (e.g., optimists may use more standards I set for myself’’). Negative self-oriented per-
active coping methods) when they encounter life stressors formance perfectionism is defined by high personal stan-
(Nes and Segerstrom 2006). dards of performance that involve negative outcome
cognitions for the individual (e.g., ‘‘I accomplish nothing
when I try to meet the high standards I set for myself’’).
Positive socially prescribed performance perfectionism is
Performance Perfectionism: An Integration of Positive defined by perceived high standards of performance placed
and Negative Outcome Cognitions to High Standards on an individual by others that involve positive outcome
of Performance cognitions for the individual (e.g., ‘‘My performance is
always made better when others expect more from me’’).
Most perfectionism researchers have assumed that they are Negative socially prescribed performance perfectionism is
studying a personality variable that is essentially mal- defined by perceived high standards of performance placed
adaptive in nature (see Chang 2003, for a review). But on an individual by others that involve negative outcome
recently a growing number of investigators have raised the cognitions for the individual (e.g., ‘‘My work suffers when
possibility that perfectionism may involve both adaptive others push me to meet their high standards’’). Consistent
and maladaptive aspects (e.g., Fedewa et al. 2005; Rice and with this conceptualization, results from a factor analysis
Ashby 2007; Slade and Owens 1998; Slaney et al. 2001; conducted on items used to develop the Performance Per-
Stoeber and Otto 2006). For example, Frost et al. (1993) fectionism Scale or PPS indicated a four-factor solution
found that certain dimensions of perfectionism (e.g., per- reflecting these respective four dimensions (Chang 2006,
sonal standards) were unrelated to depressive symptoms Study 1b). Moreover, scores on the PPS were not found to
and negative affect and positively related to positive affect. be redundant with scores obtained from alternative mea-
Other studies have drawn links between ‘‘healthy’’ per- sures of personality, including measures of neuroticism,
fectionism and traits like extraversion and conscientious- optimism, and social desirability (Chang 2006, Study 2).
ness (e.g., Parker and Stumpf 1995; Stumpf and Parker Accordingly, the PPS is composed of four distinguishable
2000). In considering that possibility and the absence of a (and reliable) scales that tap for positive and negative self-
theoretically driven and inclusive framework to capture oriented performance perfectionism and positive and neg-
adaptive and maladaptive dimensions of perfectionism in ative socially prescribed performance perfectionism
the literature, Chang (2006, 2009) proposed and found (Chang 2006, Study 1a and Study 1b). Furthermore, con-
empirical support across a series of studies for a novel sistent with the notion that the PPS adds value over other
construct called performance perfectionism. Performance popular perfectionism measures, Chang (2006, Study 3b)
perfectionism is thought to involve adaptive and mal- found that PPS scores accounted for significant additional
adaptive aspects based on the integration of positive and variance in measures of psychological functioning beyond
negative outcome cognitions to high standards of perfor- what was accounted for by the Frost Multidimensional
mance, respectively. Because performance perfectionism is Perfectionism Scale (Frost et al. 1990).
determined by the source with high standards of perfor- It is a topic of debate whether self-oriented or socially
mance (self-oriented vs. socially prescribed) and by the prescribed perfectionism plays a larger role in the well-being
valence (positive vs. negative) of the outcome cognitions and functioning of adults. Indeed, a review of the literature
involved, four distinguishable aspects of performance indicates a mixed pattern (Shafran and Mansell 2001).
perfectionism are thought to exist. As shown in Fig. 1, two Studies based on Hewitt and Flett’s (1991) model and
aspects of performance perfectionism are considered to be measure have generally indicated support for the greater role
adaptive (positive self-oriented performance perfectionism of socially-prescribed over self-oriented dimensions of
and positive socially prescribed performance perfection- perfectionism, whereas those based on Frost et al.’s (1990)
ism), and two aspects of performance perfectionism are model and measure have generally indicated support
considered to be maladaptive (negative self-oriented for the greater role of self-oriented over socially-prescribed

123
Cogn Ther Res (2012) 36:1–14 3

Fig. 1 A distal–proximal
model of performance
perfectionism demonstrating the
interrelations among socially
prescribed standards,
self-oriented high standards and
psychological outcomes
(reprinted from Chang et al.
2006, p. 57)

dimensions of perfectionism. Nonetheless, between socially avoidance motives, would be more strongly associated with
prescribed and self-oriented variants of performance negative outcomes and conditions (Slade and Owens 1998).
perfectionism, it is believed that self-oriented aspects of Consistent with this expectation, findings from studies
performance perfectionism play a greater role in adult involving young adults have demonstrated that positive, but
functioning than do socially prescribed aspects (Chang et al. not negative, self-oriented performance perfectionism is
2006). For example, consistent with cognitive models link- positively and strongly associated with important measures
ing self-focused attention with emotional disorders and of positive psychological functioning (Chang 2006, Study
distress in adults, results from numerous studies attempting 3b) and positive outcome (Chang 2006, Study 4). For
to identify vulnerability factors associated with maladjust- example, after partialing out the influence of the other three
ment have consistently implicated negative self-referent performance perfectionism dimensions, positive self-ori-
thoughts as a major etiological factor (Mor and Winquist ented performance perfectionism was the only dimension
2002). Accordingly, one should find self-oriented aspects, found to be significantly associated with purpose in life
and not socially prescribed aspects, of performance perfec- (pr = .44). Alternatively, recent findings involving young
tionism to matter most in determining important outcomes adults have demonstrated that negative, but not positive, self-
and conditions. Consistent with this expectation, findings oriented performance perfectionism is positively and
from recent adult studies have demonstrated that self- strongly associated with important measures of negative
oriented aspects, but not socially prescribed aspects, of psychological functioning (Chang 2006, Study 3a). For
performance perfectionism have unique associations with a example, after partialing out the influence of other three
number of important measures of negative psychological performance perfectionism dimensions, negative self-ori-
functioning (e.g., depressive symptoms, worry, and anxiety; ented performance perfectionism was the only dimension
Chang 2006, Study 3a), positive psychological functioning found to be significantly associated with depressive symp-
(e.g., life satisfaction, self-acceptance, personal growth, and toms (pr = .24). However, studies on performance perfec-
purpose in life; Chang 2006, Study 3b), and positive out- tionism have thus far been limited to the study of Westerners
come (e.g., exam performance and final course grade; Chang (i.e., European Americans). This represents a critical limi-
2006, Study 4). tation for understanding the general usefulness of perfor-
Moreover, between positive and negative self-oriented mance perfectionism given that potential cultural differences
aspects of performance perfectionism it would seem rea- between Westerners and Easterners have been implicated in
sonable to expect that the former, involving approach past studies of general models of (maladaptive) perfection-
motives, would be more strongly associated with positive ism (Chang 1998; Kawamura et al. 2002) and of outcome
outcomes and conditions, whereas the latter, involving cognitions (Chang 1996, 2002a; Zane et al. 1991).

123
4 Cogn Ther Res (2012) 36:1–14

Cultural Variations in Performance Perfectionism expectations about one’s high standards), whereas because
and Relations to Depressive Symptoms Easterners have a pessimistic and self-critical bias, they
may be more likely to have high levels of negative self-
In recent years, researchers have become increasingly oriented performance perfectionism (which involves neg-
aware that there may be important cultural differences ative outcome expectations about one’s standards).
between individuals from predominantly Western versus Beyond an examination of mean differences on perfor-
predominantly Eastern cultures (Kim et al. 1994; Triandis mance perfectionism, the present study sought to investi-
1995). Several studies have indicated that Easterners tend gate the relationship between performance perfectionism
to be more collectivistic and have a self-construal that and depressive symptomatology. Although past studies
encompasses the members of the ‘‘in-group’’ to which they have indicated an important link between general (mal-
belong, whereas Westerners tend to be more individualistic adaptive) perfectionism and depressive symptomatology
and have a self-construal that is autonomous of their social (Shafran and Mansell 2001), findings from these studies
context (Markus and Kitayama 1991; Oyserman et al. have been based on using measures of perfectionism that
2002). In traditional Eastern culture, much emphasis is often already refer to some expression of negative affect
placed on maintaining interpersonal harmony, which has (e.g., frustration, anger, irritableness). In contrast, perfor-
led some researchers to suggest that Easterners may have mance perfectionism is based on a model and a measure
higher levels of self-criticism and pessimism compared that does not explicitly focus on the specific involvement of
with Westerners because Eastern culture encourages neg- negative affective conditions within high standards of
ative self-cognitions to the end of improving oneself and performance (Chang 2006). Given this important concep-
decreasing one’s chances of offending other individuals tual and methodological difference, it is worth noting that
(Chang and Chang 2009; Masuda and Nisbett 2001; depressive symptomatology has been found in one study to
Miyamoto et al. 2006). In a recent set of studies investi- have the strongest association with negative self-oriented
gating cultural variations in comparative predictions made performance perfectionism (Chang 2006, Study 3a). Yet,
for future positive and negative life events between this finding was based on a predominantly European
Westerners and Easterners, Chang and Asakawa (2003) American sample. Therefore, it is unclear if the association
found that European Americans held an optimistic bias for between negative self-oriented perfectionism and depres-
positive outcomes. Specifically, European Americans pre- sive symptomatology also exists in Easterners. Findings
dicted that positive events (e.g., ‘‘Meet someone new with from some studies have shown that maladaptive aspects of
whom you expect to be close friends for year’’) were more perfectionism that focus on self-critical thinking may serve
likely to occur to them than to a similarly aged sibling. This a similar function across Easterners and Westerners. In that
is consistent with the general notion of self-enhancement in regard, it is worth mentioning findings obtained by Sumi
the West (Chang et al. 2001; Heine and Lehman 1995; and Kanda (2002), in their study of male Japanese college
Heine et al. 1999; Held 2002; Kitayama et al. 1997; Taylor students looking at maladaptive perfectionism (measured
and Brown 1988). In contrast, these investigators found by the Burns Perfectionism Scale; BPS; Burns 1980) as a
that Japanese held a pessimistic bias for negative outcomes. predictor of concurrent and prospective depressive symp-
That is, Japanese predicted that negative events (e.g., ‘‘Fail toms. Specifically, these researchers found that scores on
a test’’) were more likely to occur to them than to a simi- the BPS, which relate to self-critical thoughts (similar to
larly aged sibling. This is consistent with the general notion what is focused on by the negative self-oriented perfor-
of self-criticism in the East (Heine et al. 2000; Kitayama mance perfectionism dimension), were not only correlated
et al. 1997). Because Chang and Asakawa’s (2003) findings with depressive symptoms, but they were also predictive of
revealed significant cultural differences in expectancies for depressive symptoms 6 weeks later even after controlling
experiencing positive and negative outcomes between for initial symptoms. Unfortunately, their study had some
Westerners and Easterners, one might expect cultural dif- key limitations. For example, their study only focused on
ferences in performance perfectionism. For example, given male students, the internal consistency of the BPS in their
the presumed self-enhancement bias of most Westerners, study was not high (a = .63), and depressive symptoms
one might predict that the self-system of European Amer- was measured by a 10-item subscale taken from a larger
icans, compared to that of Japanese, would involve greater symptoms checklist. Only a few other studies have exam-
expressions of positive self-oriented performance perfec- ined perfectionism in a Japanese sample, and of those, only
tionism and positive socially prescribed performance per- two have specifically examined the link between perfec-
fectionism. In other words, because Westerners have an tionism and psychological well-being (e.g., Kobori and
optimistic and self-enhancing bias, they may be more Tanno 2005; Stoeber et al. 2010). Further research on the
likely to have high levels of positive self-oriented perfor- cultural manifestations of perfectionism is therefore
mance perfectionism (which involves positive outcome necessary.

123
Cogn Ther Res (2012) 36:1–14 5

Beyond a consideration of past research, additional both Easterners and Westerners (e.g., Chang 1996, 2002a).
theoretical considerations are worth noting. According to Thus, both negative self-oriented and negative socially
the developmental model of performance perfectionism prescribed performance perfectionism were expected to
presented by Chang et al. (2006; see Fig. 1), socially pre- have positive associations with depressive symptomatology
scribed dimensions of performance perfectionism are in both groups, in accordance with past research (Sumi and
generally believed to play a strong role during childhood, Kanda 2002). However, consistent with performance per-
helping to shape and establish self-oriented dimensions of fectionism theory and past research (Chang 2006; Shafran
performance perfectionism as the child slowly moves into and Mansell 2001) negative self-oriented performance
adulthood. Yet, these self-oriented dimensions of perfor- perfectionism was expected in both groups to be most
mance perfectionism are not stagnant in adults. Rather, strongly associated with both concurrent and prospective
they are dynamic attributes and are believed to impact (as depressive symptomatology. Furthermore, as a potential
well as be impacted by) important outcomes (Chang et al. causal variable, negative self-oriented performance per-
2006). Chang (2006) showed that negative self-oriented fectionism was expected for both groups to most strongly
performance perfectionism was the only performance per- predict changes in depressive symptomatology across time.
fectionism dimension to be uniquely associated with
depressive symptoms. However, to have stronger support
for the general applicability of performance perfectionism Method
theory, we would have to find that negative self-oriented
performance perfectionism predicts changes in depressive Participants
symptoms across time in both Easterners and Westerners.
Participants were 177 (42 male and 135 female) European
American college students attending a public university in
Purpose of the Present Study the US (Midwest) and 155 (71 male and 84 female)
Japanese college students attending a public university in
The objectives of the present study were to (a) examine if Japan. All participants were enrolled in a psychology
there are cultural variations on performance perfectionism course and received extra credit for participation. For
between and within European American and Japanese European Americans, ages ranged from 18 to 23 years of
college students; (b) examine if performance perfectionism age, with a mean age of 19.7. For Japanese, ages ranged
is related to both concurrent and prospective depressive from 18 to 23 years of age, with a mean age of 20.0.
symptomatology for both groups; and (c) determine if
performance perfectionism is able to predict changes in Measures
depressive symptomatology after controlling for concurrent
depressive symptomatology in both groups. Performance Perfectionism
Given presumed cultural differences in self-enhance-
ment versus self-criticism, European Americans, compared To assess for performance perfectionism, the Performance
to Japanese, were expected to report greater adaptive per- Perfectionism Scale (PPS; Chang 2006) was used. The PPS
formance perfectionism (as indicated by higher scores on is a 32-item multidimensional measure of performance
positive self-oriented performance perfectionism and on perfectionism. Respondents are asked to indicate the extent
positive socially prescribed performance perfectionism). In to which each item is ‘‘generally true of you’’ using a
contrast, Japanese, compared to European Americans, were 5-point Likert-type scale, ranging from 1 (extremely untrue
expected to report greater maladaptive performance per- of me) to 5 (extremely true of me). A factor analysis indi-
fectionism (as indicated by higher scores on negative self- cated that the PPS taps four distinguishable dimensions of
oriented performance perfectionism and on negative performance perfectionism (Chang 2006, Study 1b). These
socially prescribed performance perfectionism). Despite dimensions are tapped by a Positive Self-Oriented Perfor-
these anticipated differences reflecting self-enhancement mance Perfectionism scale (e.g., ‘‘My performance is
and self-criticism in European Americans and Japanese, always made better by the high standards I set for myself’’),
respectively, the function of performance perfectionism the Negative Self-Oriented Performance Perfectionism
was expected to be quite similar across the two groups. scale (e.g., ‘‘I accomplish nothing when I try to meet the
Findings from past studies indicate that some of the ele- high standards I set for myself’’), the Positive Socially
ments that are reflected in performance perfectionism may Prescribed Performance Perfectionism scale (e.g., ‘‘My
fulfill similar functions in Easterners and Westerners. For performance is always made better when others expect
example, negative outcome expectancies have been found more from me’’), and the Negative Socially Prescribed
to be associated with greater depressive symptomatology in Performance Perfectionism scale (e.g., ‘‘My work suffers

123
6 Cogn Ther Res (2012) 36:1–14

when others push me to meet their high standards’’). The native language. The English version was translated into
two positive dimensions represent adaptive aspects of Japanese and then independently back-translated into
performance perfectionism. The two negative dimensions English to ensure accuracy and consistency with the ori-
represent maladaptive aspects of performance perfection- ginal English items. Of the initial 177 European American
ism. In support for construct validity, the PPS was found to students who completed Time 1 measures, 9 students did
be related to, but not redundant with, measures of general not complete the BDI at Time 2. This left a total of 168 (39
(maladaptive) perfectionism (Frost Multidimensional Per- male and 129 female) participants for the European
fectionism Scale; Frost et al. 1990; Chang 2006, Study 3b), American group (mean age of 19.6). Of the initial 155
optimism and pessimism (Revised Life Orientation Test; Japanese students who completed Time 1 measures, 4
Scheier et al. 1994; Chang 2006, Study 2), and a measure students did not complete the BDI at Time 2. This left a
of the five-factor model of personality (NEO Five-Factor total of 151 (69 male and 82 female) participants for the
Inventory; Costa and McCrae 1985; Chang 2006, Study 2), Japanese group (mean age of 20.0).
but unrelated to a measure of social desirability (Social Participants were not made aware of the purpose of the
Desirability Scale; Crowne and Marlowe 1960; Chang study until after they had completed all measures. To
2006, Study 2). In addition, the PPS has been found to be protect the participants’ anonymity, only subject numbers
related to measures of negative psychological functioning were placed on the instruments.
(e.g., depressive symptomatology as measured by the Beck
Depression Inventory; Beck et al. 1961; anxiety as mea-
sured by the Brief Symptom Inventory; Derogatis and Results
Spencer 1982; negative affect as measured by the Positive
and Negative Affect Schedule; Watson et al. 1988; Chang To explore for possible gender influences, we conducted a
2006, Study 3a), positive psychological functioning (e.g., 2 (male vs. female) 9 2 (European American vs. Japanese)
life satisfaction as measured by the Satisfaction with Life MANOVA on the depressive symptomatology measures.
Scale; Diener et al. 1985; purpose in life as measured by Results indicated a main effect for culture, Wilk’s Lambda
the Scales of Psychological Well-Being; Ryff 1989; Chang (2, 313) = .92, p \ .000001, and a gender 9 culture
2006, Study 3b), and prospective academic achievement in interaction effect, Wilk’s Lambda (2, 313) = .96, p \ .01.
the classroom (Chang 2006, Study 4). Six-week test–retest The main effect for gender was not significant, Wilk’s
reliabilities for the four PPS subscales have been found to Lambda (6, 312) = .99, n.s. Results of conducting uni-
range from .63 to .70 (Chang 2006, Study 1b). variate 2 9 2 ANOVAs indicated that this pattern was
consistent across the two depressive symptomatology
Depressive Symptomatology measures. Specifically, for European Americans, but not
Japanese, women reported greater depressive symptom-
To assess for depressive symptomatology, the Beck atology than men at Time 1 (M = 7.27, SD = 9.57 vs.
Depression Inventory or BDI (Beck et al. 1961) was used. M = 2.56, SD = 2.54, respectively, t[165] = 3.03, p \
The BDI is a widely used 21-item self-report measure of .01), and at Time 2 (M = 8.78, SD = 8.18 vs. M = 3.74,
depressive symptomatology. Respondents are asked to rate SD = 3.69, respectively, t[165] = 3.73, p \ .001). None-
the extent to which they have experienced in the past week, theless, given that the primary focus of the present study
including today, specific depressive symptoms across a was to examine for the role of culture on performance
4-point scale (for example, ‘‘0 = I do not feel sad’’ to perfectionism, and given unequal gender composition of
‘‘3 = I am so sad or unhappy that I can’t stand it’’). When the two cultural groups, we controlled for gender in all
used in a non-clinical sample, scores on the BDI are typ- subsequent data analyses.
ically taken to measure for dysphoria (or at most, sub-
threshold levels of depression). Between-Cultures Differences on Performance
Perfectionism
Procedure
We first sought to determine if there were any meaningful
At Time 1, in addition to the PPS, all participants com- cultural differences to be examined. Results of computing
pleted the BDI (Beck et al. 1961). The PPS and BDI were partial correlations between scores on the PPS subscales,
presented in randomized order at Time 1. At Time 2, controlling for gender, indicated a comparable pattern
2 months later (within 2–3 weeks from the end of the between European Americans and Japanese (see Table 1).
academic term or of the academic year), all participants However, results of conducting a MANOVA between
completed the BDI again. Measures were in both English European Americans and Japanese on the present set of
and Japanese, and participants completed them in their measures indicated highly significant cultural differences,

123
Cogn Ther Res (2012) 36:1–14 7

Table 1 Partial correlations of positive self-oriented performance perfectionism (PSPPP), and negative socially prescribed performance
perfectionism (PSOPP), negative self-oriented performance perfec- perfectionism (NSPPP) scales for European Americans and Japanese
tionism (NSOPP), positive socially prescribed performance controlling for gender
PSOPP NSOPP PSPPP NSPPP

PSOPP – -.48*** .65*** -.28***

NSOPP -.52*** – -.41*** .68***
PSPPP .62*** -.37*** – -.41***
NSPPP -.49*** .74*** -.48*** –

Partial correlations below the diagonal are for European Americans (n = 168). Partial correlations above the diagonal are for Japanese (n = 151)
*** p \ .001

Wilk’s Lambda (6, 312) = .61, p \ .000001. Accordingly, (see Table 3). In support of using the adaptive and mal-
we next conducted a series of one-way ANOVAs exam- adaptive distinction for both European Americans and
ining for between-cultures differences on performance Japanese, the different aspects of performance perfection-
perfectionism with gender entered as a covariate (see ism were associated with depressive symptomatology in
Table 2). As the table shows, positive self-oriented per- the expected direction for both groups. For example,
formance perfectionism was found to be greater in Euro- positive self-oriented performance perfectionism was neg-
pean Americans than in Japanese, whereas negative self- atively associated with concurrent depressive symptom-
oriented performance perfectionism was found to be atology in both European Americans (pr = -.27,
greater in Japanese than in European Americans. Similarly, p \ .001) and in Japanese (pr = -.19, p \ .05). Further-
positive socially prescribed performance perfectionism was more, results of computing additional zero-order correla-
found to be greater in European Americans than in Japa- tions indicated that the different aspects of performance
nese, whereas negative socially prescribed performance perfectionism were found to be associated with each other
perfectionism was found to be greater in Japanese than in in the expected direction for both groups. For example,
European Americans. Consistent with past findings (e.g., positive self-oriented performance perfectionism was
Chang and Asakawa 2003), depressive symptomatology found to be negatively associated with negative self-ori-
was found to be greater in Japanese than in European ented performance perfectionism in both European Amer-
Americans at both Time 1 and Time 2. Noteworthy, icans (pr = -.43, p \ .001) and in Japanese (pr = -.47,
internal consistencies for all study measures were found to p \ .001).
be good across both cultural groups (see Table 2). Partial correlations that controlled for gender and
To gain a better appreciation of the magnitude of these covariation among the PPS subscales were also computed
between-cultures differences, effects sizes using Cohen’s to identify the most robust associations involving perfor-
d (Cohen 1977) were computed. In these computations, mance perfectionism. Results of computing these more
sigma was adjusted given the unequal standard deviations stringent partial correlations indicated that for European
underlying the two samples. Using Cohen’s (1977) con- Americans only negative self-oriented performance per-
vention for small (d = .20), medium (d = .50), and large fectionism was associated with concurrent depressive
effects (d = .80), the table shows that all of the significant symptomatology (pr = .24, p \ .01) and with depressive
differences found between European Americans and Jap- symptomatology 2 months later (pr = .35, p \ .001).
anese on the four PPS subscales were in the large effect Likewise, for Japanese, results indicated that only negative
size range. In contrast, the significant differences found self-oriented performance perfectionism was associated
between European Americans and Japanese on depressive with concurrent depressive symptomatology (pr = .23,
symptomatology at Time 1 and at Time 2 were in the small p \ .01) and with depressive symptomatology 2 months
effect size range. later (pr = .24, p \ .01).

Relations Between Performance Perfectionism Performance Perfectionism as a Predictor of Changes

and Depressive Symptomatology in Depressive Symptoms Across Time

To examine how performance perfectionism relates to To determine if performance perfectionism predicts chan-
depressive symptomatology across European Americans ges in depressive symptomatology across time for both
and Japanese, partial correlations controlling for gender European Americans and Japanese, a hierarchical regres-
were computed between scores on the PPS subscales and sion analysis was conducted for each group with BDI
scores on the BDI at Time 1 and Time 2 for each group scores at Time 2 as the dependent variable. For each

123
8 Cogn Ther Res (2012) 36:1–14

Table 2 Cultural variations of positive self-oriented performance perfectionism (NSPPP) scales and of the beck depression inventory
perfectionism (PSOPP), negative self-oriented performance perfec- (BDI) at time 1 and time 2 between European Americans and
tionism (NSOPP), positive socially prescribed performance perfec- Japanese controlling for gender
tionism (PSPPP), and negative socially prescribed performance
Scale Cultural group
European American Japanese
M SD a M SD a F(1, 317) Cohen’s d

PSOPP 32.63 5.39 .91 25.41 5.94 .88 120.13*** 1.27

NSOPP 14.65 5.28 .87 21.32 5.42 .85 125.33*** 1.25
PSPPP 27.83 5.65 .91 23.27 5.63 .87 48.31*** .81
NSPPP 14.44 5.35 .90 21.97 6.14 .89 140.67*** 1.31
BDI—Time 1 6.16 8.68 .91 9.02 8.18 .90 10.44** .34
BDI—Time 2 7.61 7.67 .93 11.28 8.45 .91 18.62*** .45

For European Americans, n = 168. For Japanese, n = 151. At Time 1, participants completed the Performance Perfectionism Scale along with
the BDI. At Time 2 (2 months later), participants again completed the BDI
** p \ .01; *** p \ .001

Table 3 Partial correlations of positive self-oriented performance perfectionism (PSPPP), and negative socially prescribed performance
perfectionism (PSOPP), negative self-oriented performance perfec- perfectionism (NSPPP) scales with the beck depression inventory
tionism (NSOPP), positive socially prescribed performance (BDI) at time 1 and time 2 for European Americans and Japanese
Scale PSOPP NSOPP PSPPP NSPPP
a b a b a b
pr pr pr pr pr pr pra prb

European Americans
BDI—Time 1 -.27*** -.09 .29*** .24** -.27*** -.15 .24** -.06
BDI—Time 2 -.25*** .01 .38*** .35*** -.26*** -.15 .26*** -.12
Japanese
BDI—Time 1 -.19* .00 .30*** .23** -.22** -.11 .25*** .05
BDI—Time 2 -.30*** -.12 .36*** .24** -.28*** -.08 .30*** .08

For European Americans, n = 168. For Japanese, n = 151. At Time 1, participants completed the Performance Perfectionism Scale (PPS) along
with the BDI. At Time 2 (2 months later), participants completed the BDI again
a
Partial correlations controlling for gender
b
Partial correlations controlling for gender and PPS subscales
* p \ .05; ** p \ .01; *** p \ .001

analysis, gender was entered as a covariate in Step 1, fol- 2, F(4, 161) = 4.56, p \ .01. For Japanese, depressive
lowed by BDI scores at Time 1 entered in Step 2, and symptomatology at Time 1 accounted for a large
scores on the four PPS subscales entered as a set in Step 3. (f2 = 1.38) and significant 58% of the variance in depres-
Results are presented in Table 4. To examine whether sive symptomatology at Time 2, F(1, 148) = 219.49,
performance perfectionism accounted for a small, medium, p \ .001. The performance perfectionism set was found to
or large amount of the variance in changes in depressive account for an additional small (f2 = .03) albeit significant
symptomatology, we used Cohen’s (1977) convention for 3% of the variance in depressive symptomatology at Time
small (f2 = .02), medium (f2 = .15), and large effects 2, F(4, 144) = 3.14, p \ .05. It is worth noting that neg-
(f2 = .35). As this table shows, for European Americans, ative self-oriented performance perfectionism emerged as
depressive symptomatology at Time 1 accounted for a the only significant predictor within the performance per-
modest (f2 = .19) yet significant 16% of the variance in fectionism set for both European Americans and Japanese
depressive symptomatology at Time 2, F(1, 165) = 34.01, (bs = .27 and .20, respectively). Thus, for both groups,
p \ .001. The performance perfectionism set was found to negative self-oriented performance perfectionism was able
account for an additional small (f2 = .05) albeit significant to predict unique changes in depressive symptomatology
4% of the variance in depressive symptomatology at Time across time.

123
Cogn Ther Res (2012) 36:1–14 9

Table 4 Hierarchical
Group and predictor b R2 DR2 df F
regression analyses showing
amount of variance in beck European Americans
depression inventory (BDI)
scores at time 2 accounted for BDI—Time 2
by scores on the performance Step 1: Gender .28*** .08 1, 166 14.16***
perfectionism scale (PPS) over Step 2: BDI—Time 1 .41*** .23 .16 1, 165 34.01***
BDI scores at time 1 controlling
Step 3: PPS .24 .04 4, 161 4.56**
for gender
PSOPP .01
NSOPP .27***
PSPP -.11
For European Americans, NSPP .07
n = 168. For Japanese, n = 151
Japanese
PSOPP positive self-oriented
BDI—Time 2
performance perfectionism,
NSOPP negative self-oriented Step 1: Gender -.08 .01 1, 149 .98
performance perfectionism, Step 2: BDI—Time 1 .77*** .59 .58 1, 148 219.49***
PSPPP positive socially Step 3: PPS .62 .03 4, 144 3.14*
prescribed performance
perfectionism, NSPPP negative PSOPP -.10
socially prescribed performance NSOPP .20**
perfectionism PSPP -.00
* p \ .05; ** p \ .01; NSPP .05
*** p \ .001

Discussion strivings. The effects were similar even when the high
expectations derived from an external source. It is worth
One goal of the present study was to examine cultural noting that these findings were found to be large and
variations in positive and negative performance perfec- robust. Accordingly, these group differences findings are
tionism between European Americans and Japanese. generally consistent with the notion of greater self-criti-
Another goal of the present study was to examine the cism in the East and greater self-enhancement in the West
function of performance perfectionism as a predictor of (Chang 2007).
depressive symptomatology across European American A question might be raised by some regarding the
and Japanese. Accordingly, findings relative to these two validity of using a perfectionism scale that separates self-
goals are discussed separately below. oriented from socially prescribed expectations on Japanese
participants, given that Easterners come from cultures
Do Self-Enhancement Motives Foster Positive defined strongly by collectivist values (Doi 1971/1973) and
Performance Perfectionism in the West the self-oriented items may consequently not be as relevant
and Self-Criticism Motives Foster Negative to Japanese participants. But findings from several recent
Performance Perfectionism in the East? studies indicate that important cross-cultural variations
may be influenced by the specific reference group used by
Performance perfectionism involves two adaptive aspects Westerners and Easterners in making subjective judgments
(positive self-oriented performance perfectionism and (e.g., Chang and Asakawa 2003; Heine et al. 2002). The
positive socially prescribed performance perfectionism) PPS allows participants to define for themselves what
and two maladaptive aspects (negative self-oriented per- constitutes the ‘‘self’’ in the self-oriented items and what
formance perfectionism and negative socially prescribed constitutes ‘‘others’’ in the socially prescribed items, so it is
performance perfectionism). Between-cultures findings conceivable that Japanese may define the ‘‘self’’ as related
indicated that European Americans, compared to Japanese, to or composed of the identities and goals of close others
scored higher on scales measuring for adaptive aspects of (e.g., parents, siblings, friends). Thus, for example, despite
performance perfectionism, whereas Japanese, compared to the present cross-cultural similarity found in the function of
European Americans, scored higher on scales measuring negative self-oriented perfectionism in predicting depres-
for maladaptive aspects of performance perfectionism. sive symptoms, there may be strong and important differ-
More specifically, European Americans were more likely ences to be found when looking at sources of self-oriented
than Japanese to associate positive outcomes with their and socially prescribed performance perfectionism between
strivings, and Japanese were more likely than European Japanese and European Americans. Clearly, this possibility
Americans to associate negative outcomes with their warrants future empirical examination.

123
10 Cogn Ther Res (2012) 36:1–14

In general, the present cultural variations found on for both groups are also consistent with the more specific
performance perfectionism indicate that there are large and tenet that negative self-oriented performance perfectionism
important normative differences between Westerners and represents the strongest determinant of negative outcomes.
Easterners on performance perfectionism that should be In more direct support of this specific tenet, negative self-
taken into account. However, researchers are cautioned not oriented performance perfectionism was uniquely found to
to take these differences diagnostically. For example, in predict prospective changes in depressive symptomatology
traditional Eastern cultures, excessive self-enhancement is across a 2-month time period for both European Americans
not socially encouraged. As several researchers have noted, and Japanese. Taken together, these findings go beyond
it is the attainment of personal happiness, rather than group past findings pointing to the potential value of studying
happiness that appears to be highly regarded and sought maladaptive perfectionism in Westerners and Easterners
after among Westerners (Waterman 1984; Weisz et al. (e.g., Sumi and Kanda 2002), by providing support for a
1984). The opposite pattern appears to be more telling of specific aspect of perfectionism, namely, negative self-
Easterners (Doi 1971/1973; Markus and Kitayama 1991). oriented performance perfectionism, as an important
Thus, moderate or realistic levels of self-enhancement may determinant of behavior across different cultural groups.
actually be more socially adaptive for Japanese. Interest- Furthermore, relative to other models and measures of
ingly, there has been some growing appreciation for the maladaptive perfectionism, the present findings may be
distinguishing between adaptive and maladaptive variants taken to suggest that a unidimensional model of clinical
of self-criticism (Bergner 2007). perfectionism may not only be plausible, but also sufficient
(Shafran et al. 2002). Alternatively, our findings also raise
some questions about the presumed dominance of socially
Maladaptive Perfectionism as a Predictor of Depressive prescribed perfectionism in psychopathology when other
Symptoms: Some Evidence for the Universal Function variants of perfectionism are considered.
of Negative Self-Oriented Performance Perfectionism Another interesting note relates to the lack of a gender
Across Westerners and Easterners difference between Japanese males and females on
depressive symptomatology. Given what some researchers
Findings based on examining the relationship between have suggested about gender differences in depression,
performance perfectionism and depressive symptomatol- namely, that women are more prone to depression because
ogy in both European Americans and Japanese provided gender inequity deprives them of power and control in their
support for the generalizability of the different functions own lives (Nolen-Hoeksema et al. 1999), one would expect
of performance perfectionism. First, consistent with the Japanese women to have higher rates of depression, or
notion that there are adaptive and maladaptive aspects of greater depressive symptomatology, compared with men
performance perfectionism, positive aspects of perfor- considering the higher level of gender inequity in Japanese
mance perfectionism (e.g., positive self-oriented perfor- culture. As far as we can tell, there are two potential rea-
mance perfectionism) were found for both groups to be sons for the lack of differences in our data. The first pos-
negatively related to depressive symptomatology, whereas sibility is that men in Japan face as many psychosocial
negative aspects of performance perfectionism (e.g., neg- stressors as women do, and as a result they also suffer from
ative self-oriented performance perfectionism) were found depressive symptomatology as often. Japanese men may
for both groups to be positively related to depressive feel a great deal of pressure from society to fulfill their
symptomatology. duties as breadwinners, top students, and so forth, which in
It is worth noting that after controlling for covariation turn places them more at risk for depression. Another
among the different aspects of performance perfectionism, possibility is that although there is more gender inequality
the initial significant associations of socially prescribed in Japan compared with the U.S. that gender inequality is in
aspects of performance perfectionism with depressive actuality responsible for the decreased gender gap in
symptomatology were found to become nonsignificant for depressive symptomatology. One recent study examining
both European Americans and Japanese. In contrast, it is the gender gap in depression across 29 countries found that
worth noting that after controlling for covariation, the countries with high gender equity had a bigger gender gap
initial significant associations of negative self-oriented in depression compared with countries with low gender
performance perfectionism with depressive symptomatol- equity (Hopcroft and Bradley 2007). It may be argued that
ogy were found to remain significant for both groups. Japanese women don’t experience more depression than
These findings for both European Americans and Japanese Japanese men do because they don’t have the expectations
are consistent with the general tenet that self-oriented for gender equity that European American women do, and
aspects, and not socially prescribed aspects, matter most in as a result they do not feel disappointed when the expec-
predicting important outcomes. In addition, these findings tations fall through. In our data, gender made a significant

123
Cogn Ther Res (2012) 36:1–14 11

contribution to the prediction of depressive symptomatol- besides stress that may moderate the link between perfor-
ogy at Time 2 in European Americans, and whereas Time 1 mance perfectionism and depressive symptomatology. For
depressive symptomatology made a larger contribution to example, findings from past studies on general perfec-
the prediction of depressive symptomatology at Time 2 in tionism have shown that vulnerability factors such as
Japanese. Again, it is conceivable that the greater influence negative attributional style can increase the association
of gender in the European American sample reflects the between perfectionism and depressive symptomatology
stress many European American women experience from (Chang and Sanna 2001), whereas buffering factors such as
frustrated expectations of gender inequity, or perhaps there social problem-solving ability can decrease the association
is a greater cultural acceptance of the expression of sadness between general perfectionism and depressive symptom-
in women relative to men in European American culture. atology (Chang 2002b). Clearly, it would to useful in future
The greater contribution of Time 1 depressive symptom- studies to consider the role of potential mediator variables
atology in Japanese may simply reflect the greater contribution and the impact of potential moderator variables for
of gender in predicting Time 2 depressive symptomatology in understanding the link between performance perfectionism
European Americans. Interestingly, and depressive symptomatology. Indeed, because cultural
Because a number of factors may causally account for differences have been indicated in studies of variables like
(mediate) the associations of performance perfectionism pessimism (Chang 1996, 2002a), social problem solving
with important outcomes, it may be useful to consider (Chang 1998), and accessibility experiences (Chang and
variables that account for the association between perfor- Asakawa 2003; Sanna et al. 2009), it would also be
mance perfectionism and depressive symptomatology. One important in future studies to determine if they operate or
notable variable that has been found in past studies to impact the association between performance perfectionism
partially mediate the associations of general perfectionism and depressive symptomatology in the same or similar
with important outcomes is stress (e.g., Chang 2000; Chang manner across different cultural groups.
et al. 2004). Consistent with a stress-generation model
(Hewitt et al. 1996), negative self-oriented performance
perfectionism may lead to the generation of greater stress
which, in turn, leads to the generation of greater depressive Limitations of This Study
symptomatology over time. Another possible mechanism
worth considering is pessimism. Negative self-oriented It is worth noting some limitations of the present study.
performance perfectionism may lead to the generation of First, the present study focused on the prediction of
greater pessimism, leading to the generation of greater depressive symptomatology. Although this is an important
depressive symptomatology over time. However, this is not cross-cultural outcome examined in many theories of
to say that these variables can sufficiently account for the negative psychological functioning (e.g., Anderson 1999),
associations of performance perfectionism with important the present study did not focus on examining if adaptive
outcomes. aspects of performance perfectionism (e.g., positive self-
Similarly, a number of factors may influence the oriented performance perfectionism) represent useful pre-
strength of (moderate) the associations of performance dictors of positive outcomes and conditions. Indeed, only a
perfectionism with important outcomes. Accordingly, it handful of previous studies have even sought to look at the
may be useful to consider complex interactive models possible links between perfectionism and positive out-
involving performance perfectionism and other relevant comes and conditions (e.g., Chang 2006; Flett et al. 1991).
variables in predicting depressive symptomatology. As Therefore, it would be important to determine if support for
findings from past studies on general perfectionism have adaptive performance perfectionism can be found across
shown (e.g., Chang and Rand 2000; Hewitt et al. 1996; Westerners and Easterners in the prediction of positive
Joiner and Schmidt 1995), a diathesis-stress model may psychological functioning (e.g., life satisfaction, positive
provide additional insights into how performance perfec- relations with others).
tionism may interact with stress to predict depressive Second, findings from a recent study of Black and White
symptomatology. For example, one might predict a sig- adult females indicate that there may be other important
nificant Negative Self-Oriented Performance Perfectionism racial variations that need to be considered when studying
Stress interaction in the prediction of depressive symp- performance perfectionism. In a study that looked at the
tomatology. This is to say, individuals with high (vs. low) role of stress as a mediator of the associations of a general
negative self-oriented performance perfectionism experi- measure of maladaptive perfectionism with several key
encing high (vs. low) stress levels may be most apt to measures of psychological functioning, Chang et al. (2004)
report the greatest levels of depressive symptomatology. found that stress fully mediated the associations of mal-
Alternatively, it may be worth considering other variables adaptive perfectionism with all of the study measures of

123
12 Cogn Ther Res (2012) 36:1–14

psychological functioning for Blacks, but not for Whites. Concluding Comment
Additionally, a recent study found that certain adaptive
components of perfectionism may be more strongly related In looking at possible cultural variations on performance
to self-esteem in African Americans than in European perfectionism between European Americans and Japanese,
Americans (Mobley et al. 2005). Accordingly, further findings from the present study provided support for more
evaluation of the relation between performance perfec- positive, self-enhancing aspects of perfectionism in the
tionism and psychological functioning across different West, and for more negative, self-critical aspects of per-
cultural and racial groups would be useful in future fectionism in the East. Moreover, consistent with the
research (Chang et al., in press). notion that performance perfectionism may represent a
Third, although results from the present study found that useful universal construct, negative self-oriented perfor-
performance perfectionism predicts changes in depressive mance perfectionism was found to predict changes in
symptomatology, they do not indicate how depressive depressive symptomatology across time in both European
symptomatology may predict changes in performance Americans and Japanese. Thus, negative self-oriented
perfectionism. Because self-oriented aspects of perfor- performance perfectionism appears to be an important
mance perfectionism are believed to have a reciprocal marker of negative psychological functioning for both
relationship with important outcomes (Chang 2006), Westerners and Easterners. Taken together, these findings
experiences of depressive symptomatology may them- provide important support for the study of perfectionism
selves lead to greater negative self-oriented performance across cultures. More studies are now needed to examine
perfectionism. Thus, it would be useful to evaluate the performance perfectionism as a useful predictor of both
extent to which important outcomes and conditions predict positive and negative outcomes and conditions across dif-
performance perfectionism. And although the present ferent cultural and racial groups.
findings provide some promising support for the usefulness
of the PPS in studying different cultural groups, more Acknowledgments This research was supported by a Faculty
Research Grant awarded to Edward C. Chang from the Center of
research is needed to clarify the psychometric properties of Japanese Studies, University of Michigan. We would like to thank
the PPS (e.g., test–retest reliability) when used in more Kiyoshi Asakawa for collecting the Japanese data. The first author
diverse populations. would also like to thank Chang Suk-Choon and Tae Myung-Sook for
More broadly speaking, one potential problem with the their encouragement and support throughout this project.
current study is that we are assuming an equivalency of
constructs between European Americans and Japanese. By References
using a measure developed on European Americans, we are
assuming that the two groups think about the ‘‘self’’ (and Anderson, C. A. (1999). Attributional style, depression, and loneli-
by extension ‘‘self-oriented’’ perfectionism) in similar ness: A cross-cultural comparison of American and Chinese
ways. But as Triandis (1996) has argued, the definition of students. Personality and Social Psychology Bulletin, 25,
482–499.
the ‘‘self’’ can be very different depending on the culture to Bandura, A. (1977). Self-efficacy: Toward a unifying theory of
which one belongs, and one can further argue that Japanese behavioral change. Psychological Review, 84, 191–215.
may not be able to meaningfully separate the ‘‘self’’ Beck, A. T., Ward, C. H., Mendelson, M., Mock, L., & Erbaugh, J.
from the social context. Given that for collectivists, the (1961). An inventory for measuring depression. Archives of
General Psychiatry, 4, 561–571.
in-group’s goals are often adopted as one’s own goals and Bergner, R. M. (2007). On promoting adaptive self-criticism in
even preferred over one’s own goals if there is a mis- psychotherapy. In E. C. Chang (Ed.), Self-criticism and
alignment, one can also argue that what the Japanese call self-enhancement: Theory, research, and clinical implications
personal standards may in fact be social standards. The (pp. 225–245). Washington, DC: American Psychological
Association.
scope of the current study was not fit to address this Burns, D. D. (1980). The perfectionist’s script for self-defeat.
broader theoretical issue, but future studies may look Psychology Today, 14, 34–52.
toward developing emic models of perfectionism in col- Chang, E. C. (1996). Cultural differences in optimism, pessimism,
lectivist cultures. Indeed, although we did not expect to and coping: Predictors of subsequent adjustment in Asian
American and Caucasian American college students. Journal
find cultural differences in the function of performance of Counseling Psychology, 43, 113–123.
perfectionism in predicting depressive symptoms between Chang, E. C. (1998). Cultural differences, perfectionism, and suicidal
European Americans and Japanese, it would be interesting risk: Does social problem solving still matter? Cognitive
to determine if culture is an important moderator of the link Therapy and Research, 22, 237–254.
Chang, E. C. (2000). Perfectionism as a predictor of positive and
between performance perfectionism and depressive symp- negative psychological outcomes: Examining a mediation model
toms when performance is link to achieving individualistic in younger and older adults. Journal of Counseling Psychology,
versus collectivistic goals. 47, 18–26.

123
Cogn Ther Res (2012) 36:1–14 13

Chang, E. C. (Ed.). (2001). Optimism and pessimism: Implications for Costa, P. T., & McCrae, R. R. (1985). NEO Five-Factor Inventory:
theory, research, and practice. Washington, DC: American Form S. Odessa, FL: Psychological Assessment Resources.
Psychological Association. Crowne, D. P., & Marlowe, D. (1960). A new scale of social
Chang, E. C. (2002a). Cultural differences in psychological distress in desirability independent of psychopathology. Journal of Con-
Asian and Caucasian American college students: Examining the sulting Psychology, 24, 349–354.
role of cognitive and affective concomitants. Journal of Coun- Derogatis, L. R., & Spencer, P. M. (1982). The Brief Symptom
seling Psychology, 49, 47–59. Inventory (BSI): Administration, scoring and procedures man-
Chang, E. C. (2002b). Examining the link between perfectionism and ual-I. Baltimore: Clinical Psychometrics Research.
psychological maladjustment: Social problem solving as a Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The
buffer. Cognitive Therapy and Research, 26, 581–595. Satisfaction with Life Scale. Journal of Personality Assessment,
Chang, E. C. (2003). On the perfectibility of the individual: Going 49, 71–75.
beyond the dialectic of good versus evil. In E. C. Chang & Doi, T. (1973). The anatomy of dependence. Tokyo: Kodansha.
L. J. Sanna (Eds.), Virtue, vice, and personality: The complexity J. Bester, Trans., Original work published in 1971.
of behavior (pp. 125–143). Washington, DC: American Psycho- Enns, M. W., & Cox, B. J. (1999). Perfectionism and depression
logical Association. symptom severity in major depressive disorder. Behaviour
Chang, E. C. (2006). Conceptualization and measurement of adaptive Research and Therapy, 37, 783–794.
and maladaptive aspects of performance perfectionism: Rela- Fedewa, B. A., Burns, L. R., & Gomez, A. A. (2005). Positive and
tions to personality, psychological functioning, and academic negative perfectionism and the shame/guilt distinction: Adaptive
achievement. Cognitive Therapy and Research, 30, 677–697. and maladaptive characteristics. Personality and Individual
Chang, E. C. (2007). Introduction to self-criticism and self-enhance- Differences, 38, 1609–1619.
ment: Views from ancient Greece to the modern world. In E. Flett, G. L., Hewitt, P. L., Blankstein, K. R., & O’Brien, S. (1991).
C. Chang (Ed.), Self-criticism and self-enhancement: Theory, Perfectionism and learned resourcefulness in depression and
research, and clinical implication (pp. 3–15). Washington, DC: self-esteem. Personality and Individual Differences, 12, 61–68.
American Psychological Association. Frost, R. O., Heimberg, R. G., Holt, C. S., Mattia, J. I., & Neubauer,
Chang, E. C. (2009). An examination of optimism, pessimism, and A. L. (1993). A comparison of two measures of perfectionism.
performance perfectionism as predictors of positive psycholog- Personality and Individual Differences, 14, 119–126.
ical functioning in middle-aged adults: Does holding high Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The
standards of performance matter beyond generalized outcome dimensions of perfectionism. Cognitive Therapy and Research,
expectancies? Cognitive Therapy and Research, 33, 334–344. 14, 449–468.
Chang, E. C., & Asakawa, K. (2003). Cultural variations on optimistic Hardin, E. E., & Leong, F. T. L. (2005). Optimism and pessimism as
and pessimistic bias for self versus a sibling: Is there evidence mediators of the relations between self-discrepancies among
for self-enhancement in the West and self-criticism in the East Asian and European Americans. Journal of Counseling Psy-
when the referent group is specified? Journal of Personality and chology, 52, 25–35.
Social Psychology, 84, 569–581. Hart, B. A., Gilner, F. H., Handal, P. J., & Gfeller, J. D. (1998). The
Chang, E. C., Asakawa, K., & Sanna, L. J. (2001). Cultural variations relationship between perfectionism and self-efficacy. Personality
in optimistic and pessimistic bias: Do Easterners really expect and Individual Differences, 24, 109–113.
the worst and Westerners really expect the best when predicting Heine, S. J., & Lehman, D. R. (1995). Cultural variation in unrealistic
future life events? Journal of Personality and Social Psychology, optimism: Does the West feel more invulnerable than the East?
81, 476–491. Journal of Personality and Social Psychology, 68, 595–607.
Chang, R., & Chang, E. C. (2009). Effects of socially prescribed Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999). Is
expectations on emotions and cognitions in Asian and European there a universal need for positive self-regard? Psychological
Americans. Cognitive Therapy and Research, 33, 272–282. Review, 106, 766–794.
Chang, E. C., Downey, C. A., & Lin, E. Y. (2006). Pursuing the path Heine, S. J., Lehman, D. R., Peng, K., & Greenholtz, J. (2002).
to a more perfect future: On causes, correlates, and consequences What’s wrong with cross-cultural comparisons of subjective
of perfectionism. In L. J. Sanna & E. C. Chang (Eds.), Judgments Likert scales?: The reference-group effect. Journal of Person-
over time: The interplay of thoughts, feelings, and behaviors ality and Social Psychology, 82, 903–918.
(pp. 47–66). New York: Oxford University Press. Heine, S. J., Takata, T., & Lehman, D. R. (2000). Beyond self-
Chang, E. C., Hirsch, J. K., Sanna, L. J., Jeglic, E. L., & Fabian, C. G. presentation: Evidence for self-criticism among Japanese. Per-
(2011). A preliminary study of perfectionism and loneliness as sonality and Social Psychology Bulletin, 26, 71–78.
predictors of depressive and anxious symptoms in Latinas: A top- Held, B. S. (2002). The tyranny of the positive attitude in America:
down test of a model. Journal of Counseling Psychology (in press). Observation and speculation. Journal of Clinical Psychology, 58,
Chang, E. C., & Rand, K. L. (2000). Perfectionism as a predictor of 965–991.
subsequent adjustment: Evidence for a specific diathesis-stress Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and
mechanism among college students. Journal of Counseling social contexts: Conceptualization, assessment, and association
Psychology, 47, 129–137. with psychopathology. Journal of Personality and Social
Chang, E. C., & Sanna, L. J. (2001). Negative attributional style as a Psychology, 60, 456–470.
moderator of the link between perfectionism and depressive Hewitt, P. L., Flett, G. L., & Ediger, E. (1996). Perfectionism and
symptoms: Preliminary evidence for an integrative model. depression: Longitudinal assessment of a specific vulnerability
Journal of Counseling Psychology, 48, 490–495. hypothesis. Journal of Abnormal Psychology, 105, 276–280.
Chang, E. C., Watkins, A. F., & Banks, K. H. (2004). How adaptive Hewitt, P. L., Flett, G. L., & Weber, C. (1994). Dimensions of
and maladaptive perfectionism relate to positive and negative perfectionism and suicide ideation. Cognitive Therapy and
psychological functioning: Testing a stress-mediation model in Research, 18, 439–460.
Black and White female college students. Journal of Counseling Hopcroft, R. L., & Bradley, D. B. (2007). The sex difference in
Psychology, 51, 93–102. depression across 29 countries. Social Forces, 85, 1483–1507.
Cohen, J. (1977). Statistical power analysis for the behavioral Joiner, T. E., Jr., & Schmidt, N. B. (1995). Dimensions of
sciences (Revised ed.). New York: Academic Press. perfectionism, life stress, and depressed and anxious symptoms:

123
14 Cogn Ther Res (2012) 36:1–14

Prospective support for diathesis-stress but not specific vulner- Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health:
ability among male undergraduates. Journal of Social and Assessment and implications of generalized outcome expectan-
Clinical Psychology, 14, 165–183. cies. Health Psychology, 4, 219–247.
Kawamura, K. Y., Frost, R. O., & Harmatz, M. G. (2002). The Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguish-
relationship of perceived parenting styles to perfectionism. ing optimism from neuroticism (and trait anxiety, self-mastery,
Personality and Individual Differences, 32, 317–327. and self-esteem): A reevaluation of the life orientation test.
Kim, U., Triandis, H. C., Kagitçibasi, C., Choi, S. C., & Yoon, G. Journal of Personality and Social Psychology, 67, 1063–1078.
(Eds.). (1994). Individualism and collectivism: Theory, method, Shafran, R., Cooper, Z., & Fairburn, C. G. (2002). Clinical
and applications. Newbury Park, CA: Sage. perfectionism: A cognitive-behavioural analysis. Behaviour
Kitayama, S., Markus, H. R., Matsumoto, H., & Norasakkunkit, V. Research and Therapy, 40, 773–791.
(1997). Individual and collective processes in the construction of Shafran, R., & Mansell, W. (2001). Perfectionism and psychopathol-
the self: Self-enhancement in the United States and self-criticism ogy: A review of research and treatment. Clinical Psychology
in Japan. Journal of Personality and Social Psychology, 72, Review, 21, 879–906.
1245–1267. Slade, P. D., & Owens, R. G. (1998). A dual process model of
Kobori, O., & Tanno, Y. (2005). Self-oriented perfectionism and its perfectionism based on reinforcement theory. Behavior Modifi-
relationship to positive and negative affect: The mediation of cation, 22, 372–390.
positive and negative perfectionism cognitions. Cognitive Ther- Slaney, R. B., Rice, K. G., Mobley, M., & Trippi, J. (2001). The
apy and Research, 29, 555–567. revised almost perfect scale. Measurement and Evaluation in
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Counseling and Development, 34, 130–145.
Implications for cognition, emotion, and motivation. Psycholog- Snyder, C. R. (2000). Handbook of hope: Theory, measures, and
ical Review, 98, 224–253. applications. San Diego: Academic Press.
Marshall, G. N., Wortman, C. B., Kusulas, J. W., Hervig, L. K., & Snyder, C. R., Harris, C., Anderson, J. R., Holleran, S. A., Irving, L.
Vickers, R. R., Jr. (1992). Distinguishing optimism from M., Sigmon, S. T., et al. (1991). The will and the ways:
pessimism: Relations to fundamental dimensions of mood and Development and validation of an individual-differences mea-
personality. Journal of Personality and Social Psychology, 62, sure of hope. Journal of Personality and Social Psychology, 60,
1067–1074. 570–585.
Masuda, T., & Nisbett, R. E. (2001). Attending holistically versus Stoeber, J., Kobori, O., & Tanno, Y. (2010). The Multidimensional
analytically: Comparing the context sensitivity of Japanese and Perfectionism Cognitions Inventory-English (MCPI-E): Reli-
Americans. Journal of Personality and Social Psychology, 81, ability, validity, and relationships with positive and negative
922–934. affect. Journal of Personality Assessment, 92, 16–25.
Miyamoto, Y., Nisbett, R. E., & Masuda, T. (2006). Culture and the Stoeber, J., & Otto, K. (2006). Positive conceptions of perfectionism:
physical environment: Holistic versus analytic perceptual affor- Approaches, evidence, challenges. Personality and Social
dances. Psychological Science, 17, 113–119. Psychology Review, 10, 295–319.
Mobley, M., Slaney, R. B., & Rice, K. G. (2005). Cultural validity of Stumpf, H., & Parker, W. D. (2000). A hierarchical structural analysis
the Almost Perfect Scale—Revised for African American college of perfectionism and its relation to other personality character-
students. Journal of Counseling Psychology, 52, 629–639. istics. Personality and Individual Differences, 28, 837–852.
Mor, N., & Winquist, J. (2002). Self-focused attention and negative Sumi, K., & Kanda, K. (2002). Relationship between neurotic
affect: A meta-analysis. Psychological Bulletin, 128, 638–662. perfectionism, depression, anxiety, and psychosomatic symp-
Nes, L. S., & Segerstrom, S. C. (2006). Dispositional optimism and toms: A prospective study among Japanese men. Personality and
coping: A meta-analytic review. Personality and Social Psy- Individual Differences, 32, 817–826.
chology Review, 10, 235–251. Taylor, S. E., & Brown, J. D. (1988). Illusion and well being: A social
Nolen-Hoeksema, S., Larson, J., & Grayson, C. (1999). Explaining psychological perspective on mental health. Psychological
the gender difference in depressive symptoms. Journal of Bulletin, 103, 193–210.
Personality and Social Psychology, 77, 1061–1072. Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO:
Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking Westview Press.
individualism and collectivism: Evaluation of theoretical Triandis, H. C. (1996). The psychological measurement of cultural
assumptions and meta-analyses. Psychological Bulletin, 128, syndromes. American Psychologist, 51, 407–415.
3–72. Waterman, A. S. (1984). The psychology of individualism. New York:
Parker, W. D., & Stumpf, H. (1995). An examination of the multidi- Praeger.
mensional perfectionism scale with a sample of academically Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and
talented children. Journal of Psychoeducational Assessment, 13, validation of brief measures of positive and negative affect: The
372–383. PANAS scales. Journal of Personality and Social Psychology,
Rice, K. G., & Ashby, J. S. (2007). An efficient method for classifying 54, 1063–1070.
perfectionists. Journal of Counseling Psychology, 54, 72–85. Weisz, J. R., Rothbaum, F. M., & Blackburn, T. C. (1984). Standing
Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on out and standing in: The psychology of control in America and
the meaning of psychological well-being. Journal of Personality Japan. American Psychologist, 39, 955–969.
and Social Psychology, 57, 1069–1081. Zane, N. W. S., Sue, S., Hu, L.-T., & Kwon, J.-H. (1991). Asian-
Sanna, L. J., Kennedy, L. A., Chang, E. C., & Miceli, P. M. (2009). American assertion: A social learning analysis of cultural
When thoughts don’t feel like they used to: Changing feelings of differences. Journal of Counseling Psychology, 38, 63–70.
subjective ease in judgments of the past. Journal of Experimental
Social Psychology, 45, 940–946.

123