Journal of Counseling Psychology Copyright 2004 by the American Psychological Association, Inc.

2004, Vol. 51, No. 1, 93–102 0022-0167/04/$12.00 DOI: 10.1037/0022-0167.51.1.93

How Adaptive and Maladaptive Perfectionism Relate to Positive and

Negative Psychological Functioning: Testing a Stress-Mediation Model in
Black and White Female College Students

Edward C. Chang Angela F. Watkins

University of Michigan Spelman College

Kira Hudson Banks

University of Michigan
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

This study assessed racial variations in how adaptive and maladaptive perfectionism relate to psycho-
logical functioning in a sample of 150 Black and 150 White female college students. Comparative results
indicated that Black women, as compared with White women, reported less adaptive perfectionism, less
life satisfaction, greater stress, and greater negative affect. Correlational results indicated that for both
groups, maladaptive perfectionism, but not adaptive perfectionism, was associated with stress. Accord-
ingly, a model in which stress mediates the link between maladaptive perfectionism and psychological
functioning was tested. Overall, path-analytic results indicated that stress completely or partially
mediated the link between maladaptive perfectionism and psychological functioning for both Black and
White women. Potential implications of the present findings for counseling Black and White women are
discussed.

In recent decades, there has been growing interest in studying in research published from 1970 to 1989 reflected in half a dozen
the relationship between perfectionism and psychological func- leading American Psychological Association (APA) journals. Re-
tioning in adults (Flett & Hewitt, 2002). Recent reviews of per- sults from her analyses indicated a steady decline in both the
fectionism studies (e.g., Blatt, 1995; Chang, 2003; Shafran & number and the percentage of articles that focused on the experi-
Mansell, 2001) indicate that perfectionism is a key variable asso- ences of Blacks. Thus, it is not surprising that of the hundreds of
ciated with negative psychological functioning, including greater studies published on perfectionism over the past decade, less than
depressive symptomatology, greater anxiety, greater stress, greater half a dozen of these looked at perfectionism in Blacks (e.g.,
suicidal risk, greater eating disturbances, and poorer outcomes in Striegel-Moore et al., 2000). Even among those studies that have
the treatment of clinical depression. However, several key ques- been conducted, some have used a measure of perfectionism tied
tions have yet to be resolved in studying the relationship between to eating disturbances (Garner, Olmstead, & Polivy, 1983) rather
perfectionism and psychological functioning. than a measure related to general psychological functioning. In
fact, few studies have looked at perfectionism, broadly defined, in
On the (Unknown) Generalizability of Perfectionism Black young adults (e.g., Castro & Rice, 2003), and only one study
Theory and Research to Diverse Racial Groups: Most of has examined perfectionism comparatively between Black and
the Participants Were White, and They Still Are a Decade White, male and female young adults.
Later! In a study involving 118 Black and 299 White college students,
Nilsson, Paul, Lupini, and Tatem (1999) found variations between
Gender, race, and culture have been neglected in mainstream
Black and White students on levels of perfectionism. Using Frost,
psychological research (Sue, Bingham, Porché-Burke, & Vasquez,
Marten, Lahart, and Rosenblate’s (1990) Multidimensional Perfec-
1999). More than a decade ago, Graham (1992) analyzed the trend
tionism Scale (MPS), these investigators found that Black stu-
dents, as compared with White students, reported greater parental
expectations. In contrast, White students, in comparison with
Edward C. Chang and Kira Hudson Banks, Department of Psychology, Black students, reported greater concern over mistakes and paren-
University of Michigan; Angela F. Watkins, Department of Psychology, tal criticism. When the two groups were compared on the basis of
Spelman College. gender, Nilsson et al. found that Black female students in com-
The first author wishes to acknowledge Myung-Sook Tae and Suk-
parison with White female students reported greater parental ex-
Choon Chang for their support and encouragement in making this study
possible. pectations, whereas White female students reported greater con-
Correspondence concerning this article should be addressed to Edward cern over mistakes and parental criticism. Among male students,
C. Chang, Department of Psychology, University of Michigan, 525 East Blacks, in comparison with Whites, reported greater parental ex-
University Avenue, Ann Arbor, MI 48109. E-mail: changec@umich.edu pectations only. Thus, racial variations on perfectionism appeared

93
 94 CHANG, WATKINS, AND BANKS

to be greater for female students than for male students. Although ous indices of psychological functioning remains quite small.
Nilsson et al.’s findings provide a warrant for conducting addi- Clearly, when examining perfectionism in Black and White fe-
tional investigations of perfectionism between Black and White males, it would be useful to determine how adaptive and maladap-
female college students, their findings do little more than indicate tive perfectionism relate to key indices of psychological
normative differences on levels of perfectionism across the two functioning.
groups. To determine whether racial variations on perfectionism
between Black and White female college students translate into Additional Theoretical Considerations in Studying Black
differences in the function of perfectionism between the two and White Women: Positive and Negative Psychological
groups, it would be necessary to also examine how perfectionism Functioning and the Role of Stress as a Mediator of the
relates to various indices of psychological functioning in Black Relationship Between Perfectionism and Psychological
and White female students.
Functioning
Most studies have focused on the relationship between perfec-
Hamachek’s (1978) Distinction Between Adaptive and
tionism and negative psychological functioning. Beyond the Frost
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Maladaptive Perfectionism
This document is copyrighted by the American Psychological Association or one of its allied publishers.

et al. (1993) study described earlier, only two additional studies

Decades before researchers conducted numerous studies exam- have focused on how perfectionism relates to indices of both
ining the relationship between perfectionism and psychological positive psychological functioning (e.g., life satisfaction, positive
functioning, Hamachek (1978) made a useful distinction between affect) and negative psychological functioning (e.g., negative af-
normal or adaptive and neurotic or maladaptive perfectionism: fect, worry). In a study of young and middle-age adults, Chang
(2000) found that for both groups, total MPS scores were associ-
Persons who might fit under the label “normal perfectionists” (whom ated with less life satisfaction but with greater negative affect and
we could just as easily refer to as skilled artists or careful workers or worry. For both groups, the total MPS score was found to be
masters of their craft) are those who derive a very real sense of unrelated to positive affect. In addition, greater perfectionism was
pleasure from the labors of a painstaking effort and who feel free to be associated with greater stress in both groups. Subsequent path
less precise as the situation permits. . . . This is not, however, apt to
analyses indicated that stress wholly mediated the association of
be true for neurotic perfectionists. Here we have the sort of people
whose efforts— even their best ones—never seem quite good enough,
perfectionism with life satisfaction but only partially mediated the
at least in their own eyes. . . . They are unable to feel satisfaction associations of perfectionism with negative affect and worry. In a
because in their own eyes they never seem to do things good enough more recent study involving college students, Dunkley et al.
to warrant that feeling. (p. 27) (2003) found that maladaptive perfectionism (self-critical perfec-
tionism) was associated with greater negative affect and with less
Only recently have efforts been made by researchers to integrate positive affect. Alternatively, these investigators found that adap-
models of perfectionism within Hamachek’s (1978) framework tive perfectionism (personal standards perfectionism) was associ-
(Slade & Owens, 1998). Initially, for example, Frost et al. (1990) ated with greater negative affect but was unrelated to positive
argued that perfectionism represents a multidimensional construct affect. Maladaptive perfectionism, but not adaptive perfectionism,
involving concern over mistakes, personal standards, parental crit- was also found to be associated with greater stress, which in turn
icism, parental expectations, doubts about actions, and organiza- (along with several other variables) was found to partially mediate
tion. Respective MPS subscales measure these perfectionist ten- the association between maladaptive perfectionism and negative
dencies. More recently, however, Frost, Heimberg, Holt, Mattia, affect. Unfortunately, these studies have notable limitations. For
and Neubauer (1993) showed that the MPS subscales, along with example, in reporting on the total MPS score, Chang’s (2000)
those from another multidimensional perfectionism measure, map findings do not clarify the extent to which the associations found
onto an adaptive dimension (positive strivings) and a maladaptive involved adaptive perfectionism, maladaptive perfectionism, or
dimension (maladaptive evaluation concern). According to these both. Likewise, Dunkley et al.’s intriguing finding of a positive
investigators, scores on Personal Standards and Organization are association between a measure of adaptive perfectionism and neg-
aligned with adaptive perfectionism, whereas scores on Concern ative affect seems to go against past research and theory on
Over Mistakes, Parental Expectations, Parental Criticism, and adaptive perfectionism. However, in our view, the greatest limita-
Doubts About Actions are aligned with maladaptive perfectionism. tion across the Frost et al. (1993), Chang (2000), and Dunkley et
Consistent with this view, Frost et al. (1993) found that adaptive al. (2003) studies is that the majority of participants were White.
perfectionism was positively related to positive affect but was Thus, it remains unclear how adaptive and maladaptive perfection-
unrelated to depressive symptoms and negative affect. Alterna- ism relate to indices of positive and negative psychological func-
tively, they found that maladaptive perfectionism was positively tioning in non-White populations.
related to depressive symptoms and negative affect. Similar find- Positive psychological functioning can be defined by two indi-
ings have been reported by other researchers who have made a ces, namely, positive affect and life satisfaction (Ryff & Keyes,
conceptual distinction between adaptive and maladaptive perfec- 1995). Alternatively, negative psychological functioning can be
tionism (e.g., Dunkley, Zuroff, & Blankstein, 2003; Rice, Ashby, defined by negative affect, which taps many of the emotional
& Slaney, 1998; Slaney, Rice, Mobley, & Trippi, 2001; Terry- qualities found in most measures of psychological maladjustment,
Short, Owens, Slade, & Dewey, 1995). Unfortunately, relative to including measures of depression and anxiety (Watson, Clark, &
the number of studies conducted on perfectionism in the past Tellegen, 1988). Because life satisfaction, positive affect, and
decade, the number of studies focusing on distinguishing the negative affect are all generally believed to have universal rele-
relations of adaptive versus maladaptive perfectionism with vari- vance (Keyes, Shmotkin, & Ryff, 2002), they should represent key
 PERFECTIONISM IN BLACK AND WHITE WOMEN 95

measures of positive and negative psychological functioning in Purpose of the Present Research
both Black and White females. Examining how adaptive and
maladaptive perfectionism relate to positive and negative psycho- Given the aforementioned limitations and concerns associated
logical functioning in Black and White females would help clarify with previous research on perfectionism, the goals of the present
study were (a) to examine whether there are mean differences on
past findings and identify any significant racial variations. More-
measures of adaptive and maladaptive perfectionism, stress, and
over, in considering key measures of negative psychological func-
psychological functioning between Black and White females; (b)
tioning in female college students, there are at least three reasons
to examine whether there are differences in how adaptive and
to also look at suicide ideation. First, women make three times
maladaptive perfectionism relate to positive psychological func-
more suicide attempts than men (Canetto & Lester, 1995), and thus
tioning (positive affect and life satisfaction) and negative psycho-
suicide ideation is likely to be more pronounced in women than in logical functioning (negative affect and suicide ideation) in Black
men. Second, studies have shown that suicide-related behaviors, and White females; and (c) to determine whether stress mediates
including suicide ideation, are not uncommon in college popula- the associations of adaptive and maladaptive perfectionism with
tions. For example, Westefeld and Furr (1987) found the preva- positive and negative psychological functioning differently in
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

lence of suicide ideation across three samples of college students Black and White females.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

to range from 24% to 46%. Third, a link between perfectionism Given the preliminary nature of past comparative findings be-
and suicide potential, including suicide ideation, has been identi- tween Black and White females (Nilsson et al., 1999), and given
fied in several recent studies (e.g., Chang, 1998, 2002; Hewitt, our emphasis on examining the generalizability of previous find-
Flett, & Weber, 1994). Yet, it is not known how adaptive and ings based on Whites to Blacks, we hypothesized that there would
maladaptive perfectionism relate to suicide ideation or whether be no normative difference between Black and White women on
any associations that exist are similar or different between Black adaptive perfectionism or on maladaptive perfectionism. In a sim-
and White females. Resolving this question may be helpful for ilar vein, insofar as the distinction between adaptive and maladap-
understanding potential variations regarding which variables may tive perfectionism represents a useful theoretical framework that is
contribute to the greater suicide risk in Black compared with White applicable to all adults (Hamachek, 1978), we also predicted that
females and what the mechanisms may be (Jackson, 1990). for both Black and White women, adaptive perfectionism would be
Furthermore, it would be useful to test the generalizability of a positively associated with positive psychological functioning and
more complex model, namely, one involving stress as a mediator negatively related to negative psychological functioning. Alterna-
of the link between perfectionism and psychological functioning tively, for both Black and White women, we predicted that mal-
among Black and White females. Such a mediation model is adaptive perfectionism would be negatively associated with posi-
tive psychological functioning and positively associated with
illustrated in Figure 1. Although Hewitt, Flett, and Ediger (1996)
negative psychological functioning. Finally, consistent with a
have tested for stress as a mediator, they have also contended that
stress-mediation hypothesis (Hewitt et al., 1996), we predicted that
“perfectionistic behavior can generate stress that stems, in part,
for both Black and White females, the associations of maladaptive
from the tendency for perfectionists to evaluate stringently, focus
perfectionism— but not adaptive perfectionism (Dunkley et al.,
on negative aspects of performance, and experience little satisfac- 2003)—with positive and negative psychological functioning
tion” (p. 276). In a recent test of stress as a mediator, Chang would be mediated by stress.
(2000), for example, did not find stress to fully mediate the
associations of perfectionism with psychological functioning, yet a
stress-mediation model may still be useful. Recall that Chang Method
(2000) did not distinguish between adaptive and maladaptive per- Participants
fectionism. However, there is some reason to suspect that stress
may be more likely to serve as a mediator of the link between Participants were 150 Black female college students attending a small
maladaptive perfectionism than between adaptive perfectionism historically Black college and 150 White female college students attending
a large predominantly White university matched on age and class (fresh-
and psychological functioning (Dunkley et al., 2003). Thus, for
man, sophomore, etc.). Initially, the Black female sample was composed of
both Black and White females, the link between maladaptive 152 college students. However, the responses provided by 1 participant
perfectionism and psychological functioning may be mediated by were dropped because of an incomplete survey, and the responses provided
stress. by 1 mature adult participant were also dropped because of failure to
identify a same-age White participant. All participants were enrolled in a
psychology course in their respective schools and received extra credit for
participating. For both racial groups, ages ranged from 18 to 23 years (M ⫽
19.66, SD ⫽ 0.99).

Measures
Adaptive and maladaptive perfectionism. The Multidimensional Per-
fectionism Scale (MPS; Frost et al., 1990) is a 35-item multidimensional
measure of perfectionism consisting of the following six subscales: Con-
cern Over Mistakes (e.g., “People will probably think less of me if I make
Figure 1. A hypothesized model of how stress may mediate the associ- a mistake”), Personal Standards (e.g., “I set higher goals than most peo-
ation between perfectionism and psychological functioning. ple”), Parental Expectations (e.g., “My parents have expected excellence
 96 CHANG, WATKINS, AND BANKS

from me”), Parental Criticism (e.g., “I never felt like I could meet my reported in Watson et al. (1988). Test–retest reliability (8 weeks) for the
parents’ standards”), Doubts About Actions (e.g., “Even when I do some- Negative Affect scale has been reported to be .71 (Watson et al., 1988).
thing very carefully, I often feel that it is not quite right”), and Organization The Adult Suicide Ideation Questionnaire (ASIQ; Reynolds, 1991a) is a
(e.g., “Neatness is very important to me”). Respondents are asked to rate 25-item measure designed to assess for thoughts related to suicide in adult
items across a 5-point Likert-type scale ranging from 1 (strongly disagree) populations. Each item measures a specific suicidal thought or behavior
to 5 (strongly agree). On the basis of the framework proffered by Frost et (e.g., “I thought about killing myself”). Respondents are asked to rate the
al. (1993), to obtain a measure of adaptive perfectionism, we aggregated presence and frequency of experiencing a specific suicidal thought in the
scores from the Personal Standards and Organization subscales to form the past month across a 7-point scale ranging from 0 (“I have never had this
MPS Adaptive subscale. Similarly, to obtain a measure of maladaptive thought”) to 6 (“Almost every day”). In support of its validity, ASIQ scores
perfectionism, we aggregated scores from the Concern Over Mistakes, have been found to be positively correlated with prior suicide attempts
Parental Expectations, Parental Criticism, and Doubts About Actions sub- among college students (Reynolds, 1991a). In general, higher scores on the
scales to form the MPS Maladaptive subscale. ASIQ are indicative of greater severity and frequency of suicidal thinking
Higher scores on the MPS Adaptive subscale reflect greater levels of (Reynolds, 1991a). Test–retest reliability (2 weeks) for the ASIQ in a
adaptive perfectionism, whereas higher scores on the MPS Maladaptive college student sample has been reported to be .86 (Reynolds, 1991b).
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

subscale reflect greater levels of maladaptive perfectionism. Test–retest

This document is copyrighted by the American Psychological Association or one of its allied publishers.

reliabilities (10 weeks) for the different MPS subscales have been reported Procedure
to range from .63 to .88 (Rice & Dellwo, 2001).
Stress. The Perceived Stress Scale (PSS; S. Cohen, Kamarck, & Mer- Surveys given to Black female participants were administered and
melstein, 1983) is a 14-item measure of self-appraised life stress in the past collected by a Black female experimenter. Surveys given to White female
participants taking a large class were administered and collected by an
month (e.g., “In the last month, how often have you been upset because of
Asian American male experimenter. The responses provided by all 150
something that happened unexpectedly?”). We selected the PSS to assess
Black female participants were matched on age and college class standing
for stress over life events or hassles surveys because the PSS does not
with those obtained from 150 White female participants randomly drawn
constrain respondents to a specific list or number of stressors. In addition,
from a larger pool of 214 White female participants. We used this matching
we wanted to build on Chang’s (2000) examination of the proposed
procedure to control for variations between Black and White female
stress-mediation model, which also relied on the PSS as a stress measure.
participants on these two developmental variables. As past findings have
Respondents are asked to rate the frequency of these items across a 5-point
indicated, levels of perfectionism, stress, and psychological functioning
Likert-type scale ranging from 0 (never) to 4 (very often). Higher scores can all be strongly influenced by developmental factors (Chang, 2000).
reflect greater perceived stress in the past month. Evidence for construct All study measures were administered to participants in the form of a
validity of the PSS with life events measures has been reported in S. Cohen take-home survey that was to be returned the next day of class. For our
et al. (1983). Test–retest reliability (6 weeks) for the PSS has been reported Black female sample, 153 surveys were initially distributed to Black
to be .55 (S. Cohen et al., 1983). female students attending one of two psychology classes at the historically
Positive psychological functioning. For the present study, two mea- Black college, resulting in a response rate of 99%. For our White female
sures were used to assess for positive psychological functioning, namely, sample, 306 surveys were initially distributed to students attending a large
positive affect and life satisfaction. Positive affect was assessed with the psychology class at a predominantly White university. Two hundred
Positive Affect scale of the Positive and Negative Affect Schedule seventy-five responses were returned, resulting in a general response rate
(PANAS; Watson et al., 1988). The PANAS is a 20-item self-report of 90%. Because we were unable to identify the gender or race of the 31
measure of positive and negative affect, with 10 items assessing for nonrespondents from the initial participant pool, we assumed that all were
positive affect (e.g., “enthusiastic”) and 10 items assessing for negative White women in calculating response rate for this group. Thus, we ob-
affect (e.g., “irritable”). For each item, respondents are asked to rate how tained a very conservative response rate estimate of 87% for White women.
they felt “during the past week” across a 5-point Likert-type scale ranging Participants were not made aware of the purpose of the study until after the
from 1 (very slightly) to 5 (extremely). Higher scores on the Positive Affect study was completed. To protect the participants’ anonymity, only partic-
scale reflect greater positive affect. Evidence for the construct validity of ipant identification numbers were placed on the instruments. In addition,
the Positive Affect scale has been reported in Watson et al. (1988). all participants signed consent forms indicating that all test data would be
Test–retest reliability (8 weeks) for the Positive Affect scale has been kept strictly confidential.
reported to be .68 (Watson et al., 1988).
We used the Satisfaction With Life Scale (Diener, Emmons, Larsen, & Results
Griffin, 1985) to assess life satisfaction. The Satisfaction With Life Scale
is a 5-item measure of global life satisfaction (e.g., “I am satisfied with my First we conducted a multivariate analysis of variance
life”), or a person’s satisfaction with life as a whole, rather than with any (MANOVA) between Black and White women on the present set
specific domain (Pavot & Diener, 1993). Respondents are asked to rate the of measures. A significant multivariate difference was found be-
extent of their agreement with these items across a 7-point Likert-type tween Black and White women on the present measures, ⌳(7,
scale ranging from 1 (strongly disagree) to 7 (strongly agree). Higher 292) ⫽ .85, p ⬍ .001. Accordingly, we conducted univariate
scores on the scale reflect greater life satisfaction. Test–retest reliability (8 analyses between Black and White women for each of the present
weeks) for the scale has been reported to be .82 (Diener et al., 1985).
measures to determine where significant differences existed.
Negative psychological functioning. For the present study, two mea-
sures were used to assess for negative psychological functioning, namely,
negative affect and suicide ideation. As noted above, the PANAS is a Do Black and White Women Differ on Measures of
20-item self-report measure of affect. The 10-item Negative Affect scale of Adaptive and Maladaptive Perfectionism, Stress, and
the PANAS was used to assess for negative affect (Watson et al., 1988). Positive and Negative Psychological Functioning?
Respondents are asked to rate how they felt for each item “during the past
week” across a 5-point Likert-type scale ranging from 1 (very slightly) to Before analyzing the results of our univariate analyses, we
5 (extremely). Higher scores on the Negative Affect scale reflect greater assessed whether our operationalization of adaptive and maladap-
negative affect. Evidence for the construct validity of the scale has been tive perfectionism was empirically valid for both Black and White
 PERFECTIONISM IN BLACK AND WHITE WOMEN 97

women. We conducted an exploratory factor analysis on the six .20), medium (d ⫽ .50), and large effects (d ⫽ .80), we found that
MPS subscales, using principal-axis factoring with varimax rota- the difference in the mean MPS Adaptive subscale score between
tion (Frost et al., 1993). Results of this analysis are presented in Black and White women reflected a small effect (d ⫽ .29). The
Table 1. As shown in the table, results for Black women indicated difference in the mean PSS score between Black and White
two clear factors with eigenvalues greater than 1. Factor 1 was women was also found to reflect a small effect (d ⫽ .38). Simi-
composed of Concern Over Mistakes, Parental Expectations, Pa- larly, the difference in the mean Satisfaction With Life Scale
rental Criticism, and Doubts About Actions (factor loadings scores between Black and White women was found to reflect a
ranged from .68 to .79), which accounted for 42% of the variance. small effect (d ⫽ .48). The difference in the mean Negative Affect
Factor 2 was composed of Personal Standards and Organization scores between Black and White women was found to reflect a
(factor loadings were .78 and .81, respectively), which accounted medium effect (d ⫽ .52). Finally, the difference in the mean ASIQ
for an additional 23% of the variance. Results for White women scores between Black and White women reflected a small effect
also indicated two clear factors with eigenvalues greater than 1. (d ⫽ .21).
Again, Factor 1 was composed of the Concern Over Mistakes,
Parental Expectations, Parental Criticism, and Doubts About Ac-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

How Do Adaptive and Maladaptive Perfectionism Relate

This document is copyrighted by the American Psychological Association or one of its allied publishers.

tions (factor loadings ranged from .57 to .83), which accounted for to Stress and to Positive and Negative Psychological
36% of the variance. Similarly, Factor 2 was composed of Personal Functioning in Black and White Women?
Standards and Organization (factor loadings were .79 and .83,
respectively), which accounted for an additional 29% of the vari- To examine how adaptive and maladaptive perfectionism relates
ance. However, unlike the case for Black women, Parental Expec- to stress and psychological functioning, we computed correlations
tations and Parental Criticism for White women were found to between the present set of variables for Black and White women
have modest loadings of .38 and .36, respectively, on Factor 2. It (see Table 3). As the table shows, for Black women, adaptive
is worth noting that these results not only support the validity of perfectionism was associated with less suicide ideation. For White
our operationalization of adaptive and maladaptive perfectionism women, adaptive perfectionism was associated with greater posi-
for both Black and White women, but they also join Stumpf and tive affect and life satisfaction. For both Black and White women,
Parker’s (2000) factor-analytic results to suggest that the MPS, maladaptive perfectionism was associated with greater stress, neg-
without the inclusion of other measures, can be used to assess for ative affect, and suicide ideation and with less positive affect. In
adaptive and maladaptive dimensions of perfectionism. addition, for White women, maladaptive perfectionism was also
Results of our univariate analyses are presented in Table 2. As associated with less life satisfaction. Consistent with past research
the table shows, Black women, as compared with White women, (e.g., Frost et al., 1993), adaptive and maladaptive perfectionism in
were found to report significantly greater stress and negative both Black and White women were not found to be strongly
affect. Alternatively, White women in comparison with Black redundant with each other. Thus, Hamachek’s (1978) notion of
women were found to report significantly greater adaptive perfec- adaptive and maladaptive perfectionism may serve not only to
tionism and life satisfaction. However, because statistical values distinguish between individuals but also to distinguish perfection-
do not directly convey the magnitude of the differences found, we istic attributes within the individual.
computed effect sizes using Cohen’s d (J. Cohen, 1977). For these
analyses, sigma was adjusted given the unequal standard devia- Does Stress Account for the Relations of Adaptive and
tions underlying the two samples. The results of these analyses are Maladaptive Perfectionism With Positive and Negative
presented in Table 1. Using J. Cohen’s convention for small (d ⫽ Psychological Functioning in Black and White Women?
Examining a Stress-Mediation Model
Table 1 Because maladaptive perfectionism, but not adaptive perfection-
Results of Principal-Axis Factor Analysis (Varimax Rotation) ism, was found to be associated with stress and psychological
Conducted on the Subscales of the Multidimensional functioning in both Black and White women, we tested a model
Perfectionism Scale for Black and White Female Students involving stress as a mediator of the associations of maladaptive
perfectionism with psychological functioning. Following the gen-
Factor 1 Factor 2 eral guidelines of Baron and Kenny (1986), to establish evidence
for the proposed mediation model by conducting a series of re-
Perfectionism Black White Black White
gression analyses, it would be necessary to meet three conditions.
Maladaptive perfectionism First, results of regressing stress on maladaptive perfectionism
Concern Over Mistakes .79 .83 ⫺.20 ⫺.11 must show that maladaptive perfectionism is significantly associ-
Parental Expectations .68 .57 .12 .38 ated with stress. Second, results of regressing psychological func-
Parental Criticism .79 .68 ⫺.14 .36 tioning on maladaptive perfectionism must show that maladaptive
Doubts About Actions .75 .58 .03 .20
Adaptive perfectionism perfectionism is significantly associated with psychological func-
Personal Standards .15 .26 .78 .79 tioning. Third, results of regressing psychological functioning on
Organization .20 .13 .81 .83 both maladaptive perfectionism and stress should result in a
Eigenvalue 2.50 2.31 1.37 1.46 weaker association between maladaptive perfectionism and psy-
Percentage of variance 41.6 35.9 22.9 29.0
chological functioning. Complete mediation would be indicated if
Note. n ⫽ 150 Black female students; n ⫽ 150 White female students. the association between maladaptive perfectionism and stress and
Significant factor loadings (ⱖ .57) are in boldface type. the association between stress and psychological functioning were
 98 CHANG, WATKINS, AND BANKS

Table 2
Racial Variations in Adaptive and Maladaptive Perfectionism, Stress, and Positive and Negative
Psychological Functioning Among Black and White Female Students

Black (n ⫽ 150) White (n ⫽ 150)

Scale M SD ␣ M SD ␣ t(1, 298) Cohen’s d

MPS-AS 47.92 7.55 .86 50.01 7.13 .88 ⫺2.46* .29

MPS-MS 55.51 12.93 .90 54.89 12.40 .90 0.42 .05
PSS 26.91 7.93 .89 24.19 6.38 .86 3.27** .38
PA 35.36 5.77 .84 35.09 5.17 .83 0.27 .05
SWLS 24.21 6.49 .93 27.09 5.27 .94 ⫺4.22*** .48
NA 21.71 6.53 .85 18.65 5.13 .84 4.52*** .52
ASIQ 12.08 19.36 .97 8.10 19.39 .98 1.78 .21
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Note. MPS-A ⫽ Multidimensional Perfectionism Scale—Adaptive subscale; MPS-M ⫽ Multidimensional

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Perfectionism Scale—Maladaptive subscale; PSS ⫽ Perceived Stress Scale; PA ⫽ Positive Affect; SWLS ⫽
Satisfaction With Life Scale; NA ⫽ Negative Affect; ASIQ ⫽ Adult Suicide Ideation Questionnaire.
* p ⬍ .05. ** p ⬍ .01. *** p ⬍ .001.

significant, but the previously significant association between mal- link between maladaptive perfectionism and positive affect for
adaptive perfectionism and psychological functioning became White women (␤ ⫽ ⫺.31) became nonsignificant (␤ ⫽ ⫺.03) and
nonsignificant after controlling for stress. Alternatively, partial was also found to be mediated by stress (⌬␤ ⫽ .28), indicating a
mediation would be indicated if the association between maladap- reduction of 90.3% of the variance in positive affect (Sobel’s
tive perfectionism and psychological functioning became signifi- statistic ⫽ ⫺5.05, p ⬍ .001). For this group, the resulting model
cantly reduced, but not nonsignificant, after controlling for stress. involving maladaptive perfectionism and stress accounted for
As suggested by Baron and Kenny, we used Sobel’s statistic to test 35.3% of the variance in positive affect, F(2, 147) ⫽ 40.09, p ⬍
the magnitude and significance of the reduction in variance result- .001. For life satisfaction, as Figure 2B shows, the previously
ing from the inclusion of stress as a mediator in our analyses. significant link between maladaptive perfectionism and life satis-
Because maladaptive perfectionism was not related to life satis- faction for Whites (␤ ⫽ ⫺.42) was not completely mediated by
faction for Black women, we did not test a mediation model for life stress (⌬␤ ⫽ .22). However, evidence supporting partial mediation
satisfaction for this group. was found, indicating a reduction of 52.4% of the variance in life
Results of our analyses for Black and White women for each of satisfaction (Sobel’s statistic ⫽ ⫺4.50, p ⬍ .001). For this group,
the two indices of positive and negative psychological functioning the resulting model involving maladaptive perfectionism and stress
are presented in Figures 2A through 2D. As Figure 2A shows, the accounted for 34.2% of the variance in life satisfaction, F(2,
previously significant link between maladaptive perfectionism and 147) ⫽ 38.18, p ⬍ .001.
positive affect for Black women (␤ ⫽ ⫺.23) became nonsignifi- As shown in Figure 2C, the previously significant link between
cant (␤ ⫽ ⫺.06) and was found to be completely mediated by maladaptive perfectionism and negative affect for Black women
stress (⌬␤ ⫽ .17), indicating a reduction of 73.9% of the variance (␤ ⫽ .27) became nonsignificant (␤ ⫽ .04) and was found to be
in positive affect (Sobel’s statistic ⫽ ⫺5.06, p ⬍ .001). For this completely mediated by stress (⌬␤ ⫽ .23), indicating a reduction
group, the resulting model involving maladaptive perfectionism of 85.2% of the variance in negative affect (Sobel’s statistic ⫽
and stress accounted for 34.9% of the variance in positive affect, 4.43, p ⬍ .001). For this group, the resulting model involving
F(2, 147) ⫽ 39.36, p ⬍ .001. Similarly, the previously significant maladaptive perfectionism and stress accounted for 26.2% of the

Table 3
Zero-Order Correlations Between All Study Measures for Black and White Female Students

Measure MPS-AS MPS-MS PSS PA SWLS NA ASIQ

MPS-AS — .02 ⫺.07 .15 .04 ⫺.07 ⫺.17*

MPS-MS .18* — .47*** ⫺.23** ⫺.08 .27*** .20**
PSS ⫺.11 .49*** — ⫺.59*** ⫺.50*** .51*** .44***
PA .32*** ⫺.31*** ⫺.59*** — .58*** ⫺.42*** ⫺.36***
SWLS .22** ⫺.42*** ⫺.56*** .44*** — ⫺.57*** ⫺.44***
NA ⫺.09 .45*** .60*** ⫺.43*** ⫺.31*** — .47***
ASIQ ⫺.06 .40*** .37*** ⫺.34*** ⫺.41*** .29*** —

Note. Correlations for Black female students (n ⫽ 150) are above the diagonal and correlations for White female students (n ⫽ 150) are below the
diagonal. MPS-AS ⫽ Multidimensional Perfectionism Scale-Adaptive subscale; MPS-MS ⫽ Multidimensional Perfectionism Scale-Maladaptive subscale;
PSS ⫽ Perceived Stress Scale; PA ⫽ Positive Affect; SWLS ⫽ Satisfaction With Life Scale; NA ⫽ Negative Affect; ASIQ ⫽ Adult Suicide Ideation
Questionnaire.
* p ⬍ .05. ** p ⬍ .01. *** p ⬍ .001.
 PERFECTIONISM IN BLACK AND WHITE WOMEN 99
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 2. Results of the path analyses delineating those paths found to be significant for at least one group in
looking at positive and negative indices of psychological functioning. The path found to be nonsignificant for
both groups is indicated by a broken line. All numbers represent standardized beta weights. Numbers outside of
parentheses are for Black female students (n ⫽ 150). Numbers within parentheses are for White female students
(n ⫽ 150). *p ⬍ .05. **p ⬍ .01. ***p ⬍ .001.

variance in negative affect, F(2, 147) ⫽ 26.04, p ⬍ .001. For .35), we found that the prediction model involving maladaptive
White women, the previously significant link between maladaptive perfectionism and stress in Black women accounted for a medium
perfectionism and negative affect (␤ ⫽ .45) was not completely amount of the variance in suicide ideation ( f 2 ⫽ .25) and for a
mediated by stress (⌬␤ ⫽ .24). However, evidence supporting large amount of the variance in both positive affect ( f 2 ⫽ .54) and
partial mediation was found, indicating a reduction of 53.3% of the negative affect ( f 2 ⫽ .36). For White women, the same prediction
variance in negative affect (Sobel’s statistic ⫽ 4.80, p ⬍ .001). For model accounted for a medium amount of the variance in suicide
this group, the resulting model involving maladaptive perfection- ideation ( f 2 ⫽ .26) and for a large amount of the variance in
ism and stress accounted for 39.4% of the variance in negative positive affect ( f 2 ⫽ .56), life satisfaction ( f 2 ⫽ .52), and nega-
affect, F(2, 147) ⫽ 47.77, p ⬍ .001. For suicide ideation, as Figure tive affect ( f 2 ⫽ .65). Hence, on the basis of effect size, and
2D shows, the previously significant link between maladaptive despite the unmediated associations for White participants, mal-
perfectionism and suicide ideation for Black women (␤ ⫽ .20) adaptive perfectionism and stress together appeared to account for
became nonsignificant (␤ ⫽ .01) and was completely mediated by a similar amount of the variance in positive affect, negative affect,
stress (⌬␤ ⫽ .19), indicating a reduction of 95.0% of the variance and suicide ideation between Black and White women.
in suicide ideation (Sobel’s statistic ⫽ 4.13, p ⬍ .001). For this
group, the resulting model involving maladaptive perfectionism Discussion
and stress accounted for 19.7% of the variance in suicide ideation,
F(2, 147) ⫽ 18.06, p ⬍ .001. For White students, the previously The present study tested the hypothesis that there would be no
significant link between maladaptive perfectionism and suicide differences on adaptive and maladaptive perfectionism between
ideation (␤ ⫽ .40) was not completely mediated by stress (⌬␤ ⫽ Black and White women. We found that White women, as com-
.11). However, evidence supporting partial mediation was found, pared with Black women, reported greater adaptive perfectionism.
indicating a reduction of 27.5% of the variance in suicide ideation In contrast, we found no difference between Black and White
(Sobel’s statistic ⫽ 2.78, p ⬍ .01). For this group, the resulting women on maladaptive perfectionism. In comparing Black and
model involving maladaptive perfectionism and stress accounted White women, we also looked at differences in stress and psycho-
for 20.4% of the variance in suicide ideation, F(2, 147) ⫽ 18.86, logical functioning. We found that Black women, in comparison
p ⬍ .001. Overall these results indicate racial differences in how with White women, reported greater stress and negative affect and
the association between maladaptive perfectionism and psycholog- reported less life satisfaction. As with the findings for adaptive
ical functioning may be mediated by stress. perfectionism, these differences indicate that there are notable
To appreciate the magnitude of these results, we computed normative differences between Black and White women on these
effect sizes using f 2 (J. Cohen, 1977). Following the convention measures also. These normative differences should not be taken to
for small ( f 2 ⫽ .02), medium ( f 2 ⫽ .15), and large effects ( f 2 ⫽ imply, for example, that Black women are somehow substantively
 100 CHANG, WATKINS, AND BANKS

more distressed and maladjusted than White women. The most tionism and psychological functioning in both Black and White
parsimonious interpretation of these comparative differences is women, our results also indicated that stress plays a stronger
that what is “normal” in Black women may simply be different mediating role in Black women than in White women. Accord-
from what is normal in White women and vice versa. ingly, interventions that target both an alleviation of stress (e.g.,
We also examined whether there were differences between Meichenbaum, 1985) and a reduction in maladaptive perfectionism
Black and White women on how adaptive and maladaptive per- (e.g., Ferguson & Rodway, 1994) may offer the most promise in
fectionism relate to psychological functioning. Results of correla- working with White female students experiencing poor positive
tional analyses indicated that adaptive perfectionism was associ- psychological functioning, excessive negative psychological func-
ated with less suicide ideation for Black women. In contrast, tioning, or both. In contrast, our path-analytic findings raise the
adaptive perfectionism was associated with greater positive affect possibility that stress, relative to maladaptive perfectionism, may
and life satisfaction for White women. For both groups, adaptive be a proximal and sufficient determinant of psychological func-
perfectionism was unrelated to stress and negative affect. Thus, the tioning in Black women. Thus, it may be useful for counselors to
function of adaptive perfectionism in Black and White women in focus more immediately on the assessment of stress experiences
relation to psychological functioning appears to encompass simi- than on perfectionistic tendencies in working with Black female
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

larities and differences. For both groups, however, maladaptive students experiencing poor psychological functioning.
perfectionism was associated with greater stress, negative affect, As several studies have shown, identifying the source of stress
and suicide ideation and with less positive affect. Greater mal- experiences unique to Black adults is critical (e.g., Mays,
adaptive perfectionism was also associated with less life satisfac- Coleman, & Jackson, 1996; Utsey & Ponterotto, 1996). For ex-
tion, but this was found for White women only. Thus, greater ample, although studies conducted on predominantly White adult
maladaptive perfectionism does not appear to necessarily be asso- samples indicate that everyday stress is associated with various
ciated with greater dissatisfaction with life for all groups (cf. indexes of negative psychological functioning, including suicide
Hamachek, 1978). ideation (e.g., Dixon, Rumford, Heppner, & Lips, 1992), a study
Finally, we tested the validity and generalizability of a model conducted by Klonoff, Landrine, and Ullman (1999) on a large
involving stress as a mediator of the link between perfectionism sample of Black adults indicated that experiences of racial dis-
and psychological functioning in Black and White women. Be- crimination accounted for 6% to 10% of unique variance in mea-
cause adaptive perfectionism was not found to be associated with sures of psychiatric symptoms after controlling for other relevant
stress in Black and White women, whereas maladaptive perfec- variables (e.g., gender, experiences of general stress). However,
tionism was, our findings suggest that the role of stress as a we agree with Mays’s (1988) contention that the experiences of
potential mediator of the link between perfectionism and psycho- Black women need to be distinguished from those of Black men
logical functioning may be restricted to situations involving mal- given that the experiences of each group are likely to be quite
adaptive perfectionism (Dunkley et al., 2003). different. Hence, we believe that identifying the source of stress
Results of a series of path analyses conducted to test our experiences unique among Black women remains a crucial topic
mediation model indicated that stress plays an important role in for future investigation.
both Black and White women. For Black women, stress com- Although a number of specific interventions have been proposed
pletely mediated the association between maladaptive perfection- for reducing stress in working with Black women, Mays (1986)
ism and psychological functioning in three out of four cases. Thus, found that a self-help group intervention that directly focuses on
for Black women, some support was found for Hewitt et al.’s building group cohesion among Black women, presenting Black
(1996) contention that the association between perfectionism and women with a useful model of facilitation, and promoting adaptive
psychological functioning may involve stress generated by the coping patterns among Black women was effective in decreasing
former. For White women, stress partially mediated the association experiences of stress and enhancing sense of social support and
between maladaptive perfectionism and psychological functioning interpersonal relationships. Mays’s (1986) study involved a small
in three out of four cases and completely mediated the association sample of working Black middle-aged women, and thus the appli-
in the one remaining case. Thus, we were able to also find support cability of her findings for working with distressed young adult
for Hewitt et al.’s stress-generation hypothesis in this group. The Black female college students remains unknown. Nevertheless, we
partial mediation results for White women are consistent with believe that there may be considerable value in extending Mays’s
those obtained in another study based on predominantly White promising intervention to situations in which women of similar
adult samples (Chang, 2000). racial backgrounds have a context in which to discuss and work
through their unique concerns and issues.
Promoting Psychological Functioning in Black and White
Female College Students: Targeting Stress in Black Some Limitations of This Study
Women and Maladaptive Perfectionism in White Women?
It is worth noting several potential limitations to the present
Because our path models accounted for a moderate amount of findings. First, this study focused on how adaptive and maladap-
variance in suicide ideation, and a large amount of the variance in tive perfectionism relates to broad indexes of positive and negative
positive affect, life satisfaction, and negative affect, we would be psychological functioning in Black and White female college
remiss if we did not discuss potential implications of our findings students. Insofar as perfectionism has been linked in adult popu-
for working with Black and White female students. In addition, it lations to other key indexes of psychological functioning (e.g.,
is important to note that although our path-analytic results indi- depressive symptoms, eating disturbances), it would be useful to
cated that stress mediates the associations of maladaptive perfec- expand on the present findings by including other measures of
 PERFECTIONISM IN BLACK AND WHITE WOMEN 101

psychological functioning in future studies. Second, and relatedly, Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of
it would be useful in future studies to consider other measures of perceived stress. Journal of Health and Social Behavior, 24, 386 –396.
stress (e.g., negative life events) in replicating the present findings Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The
as well as to consider testing for other potential mediators (e.g., Satisfaction With Life Scale. Journal of Personality Assessment, 49,
social problem-solving ability; Chang, D’Zurilla, & Sanna, in 71–75.
press). Third, the present findings are limited to young adults. Dixon, W. A., Rumford, K. G., Heppner, P. P., & Lips, B. J. (1992). Use
of different sources of stress to predict hopelessness and suicide ideation
Thus, for example, it remains useful to examine the form and
in a college population. Journal of Counseling Psychology, 39, 342–349.
function of adaptive and maladaptive perfectionism in more ma-
Dunkley, D. M., Zuroff, D. C., & Blankstein, K. R. (2003). Self-critical
ture Black and White females to identify possible developmental perfectionism and daily affect: Dispositional and situational influences
variations in adulthood. Last, given that our study involved a on stress and coping. Journal of Personality and Social Psychology, 84,
cross-sectional design, we cannot draw strong inferences about 234 –252.
cause and effect from the findings. Ferguson, K. L., & Rodway, M. R. (1994). Cognitive behavioral treatment
of perfectionism: Initial evaluation studies. Research on Social Work
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Practice, 4, 283–308.
Summary and Concluding Comment
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Flett, G. L., & Hewitt, P. L. (2002). Perfectionism: Theory, research, and

In conclusion, the present study addressed the need for addi- treatment. Washington, DC: American Psychological Association.
Frost, R. O., Heimberg, R. G., Holt, C. S., Mattia, J. I., & Neubauer, A. L.
tional investigation of the functioning of adaptive and maladaptive
(1993). A comparison of two measures of perfectionism. Personality
perfectionism across different racial groups. Our findings indicate
and Individual Differences, 14, 119 –126.
that results obtained in studying perfectionism in Black adults may Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimen-
be similar to, but not identical with, those obtained in studying sions of perfectionism. Cognitive Therapy and Research, 14, 449 – 468.
White adults. In addition, our path-analytic results for both Black Garner, D. M., Olmstead, M. P., & Polivy, J. (1983). Development and
and White participants point to the usefulness of considering stress validation of a multidimensional eating disorder inventory for anorexia
as a mechanism that may account for the association between nervosa and bulimia. International Journal of Eating Disorders, 2,
perfectionism and psychological functioning. Clearly, if research- 385–389.
ers seek to attain a more inclusive understanding of the value of Graham, S. (1992). “Most of the subjects were White and middle class”:
adaptive and maladaptive perfectionism in people’s lives, then Trends in published research on African Americans in selected APA
more studies will be needed to clarify potential similarities and journals, 1970 –1989. American Psychologist, 47, 629 – 639.
differences in the form and function of perfectionism across di- Hamachek, D. E. (1978). Psychodynamics of normal and neurotic perfec-
verse groups. tionism. Psychology: A Journal of Human Behavior, 15, 27–33.
Hewitt, P. L., Flett, G. L., & Ediger, E. (1996). Perfectionism and depres-
sion: Longitudinal assessment of a specific vulnerability hypothesis.
References Journal of Abnormal Psychology, 105, 276 –280.
Hewitt, P. L., Flett, G. L., & Weber, C. (1994). Dimensions of perfection-
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable ism and suicide ideation. Cognitive Therapy and Research, 18, 439 –
distinction in social psychological research: Conceptual, strategic, and 460.
statistical considerations. Journal of Personality and Social Psychology, Jackson, J. J. (1990). Suicide trends of Blacks and Whites by sex and age.
51, 1173–1182. In D. S. Ruiz (Ed.), Handbook of mental health and mental disorder
Blatt, S. J. (1995). The destructiveness of perfectionism: Implications for among Black Americans (pp. 95–109). New York: Greenwood Press.
the treatment of depression. American Psychologist, 50, 1003–1020. Keyes, C. L. M., Shmotkin, D., & Ryff, C. D. (2002). Optimizing well-
Canetto, S. S., & Lester, D. (1995). Women and suicidal behavior. New being: The empirical encounter of two traditions. Journal of Personality
York: Springer.
and Social Psychology, 82, 1007–1022.
Castro, J. R., & Rice, K. G. (2003). Perfectionism and ethnicity: Implica-
Klonoff, E. A., Landrine, H., & Ullman, J. B. (1999). Racial discrimination
tions for depressive symptoms and self-reported academic achievement.
and psychiatric symptoms among Blacks. Cultural Diversity & Ethnic
Cultural Diversity & Ethnic Minority Psychology, 9, 64 –78.
Minority Psychology, 5, 329 –339.
Chang, E. C. (1998). Cultural differences, perfectionism, and suicidal risk:
Mays, V. M. (1986). Black women and stress: Utilization of self-help
Does social problem solving still matter? Cognitive Therapy and Re-
groups for stress reduction. Women and Therapy, 4, 67–79.
search, 22, 237–254.
Mays, V. M. (1988). Even the rat was White and male: Teaching the
Chang, E. C. (2000). Perfectionism as a predictor of positive and negative
psychological outcomes: Examining a mediation model in younger and psychology of Black women. In P. Bronstein & K. Quina (Eds.),
older adults. Journal of Counseling Psychology, 47, 18 –26. Teaching a psychology of people: Resources for gender and social
Chang, E. C. (2002). Examining the link between perfectionism and awareness (pp. 142–146). Washington, DC: American Psychological
psychological maladjustment: Social problem solving as a buffer. Cog- Association.
nitive Therapy and Research, 26, 581–595. Mays, V. M., Coleman, L. M., & Jackson, J. S. (1996). Perceived race-
Chang, E. C. (2003). On the perfectibility of the individual: Going beyond based discrimination, employment status, and job stress in a national
the dialectic of good versus evil. In E. C. Chang & L. J. Sanna (Eds.), sample of Black women: Implications for health outcomes. Journal of
Virtue, vice, and personality: The complexity of behavior (pp. 125–143). Occupational Health Psychology, 1, 319 –329.
Washington, DC: American Psychological Association. Meichenbaum, D. H. (1985). Stress inoculation training. Elmsford, NY:
Chang, E. C., D’Zurilla, T. J., & Sanna, L. J. (Eds.). (in press). Social Pergamon Press.
problem solving: Theory, research, and training. Washington, DC: Nilsson, J. E., Paul, B. D., Lupini, L. N., & Tatem, B. (1999). Cultural
American Psychological Association. differences in perfectionism: A comparison of African American and
Cohen, J. (1977). Statistical power analysis for the behavioral sciences White college students. Journal of College Student Development, 40,
(Rev. ed.). New York: Academic Press. 141–150.
 102 CHANG, WATKINS, AND BANKS

Pavot, W., & Diener, E. (1993). Review of the Satisfaction With Life E., & Rodin, J. (2000). Eating disorder symptoms in a cohort of 11- to
Scale. Psychological Assessment, 5, 164 –172. 16-year-old Black and White girls: The NHLBI Growth and Health
Reynolds, W. M. (1991a). Adult Suicidal Ideation Questionnaire: Profes- Study. International Journal of Eating Disorders, 27, 49 – 66.
sional manual. Odessa, FL: Psychological Assessment Resources. Stumpf, H., & Parker, W. D. (2000). A hierarchical structural analysis of
Reynolds, W. M. (1991b). Psychometric characteristics of the Adult Sui- perfectionism and its relation to other personality characteristics. Per-
cidal Ideation Questionnaire with college students. Journal of Person- sonality and Individual Differences, 28, 837– 852.
ality Assessment, 56, 289 –307. Sue, D. W., Bingham, R. P., Porché-Burke, L., & Vasquez, M. (1999). The
Rice, K. G., Ashby, J. S., & Slaney, R. B. (1998). Self-esteem as a mediator diversification of psychology: A multicultural revolution. American
between perfectionism and depression: A structural equations analysis. Psychologist, 54, 1061–1069.
Journal of Counseling Psychology, 45, 304 –314. Terry-Short, L. A., Owens, R. G., Slade, P. D., & Dewey, M. E. (1995).
Rice, K. G., & Dellwo, J. P. (2001). Within-semester stability and adjust- Positive and negative perfectionism. Personality and Individual Differ-
ment correlates of the Multidimensional Perfectionism Scale. Measure- ences, 18, 663– 668.
ment and Evaluation in Counseling and Development, 34, 146 –156. Utsey, S. O., & Ponterotto, J. G. (1996). Development and validation of the
Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological Index of Race-Related Stress (IRRS). Journal of Counseling Psychol-
well-being revisited. Journal of Personality and Social Psychology, 69, ogy, 43, 490 –501.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

719 –727.
Shafran, R., & Mansell, W. (2001). Perfectionism and psychopathology: A dation of brief measures of positive and negative affect: The PANAS
review of research and treatment. Clinical Psychology Review, 21, scales. Journal of Personality and Social Psychology, 54, 1063–1070.
879 –906. Westefeld, J. S., & Furr, S. R. (1987). Suicide and depression among
Slade, P. D., & Owens, R. G. (1998). A dual process model of perfection- college students. Professional Psychology: Research and Practice, 18,
ism based on reinforcement theory. Behavior Modification, 22, 372–390. 119 –123.
Slaney, R. B., Rice, K. G., Mobley, M., & Trippi, J. (2001). The Revised
Almost Perfect Scale. Measurement and Evaluation in Counseling and Received January 2, 2003
Development, 34, 130 –145. Revision received June 2, 2003
Striegel-Moore, R. H., Schreiber, G. B., Lo, A., Crawford, P., Obarzanek, Accepted June 6, 2003 䡲