Content uploaded by Martin M Smith
Author content
All content in this area was uploaded by Martin M Smith on Jun 06, 2016
Content may be subject to copyright.
Are Perfectionism Dimensions Vulnerability Factors for Depressive Symptoms
After Controlling for Neuroticism? A Meta-analysis of 10 Longitudinal Studies
MARTIN M. SMITH
1
*, SIMON B. SHERRY
2
, KATERINA RNIC
1
, DONALD H. SAKLOFSKE
1
, MURRAY ENNS
3
and TARA GRALNICK
4
1
Department of Psychology, University of Western Ontario, London, Ontario, Canada
2
Department of Psychology, Dalhousie University, Halifax, Nova Scotia, Canada
3
Department of Psychiatry, University of Manitoba, Winnipeg, Manitoba, Canada
4
Department of Psychology, University of Toronto, Toronto, Ontario, Canada
Abstract: Extensive evidence suggests neuroticism is a higher-order personality trait that overlaps substantially with
perfectionism dimensions and depressive symptoms. Such evidence raises an important question: Which perfectionism
dimensions are vulnerability factors for depressive symptoms after controlling for neuroticism? To address this, a meta-
analysis of research testing whether socially prescribed perfectionism, concern over mistakes, doubts about actions,
personal standards, perfectionistic attitudes, self-criticism and self-oriented perfectionism predict change in depressive
symptoms, after controlling for baseline depression and neuroticism, was conducted. A literature search yielded 10
relevant studies (N = 1,758). Meta-analysis using random-effects models revealed that all seven perfectionism dimen-
sions had small positive relationships with follow-up depressive symptoms beyond baseline depression and neuroticism.
Perfectionism dimensions appear neither redundant with nor captured by neuroticism. Results lend credence and
coherence to theoretical accounts and empirical studies suggesting perfectionism dimensions are part of the premorbid
personality of people vulnerable to depressive symptoms. Copyright © 2016 European Association of Personality
Psychology
Key words: perfectionism; neuroticism; depression; meta-analysis; longitudinal
Neuroticism is a dispositional tendency to experience
negative emotional states. This higher-order personality
dimension encapsulates several lower-order characteristics
(e.g. anxiety, hostility, impulsivity and vulnerability) and is
robustly predictive of numerous mental-health problems
(Lahey, 2009), including depressive symptoms (e.g. sadness,
loneliness, anhedonia, apathy, hopelessness, helplessness
and suicidal ideation; Békés et al., 2015; Graham et al.,
2010). Given that neuroticism shares substantial variance
with depressive symptoms, researchers have legitimately
questioned whether lower-order personality traits such as
perfectionism predict depressive symptoms beyond higher-
order vulnerability factors such as neuroticism (Coyne &
Whiffen, 1995; Enns & Cox, 1997; Enns, Cox, & Clara,
2005). The present meta-analysis of 10 longitudinal studies
(N= 1758) represents the most comprehensive examination
to date of the relationship between perfectionism and depres-
sive symptoms after controlling for baseline neuroticism.
Perfectionism Dimensions, Neuroticism and Depressive
Symptoms
Extensive evidence suggests two higher-order factors under-
lie and account for the shared variance amongst core
perfectionism dimensions: perfectionistic concerns and per-
fectionistic strivings (see Stoeber & Otto, 2006, for review).
Perfectionistic concerns are composed of a family of traits,
including socially prescribed perfectionism (i.e. perceiving
others as demanding perfection of oneself; Hewitt & Flett,
1991), concern over mistakes (i.e. adverse reactions to fail-
ures; Frost, Marten, Lahart, & Rosenblate, 1990), doubts
about actions (i.e. doubts about performance abilities; Frost
et al., 1990) and self-criticism (i.e. the tendency to assume
blame and feel self-critical towards the self; Blatt, D’Afflitti,
& Quinlan, 1976). Perfectionistic strivings encompass a con-
stellation of traits, including self-oriented perfectionism
(i.e. demanding perfection of oneself; Hewitt & Flett, 1991)
and personal standards (i.e. setting unreasonably high personal
standards and goals; Frost et al., 1990). In the present study,
perfectionistic attitudes also receive attention. Beck and asso-
ciates’(e.g. Imber et al., 1990) treat perfectionism as a uni-
tary cognitive style that we label perfectionistic attitudes.
These attitudes include cognitive distortions with perfectionis-
tic themes (e.g. black-and-white, dichotomous thinking) and
social difficulties with perfectionistic themes (e.g. social-
evaluative concerns). Perfectionistic attitudes align more
closely with perfectionistic concerns (versus perfectionistic
strivings; Sherry, Hewitt, Flett, & Harvey, 2003).
Accumulated evidence suggests perfectionistic concerns
exacerbate the effect of stress on depressive symptoms across
clinical (Békés et al., 2015; Enns & Cox, 2005; Hewitt, Flett,
& Ediger, 1996) and non-clinical samples (Flett, Hewitt,
*Correspondence to: Martin M. Smith, Department of Psychology, Univer-
sity of Western Ontario, 1151 Richmond Street, London, Ontario, Canada
N6A3K7. E-mail: msmit454@uwo.ca
European Journal of Personality,Eur. J. Pers. 30: 201–212 (2016)
Published online 11 March 2016 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/per.2053
Received 27 October 2015
Revised 28 January 2016, Accepted 5 February 2016
Copyright © 2016 European Association of Personality Psychology
Blankstein, & Mosher, 1995; Sherry, Gautreau, Mushquash,
Sherry, & Allen, 2014). Likewise, prior research suggests
perfectionistic concerns confer vulnerability to depressive
symptoms through negative social situations (e.g. hostile
interactions), social cognitions (e.g. perceiving others as un-
caring), maladaptive coping (e.g. avoidance), negative life
events (e.g. romantic breakups) and daily hassles (Dunkley
& Blankstein, 2000; Dunkley, Blankstein, Halsall, Williams,
& Winkworth, 2000; Dunkley, Sanislow, Grilo, &
McGlashan, 2006; Hewitt & Flett, 1993; Sherry et al.,
2012). In contrast, perfectionistic strivings are inconsistent
predictors of depressive symptoms, with some research sug-
gesting they are vulnerability factors (Békés et al., 2015;
Hewitt et al., 1996; Joiner & Schmidt, 1995), and other
research suggesting they are resiliency factors (Enns et al.,
2005). On the one hand, perfectionistic strivings confer
vulnerability to depressive symptoms in the presence of
ego-involving achievement stressors (e.g. failing a test; Békés
et al., 2015; Hewitt et al., 1996). On the other hand, perfec-
tionistic strivings are occasionally associated with positive
outcomes (e.g. resourcefulness and task-oriented coping;
Dunkley, Zuroff, & Blankstein, 2003; Stoeber & Otto,
2006) and, after controlling for perfectionistic concerns, are
sometimes negatively associated with depressive symptoms
(Smith, Saklofske, Yan, & Sherry, 2015; see Stoeber & Otto,
2006, for review).
Aside from perfectionistic strivings’status as a vulnerabil-
ity factor, some investigators also question whether the appar-
ent link between perfectionism dimensions and depressive
symptoms stem from overlap with the ‘third variable’neurot-
icism (Enns et al., 2005). Indeed, a long-standing debate in
psychology centres on whether lower-order characteristics,
such as perfectionism dimensions, predict change in
outcomes (e.g. depressive symptoms) beyond higher-order
traits such as neuroticism (Coyne & Whiffen, 1995; Zuroff,
Mongrain, & Santor, 2004). Research on the incremental
explanatory power of perfectionism dimensions beyond neu-
roticism is particularly important given that depression and
several perfectionism dimensions have strong positive associ-
ations with neuroticism (Dunkley, Blankstein, & Berg, 2012;
Dunkley, Sanislow, Grilo, & McGlashan, 2009; Lahey,
2009), and because vulnerability is a fundamental component
of neuroticism (Costa & McCrae, 1992). Moreover, neuroti-
cism is a robust predictor of change in depressive symptoms
across both clinical and non-clinical samples (Lahey, 2009).
However, while many perfectionism dimensions are concep-
tually and empirically related to neuroticism, perfectionism
dimensions also have unique components that distinguish
them from neuroticism, such as a profound sense that one
is making irreconcilable mistakes, as well as feeling as
though others impose unfair demands on the self to be per-
fect (Flett & Hewitt, 2015). Additionally, Dunkley et al.
(2012) found that perfectionistic concerns are distinguish-
able from neuroticism in terms of lower agreeableness.
Nonetheless, there are notable between-study inconsis-
tencies concerning the status of perfectionism as a vulnera-
bility factor that predicts incremental change in depressive
symptoms beyond neuroticism (Békés et al., 2015; Dunkley
et al., 2009; Enns et al., 2005; Sherry, Mackinnon, Macneil,
& Fitzpatrick, 2013). Given that neuroticism overlaps with
many perfectionism constructs (Dunkley et al., 2012; Enns
et al., 2005; Graham et al., 2010), it is crucial that researchers
determine which, if any, perfectionism dimensions are
vulnerability factors for depressive symptoms after control-
ling for baseline neuroticism. The apparent link between
perfectionism and depressive symptoms may otherwise be
an artefact arising from shared variance with the ‘third-
variable’neuroticism. By controlling for this covariate, our
study represents a rigorous test of the perfectionism–
depressive symptoms relationship.
Advancing Research on Perfectionism and Depressive
Symptoms Using Meta-analysis
A quantitative synthesis may clarify between-study inconsis-
tencies concerning the status of perfectionism as a vulnera-
bility factor for depressive symptoms (Enns, Cox, &
Inayatulla, 2003; Sherry, Mackinnon, et al., 2013; Sherry,
Nealis, et al., 2013), allowing an overall conclusion to be
reached. Given that the majority of studies suggest perfec-
tionism has a small-to-moderate effect on depressive symp-
toms, it is likely that they are underpowered (Enns, Cox,
Sareen, & Freeman, 2001). Advantages of a meta-analysis
will help overcome limitations of small sample sizes (Card,
2012), bringing greater clarity to our understanding of the
longitudinal effects of perfectionism dimensions on depres-
sive symptoms. The consequences of perfectionistic strivings
on depressive symptoms are also contentiously debated, with
researchers either arguing they are vulnerability (Békés et al.,
2015) or resiliency (Enns et al., 2005) factors for change in
depressive symptoms. Meta-analysis will provide a more
encompassing and generalizable statement about the longitu-
dinal effects of perfectionistic strivings on depressive
symptoms, which is difficult to establish through any single
longitudinal study.
Objectives and Hypotheses
Are perfectionism dimensions part of a premorbid personal-
ity structure that reliably increases the risk of experiencing
depressive symptomology above and beyond the effects of
baseline neuroticism and baseline depression? Do only cer-
tain perfectionism dimensions confer vulnerability to depres-
sive symptoms? This study addressed these contentiously
debated questions by comprehensively meta-analysing extant
research.
Based on theory and empirical evidence, we hypothe-
sized that baseline socially prescribed perfectionism would
predict follow-up depressive symptoms after controlling for
baseline neuroticism and baseline depressive symptoms. A
similar hypothesis was made for the other perfectionistic
concern dimensions, including concern over mistakes,
doubts about actions, perfectionistic attitudes and self-
criticism. We also explored whether perfectionistic striving
dimensions (self-oriented perfectionism and personal stan-
dards) predict change in follow-up depressive symptoms be-
yond neuroticism. Additionally, we investigated the effects
of perfectionistic strivings on depressive symptoms after
202 M. M. Smith et al.
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
controlling for perfectionistic concerns, baseline neuroticism
and baseline depression.
METHOD
Selection of studies
A literature search on PsycINFO was conducted using the
keywords and Boolean search terms ‘perfection*’OR ‘self-
criticism’AND ‘longitudinal*’OR ‘prospective’. Disserta-
tions and non-English language articles were excluded. This
search yielded 241 studies. The first and third author
reviewed the abstract and method of all studies identified
from this broad search selecting studies that met inclusion
criteria. Journal articles were included if the following
criteria were met: (i) the study used a longitudinal design;
(ii) depressive symptoms were assessed on at least two
measurement occasions; (iii) perfectionism was assessed
alongside depression in one of the measurement occasions
preceding the final assessment of depression; and (iv)
neuroticism was assessed alongside depression and perfec-
tionism at one of the measurement occasions preceding the
final assessment of depression.
The literature search yielded a total of 12 articles for in-
clusion. Interrater agreement on inclusion or exclusion in
the meta-analysis was high (100%). Following the literature
search, the reference lists of the included articles were exam-
ined in an attempt to locate other relevant studies (Card,
2012). If a study did not report information needed to com-
pute effect sizes, the authors were contacted. All authors
contacted (N= 3) provided the requested information. On 5
October 2015, we terminated all search strategies and insti-
gated data reduction and analysis. We elected to exclude
Mushquash and Sherry (2013) as it used the same sample
and measure of depression (the Profile of Mood States De-
pression Subscale; McNair, Lorr, & Droppleman, 1992) as
Sherry et al. (2014). We also excluded Enns et al. (2003) as
it was a treatment study. Finally, one study (Mackinnon
et al., 2012) reported data on couples. In this case, females
and males in the dyad were treated as unique studies. Thus,
the final sample of selected studies was comprised of 10 arti-
cles with 11 samples (see Table 1 for sample characteristics).
Coding of studies
The first and third author coded each study based on 10 char-
acteristics: sample size at baseline, sample type, mean age of
participants at baseline, percent of female participants at
baseline, percent of Caucasian participants at baseline, time
lag between assessments, percent attrition, measure used to
assess perfectionism, measure used to assess neuroticism
and measure used to assess depressive symptoms.
Meta-analytic procedure
Random-effects analyses were performed using COMPREHEN-
SIVE META-ANALYSIS software (Version 3.3; Borenstein,
Hedges, Higgins, & Rothenstein, 2005). We chose random-
effects models, over fixed-effects models, as the 10 selected
studies varied widely in design (Table 1). Moreover,
random-effects models are generally preferable to fixed-
effects models, as they allow for generalizations beyond the
set of selected studies to future studies (Card, 2012).
Weighted mean effect sizes were computed following the
procedure recommended by Hunter and Schmidt (1990).
This allowed for estimation of mean effect sizes and the var-
iance in observed scores after considering sampling error
(Card, 2012). Effect size estimates were weighted by sample
size and aggregated. We chose to weight effects by sample
size as studies with larger sample sizes, relative to studies
with smaller sample sizes, have greater precision. To exam-
ine the relationship between perfectionism dimensions and
depressive symptoms, after controlling for baseline neuroti-
cism and baseline depression, standardized betas were com-
puted for each of the 11 samples using MPLUS 6 (Muthén
& Muthén, 1998–2012). In studies that included more than
one measure of depressive symptoms, effect sizes obtained
using various measures of depression were averaged such
that one effect size was included in the analysis (Card,
2012). This commonly used meta-analytic strategy guards
against overrepresentation of studies that include multiple ef-
fects. Prior to averaging, correlations were transformed into
Fisher’sZ(Card, 2012)
.
When studies included more than
two waves of data collection, the time points whereby the
necessary measures were administered (depressive symp-
toms, neuroticism, perfectionism at one time point and de-
pressive symptoms at a subsequent time point), and that
correspond to the longest time lag between measurement oc-
casions, were selected to compute effect sizes. Selection of
the longest possible time lag provided the most conservative
test of the perfectionism-depressive symptoms link. To facil-
itate interpretation, weighted mean effect size correlations, as
well as 95% confidence intervals, are reported in Table 2.
For each analysis, the total heterogeneity of weighted
mean effect sizes (Q
T
) was assessed (Table 3). If Q
T
is signif-
icant, it indicates the variance evident in the weighted
mean effect sizes is greater than would be expected by sam-
pling error (Card, 2012). A non-significant Q
T
suggests a
weak basis for moderation. The inconsistency in observed re-
lationships across studies (I
2
) was also computed for each
analysis. I
2
indicates the percentage of total variation across
studies due to heterogeneity: values of 25%, 50% and 75%
correspond to low, medium and high heterogeneity, respec-
tively (Higgins & Thompson, 2002). Unlike Q
T
,I
2
is not
adversely influenced by the number of included studies. To
ensure accuracy, the first and third author computed effect
sizes independently. No discrepancies in reported effect sizes
were found.
RESULTS
Description of studies
Our literature search identified 10 studies and 11 samples that
contained relevant effect size data (Table 1). The total num-
ber of participants pooled across studies was 1758. Studies
Perfectionism, neuroticism, and depressive symptoms 203
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
were published between 2001 and 2015, and the median year
of publication was 2012. Studies varied considerably. Sample
size varied between 47 and 240, with a median of 152. The av-
erage percent of female participants was 65.2%; the average
percent of Caucasian participants was 83.9%. The mean age
of the participants at baseline was 28.4 years (SD = 10.3;
range: 18.3–50.1). The time lag between assessments varied
between 2 and 192 weeks (M= 40.04, SD = 68.7). A total of
three samples contained undergraduates, one sample
contained community members, two samples used psychiatric
patients, two samples used medical students, one sample used
depressed outpatients, and two samples contained a mix of un-
dergraduates, graduate students and community members.
The average percent attrition was 14.5%. Perfectionism was
assessed using four measures (Table 1). Neuroticism was
assessed using four measures (Table 1). Depressive symptoms
were assessed using 11 measures (Table 2).
Overall effect sizes
The weighted mean effect sizes between perfectionism at
baseline and depressive symptoms at follow-up, while con-
trolling for neuroticism and depressive symptoms at baseline,
are reported in Table 2. Following Cohen’s (1992) guidelines
for small, medium, and large effect sizes (r= .10, .30, .50, re-
spectively), all longitudinal perfectionism–depression effects
were small in magnitude. For socially prescribed perfection-
ism, a positive effect (β= .13, p<.001) was observed
Table 1. Characteristics of longitudinal studies included in the meta-analysis
Sample Measurement
N
Sample
type
Mean
age
Time
lag
Attrition
(%)
Female
(%)
Caucasian
(%) Neuroticism
Perfectionistic
concerns
Perfectionistic
strivings
Békés et al. (2015) 47 Psychiatric
1
45.5 50.9 —70.2 75.0 NEOPIR-N DAS-P FMPS-PS
DEQ-SC HFMPS-SOP
FMPS-COM
HFMPS-SPP
Dunkley et al. (2006) 96 Psychiatric
1
34.3 158.6 —62.5 84.0 NEOPIR-N DAS-P —
Dunkley et al. (2009) 107 Psychiatric
1
34.4 192.0 —60.7 82.0 NEOPIR-N DAS-P —
Enns et al. (2001) 96 Medical
2
25.1 24.0 39.6 41.7 —NEOFFI-N FMPS-COM FMPS-PS
FMPS-DAA HFMPS-SOP
HFMPS-SPP
Enns et al. (2005) 206 Medical
2
24.0 20.0 32.5 44.2 —NEOFFI-N FMPS-COM FMPS-PS
FMPS-DAA HFMPS-SOP
HFMPS-SPP
Graham et al. (2010) 240 Undergrad
3
20.0 3.0 3.3 83.3 86.7 BFI-N FMPS-SF-COM FMPS-SF-PS
FMPS-DAA HFMPS-SF-SOP
HFMPS-SF-SPP
Mackinnon and
Sherry (2012)
127 Undergrad
3
18.3 19.0 9.4 77.9 81.1 BFI-N FMPS-SF-COM FMPS-SF-PS
FMPS-DAA HFMPS-SF-SOP
HFMPS-SF-SPP
Mackinnon et al. (2012) 226 Mixed
4
22.4 4.0 2.7 0.0 88.5 BFI-N DEQ-SF-SC —
FMPS-SF-COM
HFMPS-SF-SPP
Mackinnon et al. (2012) 226 Mixed
4
21.5 4.0 2.2 100.0 88.5 BFI-N DEQ-SF-SC —
FMPS-SF-COM
HFMPS-SF-SPP
Sherry, Nealis, et al. (2013) 155 Undergrad
3
20.7 4.3 1.9 76.8 70.3 BFI-N DEQ-SF-SC —
HFMPS-SF-SPP
FMPS-SF-COM
FMPS-DAA
Sherry et al. (2014) 232 Community 50.1 3.3 9.2 100.0 90.4 IPIP-N DEQ-SF-SC FMPS-SF-PS
FMPS-SF-COM HFMPS-SF-SOP
FMPS-DAA
HFMPS-SF-SPP
Note: Time lag in weeks. COM, concern over mistakes; DAA, doubts about actions; PS, personal standards; SC, self-criticism; SOP, self-oriented perfectionism;
SPP, socially prescribed perfectionism; D, depression; P, perfectionism; N, neuroticism; NA, negative affect; DAS, Weissman and Beck’s (1978) Dysfunc-
tional Attitude Scale; DEQ-SC, Blatt et al.’s (1976) Depressive Experiences Questioner Self-Criticism; DEQ-SF-SC, Depressive Experiences Questionnaire
Self-Criticism Short Form (Bagby, Parker, Joffe, & Buis, 1994); FMPS, Frost et al.’s (1990) Multidimensional Perfectionism Scale; FMPS-SF, Frost et al.’s
Multidimensional Perfectionism Scale Short Form (Cox, Enns, & Clara, 2002); HFMPS, Hewitt and Flett’s (1991) Multidimensional Perfectionism Scale;
HFMPS-SF, Hewitt and Flett’s Multidimensional Perfectionism Scale Short Form (Hewitt, Habke, Lee-Baggley, Sherry, & Flett, 2008); BFI, Benet-Martínez
and John’s (1998) Big Five Inventory; IPIP, Donnellan, Oswald, Baird, and Lucas’(2006) Mini International Personality Item Pool; NEOFFI, Costa and
McCrae’s (1992a) NEO Five-Factor Inventory; NEOPIR, Costa and McCrae’s (1992b) Revised NEO Personality Inventory.
1
Psychiatric patients
2
Medical students
3
Undergraduates
4
Undergraduates, graduate students and community members
204 M. M. Smith et al.
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
Table 2. Relationships between perfectionism dimensions, neuroticism and depressive symptoms
Concern over mistakes
Outcome
r
COM
1
,N
1
r
COM
1
,DEP
1
r
N
1
,DEP
1
r
COM
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Békés et al. (2015) BDI .43 .20 .32 .08 –.07 .50
HAM-D .43 .08 .24 .26 .05 .24
Overall .43 .14 .28 .17 –.01 .38
Enns et al. (2001) BDI-SF .52 .33 .57 –.07 .40 .22
Overall .52 .33 .57 –.07 .40 .22
Enns et al. (2005) BDI .54 .48 .60 .09 .03 .57
PANAS-NA .54 .42 .55 .21 .20 .28
Overall .54 .45 .58 .15 .12 .44
Graham et al. (2010) CES-D-SF .48 .43 .55 .06 .08 .65
DASS-D .48 .41 .48 .10 .08 .49
SCLR-D .48 .44 .52 .09 .06 .65
Overall .48 .43 .52 .08 .07 .60
Mackinnon and Sherry (2012) CES-D .42 .55 .63 .17 .00 .45
PANAS-NA .42 .30 .61 .25 .02 .41
POMS-D .42 .52 .53 .21 –.12 .54
Overall .42 .46 .59 .21 –.03 .47
Mackinnon et al. (2012) men CES-D .18 .28 .54 .04 .19 .57
Overall .18 .28 .54 .04 .19 .57
Mackinnon et al. (2012) women CES-D .16 .15 .54 .08 .11 .60
Overall .16 .15 .54 .08 .11 .60
Sherry, Nealis, et al. (2013) CES-D .30 .14 .50 .17 .12 .41
DASS-D .30 .42 .46 .12 .04 .38
SCL90R-D .30 .48 .52 .10 .19 .40
Overall .30 .35 .49 .13 .12 .40
Sherry et al. (2014) DACLE .37 .54 .48 .11 .20 .51
DACLG .37 .50 .47 .14 .20 .47
POMS-D .37 .52 .47 .11 .15 .57
Overall .37 .52 .47 .12 .18 .52
Doubts about actions
Outcome
r
DAA
1
,N
1
r
DAA
1
,DEP
1
r
N
1
,DEP
1
r
DAA
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Enns et al. (2001) BDI-SF .62 .39 .57 .16 .24 .21
Overall .62 .39 .57 .16 .24 .21
Enns et al. (2005) BDI .65 .51 .60 .04 .04 .58
PANAS-NA .65 .42 .55 .10 .24 .31
Overall .65 .47 .58 .07 .14 .45
Graham et al. (2010) CES-D-SF .50 .57 .55 .07 .08 .63
DASS-D .50 .47 .48 .13 .07 .48
SCLR-D .50 .54 .52 .11 .05 .63
Overall .50 .53 .52 .10 .07 .58
Mackinnon and Sherry (2012) CES-D .53 .57 .63 .29 –.06 .42
PANAS-NA .53 .45 .61 .31 –.01 .37
POMS-D .53 .49 .53 .28 –.18 .55
Overall .53 .50 .59 .29 –.08 .45
Sherry, Nealis, et al. (2013) CES-D .36 .35 .50 .16 .10 .43
DASS-D .36 .32 .46 .10 .03 .40
SCL90R-D .36 .48 .52 .11 .18 .40
Overall .36 .38 .49 .12 .10 .41
Sherry et al. (2014) DACLE .43 .56 .48 .08 .20 .52
DACLG .43 .47 .47 .16 .18 .47
POMS-D .43 .53 .47 .09 .15 .58
Overall .43 .52 .47 .11 .18 .52
Perfectionistic attitudes
Outcome
r
PA
1
,N
1
r
PA
1
,DEP
1
r
N
1
,DEP
1
r
PA
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Békés et al. (2015) BDI .49 .24 .32 -.08 .00 .51
HAM-D .49 .02 .24 .18 .07 .26
Overall .49 .13 .28 .05 .04 .39
Dunkley et al. (2006) BDI .63 .24 .41 .27 .26 .20
Overall .63 .24 .41 .27 .26 .20
(Continues)
Perfectionism, neuroticism, and depressive symptoms 205
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
Table 2. (Continued)
Perfectionistic attitudes
Outcome
r
PA
1
,N
1
r
PA
1
,DEP
1
r
N
1
,DEP
1
r
PA
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Dunkley et al. (2009) LIFEPSPSR-D .59 .18 .38 .24 –.06 .27
PAI-D .59 .18 .38 .31 .24 .15
Overall .59 .18 .38 .28 .09 .21
Personal standards
Outcome
r
PS
1
,N
1
r
PS
1
,DEP
1
r
N
1
,DEP
1
r
PS
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Békés et al. (2015) BDI .23 .20 .32 .00 –.04 .51
HAM-D .23 –.06 .24 .16 .13 .26
Overall .23 .07 .27 .08 .09 .39
Enns et al. (2001) BDI-SF .21 .11 .57 .00 .35 .22
Overall .21 .11 .57 .00 .35 .22
Enns et al. (2005) BDI .18 .24 .60 .03 .06 .59
PANAS-NA .18 .19 .55 .19 .27 .30
Overall .18 .22 .58 .11 .17 .45
Graham et al. (2010) CES-D-SF .15 .17 .55 .05 .10 .65
DASS-D .15 .12 .48 .06 .11 .51
SCL-R-D .15 .21 .52 .04 .08 .66
Overall .15 .17 .52 .05 .10 .61
Mackinnon and Sherry (2012) CES-D .25 .23 .63 .14 .00 .52
PANAS-NA .25 .21 .61 .20 .08 .41
POMS-D .25 .17 .53 .19 –.12 .62
Overall .25 .20 .59 .18 –.01 .52
Sherry et al. (2014) DACLG .18 .34 .48 .13 .21 .51
DACLE .18 .34 .47 .19 .22 .47
POMS-D .18 .36 .48 .12 .17 .58
Overall .18 .35 .48 .15 .20 .52
Self-criticism
Outcome
r
SC
1
,N
1
r
SC
1
,DEP
1
r
N
1
,DEP
1
r
SC
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Békés et al. (2015) BDI .44 .25 .32 –.07 –.01 .51
HAM-D .44 .02 .24 .18j .12 .25
Overall .44 .14 .28 .06 .06 .38
Mackinnon et al. (2012) women CES-D .32 .37 .54 .15 .09 .57
Overall .32 .37 .54 .15 .09 .57
Mackinnon et al. (2012) men CES-D .30 .43 .54 .06 .18 .55
Overall .30 .43 .54 .06 .18 .55
Sherry, Nealis, et al. (2013) CES-D .23 .18 .50 .17 .11 .45
DASS-D .23 .14 .46 .17 .02 .42
SCL90R-D .23 .17 .52 .19 .16 .43
Overall .23 .16 .49 .18 .10 .43
Sherry et al. (2014) DACLE .46 .43 .47 .20 .16 .44
DACLG .46 .48 .47 .20 .16 .49
POMS-D .46 .51 .47 .17 .12 .54
Overall .46 .47 .47 .19 .15 .49
Self-oriented perfectionism
Outcome
r
SOP
1
,N
1
r
SOP
1
,DEP
1
r
N
1
,DEP
1
r
SOP
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Békés et al. (2015) BDI .13 .29 .32 .12 –.04 .47
HAM-D .13 .10 .24 .26 .14 .21
Overall .13 .20 .28 .19 .05 .34
Enns et al. (2001) BDI-SF .39 .18 .57 –.03 .37 .22
Overall .39 .18 .57 –.03 .37 .22
Enns et al. (2005) BDI .18 .22 .60 .07 .06 .58
PANAS-NA .18 .18 .55 .19 .27 .30
Overall .18 .20 .58 .13 .17 .44
Graham et al. (2010) CES-D-SF .14 .11 .55 –.03 .10 .66
DASS-D .14 .14 .48 .02 .11 .52
SCL-R-D .14 .16 .52 .00 .09 .67
Overall .14 .14 .52 .00 .10 .62
(Continues)
206 M. M. Smith et al.
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
between socially prescribed perfectionism at baseline and de-
pressive symptoms at follow-up, while controlling for neu-
roticism and depression at baseline. In this regard, a
positive effect (β= .10, p<.001) was found for concern over
mistakes, a positive effect (β= .13, p<.001) was found for
doubts about actions, a positive effect (β= .12, p= .027)
was found for self-criticism, a positive effect (β= .08,
p= .018) was found for self-oriented perfectionism, a posi-
tive effect (β= .10, p= .003) was found for personal stan-
dards and a positive effect (β= .24, p<.001) was found for
perfectionistic attitudes. Results suggest all perfectionism di-
mensions confer vulnerability to depressive symptoms, even
after removal of variance attributable to baseline depressive
symptoms and baseline neuroticism.
Table 2. (Continued)
Self-oriented perfectionism
Outcome
r
SOP
1
,N
1
r
SOP
1
,DEP
1
r
N
1
,DEP
1
r
SOP
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Mackinnon and Sherry (2012) CES-D .17 .13 .63 .17 .00 .52
PANAS-NA .17 .09 .61 .19 .09 .43
POMS-D .17 .09 .53 .13 –.10 .63
Overall .17 .10 .59 .16 .00 .53
Sherry et al. (2014) DACLE .18 .24 .47 .15 .21 .50
DACLG .18 .24 .48 .11 .21 .54
POMS-D .18 .21 .47 .07 .16 .61
Overall .18 .23 .47 .11 .19 .55
Socially prescribed perfectionism
Outcome
r
SPP
1
,N
1
r
SPP
1
,DEP
1
r
N
1
,DEP
1
r
SPP
1
➔DEP
2
r
N
1
➔DEP
2
r
DEP
1
➔DEP
2
Békés et al. (2015) BDI .32 .35 .32 .08 –.06 .49
HAM-D .32 .23 .24 .32 .08 .17
Overall .32 .29 .28 .20 .01 .34
Enns et al. (2001) BDI-SF .47 .25 .57 .06 .31 .21
Overall .47 .25 .57 .06 .31 .21
Enns et al. (2005) BDI .46 .39 .60 .14 .01 .57
PANAS-NA .46 .36 .55 .19 .23 .29
Overall .46 .38 .58 .17 .12 .44
Graham et al. (2010) CES-D-SF .24 .24 .55 .13 .08 .64
DASS-D .24 .14 .48 .15 .08 .51
SCLR-D .24 .22 .52 .13 .06 .65
Overall .24 .20 .52 .14 .07 .60
Mackinnon and Sherry (2012) CES-D .33 .37 .63 .09 .01 .51
PANAS-NA .33 .34 .61 .10 .10 .41
POMS-D .33 .32 .53 .08 –.10 .62
Overall .33 .34 .59 .09 .00 .52
Mackinnon et al. (2012) men CES-D .18 .27 .54 .07 .19 .56
Overall .18 .27 .54 .07 .19 .56
Mackinnon et al. (2012) women CES-D .12 .18 .54 .04 .11 .60
Overall .12 .18 .54 .04 .11 .60
Sherry, Nealis, et al. (2013) CES-D .12 .24 .50 .19 .14 .42
DASS-D .12 .28 .46 .20 .06 .36
SCL90R-D .12 .31 .52 .16 .21 .39
Overall .12 .28 .49 .18 .14 .39
Sherry et al. (2014) DACLG .35 .44 .48 .20 .18 .48
DACLE .35 .40 .47 .28 .16 .44
POMS-D .35 .38 .47 .16 .13 .58
Overall .35 .41 .47 .21 .16 .50
Note: COM, concern over mistakes; DAA, doubts about actions; PA, perfectionistic attitudes; PS, personal standards; SC, self-criticism; SOP, self-oriented per-
fectionism; SPP, socially prescribed perfectionism; N, neuroticism; x
1
, baseline variable; x
2
, follow-up variable;
r
x
1
,
r
y
1
,bivariate correlation between baseline
variables; COM
1
➔DEP
2
, standardized beta for concern over mistakes predicting follow-up depressive symptoms (controlling for baseline depressive symptoms,
neuroticism); N
1
➔DEP
2
, standardized beta for neuroticism predicting follow-up depressive symptoms (controlling for baseline depressive symptoms, concern
over mistakes); DEP
1
➔DEP
2
, standardized beta for depressive symptoms predicting follow-up depressive symptoms (controlling for baseline neuroticism, con-
cern over mistakes). D, depression; NA, negative affect; BDI, Beck, Ward, and Mendelson’s (1961) Beck Depression Inventory; BDI-SF, Beck and Beck’s
(1972) short form of the Beck Depression Inventory of Beck et al. (1961); CES, Radloff’s (1977) Center for Epidemiological Studies Scale; CES-SF, Radloff’s
(1977) Center for Epidemiological Studies Scale Short Form; DACLG, Lubin’s (1965) Depression Adjective Checklist Form G; DACLE, Lubin’s (1965) De-
pression Adjective Checklist Form E. DASS, Lovibond and Lovibond’s (1995) Depression, Anxiety, and Stress Scales; HAM-D, Hamilton’s (1960) Hamilton
Depression Rating Scale; LIFEPSPCR, the Longitudinal Interval Follow-up Evaluation of Keller et al. (1987); PAI, Morey’s (1991) Personality Assessment In-
ventory; PANAS, Watson, Clark, and Tellegen’s (1988) Positive and Depressive Affect Scale; POMS, the Profile of Mood States of McNair et al. (1992);
SCL90R, Derogatis and Lazarus’(1994) Symptom Checklist-Revised.
Perfectionism, neuroticism, and depressive symptoms 207
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
Additionally, all weighted mean effect sizes correspond-
ing to perfectionism dimensions effects on follow-up depres-
sion had non-significant Q
T
values and I
2
estimates of 0.0%
(Table 3). This suggests that the assumption of homogeneity
should be retained and indicates common study effects
(Card, 2012). The non-significant Qvalues also indicate that
differences in relevant effect sizes were not greater than
would be expected on the basis of sample variation alone.
This may be an artefact of the small sample sizes of five of
the included studies (e.g. Békés et al., 2015). In addition,
the percentage of total variance due to true heterogeneity
(i.e. I
2
) was consistently small, suggesting that variability
amongst effect sizes was not due to additional sources and
suggests a weak basis for testing the influence of potential
moderating factors (Card, 2012).
After controlling for concern over mistakes, doubts about
actions and socially prescribed perfectionism, as well as base-
line depressive symptoms and baseline neuroticism, the effect
Table 3. Summary of effect sizes for the relationship between perfectionism dimensions, neuroticism and depressive symptoms
Variable kN r
+
95% CI Q
T
I
2
(%)
Neuroticism
r
N
1
,DEP
1
11 1758 .51*** [.47, .55] 13.43 25.55
Concern over mistakes
r
COM
1
,N
1
9 1555 .38*** [.28, .47] 39.61*** 79.80
r
COM
1
,DEP
1
9 1555 .36*** [.27, .45] 30.44*** 73.72
N
1
➔DEP
2
9 1402 .13*** [.08, .19] 9.07 11.83
DEP
1
➔DEP
2
9 1402 .50*** [.42, .56] 21.38** 62.58
COM
1
➔DEP
2
9 1402 .10*** [.05, .15] 4.82 0.00
Doubts about actions
r
DAA
1
,N
1
6 1056 .52*** [.43, .60] 19.39** 74.21
r
DAA
1
,DEP
1
6 1056 .48*** [.43, .53] 5.29 5.43
N
1
➔DEP
2
6 914 .10*** [.03, .18] 6.65 6.65
DEP
1
➔DEP
2
6 914 .46*** [.38, .54] 11.83* 57.74
DAA
1
➔DEP
2
6 914 .13*** [.07, .19] 4.02 0.00
Perfectionistic attitudes
r
PA
1
,N
1
3 250 .59*** [.50, .67] 1.26 0.00
r
PA
1
,DEP
1
3 250 .19*** [.07, .31] 0.43 0.00
N
1
➔DEP
2
3 250 .15*** [.02. .27] 2.16 7.28
DEP
1
➔DEP
2
3 250 .24*** [.12, .36] 1.50 0.00
PA
1
➔DEP
2
3 250 .24*** [.11, .35] 1.95 0.00
Personal standards
r
PS
1
,N
1
6 948 .19*** [.13, .25] 1.05 0.00
r
PS
1
,DEP
1
6 948 .21*** [.13, .29] 5.53 33.58
N
1
➔DEP
2
6 809 .14*** [.06, .22] 6.76 26.05
DEP
1
➔DEP
2
6 809 .48*** [.39, .57] 7.84* 61.84
PS
1
➔DEP
2
6 809 .10** [.04, .17] 2.45 0.00
Self-criticism
r
SC
1
,N
1
5 883 .39*** [.33, .44] 7.90 49.38
r
SC
1
,DEP
1
5 883 .34*** [.22, .46] 14.90** 73.16
N
1
➔DEP
2
5 861 .14*** [.06, .21] 2.12 0.00
DEP
1
➔DEP
2
5 861 .51*** [.46, .56] 5.16 22.47
SC
1
➔DEP
2
5 861 .12*[.07, .20] 2.59 0.00
Self-oriented perfectionism
r
SOP
1
,N
1
6 948 .19*** [.12, .25] 5.37 6.88
r
SOP
1
,DEP
1
6 948 .18*** [.11, .24] 1.95 0.00
N
1
➔DEP
2
6 809 .09* [.02, .16] 4.45 0.00
DEP
1
➔DEP
2
6 809 .49*** [.38, .58] 15.95** 68.65
SOP
1
➔DEP
2
6 809 .08*[.01, .15] 3.97 0.00
Socially prescribed perfectionism
r
SPP
1
,N
1
9 1555 .28*** [.19, .36] 25.73*** 68.91
r
SPP
1
,DEP
1
9 1555 .26*** [.20, 32] 13.40 40.28
N
1
➔DEP
2
9 1402 .11*** [.06, .17] 8.12 1.42
DEP
1
➔DEP
2
9 1402 .49*** [.42, .56] 22.37** 64.24
SPP
1
➔DEP
2
9 1402 .13*** [.07, .18] 5.36 0.00
Note.k, number of studies; N, total number of participants in the ksamples; r
+
, weighted mean r; CI, confident interval; Q
T
, measure of heterogeneity of effect
sizes; I
2
, percentage of heterogeneity; COM, concern over mistakes; DAA, doubts about actions; PA, perfectionistic attitudes; PS, personal standards; SC, self-
criticism; SOP, self-oriented perfectionism; SPP, socially prescribed perfectionism; N, neuroticism; DEP, depressive symptoms; x
1
, baseline variable; x
2
, follow-
up variable;
r
x
1
,
r
y
1
,bivariate correlation between baseline variables; COM
1
➔DEP
2
, standardized beta for concern over mistakes predicting follow-up depressive
symptoms (controlling for baseline depressive symptoms, neuroticism); N
1
➔DEP
2
, standardized beta for neuroticism predicting follow-up depressive symptoms
(controlling for baseline depressive symptoms, concern over mistakes);
r
DEP
1
➔DEP
2
, standardized beta for depressive symptoms predicting follow-up depres-
sive symptoms (controlling for baseline neuroticism, concern over mistakes).
*p<.05; **p<.01; ***p<.001.
208 M. M. Smith et al.
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
of personal standards on follow-up depressive symptoms was
non-significant (β=.02, p= .504). Likewise, a similar pattern
was observed for self-oriented perfectionism (β=.00,
p= .930). Detailed statistics regarding the effects of personal
standards and self-oriented perfectionism on follow-up depres-
sive symptoms after controlling for baseline depressive symp-
toms, baseline neuroticism, baseline concern over mistakes,
baseline doubts about actions and baseline socially prescribed
perfectionism are presented in the Supporting Information.
Additionally, while outside the scope of the present paper,
the effects of concern over mistakes, personal standards, self-
oriented perfectionism and socially prescribed perfectionism
on follow up depressive symptoms, after controlling for con-
scientiousness, are available in the Supporting Information.
DISCUSSION
Empirical studies and theoretical accounts suggest that perfec-
tionism is a vulnerability factor for depressive symptoms
(Békés et al., 2015; Graham et al., 2010; Hewitt et al., 1996;
Joiner & Schmidt, 1995). However, it is unclear, the extent
to which this relationship persists after controlling for the com-
pelling covariate of neuroticism. Prior studies have shown that
when measures of depressive symptoms are highly saturated
with items assessing negative emotionality, depressive symp-
toms’relation with vulnerability factors (e.g. perfectionism)
will be largely explained by shared variance with neuroticism
(Dunkley, Blankstein, & Flett, 1997; Zuroff et al., 2004). Ac-
cordingly, this renders the present meta-analytic review of
the extant empirical literature examining if perfectionism di-
mensions continue to predict change in depressive symptoms,
after controlling for baseline neuroticism, a particularly strin-
gent test of the perfectionism–depressive symptoms link.
In our meta-analysis of 10 longitudinal studies composed
of undergraduate, community member, psychiatric patient,
outpatient and medical student samples, neuroticism was
the strongest predictor of change in depressive symptoms.
Even so, all seven perfectionism dimensions still predicted
change in depressive symptoms beyond neuroticism. Find-
ings lend credence and coherence to research and theories
suggesting that perfectionism dimensions are part of the
premorbid personality of people vulnerable to depressive
symptoms (e.g. Békés et al., 2015; Dunkley et al., 2003; Flett
et al., 1995; Hewitt & Flett, 1993; Hewitt et al., 1996).
Perfectionistic concerns
Consistent with hypotheses, socially prescribed perfection-
ism, concern over mistakes, doubts about actions, self-
criticism, and perfectionistic attitudes add incrementally to
understanding change in depressive symptoms beyond neu-
roticism. Effects were small in magnitude across a wide
range of samples, methods, and measures. Results suggest
perfectionistic concerns’constructs are lower-order per-
sonality traits neither redundant with nor captured by neurot-
icism. As prior research suggests, people high in
perfectionistic concerns appear to think, feel and behave in
ways that have depressogenic consequences (Graham et al.,
2010). Such people believe others hold lofty expectations
for them, and often feel incapable of living up to the perfec-
tion they perceive others demand. They may agonize about
perceived failures and have doubts about performance
abilities because they experience their social world as judg-
mental, pressure-filled and unyielding. Perfectionistic con-
cerns also appear to be composed of stable, underlying
traits that trigger depressive symptoms by predisposing peo-
ple to the frequent subjective experience of disappointing
others (Sherry et al., 2014). Additionally, consistent with
the diathesis-stress model, perfectionistic concerns predict
heightened depressive symptoms by predisposing people to
perceive interpersonal stressors as more ego-involving and
distressing (Békés et al., 2015; Hewitt & Flett, 1993, 2002).
Perfectionistic strivings
Do personal standards and self-oriented perfectionism protect
against depressive symptoms? Our meta-analysis offers a
resounding ‘no’to this question. Findings from our meta-
analysis are incongruent with a view of perfectionistic striv-
ings’constructs as resiliency factors that protect against
increases in depressive symptoms (Enns et al., 2005). An
over-reliance on cross-sectional studies may have clouded
the nature of the perfectionism–depressive symptoms rela-
tionship, resulting in inconsistencies in the literature
concerning the consequences of this trait. In particular,
according to the diathesis-stress model of perfectionism, per-
fectionistic strivings only promotes depressive symptoms in
the presence of ego-threatening stressors, such as achieve-
ment failures (e.g. poor performance on an exam; Békés
et al., 2015; Enns & Cox, 2005). This might render the
deleterious effects of perfectionistic strivings on depressive
symptoms elusive when assessed at only a single time point.
Additionally, our findings dovetail with past theoretical ac-
counts, case histories and empirical studies. In fact, clinicians
have long described perfectionistic strivings as a ‘Trojan
horse’, whereby self-concealment and perfectionistic self-
presentation mask perfectionistic strivings’depressogenic
effects (Blatt, 1995). Our results complement studies showing
that perfectionistic strivings’rob people of satisfaction and
positive affect (Hewitt & Flett, 1991) and amplify the risk of
suicide (Blatt, 1995; Flett, Hewitt, & Heisel, 2014) and early
mortality (Fry & Debats, 2009). Individuals with high perfec-
tionistic strivings are only satisfied when everything in their
lives suggests they are perfect; when life events inevitably sug-
gest they are not perfect, depressive symptoms follow.
Despite this, our findings also complement research
showing perfectionistic strivings confer vulnerability for de-
pressive symptoms through overlap with perfectionistic con-
cerns (Stoeber & Otto, 2006). After controlling for baseline
depression, baseline neuroticism and baseline perfectionistic
concerns, personal standards and self-oriented perfectionism
ceased to be significant predictors of follow-up depressive
symptoms. Nevertheless, we caution against over-interpretation
of this finding in light of increasing apprehension that con-
trolling for perfectionistic concerns when examining the
effects of perfectionistic strivings may change the
Perfectionism, neuroticism, and depressive symptoms 209
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
conceptual meaning of perfectionistic strivings and may well
undermine its relevance to perfectionism research (e.g., Hill,
2014; Molnar, Sadava, Flett, & Colautti, 2012; Powers,
Koestner, Zuroff, Milyavskaya, & Gorin, 2011).
Limitations of overall literature
Summarizing limitations within the extant research eluci-
dates further areas requiring examination, thereby provid-
ing direction to advance the field of study. While
conducting our literature search, it became apparent that
the majority of studies on the perfectionism–depressive
symptoms link are cross sectional in nature and do not take
neuroticism into account. This is problematic, as cross-
sectional studies fail to address temporal precedence and
thus are incapable of evaluating the extent to which perfec-
tionism dimensions predict change in depressive symptoms.
Moreover, studies that neglect to control for neuroticism
run the risk of drawing erroneous conclusions because of
the substantial overlap between perfectionism dimensions
and the ‘third-variable’neuroticism (Dunkley et al., 2012;
Enns et al., 2005). Given the importance of assessing con-
structs longitudinally, and extensive evidence suggesting
perfectionism, neuroticism and depressive symptoms are
highly correlated (Dunkley et al., 2012; Enns et al., 2005;
Graham et al., 2010), researchers in the area are advised
to move forward by using longitudinal designs that control
for neuroticism.
Moreover, the vast majority of research on the
perfectionism–depressive symptom link relies on mono-
source designs (cf. Flett, Besser, & Hewitt, 2005; Sherry,
Mackinnon, et al., 2013). Mono-source designs are problem-
atic when studying personality traits such as perfectionism
that can involve self-presentational biases (e.g. defensively
concealing imperfections from others; Klonsky & Oltmanns,
2002). Future studies can advance the literature by using al-
ternative methods of data collection (e.g. informant reports;
Sherry, Nealis et al., 2013). Finally, as five of the 10 studies
included in our meta-analysis had sample sizes below 150,
the present research suggests many longitudinal perfection-
ism studies are underpowered. Researchers are advised to
move forward by using sample sizes large enough to detect
small to moderate effects.
Limitations of the present study and future directions
Certain limitations in extant literature translate into limita-
tions in the present meta-analysis. In this regard, studies
from only three research teams met our inclusion criteria,
limiting investigator variability. Also, while the effects of
five perfectionistic concern dimensions were tested in the
current meta-analysis, only two perfectionistic striving di-
mensions were included (self-oriented perfectionism and
personal standards). Accordingly, it is likely that perfection-
istic concerns captured a more comprehensive construct,
thereby limiting our ability to accurately compare the contri-
butions of perfectionistic concerns and perfectionistic striv-
ings. Furthermore, seven of the 10 studies included used a
short-form, opposed to a long-form, measure of neuroticism.
A richer, more fine-grained analysis of the longitudinal ef-
fects of perfectionism on depressive symptoms beyond neu-
roticism’s six lower-order facets is needed. Also, findings
derived from the current meta-analysis may have limited
generalizability beyond the specific set of samples included.
Additionally, future research should explore the extent to
which perfectionism dimensions are vulnerability factors
for other forms of emotional distress such as anger and anx-
iety. Finally, the predictive utility of perfectionism in the
present meta-analysis was likely understated due to not ac-
counting for life stressors, which consistent with a
diathesis-stress model, may need to be present for perfec-
tionism’s role as a vulnerability factor to become evident
(Hewitt & Flett, 1993, 2002).
Concluding remarks
The present meta-analysis of 10 longitudinal studies (involving
11 samples and 1758 participants) represents the most
comprehensive test to date of the perfectionism–depressive
symptoms relationship. Results add substantively to the per-
fectionism and depression literature by synthesizing existing
research to demonstrate that all perfectionism dimensions
predict change in depressive symptoms beyond neuroticism.
Findings support past evidence suggesting perfectionistic
concerns and perfectionistic strivings comprise lower-order
personality traits that place individuals at risk for experienc-
ing depressive symptoms. In sum, our meta-analysis sheds
light on the experiences of people with high levels of perfec-
tionism, highlighting the importance of developing ways of
intervening when people feel that they must meet the perfec-
tionistic expectations of themselves and others.
SUPPORTING INFORMATION
Additional supporting information may be found in the online
version of this article at the publisher’s web site.
REFERENCES
References marked with an asterisk indicate studies included
in the meta-analysis.
Bagby, R. M., Parker, J. D., Joffe, R. T., & Buis, T. (1994). Recon-
struction and validation of the Depressive Experiences Question-
naire. Assessment,1,59–68.
Beck, A. T., & Beck, R. W. (1972). Screening depressed patients in
family practice: A rapid technique. Postgraduate Medical Jour-
nal,52,81–85.
Beck, A. T., Ward, C., & Mendelson, M. (1961). Beck Depression
Inventory (BDI). Archives of General Psychiatry,4, 561–571.
*Békés, V., Dunkley, D., Taylor, G., Zuroff, D., Lewkowski, M.,
Foley, J., …Westreich, R. (2015). Chronic stress and attenuated
improvement in depression over 1 year: The moderating role of
perfectionism. Behavior Therapy,46, 479–492.
Benet-Martínez, V., & John, O. P. (1998). Los Cinco Grandes
across cultures and ethnic groups: Multitrait-multimethod analy-
ses of the Big Five in Spanish and English. Journal of Personal-
ity and Social Psychology,75, 729–750.
210 M. M. Smith et al.
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
Blatt, S. J. (1995). The destructiveness of perfectionism: Implica-
tions for the treatment of depression. American Psychologist,
50, 1003–1020.
Blatt, S. J., D’Afflitti, J. P., & Quinlan, D. M. (1976). Experiences
of depression in normal young adults. Journal of Abnormal Psy-
chology,85, 383–389.
Borenstein, M., Hedges, L., Higgins, J., & Rothenstein, H. (2005).
Comprehensive meta-analysis (Version 3.3). Englewood, NJ:
Biostat.
Card, N. A. (2012). Applied meta-analysis for social science re-
search. New York: Guilford Press.
Cohen, J. (1992). A power primer. Psychological Bulletin,112,
155–159.
Costa, P. T., & McCrae, R. R. (1992). Normal personality assess-
ment in clinical practice: The NEO Personality Inventory. Psy-
chological Assessment,4,5–13.
Cox, B. J., Enns, M. W., & Clara, I. P. (2002). The multidimen-
sional structure of perfectionism in clinically distressed and col-
lege student samples. Psychological Assessment,14, 365–373.
Coyne, J. C., & Whiffen, V. E. (1995). Issues in personality diath-
esis for depression: The case of sociotropy-dependency and
autonomy-self-criticism. Psychological Bulletin,118, 358–378.
Derogatis, L. R., & Lazarus, L. (1994). SCL-90-R, Brief Symptom
Inventory, and matching clinical rating scales. In M. Maruish
(Ed.), The use of psychological testing for treatment planning
and outcome assessment (pp. 217–248). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Donnellan, M., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006).
The Mini-IPIP Scales: Tiny-yet-effective measures of the Big Five
factors of personality. Psychological Assessment,18, 192–203.
Dunkley, D. M., & Blankstein, K. R. (2000). Self-critical perfec-
tionism, coping, hassles, and current distress: A structural equa-
tion modeling approach. Cognitive Therapy and Research,24,
713–730.
Dunkley, D. M., Blankstein, K. R., & Berg, J. L. (2012). Perfection-
ism dimensions and the five-factor model of personality. Euro-
pean Journal of Personality,26, 233–244.
Dunkley, D. M., Blankstein, K. R., & Flett, G. L. (1997). Specific
cognitive-personality vulnerability styles in depression and the
five-factor model of personality. Personality and Individual Dif-
ferences,23, 1041–1053.
Dunkley, D. M., Blankstein, K. R., Halsall, J., Williams, M., &
Winkworth,G. (2000). The relation between perfectionism and dis-
tress: Hassles, coping, and perceived social support as mediators
and moderators. Journal of Counseling Psychology,47,437–453.
*Dunkley, D. M., Sanislow, C. A., Grilo, C. M., & McGlashan, T. H.
(2006). Perfectionism and depressive symptoms 3 years later:
Negative social interactions, avoidant coping, and perceived social
support as mediators. Comprehensive Psychiatry,47, 106–115.
*Dunkley, D. M., Sanislow, C. A., Grilo, C. M., & McGlashan, T.
H. (2009). Self-criticism versus neuroticism in predicting depres-
sion and psychosocial impairment for 4 years in a clinical sample.
Comprehensive Psychiatry,50, 335–346.
Dunkley, D. M., Zuroff, D. C., & Blankstein, K. R. (2003). Self-
critical perfectionism and daily affect: Dispositional and situa-
tional influences on stress and coping. Journal of Personality
and Social Psychology,84, 234–252.
Enns, M. W., & Cox, B. J. (1997). Personality dimensions and de-
pression: Review and commentary. Canadian Journal of Psychi-
atry,42, 274–284.
Enns, M. W., & Cox, B. J. (2005). Perfectionism, stressful life
events, and the 1-year outcome of depression. Cognitive Therapy
and Research,29, 541–553.
*Enns, M. W., Cox, B. J., & Clara, I. P. (2005). Perfectionism and neu-
roticism: A longitudinal study of specific vulnerability and diathesis-
stress models. Cognitive Therapy and Research,29, 463–478.
Enns, M. W., Cox, B. J., & Inayatulla, M. (2003). Personality pre-
dictors of outcome for adolescents hospitalized for suicidal idea-
tion. Journal of the American Academy of Child and Adolescent
Psychiatry,42, 720–727.
*Enns, M. W., Cox, B. J., Sareen, J., & Freeman, P. (2001). Adap-
tive and maladaptive perfectionism in medical students: A longi-
tudinal investigation. Medical Education,35, 1034–1042.
Flett, G. L., Besser, A., & Hewitt, P. L. (2005). Perfectionism,
ego defense styles, and depression: A comparison of self-
reports versus informant ratings. Journal of Personality,73,
1355–1396.
Flett, G. L., & Hewitt, P. L. (2015). Measures of perfectionism. In
G. J. Boyle, D. H. Saklofske, & G. Matthews (Eds.), Measures
of personality and social constructs (pp. 595–618). London: Ac-
ademic Press.
Flett, G. L., Hewitt, P. L., Blankstein, K. R., & Mosher, S. W. (1995).
Perfectionism, life events, and depressive symptoms: A test of a
diathesis-stress model. Current Psychology,14, 112–137.
Flett, G. L., Hewitt, P. L., & Heisel, M. (2014). The destructiveness
of perfectionism revisited: Implications for the assessment of sui-
cide risk and the prevention of suicide. Review of General Psy-
chology,18, 156–172.
Frost, R., Marten, P., Lahart, C., & Rosenblate, R. (1990). The di-
mensions of perfectionism. Cognitive Therapy and Research,
14, 449–468.
Fry, P. S., & Debats, D. L. (2009). Perfectionism and the five-factor
personality traits as predictors of mortality in older adults. Jour-
nal of Health Psychology,14, 513–524.
*Graham, A. R., Sherry, S. B., Stewart, S. H., Sherry, D. L.,
McGrath, D. S., Fossum, K. M., & Allen, S. L. (2010). The exis-
tential model of perfectionism and depressive symptoms: A
short-term, four-wave longitudinal study. Journal of Counseling
Psychology,57, 423–438.
Hamilton, M. (1960). A rating scale for depression. Journal of Neu-
rology, Neurosurgery, and Psychiatry,23,56–62.
Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and so-
cial contexts. Journal of Personality and Social Psychology,60,
456–470.
Hewitt, P. L., & Flett, G. L. (1993). Dimensions of perfectionism,
daily stress, and depression: A test of the specific vulnerability
hypothesis. Journal of Abnormal Psychology,102,58–65.
Hewitt, P. L., & Flett, G. L. (2002). Perfectionism and stress pro-
cesses in psychopathology. In G. L. Flett, & P. L. Hewitt
(Eds.), Perfectionism: Theory, research, and treatment
(pp. 255–284). Washington: American Psychological Association.
Hewitt, P. L., Flett, G. L., & Ediger, E. (1996). Perfectionism and
depression: Longitudinal assessment of a specific vulnerability
hypothesis. Journal of Abnormal Psychology,105, 276–280.
Hewitt, P. L., Habke, A., Lee-Baggley, D., Sherry, S. B., & Flett, G.
L. (2008). The impact of perfectionistic self-presentation on the
cognitive, affective, and physiological experience of a clinical in-
terview. Psychiatry,71,93–122.
Higgins, J., & Thompson, S. G. (2002). Quantifying heterogeneity
in a meta-analysis. Statistics in Medicine,21, 1539–1558.
Hill, A. P. (2014). Perfectionistic strivings and the perils of
partialling. International Journal of Sport and Exercise Psychol-
ogy,12, 302–315.
Hunter, J. E., & Schmidt, F. L. (1990). Dichotomization of contin-
uous variables: The implications for meta-analysis. Journal of
Applied Psychology,75, 334–349.
Imber, S. D., Pilkonis, P. A., Sotsky, S. M., Elkin, I., Watkins, J. T.,
Collins, J. F., …Glass, D. R. (1990). Mode-specific effects
among three treatments for depression. Journal of Consulting
and Clinical Psychology,58, 352–359.
Joiner, T. E., & Schmidt, N. B. (1995). Dimensions of perfection-
ism, life stress, and depressed and anxious symptoms: Prospec-
tive support for diathesis-stress but not specific vulnerability
among male undergraduates. Journal of Social and Clinical Psy-
chology,14, 165–183.
Keller, M. B., Lavori, P. W., Friedman, B., Nielsen, E., Endicott, J.,
McDonald-Scott, P., & Andreasen, N. C. (1987). The longitudi-
nal interval follow-up evaluation: A comprehensive method for
assessing outcome in prospective longitudinal studies. Archives
of General Psychiatry,44, 540–548.
Perfectionism, neuroticism, and depressive symptoms 211
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per
Klonsky, E. D., & Oltmanns, T. F. (2002). Informant-reports of per-
sonality disorder: Relation to self-reports and future research di-
rections. Clinical Psychology: Science and Practice,9, 300–311.
Lahey, B. B. (2009). Public health significance of neuroticism.
American Psychologist,64, 241–256.
Lovibond, P., & Lovibond, S. (1995). The structure of negative
emotional states. Behaviour Research and Therapy,33, 335–343.
Lubin, B. (1965). Adjective checklists for measurement of depres-
sion. Archives of General Psychiatry,12,57–62.
*Mackinnon, S. P., & Sherry, S. B. (2012). Perfectionistic self-
presentation mediates the relationship between perfectionistic
concerns and subjective well-being: A three-wave longitudinal
study. Personality and Individual Differences,53,22–28.
*Mackinnon, S. P., Sherry, S. B., Antony, M. M., Stewart, S. H.,
Sherry, D. L., & Hartling, N. (2012). Caught in a bad romance:
Perfectionism, conflict, and depression in romantic relationships.
Journal of Family Psychology,26, 215–225.
McNair, D. M., Lorr, M., & Droppleman, L. F. (1992). Profile of
mood states manual. San Diego, CA: Educational and Industrial
Testing Service.
Molnar, D. S., Sadava, S. W., Flett, G. L., & Colautti, J. (2012). Per-
fectionism and health: A mediational analysis of the roles of
stress, social support and health-related behaviours. Psychology
& Health,27, 846–864.
Morey, L. C. (1991). Personality Assessment Inventory—Profes-
sional Manual. Lutz, FL: Psychological Assessment Resources.
Mushquash, A. R., & Sherry, S. B. (2013). Testing the perfection-
ism model of binge eating in mother–daughter dyads: A mixed
longitudinal and daily diary study. Journal of Marriage and
Family,14, 171–179.
Muthén, L. K., & Muthén, B. O. (1998–2012). Mplus user’s guide.
Los Angeles, CA: Muthén & Muthén.
Powers, T. A., Koestner, R., Zuroff, D. C., Milyavskaya, M., &
Gorin, A. (2011). The effects of self-criticism and self-oriented
perfectionism on goal pursuit. Personality and Social Psychology
Bulletin,37, 964–975.
Radloff, L. S. (1977). The CES-D Scale: A self-report depression
scale for research in the general population. Applied Psychologi-
cal Measurement,1, 385–401.
*Sherry, S. B., Gautreau, C. M., Mushquash, A. R., Sherry, D. L., &
Allen, S. L. (2014). Self-critical perfectionism confers
vulnerability to depression after controlling for neuroticism: A
longitudinal study of middle-aged community-dwelling women.
Personality and Individual Differences,69,1–4.
Sherry, S. B., Hewitt, P. L., Flett, G. L., & Harvey, M. (2003). Per-
fectionism dimensions, perfectionistic attitudes, dependent atti-
tudes, and depression in psychiatric patients and university
students. Journal of Counseling Psychology,50, 373–386.
Sherry, S. B., Hewitt, P. L., Stewart, S. H., Mackinnon, A. L.,
Mushquash, A. R., Flett, G. L., & Sherry, D. L. (2012). Social
disconnection and hazardous drinking mediate the link
between perfectionistic attitudes and depressive symptoms.
Journal of Psychopathology and Behavioural Assessment,26,
193–201.
Sherry, S. B., Mackinnon, S. P., Macneil, M. A., & Fitzpatrick, S.
(2013). Discrepancies confer vulnerability to depressive symp-
toms: A three-wave longitudinal study. Journal of Counseling
Psychology,60, 112–126.
*Sherry, S. B., Nealis, L. J., Macneil, M. A., Stewart, S. H., Sherry,
D. L., & Smith, M. M. (2013). Informant reports add incremen-
tally to the understanding of the perfectionism-depression con-
nection: Evidence from a prospective longitudinal study.
Personality and Individual Differences,54, 957–960.
Smith, M. M., Saklofske, D. H., Yan, G., & Sherry, S. B. (2015).
Perfectionistic strivings and perfectionistic concerns interact to
predict negative emotionality: Support for the tripartite model
of perfectionism in Canadian and Chinese university students.
Personality and Individual Differences,81, 141–147.
Stoeber, J., & Otto, K. (2006). Positive conceptions of perfection-
ism: Approaches, evidence, challenges. Personality and Social
Psychology Review,10, 295–319.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and
validation of brief measures of positive and negative affect. Jour-
nal of Personality and Social Psychology,54, 1063–1070.
Weissman, A. N., & Beck, A. T. (1978). Development and valida-
tion of the Dysfunctional Attitude Scale: A preliminary investiga-
tion. Paper presented at the 86th Annual Convention of the
American Psychological Association, Toronto, Ontario, Canada.
Zuroff, D. C., Mongrain, M., & Santor, D. A. (2004). Conceptualiz-
ing and measuring personality vulnerability to depression: Com-
ment on Coyne and Whiffen (1995). Psychological Bulletin,130,
489–511.
212 M. M. Smith et al.
Copyright © 2016 European Association of Personality Psychology Eur. J. Pers. 30: 201–212 (2016)
DOI: 10.1002/per