Date of Award


Document Type


Degree Name


Organizational Unit

Morgridge College of Education, Research Methods and Information Science, Research Methods and Statistics

First Advisor

Antonio Olmos, Ph.D.

Second Advisor

Kathy Green

Third Advisor

Frederique Chevillot

Fourth Advisor

Trisha Raque-Bogdan


Depression, Meta-analysis, Perfectionism


Perfectionism has been shown to be related to depression, but perfectionism is multidimensional. Some dimensions are related to positive psychological characteristics and outcomes and other dimensions are related to negative psychological characteristics and outcomes. This study reports results of nine meta-analyses performed to investigate the association between each of nine subscales of perfectionism and depression to determine which dimensions of perfectionism are most strongly associated with depression. The two subscales that were used from the Hewitt and Flett (1991b) Multidimensional Perfectionism scale were Self-Oriented Perfectionism (SOP) and Socially-Prescribed Perfectionism (SPP). The five subscales that were used from the Frost et al. (1991) Multidimensional Perfectionism Scale were Personal Standards (PS), Doubts about Actions (DA), Concern over Mistakes (CM), Parental Expectations (PE), and Parental Criticism (PC). The two subscales that were used from the Slaney et al. (2001) Almost Perfect Scale-Revised were High Personal Standards (HS) and Discrepancy (Dis). The SPP, DA, CM, PE, PC, and DIS subscales are negative dimensions of perfectionism that form the higher-order factor Perfectionistic Concerns (PC). The SOP, PS, and HS subscales are more positive dimensions of perfectionism that form the higher-order factor Perfectionistic Strivings (PS). Knowing the strength of association between depression and various dimensions of perfectionism is important because only negative perfectionism is supposed to be strongly related to depression. Two commercial databases were searched for published studies, and conference proceedings from professional research organizations, gray literature websites, and ProQuest Dissertations and Theses were searched for non-published studies. The total sample consisted of 52 studies, and the search for studies was thorough but not exhaustive. Random-effects models were used for the meta-analyses. Correlations between perfectionism subscales and depression measures that were collected from the studies in the sample were corrected for attenuation due to measurement error.

As anticipated, the six negative dimensions/subscales of Perfectionistic Concerns were shown to be more strongly and directly correlated with depression than the three positive dimensions of Perfectionistic Strivings. Evidence of publication bias was examined using forest plots, funnel plots, statistical tests for asymmetry of funnel plots, and cumulative meta-analyses. Five out of the nine meta-analyses showed evidence of publication bias through the cumulative meta-analyses or the trim and fill procedure. However, none of the meta-analyses showed significant funnel plot asymmetry. In aggregate, results suggest some evidence of publication bias.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Gabriel Lynn Hottinger


Received from ProQuest

File Format




File Size

248 p.


Quantitative psychology, Statistics