Abstract
Several facets of perfectionism have been strongly associated with depression and anxiety. CitationDunkley and Blankstein (2000) combined these maladaptive traits with self-criticism to create a general construct labeled self-critical perfectionism. In this study, we employed confirmatory factor analysis to evaluate a model for assessing self-critical perfectionism in a clinically depressed sample using scales from 3 instruments. Participants were 356 depressed adult outpatients who completed 2 multidimensional measures of perfectionism and a measure of self-criticism. A confirmatory factor model that separated a self-critical perfectionism construct from a more adaptive, achievement-striving component of perfectionism was supported. A composite scale assessing self-critical perfectionism demonstrated much larger correlations with distress measures compared to a composite scale assessing achievement striving and also showed evidence of discriminant validity. In this study, we provided further support for the valid assessment of self-critical perfectionism and extended evidence for its assessment to a clinically depressed sample.
ACKNOWLEDGMENT
This research was supported by an operating grant from the Social Sciences and Humanities Research Council (SSHRC) awarded to Brian J. Cox.
Notes
1Our criteria for strength of associations is based on the formal definition of small, medium, and large effect sizes specified by CitationCohen (1992). Correlations of greater than 0.1, 0.3, and 0.5 reflect small, medium, and large effects, respectively. Correlations representing medium or large effect sizes are considered strong associations in this study.
2Because the lower loading of the ORG subscales on the adaptive perfectionism dimension might attenuate the relationship between adaptive perfectionism and self-critical perfectionism, we recalculated the model after removing the ORG subscale. There was only minimal change in the correlation between adaptive perfectionism and self-critical perfectionism (.60 with ORG in the model and .62 without ORG in the model).