Classification challenges in perfectionism

Abstract

High performance expectations are central to perfectionism, but because most participants endorse high standards, it becomes difficult for practitioners and researchers to accurately screen for perfectionists. We addressed problems linked to the measurement and classification of perfectionism by testing various strategies aimed at broadening the range and skew of scores on the Standards subscale from the Almost Perfect Scale-Revised (APS-R; Slaney, Mobley, Trippi, Ashby, & Johnson, 1996). Randomly assigned participants (N = 506) completed the APS-R following standard instructions or 1 of 2 variations, one prompting participants to consider their responses in light of a normal distribution of scores and another in which participants used a visual analog (slider) scale. The visual analog scale produced more differentiated scores, but range restrictions and skewed distributions remained for all 3 variations. Statistical transformations improved skew. Factor mixture modeling was conducted using transformed and nontransformed perfectionism scores along with criterion indicators of emotion regulation (reappraisal or suppression), perceived stress, and depression. Results supported a 3-class model, although more balanced distributions of classes emerged than were previously reported. Perfectionists were differentiated from nonperfectionists by their higher standards scores. Maladaptive perfectionists scored highest among the classes on most self-critical perfectionism indicators, suppression, perceived stress, and depression. Adaptive perfectionists had the lowest levels of perceived stress and depression and scored highest on reappraisal. Both perfectionist classes had generally comparable concerns about mistakes, but criterion indicators suggested those were more problematic for maladaptive perfectionists. Results supported the value of incorporating adaptive and maladaptive criterion indicators in classification models.