How should discrepancy be assessed in perfectionism research? A psychometric analysis and proposed refinement of the Almost Perfect Scale–Revised

Research on perfectionism with the Almost Perfect Scale-Revised (APS-R) distinguishes adaptive perfectionists versus maladaptive perfectionists based primarily on their responses to the 12-item unidimensional APS-R discrepancy subscale, which assesses the sense of falling short of standards. People described as adaptive perfectionists have high standards but low levels of discrepancy (i.e., relatively close to attaining these standards). Maladaptive perfectionists have perfectionistic high standards and high levels of discrepancy. In the current work, we re-examine the psychometric properties of the APS-R discrepancy subscale and illustrate that this supposedly unidimensional discrepancy measure may actually consists of more than one factor. Psychometric analyses of data from student and community samples distinguished a pure fiveitem discrepancy factor and a second four-item factor measuring dissatisfaction. The five-item factor is recommended as a brief measure of discrepancy from perfection and the four-item factor is recommended as a measure of dissatisfaction with being imperfect. Overall, our results confirm past suggestions that most people with maladaptive perfectionism are characterized jointly by chronic dissatisfaction as well as a sense of being discrepant due to having fallen short of expectations. These findings are discussed in terms of their implications for the assessment of perfectionism, as well as the implications for research and practice.