Self-report measures of depression
Date of this Version
This paper examines some aspects of the psychometric adequacy of existing self-report depression measures, so that clinicians might better appreciate their, reliability and validity. Issues addressed include the desirability of moderate rather than maximum test-retest and item homogeneity/internal consistency estimates; the distinction between- measurement at the surface syndrome level vs. that at the fundamental source state/trait level; the importance of appropriate rather than arbitrary factor analytic procedures; and the need for multivariate measures instead of the single-scale instruments so often employed to measure depression in isolation from interacting emotions such as anxiety or stress. Apart from these more general issues, a number of specific criticisms is considered, along with recommendations for better self-report measures of depression.