The nature and measurement of depression

Résumé 0

"The initial goal of this study was to determine if the cognitive andaffective components of depression, which are measured collectively by the BDI Cognitive-Affective subscale, could be identified as separate factors in a non-clinical population.A pool of 40 cognitive and affective depression items was adapted from the BDI andthree other widely used depression measures. These items were administered with bothstate and trait instructions to 251 university students, who also responded to the BDI,Zung, CES-D, and trait measures of anxiety, anger and curiosity. Contrary to the expectedfinding of cognitive and affective factors, two very strong factors were identified,which were defined by items that described the presence or absence of state and traitdepression. The best depression-present (dysthymia) and depression-absent (euthymia)items were selected to form 20-item State (S-Dep) and Trait (T-Dep) Depression scales,each with 10-item S-Dep and T-Dep Dysthymia and Euthymia subscales. The alphacoefficients for the S-Dep and T-Dep scales and subscales for the total sample, and inseparate analyses for females and males, were .90 or higher (mdn. r = .93), indicatingstrong internal consistency. The T-Dep Scale correlated highly with the BDI, Zung andCES-D (mdn. r = .80), providing impressive evidence of concurrent validity. Thecorrelations of the T-Dep Scale with all three widely used depression measures werealso substantially higher than the corresponding correlations of the S-Dep Scale (mdn.r = .66). These findings suggested that while the BDI, Zung and CES-D measure bothstate and trait depression, they appear to more accurately assess relatively persistenttrait-like characteristics."