MEASURING ADOLESCENT DEPRESSION - TESTS OF EQUIVALENT FACTORIAL STRUCTURE FOR ENGLISH AND FRENCH VERSIONS OF THE BECK DEPRESSION INVENTORY

The aims of the present study were threefold: (1) to test for the equi valency of an hierarchical 3-factor structure across English and Frenc h versions of the Beck Depression Inventory (BDI) for nonclinical adol escents; (2) given evidence of poor model fit, to validate the factori al structure of the French version across three independent samples (n (1) = 336; n(2) = 435; n(3) = 381) of adolescents using exploratory an d confirmatory factor analyses, and; (3) to compare the factorial patt ern of BDI measurement and structure across English and French version s. Although both versions were best represented by an hierarchical S-f actor structure, a differential pattern of loadings was evidenced for eight items. Except for three items, factorial measurement and structu re of the French BDI were invariant across two independent samples. Th ese findings carry important implications for the interpretation of sc ores derived from translated psychological tests in general, and more particularly for scores derived from English and French versions of th e BDI in measuring depression for nonclinical adolescents.