Item Response Theory Analysis of Two Self-Report Depression Measurements in French University Students

• In Event: Marriott Roundtable Session 33
  In Roundtable Session: Applied Psychometric Methods in Validation Research

Abstract

To contribue in measuring accurately depressive symptoms in university students, we used Generalized partial credit model (GPCM), which is based on item response theory (IRT), to examine the factor-model structure and psychometric properties of items of two self-report depression instruments: the Center for Epidemiological Studies Depression scale (CES-D) and the General Health Questionnaire (GHQ). Based on a sample of French university students (N= 1338), the results displayed acceptable fit indices supporting the original four- and three-dimensional structure, respectively. As expected, the both measures highly intercorrelated (r = 0.77). The reliability of the both measures appeared to be satisfactory. IRT GPCM has displayed a substantial number of items that were good for discriminating the latent trait of these instruments.

Authors

• KOSSIGAN KOKOU KPOLOU, University of Picardie Jules Verne

• Sunyoung Park, California Lutheran University