Dimensionality of the Chinese CES-D: Is It Stable across Gender, Time, and Samples?

Abstract

Depression is a common mental illness among Chinese adolescents. Although the Epidemiological Studies Depression Scale (CES-D) has been widely used in diverse populations, the reported factor structures are inconsistent, and its longitudinal invariance is under-researched. This study examined the psychometric properties and factorial invariance across gender and time of the CES-D among Chinese adolescents. Adolescents aged above 11 years from five schools in Chengdu responded to a questionnaire at Wave 1 (n = 5690). Among them, 4981 participants completed the same questionnaire after six months (Wave 2). The matched sample was composed of 4922 students (51.5% were girls; mean age = 13.15 years) at Wave 1. We used exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to examine the factor structure and performed multi-group CFA to test the factorial invariance across gender and time. A three-factor solution was identified, including “positive affect”, “somatic complaints”, and “depressed affect”. Results of multi-group CFA comparisons supported the factorial invariance of the resultant three-factor solution. Using a new sample of Chinese adolescents in Southwestern China, the present study reproduced earlier findings on adolescents in other areas in China. This study has implications for depression assessment and research in Chinese adolescents.