Since the revision of the Center for Epidemiologic Studies Depression Scale (CESD-20-R), few studies have examined its validity and none yet its gender invariance. This study addressed these critical issues among 502 Nigerian university students (50.4% of females, Mage = 22.10 years). Confirmatory factor analyses (CFAs) demonstrated that a one-, two- and three-dimensional structures were tenable; however, the latter model appeared to be the best-fitting model to the data reflecting cognitive-affective, physical, and self-destructive components. The estimates of the internal consistency reliability of the CESD-20-R were significant (McDonald ω = 0.94, Greatest Lower Bound = 0.96). Multi-group CFAs provided support for full configural and weak factorial invariances but partial strong and strict invariances across gender. At strong factorial invariance level, two items (#13 “I felt fidgety” and #19 “trouble getting to sleep”) lacked invariance. Further, latent mean differences showed significantly higher scores among males than females on the self-destructive component reflecting suicidal ideation items (d = −0.238). Following CESD-20-R diagnostic criteria, 6.6% met criteria for major depression episode (MDE), 2.4% for probable MDE, and 5.6% for possible MDE. In sum, the CESD-20-R, which has been developed to be conformed to current depression diagnostic criteria, yielded a meaningful factorial structure and excellent internal consistency reliability. It can be used successfully by researchers and clinicians to establish gender comparisons of depressive symptoms in the context under consideration here. However, caution is required when interpreting some specific items. Moreover, in health programs for early prevention and intervention, careful attention needs to be paid to males who have shown more vulnerability to the suicidal ideation component of depressive symptoms than females.
- internal reliability
- measurement invariance
- structural validity
- The Center for Epidemiologic Studies Depression Scale-Revised