The Center for Epidemiological Studies Depression Scale (CES-D): Measurement equivalence across gender groups in Hispanic college students

Abstract

BackgroundThis study examined among Hispanic college students the factor structure and gender measurement invariance of the Center for Epidemiologic Studies Depression (CES-D) scale. The CES-D's original 20 item four-factor model was compared to a recently proposed 14 item, three-factor model. MethodsParticipants included 858 Hispanic undergraduates at an ethnically diverse, major research, urban, public university in the Southwest United States (U.S.). Multigroup confirmatory factor analyses (CFAs) and structural equation modeling were used to examine the factor structure, gender invariance, and construct validity of the CES-D scores. ResultsMultigroup CFAs provided support for full configural and metric invariance and partial scalar invariance for the three-factor model across gender groups. Two of the 14 items (#3 and #18) demonstrated different thresholds; women scored higher than men in both items. The pattern of latent correlations of the three depression factors to four stress variables (college academic, social, financial and discrimination stress) demonstrated configural, metric and scalar invariance across gender groups. LimitationsAbout two thirds of the participants were women and most were of Mexican descent and had been born in the U.S. Therefore, findings may not generalize to more gender balanced samples and students from other Latin American countries or immigration generational status. ConclusionsPartial scalar invariance in two items may not have a strong impact in the calculation of the overall depression score. Given evidence for functional invariance, results provide support for the use of the14-item, three-factor CES-D scale to assess depression symptoms among Hispanic college students.