Reliability and Validity of 10-item CES-D among Middle Aged and Older Adults in China

Qingbo Huang,Xiaohua Wang,Gong Chen
DOI: https://doi.org/10.13342/j.cnki.cjhp.2015.07.023
2015-01-01
Abstract:Objective:To examine the reliability and validity of 10-item of Center for Epidemiologic Studies Depression Scale among older adults and test the scale invariance for middle aged and older a-dults.Methods:Data were derived from 2011 wave of China Health and Retirement Longitudinal Study and 14931 individuals with 45 or more years old.First,item discrimination analysis was used to test the discriminative ability among different people with various level;Second,among the entire sample,older a-dults and middle aged adults,internal consistency analysis was conducted to test the internal consistency, and,Confirmatory Factor Analysis( CFA) were used to test the three assumed factors:A one-factor mod-el,a two-factor model(Depressed affect,Positive affect)and a three-factor(Depressed affect,Somatic affect Positive affect)respectively.Next,multiple-group analysis was conducted to test the scale invari-ance for middle aged and older adults.Results:The discrimination analysis showed that item score differ-ences between the highest and lowest 27%percentile in total scores was statistically significant.The Cron-bach's alpha was 0.815,0.814 and 0.813 for entire sample,middle aged adults and older adults respec-tively and data support the two-factor model for the entire sample;For comparison with the two-factor unrestrictive model,the△CFI,△RMSEA and △SRMR meet the cutoff point for scale invariance.Con-clusion:10-item CESD demonstrates adequate reliability and validity,two-factor model has factorial e-quivalence across the older and middle aged adults,the scale has good discrimination and can be used to screen depressive symptoms among middle aged and older adults.
What problem does this paper attempt to address?