Ying-dong Zheng1, Jian-hui Shi, Ruo-xiang Cao. 1. Department of Epidemiology and Biostatistics, Peking University School of Public Health, Beijing 100191, China. zhengyd@tom.com
Abstract
OBJECTIVE: To evaluate construct validity of China residents health literacy questionnaire and explore the application of confirmative factor analysis (CFA) in health literacy measurements. METHODS: CFA was performed on dimension of basic healthy skills in the questionnaire. Latent variable scores and comprehensive score based on factor variance weight were calculated. Comparison among the latent variable scores, comprehensive scores and proportion of knowing was conducted via GLM or Logistic model. RESULTS: Two factors named general capacity and emergency capacity were well fitted (chi(2)=12.12, P=0.356, IFI=0.998, TLI=0.996, RMSEA=0.007). According to comprehensive score the sample qualification rate of basic health literacy was 38.5%(785/2 040), much similar to 38.2% (780/2 040) which is calculated by original item scores. Latent variable score or comprehensive score has more statistical power than that of proportion of knowing in multivariate analysis. CONCLUSION: CFA is a useful tool and valuable for applying in the field of health literacy measurement and analysis.
OBJECTIVE: To evaluate construct validity of China residents health literacy questionnaire and explore the application of confirmative factor analysis (CFA) in health literacy measurements. METHODS:CFA was performed on dimension of basic healthy skills in the questionnaire. Latent variable scores and comprehensive score based on factor variance weight were calculated. Comparison among the latent variable scores, comprehensive scores and proportion of knowing was conducted via GLM or Logistic model. RESULTS: Two factors named general capacity and emergency capacity were well fitted (chi(2)=12.12, P=0.356, IFI=0.998, TLI=0.996, RMSEA=0.007). According to comprehensive score the sample qualification rate of basic health literacy was 38.5%(785/2 040), much similar to 38.2% (780/2 040) which is calculated by original item scores. Latent variable score or comprehensive score has more statistical power than that of proportion of knowing in multivariate analysis. CONCLUSION:CFA is a useful tool and valuable for applying in the field of health literacy measurement and analysis.
Authors: Daitao Zhang; Shuangsheng Wu; Yi Zhang; Peng Yang; C Raina MacIntyre; Holly Seale; Quanyi Wang Journal: BMC Public Health Date: 2015-08-19 Impact factor: 3.295