Reliability and validity of SF-12 among adolescents in western rural areas of China

Objective To assess the reliability and validity of the 12-Item Short Form Health Survey (SF-12) among adolescents in western rural areas of China.Methods Cluster sampling method was used to select 2 871 adolescents of grade 7 and grade 10 from two rural middle schools in Zizhong county,Sichuan province and a cross-sectional questionnaire survey was performed in November 2015.The reliability of the SF-12 was assessed with internal consistency reliability and the validity of SF-12 was evaluated with convergent,discriminate,and structure validity.The discrimination of the scale was assessed among boys and girls,left-behind and non left-behind adolescents,respectively.Results The Cronbach's α for internal consistency reliability of the scale was 0.80 and the Cronbach's α for physical component summary (PCS)and mental component summary (MCS)were 0.66 and 0.74.The discriminant validity and convergent validity were 100%.Every item in SF-12 was highly related to the dimension it belongs to and the correlation coefficients of all the items were all above 0.80.There were two eigenvalues greater than 1 in factor analysis and the cumulative contribution of the eignvalues was 51.56%.The results of confirmatory factor analysis showed that the structure of the constructed model was in accord with the theoretical assumption (χ²/df=4.189,root mean square error of approximationRMSEA=0.033,incremental fit indexIFI=0.992,comparative-fit indexCFI=0.992).The PCS and MCS scores showed significant difference among the adolescents of different gender,left-behind status,disease condition,and the two-week prevalence,suggesting that the SF-12 had good discrimination capacity.Conclusion The SF-12 for measuring quality of life among adolescents in western rural areas of China has good reliability,validity and discrimination.