Reliability and validity of Chinese version of Revised Attribution Questionnaire-27 as used in medical students

Objective To develop the Chinese version of Attribution Questionnaire-27 and examine its replicability,reliability and validity.Methods The questionnaire was translated using the standard translation/back-translation method and examine its replicability,reliability and validity in 230 medical students who were selected by convenient method.Results The Cronbach α coefficient for Attribution Questionnaire-27 was 0.82 and ranging between 0.55 and 0.84 for the subscales.The test-retest reliabilities of the total scale was 0.77 and the subscales was between 0.61 to 0.81 after one week.Confirmatory factor analysis confirmed the hypothesized six factors attribution of responsibility model (x2/df=3,GFI=0.925,NFI=0.880,CFI=0.856,IFI=0.761,RMSEA=0.094) and three factors dangerous model (x2/df=1.773,GFI=0.959,NFI=0.941,CFI=0.973,IFI=0.973,RMSEA=0.058).The path analysis results support the existing theoretical assumptions.Condusions The Chinese version of Attribution Questionnaire-27 has good reliability and validity and can be used in medical student. Key words: Mental illness;  Discrimination;  Reliability;  Validity;  Psychometric studies