Psychometric properties and gender invariance of the Chinese version of the self-report pediatric quality of life inventory version 4.0: Short form is acceptable

Purpose To evaluate the psychometric properties and gender invariance of the Chinese version of the Pediatric Quality of Life Inventory (PedsQL) for 8- to 12-year-olds. Methods Psychometric testing and confirmatory factor analysis (CFA) were used with a convenience sample of 8- to 12-year-old children (n = 479) for PedsQL full and short forms. Results The internal consistency reliability was satisfactory for all subscales and total scores (Cronbach's a = 0.73-0.90), except for the school subscale (0.68 [full form], 0.62 [short form]). Test-retest reliability was 0.67-0.84. Convergent validity was supported by the correlation between the Children's Depression Inventory and PedsQL psychosocial subscale (r = -0.69). Construct validity determined using CFA showed a better model fit in the short form (RMSEA = 0.06) than in the full form (RMSEA = 0.08). Measurement invariance across gender determined using nested CFA models showed that all absolute DRMSEA values were<0.01. Conclusions The Chinese version of the PedsQL is a relatively reliable and valid instrument, and the PedsQL short form showed a better construct validity than did the full form. Measurement across gender was invariant; therefore, the comparisons of quality of life between boys and girls were appropriate.