Measurement invariance in the General Health Questionnaire-12 in young Australian adolescents.

The General Health Questionnaire-12 (GHQ-12) has been extensively validated with adults, but its validity with young adolescents is questionable. Further, the dimensionality of the survey remains in doubt. Confirmatory factor analysis was used to compare five competing models for the GHQ-12 using an adult sample (n = 235). The best model was then applied to a young adolescent dataset (n = 236, aged 11-15): all fit indices were "adequate" or "good" (i. e. Comparative Fit Index 0.97). Measurement invariance for the combined data was demonstrated for item loadings and factor covariance but not for error variance. These data provide support for the validity of the survey with young adolescents and suggest that adolescents interpret the GHQ-12 in a similar manner to adults.