Assessing the measurement invariance of the RCADS-25 questionnaire across gender and child-parent dyads in the presence of multilevel data.

This study aims to assess the measurement invariance of the short version of the Revised Child Anxiety and Depression Scale (RCADS-25) across male and female adolescents along with child self-reports and parent proxy-reports. The sample consisted of 2039 adolescents and their parents from 51 public and private schools in Shiraz (southern Iran). The measurement invariance of the RCADS-25 was investigated by the multilevel multiple-indicators multiple-causes model. In addition, the reliability of the questionnaire was examined by Cronbach's alpha coefficient, and its validity was evaluated by exploratory and categorical confirmatory factor analysis. Although the RCADS-25 includes five (20%) non-invariant items across male and female adolescents, their impact is negligible at the scale level. In addition, 10 out of 25 items (40%) were non-invariant across self- and proxy-reports, which did not have substantial effect at the scale level. Based on our findings, the instrument can be used, albeit with caution, for comparing depression and anxiety scores between male and female adolescents as well as child self-reports and parent proxy-reports.