A measurement framework to determine the construct validity of ADHD/ODD rating scales: Additional evaluations of the child and adolescent disruptive behavior inventory

摘要

A multiple group confirmatory factor analysis was used to test the invariance of an attention-deficit/hyperactivity disorder-inattention, attention-deficit/hyperactivity disorder-hyperactivity/impulsivity, oppositional defiant disorder toward adults, academic and social competence five-factor model across of independent groups of Thai adolescents with the Child and Adolescent Disruptive Behavior Inventory (CADBI). Mothers and fathers' ratings on 872 adolescents were compared to mothers and fathers' ratings on a different group of 983 adolescents. The design allowed four invariance analyses (i.e., mothers' ratings of the first group compared to mothers' ratings of the second group; fathers' ratings of first group compared to fathers' ratings of the second group; mothers' ratings of first group compared to fathers' ratings of the second group; and fathers' ratings of first group compared to mothers' ratings of second group). Support was found for the invariance of like-symptom loadings and symptom intercepts as well as like-factor variances, covariances, and factor means across the four invariance analyses. The findings further expand the construct validity of the CADBI. The paper also introduces a multiple indicators by multiple traits by multiple methods by multiple sources by multiple occasions by multiple groups' measurement matrix to guide the evaluation of the construct validity of ADHD/ODD rating scales.