Structure of ADHD/ODD Symptoms in Spanish Preschool Children: Dangers of Confirmatory Factor Analysis for Evaluation of Rating Scales

Abstract

[eng] Confirmatory factor analysis (CFA) is often used to evaluate attention-deficit/hyperactivity disorder (ADHD) and oppositional defiant disorder (ODD) symptom ratings by parents and teachers. An ADHD-inattention (IN), ADHD-hyperactivity/impulsivity (HI), and ODD three-factor model is usually the best model. Acceptable CFA models, however, can hide symptoms with poor convergent and discriminant validity. To demonstrate this issue, CFA models (i.e., cross-loadings set to zero) along with exploratory CFA models (i.e., cross-loadings allowed) were applied to parent (n = 308) and teacher (n = 258) ratings of ADHD/ODD symptoms with Spanish preschool children (M = 4.78, SD = .84, 56% boys). While the three-factor CFA model provided an acceptable-fit with moderate to substantial symptom-factor loadings, the three-factor exploratory CFA model, however, found a large number of the symptoms failed to show convergent and discriminant validity. These outcomes argue for the use of exploratory CFA procedures in the initial evaluation of ADHD/ODD rating scales.