Empirically based dimensions of externalizing symptoms in children and adolescents: a multitrait-multisource approach

The present study sought to refine knowledge about the structure underlying externalizing dimensions. From a "top-down" ICD/DSM-based perspective, externalizing symptoms can be categorized into attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), and conduct disorder (CD). From a "bottom-up" developmental theory-based perspective, disruptive behaviors can be meaningfully described as aggressive (AGG) and rule-breaking (RB) behaviors. We analyzed five large data sets comprising externalizing symptom ratings obtained with a screening instrument using different sources (parents, teachers, self-ratings) from different samples. Using confirmatory factor analyses, we evaluated several factor models (unidimensional; correlated factors; bifactor (S-1) models) derived from an ICD/DSM- and theory-based perspective. Our optimally fitting models were assessed for measurement invariance across all sources, sample settings, and sex. Following several model-based criteria (model fit indices; factor loadings; omega statistics; model parsimony), we discarded our models stepwise and concluded that both the ICD/DSM-based model with three correlated factors (ADHD, ODD, CD) and the developmental theory-based model with three correlated factors (ADHD, AGG, RB) displayed a statistically sound factor structure and allowed for straightforward interpretability. Furthermore, these two models demonstrated metric invariance across all five samples and across sample settings (community, clinical), as well as scalar invariance across sources and sex. While the dimensions AGG and RB may depict a more empirically coherent view than the categorical perspective of ODD and CD, at this point we cannot clearly determine whether one perspective really outperforms the other. Implications for model selection according to our model-based criteria and clinical research are discussed.