The Factor Structure of ADHD – Different Models, Analyses and Informants in a Bifactor Framework

The methodological approach of exploratory structural equation modelling (ESEM) has only been applied once to the construct of Attention-deficit/hyperactivity disorder (ADHD). We decided to compare bifactor models based on confirmatory factor analyses (Bi-CFA) and exploratory equation modeling (Bi-ESEM) only, as there is a growing support of a bifactor structure of ADHD. To examine the factorial validity of the construct we compared three possible bifactor models. One model with two specific factors (inattention and hyperactivity/impulsivity), another model with three specific factors (inattention, hyperactivity and impulsivity) and an alternative, incomplete model with one general ADHD and two specific factors (inattention and impulsivity). We used parent- (N = 1386; Age: M = 11.70, SD = 3.18; Sex: 74.5 % male) and teacher-ratings (N = 110; Age: M = 11.27, SD = 3.04; Sex: 77.5 % male) from clinically referred children between the age of 6 and 18. The results indicate that both methods lead to equally good model fit and for both informants the reliable variance of the specific factor hyperactivity is almost completely explained by the general factor. However, in the teacher condition cross-loadings seem to be of particular importance. Across both methods and informants covariation among ADHD symptom items can be in most part attributed to a general ADHD factor as well as to three (inattention, hyperactivity and impulsivity) or two (inattention and impulsivity) weakly defined specific factors. Further research regarding associations between the specific factors of ADHD and other disorders (e.g. conduct disorder) should be conducted.