OBJECTIVE: To determine if ADHD/ODD symptoms are better represented by a bifactor model of disruptive behavior [general disruptive behavior factor along with specific inattention (IN), specific hyperactivity/impulsivity (HI), and specific oppositional defiant disorder (ODD) factors] than an ADHD-IN, ADHD-HI, and ODD three-factor model. METHOD: Mothers' and fathers' ratings of ADHD-IN, ADHD-HI, and ODD symptoms in a community sample of 4,658 children and adolescents (53% female) from Brazil, Thailand, and the US were used to evaluate the measurement models. RESULTS: The bifactor model of disruptive behavior provided a better fit than the three factor model. The bifactor model also occurred with mothers' and fathers' ratings of male and female children and adolescents. CONCLUSIONS: Consistent with predictions derived from recently articulated dual-pathway and trait-impulsivity models of externalizing liability, and from behavioral genetics studies indicating near complete overlap in vulnerability to ADHD and ODD, ADHD and ODD symptoms arose from a single, general disruptive behavior factor, which accounted for all of the variance in HI subscale scores and over half of the variance IN and ODD subscales. Thus, IN, HI, and ODD subscale scores strongly reflect a general disruptive behavior factor – not the specific content of their respective constructs.
OBJECTIVE: To determine if ADHD/ODD symptoms are better represented by a bifactor model of disruptive behavior [general disruptive behavior factor along with specific inattention (IN), specific hyperactivity/impulsivity (HI), and specific oppositional defiant disorder (ODD) factors] than an ADHD-IN, ADHD-HI, and ODD three-factor model. METHOD: Mothers' and fathers' ratings of ADHD-IN, ADHD-HI, and ODD symptoms in a community sample of 4,658 children and adolescents (53% female) from Brazil, Thailand, and the US were used to evaluate the measurement models. RESULTS: The bifactor model of disruptive behavior provided a better fit than the three factor model. The bifactor model also occurred with mothers' and fathers' ratings of male and female children and adolescents. CONCLUSIONS: Consistent with predictions derived from recently articulated dual-pathway and trait-impulsivity models of externalizing liability, and from behavioral genetics studies indicating near complete overlap in vulnerability to ADHD and ODD, ADHD and ODD symptoms arose from a single, general disruptive behavior factor, which accounted for all of the variance in HI subscale scores and over half of the variance IN and ODD subscales. Thus, IN, HI, and ODD subscale scores strongly reflect a general disruptive behavior factor – not the specific content of their respective constructs.