" " , "

This pattern of correlations is suggestive of the high levels of comorbidity found in other community surveys (Anderson. Williams, McGee, & Silva, 1987; Bird et aI., 1988, 1993; Offord. Boyle. & Racine, 1989). Although the correlation between ADHD and CD in our analysis is somewhat lower than we might have predicted (Szatmari, Boyle, & Offord, 1989), this follows in part from the fact that ADHD was assessed only by parental informants, whereas in all the other domains the children as well as the parents provided information. Thus the correlations between ADHD and other domains will reflect the difference between a shared parental view of ADHD and a shared family view of the other domains. The correlations between the internalizing symptoms are consistent with the overlap found among these disorders (Kovacs, Feinberg, Crouse-Novak, Paulauskas, & Finkel stein, 1984; Strauss, Last, Hersen, & Kazdin, 1988). Perhaps what is most interesting is that, when we consider information that children and parents agree upon, the symptoms of the externalizing disorders are so clearly distinguished from symptoms of internalizing disorders. Although we must await the appropriate multivariate genetic analyses to interpret these results fully, we might predict that the VTSABD will confirm at least two broad vulnerability factors for symptoms of the common disorders of childhood.