Abstract
Research on the constructs of childhood depression and anxiety has typically found that the 2 traits are highly correlated. In a study of the construct validity of childhood depression, anxiety, and aggression, Epkins and Meyers (1994) concluded that, despite some mixed results, the 3 constructs demonstrated adequate convergent and discriminant validity. We reexamined the results of the Epkins and Meyers study in 2 new analyses—a focused comparison of correlational patterns among the traits and a confirmatory factor analysis—and found no support for the discriminant Validity of depression and anxiety and only weak support for their convergent validity.