Construct validity for measures of childhood depression: Application of multitrait-multimethod methodology

In Exp I, 24 emotionally disturbed inpatients (aged 7 yrs 1 mo to 15 yrs 8 mo) were administered a battery of measures of depression including the Children's Depression Inventory (CDI), the Piers-Harris Self-Concept Scale (P-H), a DSM-III rating scale, a measure of learned helplessness, and a peer-nomination inventory. Results show that the CDI was significantly related to the P-H and the measure of learned helplessness. In Exp II, 133 2nd-8th graders were used to investigate the convergent and discriminant validity of several measures of depression and anger in a multimethod-multitrait matrix. Measures of depression and anger were obtained from the CDI, peers, and teachers. Results show a method effect twice as large as the trait effect. Both convergent and discriminant validities were supported for the measures as a group. However, the method of measurement was as important a source of variance as was the trait being measured. (28 ref)