Discriminant validity of self-reported anxiety and depression in children: Negative affectivity or independent constructs?

Replicated recent investigations in the adult literature regarding the relation between anxiety and depression with a sample of children using sophisticated multivariate analyses. Nonreferred school children were administered commonly employed self-report measures of anxiety (Revised Children's Manifest Anxiety Scale and State-Trait Anxiety Inventory for Children) and depression (Children's Depression Inventory and Reynolds Child Depression Scale). Covariance structure analyses using the LISREL 7 program investigated whether a one- or two-factor model would best fit the data. A two-factor model was identified as better, based on chi-square and goodness-of-fit indices. Results suggest that the distinction between anxiety and depression sliould be maintained.