Presents maximum-likelihood confirmatory factor analysis (CFA) as a means for interpreting complete and incomplete multitrait-multimethod designs. CFA was applied to four data sets representing four designs and to an artificial data set to demonstrate its use in test validation designs to assess discriminant, convergent, and construct validity.…

Presents results from two studies that investigate the measurement of childhood depression through self-report and reports from others. Results were discussed in terms of the need to consider self-report in the discussion of depression with its covert components and in terms of the need for improved measurement instruments. (BH)