Interjudge Agreement of Draw-A-Person Diagnostic Impressions

Summary

The interjudge agreement of Draw-A-Person (DAP)-based diagnostic impressions was assessed. Twenty-four psychologists were asked to categorize 48 DAP protocols as being the productions of organics, paranoid schizophrenics, non-paranoid schizophrenics or normal controls. The mean between-judges proportion of agreement was only .41 and the mean proportion correct was a mere .28. (A proportion of .25 could be anticipated in each case by chance.) Strength of interjudge agreement did not vary with the extent to which judges utilized the test but did seem dependent upon the subjects' diagnoses. The results indicated that diagnostic impressions based on the DAP alone seem to be neither usefully valid nor impressively consistent across psychologists. It is suggested that the profession is due for a re-evaluation of its DAP training techniques and its use of the test.