Human figure drawings: validity in assessing intellectual level and academic achievement.

In a sample of 216 child and adolescent psychiatric inpatients, significant but low correlations were found between Goodenough-Harris (GH) drawing scores and both Full Scale IQs and academic achievement. The percentage of subjects correctly classified in appropriate IQ categories ranged from 35 to 44%; the percentage of misclassified subjects ranged from 56 to 65%. Consideration of visual-motor integration, using the Bender-Gestalt, did not improve the accuracy of G-H scores in predicting IQ scores in this sample. These results indicate that human figure drawings should not be substituted for other well-established intelligence and achievement tests nor used as an additional measure of these constructs when one is evaluating psychiatric patients.