This study sought to assess the validity of two projective instruments, a sentence completion test and human figure drawings, under conditions which approximate their clinical usage as closely as the application of rigorous experimental controls would permit. Specifically, the experimental conditions paralled common clinical practice in two important ways : (a) specific personality characteristics were described and predicted from the tests, and (b) these predictions were made on the basis of an over-all impressional inspection of the test record, rather than after detailed item-by-item scoring and analysis.
Most validation studies have been concerned with the identification of clinical groups, or with the prediction of broad, general characteristics, such as adjustment, improvement, college grades, field dependence, etc. Only a very few validation studies on either of these two tests have attempted to predict specific personality traits (lo, 12). In both of these studies there is a question of possible contamination of the criterion with the predictor.
Validation studies on the sentence completion test and on human figure drawings are far too numerous to be reviewed here. However, many of these studies are covered by one of several available surveys of the literature('-2. 3 , '9 *). 'The author wants to thank Dr. Edward E, Cureton for planning the statistical procedures employed in this study, Messrs. William A. Kobzeff, Marion F. Sobczynski and Louis M. Herman for serving as judges, and Harry M. Henkin and Stephen F. Zolczynski for supervising the statistical computations.