The reliability of Rorschach variables for inpatients presenting symptoms of depression and/or helplessness.

Fifty newly admitted patients presenting symptoms of depression and/or helplessness were recruited to participate in a study concerning depression and randomized into two groups. Rorschachs were administered shortly after admission and then three or four days later. The subjects in the experimental group were instructed to give responses different than they had in the first test. Approximately two-thirds of the responses given by the experimental group in test 2 were different from those in test 1, whereas the control subjects repeated 66% of their test 1 answers in test 2. Retest correlations for 28 variables critical to interpretation show that the two groups differed significantly for four. The retest correlations for four of five variables often used as indices of depressive features were significantly high for both groups. Unusual findings are noted in the retest correlations for FM and m for the Control group and questions are raised concerning the modest retest correlations for CF and C + Cn versus the more commonplace summation of CF + C.