The Ability of the Rorschach to Predict Subsequent Outcome: A Meta-Analysis of the Rorschach Prognostic Rating Scale

To evaluate the ability of the Rorschach to predict subsequent outcome, the journal literature on the Rorschach Prognostic Rating Scale (RPRS) was reviewed and a meta-analysis was conducted on 20 statistics derived from a combined sample of 752. Using outcome criteria obtained an average of l year after initial testing, the uncorrected population correlation between RPRS scores and outcome was found to be p = .443 (95% confidence interval = .39 to .50). After making corrections to determine the validity of the RPRS when all subjects, all RPRS scores, and all outcome scores were included in the final statistical analysis, the estimated parameter increased to p = .560 (90% credibility value = .53). We are not aware of any other personality scale that uniformly demonstrates such high predictive validity. To flesh out the meaning of these results, they were placed in the context of other predictor-criterion relations drawn from various fields of study. The disparity between this strong evidence for validity and the chronic criticisms leveled against the Rorschach is discussed and suggestions are made for future research.