Toward a Training Model for Rorschach Scoring Revisited: A Follow-up Study on a Training System for Interscorer Agreement

Abstract

A standardized method for training in Rorschach scoring using the Perceptanalytic system showed high interscorer reliability in a classroom setting by 1984. To assess whether this model for training would continue to yield stronger interscorer agreement with continued use in vivo data from classroom training were gathered over a six-year follow-up period (1985–1991). For the subjects, 167 students in a graduate clinical psychology training program, results indicated the system produces high interscorer agreement at both Level I (relatively clear and simple responses) and Level II (very complex or pathological responses). Over-all percent of agreement for 54,329 combined coding and summarizing operations attained a M of 90% ( SD = 6.0%). Sufficiently high interscorer reliability in Rorschach scoring can be attained in a classroom setting to support the use of the Perceptanalytic method for training doctoral students. Further research on training systems for the Rorschach as well as some speculative heuristic hypotheses about the nature of the learning process for Rorschach scoring are suggested.