This study was designed to assess the reliability of the "Psychological Rating Scale for Diagnostic classification as an instrument for determining diagnoses consistent with DSM-III criteria and nomenclature. Pairs of raters jointly interviewed a total of 50 hospital patients and then independently completed the 70-item rating scale to arrive at Axis I and Axis II diagnoses which were subsequently correlated with diagnoses obtained by standard psychometric methods, interrater agreement was 88 per cent for Axis I and 62 per cent for Axis II, with correlations of .94 and .79 respectively.

The poor reliability of the DSM diagnostic system has been a major issue of concern for many researchers and clinicians. Standardized interview techniques and rating scales have been shown to be effective in increasing interrater reliability in diagnosis and classification. This study hypothesized that the utilization of the Psychological Rating Scale for Diagnostic Classification for assessing the problematic behaviors, symptoms, or other characteristics of an individual would increase interrater reliability, subsequently leading to higher diagnostic agreement between raters and with DSM-III classification. This hypothesis was strongly supported by high overall profile reliability and individual profile reliability. Therefore utilization of this rating scale would enhance the accuracy of diagnosis and add to the educational efforts of technical personnel and those professionals in related disciplines.