In this study, symptom (item) level data were used to perform a psychometric analysis of the DSM-III-R personality disorders (PDs). Determined for each PD criteria set were convergent validity, discriminant validity, and internal consistency. The results indicated that the majority of the PD criteria sets (6 of the 11 ) possessed adequate convergent validity, although discriminant validity was problematic for most of these disorders. Internal consistency was also weak for the PD criteria sets, with only 3 of the 11 exceeding a minimum cutoff score of .70. The present study employed a methodology modeled after the one reported by Morey (1988a), and the results of the two studies were highly similar. Consistent findings across the two data sets can be taken to reflect the actual psychometric properties of the DSM-III-R PDs. The success of our replication demonstrates the potential that large-scale psychometric investigations hold for aiding the development and refinement of the DSM PDs.