Endler, Ruchton, and Roediger's (1979) method of obtaining exact publication counts for individual psychologists in order to make ratings of psychology departments was evaluated. Using the 1975 "Social Science Citation Index" (SSCI) as a data base, Endler et al. ranked departments on total citation and total publication counts. Additionally, they provided ranks based on mean citation and publication counts. Attention was focused on the Endler et al. study to determine the accuracy of publication counts obtained from the SSCI in comparison to self-reported publications. Self-reported 1975 publications were obtained from 113 psychologists, and publication counts were made by 16 categories. The 1975 SSCI "Source Index" was used to estimate what Endler and associates' count would have been for each of the 113 psychologists. Only 174 of the 260 reported publications could be found in the 1975 "Source Index." Moreover, the correlation between an individual's SSCI count and self-reported publications was .61, which is low given the large number of journals (over 3,000) referenced in the 1975 SSCI. "Psychological Abstracts" was then evaluated in a similar manner to determine if it would be a more reliable source for counting publications. The analysis yielded only a moderate correlation of .65. Based on the findings and additional analysis, it appeared that discrepancies between self-reported counts and archival sources were probably due to oversight and inaccuracies in the archival sources. The errors could significantly alter a department's rank in the Endler et al. study. (SW)