The purpose of this study was to test the predictive efficiency of the Graduate Record Examination Analytical (GRE-A) score when used in combination with other predictors for doctoral candidacy decisions in an educational psychology department. It was hypothesized that the GRE-A would be a more efficient predictor than faculty ratings on a qualifying examination, oral examination, and/or folder ratings. It was also hypothesized that the GRE-A's predictive efficiency would be of a similar magnitude regardless of area of specialization, sex, and ethnicity. Data were gathered across two years from 68 subjects. The hypotheses were not supported to the extent expected. To the extent the GRE-A is used in combination with other predictor variables the hypotheses were affirmed, providing sufficient evidence regarding the utility of the GRE-A as a predictive measure to warrant the continued use of this score in combination with the usual predictive variables. (BW)