The Monte Carlo study compared the performance of predictive discriminant analysis (PDA) and that of logistic regression (LR) for the two-group classification problem. Prior probabilities were used for classification, but the cost of misclassification was assumed to be equal. The study used a fully crossed three-factor experimental design (with 200 replications in each cell) manipulating sample size, prior probabilities, and equal/unequal covariance matrices. Two data patterns were simulated to provide a replication mechanism within the study. The major findings are: (1) PDA and LR have comparable performance for two groups with equal prior probabilities; and (2) for two groups with unequal prior probabilities, LR minimizes the error rate for the smaller group, and PDA minimizes the error rate of the larger and total sample. Consistency was observed across the two data patterns. The findings reveal a picture of PDA and LR that seems to be more complicated than that typically portrayed in the literature. Limitations of the study are noted, and future directions are suggested. (Contains 2 figures, 5 tables, and 29 references.) (Author/SLD)