The purpose of this study was to present and evaluate Bayesian-type models for estimating probabilities of program completion and for predicting first quarter grade point averages of community college students entering certain allied health fields. Two Bayesian models were tested. Bayesian Model 1--Estimating Probabilities of Program Completion--was developed from discrete cases of Bayes' formula with counselors' inputs as a priori probabilities and posterior probabilities of graduate status of the discriminant analysis function as likelihoods. The a priori probabilities and likelihood probabilities were combined in Bayes' Theorem to produce posterior probabilities of successful program completion. Bayesian Model 2--Predicting First Quarter Grade Point Averages--which was an application and specialization of the Bayesian linear model developed by Lindley and Smith, involved the assumption of homogeneity of regression coefficients (but not intercepts) across groups. The efficiencies for both Bayesian models were compared and evaluated in terms of two counselor-selection models and two classical statistical models with the results showing that Bayesian Model 1 was most efficient in selecting successful students while no significant difference in efficiency was found for Bayesian Model 2 as compared to a classical statistical model. (Author/JDS)