In an attempt to determine how certain demographic, employment, prior criminal history, and probation sentence variables are associated with probation outcome, this study investigated probation terminations of 266 felony adult probationers in Tennessee whose cases had been terminated between January 1, 1980 and December 31, 1989 by completion of probation, revocation, or absconding. Probation outcome, the dependent variable, was measured as success or failure. Hypotheses predicting significant relationships between key variables including race, sex, prior criminal history, marital status, and employment status were tested for statistical significance and strength. Chi square was used to determine the statistical significance. All of the predicted relationships were in the right direction; with the exception of the variables race and age, all of the relationships were statistically significant. Logit regression was used to determine the best predictors of probation outcome. Four models were estimated, with statistically insignificant variables being deleted. The final model contained the variables sex, work status, marital status, prior felonies, and conviction offense, which were statistically significant and cited as the best predictors of probation outcome.