First-time nurse candidates from a four-year baccalaureate nursing program were examined to identify predictors of success on NCLEX-RN. Five logistic regression models were tested to see if specific variables increased nurse candidates' odds for success or failure. The use of admission criteria was the poorest model in predicting performance. Cumulative grade point averages (GPAs) suggested that the end of the sophomore year was the best time for predicting success and the end of the junior year was best for predicting failure. Age was inversely related to successful performance in three of the five models. Using cumulative nursing GPAs, the likelihood of predicting success on NCLEX-RN increased at the end of each academic year.