The purpose of this study was threefold: (1) to investigate variables associated with learning and performance as measured by the National Council Licensure Examination for Registered Nurses (NCLEX-RN), (2) to validate the predictive value of the Assessment Technologies Institute (ATI) achievement exit exam, and (3) to provide a model that could be used to predict performance on the NCLEX-RN, with implications for admission and curriculum development. The study was based on school learning theory, which implies that acquisition in school learning is a function of aptitude (preadmission measures), opportunity to learn (program length), and quality of instruction (program measures). Participants were 298 graduates of an associate degree nursing program in southeastern U.S. Of the 298 graduates, 142 were Hispanic, 87 Black, non-Hispanic, 54 White, non-Hispanic, and 15 others. The graduates took the NCLEX-RN for the first time (2003-2005). This study was a predictive, correlational design that relied upon retrospective data. Point biserial correlations and chi-square analyses were used to investigate relationships between 19 selected predictor variables and a dichotomous criterion variable, NCLEX-RN. Correlation and chi-square findings indicated men did better than women on the NCLEX-RN; Blacks had the highest failure rates, followed by Hispanics; older students were more likely to pass the exam than younger students; and students who passed the exam started and completed the nursing program with a higher grade point average than those who failed the exam. (Contains 2 tables and 3 figures.)