The ability of universities and colleges to predict the success of admitted students continues to be a key concern of higher education officials. Apart from a desire to see students have successful academic careers, there is also the fiscal reality of greater tuition revenues providing needed support for university budgets. Using administrative data, this article introduces a relatively new empirical approach to estimating the determinants of student success in post-secondary institutions. Using Ordered Logit and Generalized Ordered Logit estimators, we estimate the role a number of key factors play in influencing student success. As a test of robustness we also use the "feologit" estimator which is designed to fit fixed effects ordered logit models. An important feature of our approach to determining student success is that it can be conducted using readily available administrative data. While the results are based on one institution, we feel there are useful lessons for other institutions facing similar student performance issues.