Restructuring Curricular Patterns Using Bayesian Networks

Recent studies proved the existence of a relationship between the complexity of university curricula and graduation rates. As a result, extensive efforts have been done in an attempt to restructure curricula in order to improve graduation rates. In this paper, we propose a new model for evaluating and quantifying the impact of restructuring curricula on graduation rates using a Bayesian network framework. We validate our model by analyzing a common curricular pattern found in most of the engineering programs. We demonstrate its usefulness using actual data for students at the University of New Mexico. We also extend this model to include a helpful tool that can be used to predict student performance. The advantage of our work is characterized by its data-driven nature which makes it more reliable than other proposed models.