Work in progress - a decision tree approach to predicting student performance in a high-enrollment, high-impact, and core engineering course

Abstract

This paper aims at developing a decision tree model to predict student performance in engineering dynamics - a high-enrollment, high-impact, and core engineering course. This study is innovative because no prior literature exists on the same topic. Three research contributions are made: 1) Nine “if-then” decision rules were generated to predict student performance in engineering dynamics. 2) It is revealed that a student's score in Statics and cumulative GPA play a significant role in governing student performance in engineering dynamics. 3) It is revealed that the decision tree predictions are more accurate than the predictions from the traditional multivariate linear regression technique.