Grade Prediction Model Using Regression Analysis: an Implementation in Engineering Mechanics

Abstract

The ability to predict student performance and identify those who are "academically at-risk" creates opportunities to improve educational outcomes and enhance retention. This research aimed at developing and implementing a predictive model to generate an alert of potential student failure in engineering mechanics courses. A prediction model is developed based on regression analysis and is used to predict students' final course grades based on their activities early in the semester. In general, a predictive model needs a refined and standardized data set in order to achieve meaningful results. The data set used for the grade prediction includes students' performance in the formative and summative assessments during the first four weeks of the semester as well as their class participation and homework completion. The training data was collected in two semesters where the course was offered in-person in large sections with more than 100 students per section; then the model was tested and validated in small and large sections in four consecutive semesters, two of which involved online teaching mode. The analysis of all collected data shows that the accuracy of the model is 88.0%. In particular, the model successfully predicted students who were in danger of failure. The accuracy of prediction was acceptable in both in-person and online teaching modes. The ultimate goal of developing and implementing the model is utilizing a reliable tool that is capable of predicting student performance with an acceptable accuracy to identify "academically at-risk" students. The intervention process is beyond the scope of this paper and will be discussed in a separate publication. The results of this study will provide all parties involved in student success (e.g. instructors, advisors, counselors, etc.) with a tool to identify at-risk students early in the semester enabling proactive intervention approaches before it is too late, to increase student success and retention.