Applying Machine Learning Techniques to Identify the Influential Factors of Students’ Abilities to Apply Statistics Mathematics and Engineering Knowledge

This Research to Practice Full Paper deals with the influential factors related to application of courses by the students. There are a large number of factors affecting students' abilities to apply knowledge of statistics, mathematics, and engineering. Even though the literature is replete with various methods to identify such factors, there is little research dedicated to exploring the influential factors using machine learning methods. Based on 200+ students' responses to senior exit survey, this study explored the influential factors using random forest approach. The variable of `time spent at CPP (Cal Poly Pomona) towards obtaining a degree' turned out to be the most important in influencing the students' abilities at concerned subjects. This was followed by the time of sitting for EIT (Engineer in training) exam and the engagement of students in different clubs. Clearly, as per the self-assessment of students, their time devoted at CPP and their self-confidence in undertaking the EIT significantly influence their perception of statistics, mathematics and engineering knowledge. This study demonstrated that the random forest approach may be beneficial in such analysis and may be combined with the conventional methods such as multinomial logit regression to devise more informed strategies by the faculty and administration to improve student outcomes.