A standards-based grading model to predict students' success in a first-year engineering course
Using predictive modeling methods, it is possible to identify at-risk students early in the semester and inform both the instructors and the students. While some universities have started to use standards-based grading, which has educational advantages over common score-based grading, at–risk prediction models have not been adapted to reap the benefits of standards-based grading. In this study, seven prediction models were compared to identify at-risk students in a course that used standards-based grading. When identifying at-risk students, it is important to minimize false negative (i.e., type II) errors while not increasing false positive (i.e., type I) errors significantly. To increase the generalizability of the models and accuracy of the predictions, feature selection methods were used to reduce the number of variables used in each model. The Naive Bayes Classifier and an Ensemble model using a combination of models (i.e., Support Vector Machine, K-Nearest Neighbors, and Naive Bayes Classifier) had the best results among the seven tested models. This study identified possible threshold concepts and learning objectives that are important to students’ success in the course, and learning objectives that are not correlated with student success in the course.
Madhavan, Purdue University.
Educational evaluation|Education|Engineering|Educational technology|Higher education
Off-Campus Purdue Users:
To access this dissertation, please log in to our