Using AutoML to Analyze the Effect of Attendance and Seat Location on University Student Grades

A common claim is that class attendance and sitting at the front of a classroom may improve student grades. This study employs Automated Machine Learning (AutoML) to analyze this claim. The data used in this study came from an attendance-tracking system from a private university in Tennessee, USA. The correlation analysis in Microsoft Azure’s Machine Learning workspace was performed by training regression models. No correlation was found between student attendance and seat choice and final course grades. The K-means clustering algorithm was used to train clustering models in Microsoft Azure. At \(k=2\) clusters, a cluster with perfect attendance shows a higher average grade than a cluster with a late attendance average. Seat choice within the classroom does not prove important to the clustering models.