Finding Bottlenecks: Predicting Student Attrition with Unsupervised Classifier

TL;DR

This paper employs unsupervised clustering to predict student graduation outcomes and identify bottleneck courses, aiding early intervention in higher education to improve graduation rates.

Contribution

It introduces an unsupervised approach to predict student attrition and detect bottleneck courses using minimal course data, enhancing early warning systems.

Findings

Clusters effectively predict graduation status

Bottleneck courses identified within clusters

Method applicable to multiple departments

Abstract

With pressure to increase graduation rates and reduce time to degree in higher education, it is important to identify at-risk students early. Automated early warning systems are therefore highly desirable. In this paper, we use unsupervised clustering techniques to predict the graduation status of declared majors in five departments at California State University Northridge (CSUN), based on a minimal number of lower division courses in each major. In addition, we use the detected clusters to identify hidden bottleneck courses.