Predicting Higher Education Throughput in South Africa Using a Tree-Based Ensemble Technique

TL;DR

This study employs gradient boosting and logistic regression to predict university throughput in South Africa, emphasizing socio-economic and field of study factors, and offers intervention strategies based on these insights.

Contribution

It introduces a predictive model combining machine learning techniques with socio-economic and academic data specific to South African higher education.

Findings

Socio-economic factors significantly influence throughput initially.

Field of study becomes a stronger predictor over time.

Recommendations include academic, psychosocial, and financial support interventions.

Abstract

We use gradient boosting machines and logistic regression to predict academic throughput at a South African university. The results highlight the significant influence of socio-economic factors and field of study as predictors of throughput. We further find that socio-economic factors become less of a predictor relative to the field of study as the time to completion increases. We provide recommendations on interventions to counteract the identified effects, which include academic, psychosocial and financial support.