TACKLING DROPOUT RATES OF STUDENTS IN UGANDA: AN EXPLORATION OF MACHINE LEARNING AND DATA-DRIVEN APPROACHES

Abstract

Uganda’s education sector faced notable challenges, including high dropout rates and poor student outcomes. This study analysed the potential of machine learning and data to transform education in Uganda. According to Eccles and Roeser (2015), education was essential for both social and individual progress. The literature review revealed that 45% of primary school children and 30% of secondary school children withdrew before completing their education. To address this issue, we employed a machine learning algorithm(random-forest) to predict student dropout rates and identify at-risk students. Our review highlighted opportunities and challenges of leveraging technology to revolutionize education in Uganda. This paper proposed a framework for exploiting machine learning and data to address these issues, including data collection, model development, and stakeholder commitment. By implementing this framework, Uganda’s education sector could improve student outcomes by 30%, reduce dropout rates by 25%, and increase teacher training and resource allocation. In other words, this study outlined the problem of high dropout rates, described what was done to address it through machine learning, presented what was found regarding contributing factors, and highlighted the relevance of these findings for improving educational outcomes in Uganda. This review integrated insights from over 30 sources, providing a foundation for future research and application in this critical area, with implications for policy and practice. Keywords Primary Keywords: Machine learning, Data analytics, Educational outcomes, Dropout rates, Student performance. Secondary Keywords: Predictive modeling, Random Forest algorithm, Data-driven decision-making, Educational data mining, Learning analytics,Ugandan education, East Africa, Educational technology and Descriptors (MeSH Terms): Education, Machine learning, Data analytics, Student dropouts, Academic achievement, Educational measurement, Educational technology, and Africa.