Leveraging Data-Driven Decisions: A Predictive Model to Enhance Student Success in Higher Education

Abstract

Higher education institutions often face challenges in achieving key success metrics, such as on-time graduation rates, particularly when supporting students with diverse backgrounds, needs, and barriers. This study examines a detailed dataset containing student demographics and academic records from College of Business at a large public university spanning ten years. Our results highlight significant achievement gaps based on factors such as gender, age, race, family income, and parents’ education levels. These gaps persist even among students with similar academic histories. We develop prediction models that enable institutions to identify high-risk students early, facilitating timely intervention and providing a framework for comparing institutional efforts to improve student success. This study highlights how data-driven approaches can enhance institutional management by enabling proactive identification of challenges, optimizing resource allocation, and supporting strategies that improve student outcomes.