Identifying at-Risk Students with Data Analytics and Machine Learning: Insights from a Systematic Review

Abstract

The purpose of this systematic review was to evaluate how data analytics and machine learning methods are used in Higher Education Institutions (HEIs) to forecast student achievement and identify students who may face academic difficulties. Although these methods have been studied before, there has been inadequate analysis of their strengths and limitations. Furthermore, considering the worldwide commitment to accomplishing the Sustainable Development Goals (SDGs), especially SDG 4, which prioritises inclusive and equitable quality education, tackling problems like student attrition and unequal support is essential. To fill this research gap, a systematic literature review including four major databases was carried out in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria. Studies that used data analytics and machine learning methods to detect at-risk students or identify student achievement in higher education settings were included. To guarantee the rigour of the included studies, a quality assessment utilising the Critical Appraisal Skills Programme (CASP) Checklist was used. The review revealed a wide variety of strategies, including more sophisticated approaches like ensemble methods and neural networks, as well as more conventional ones like logistic regression and support vector machines. These methods were used on a variety of data sources, such as survey data, administrative data, and data from learning management systems. The results demonstrated how data analytics and machine learning may transform higher education by making it easier to identify at-risk students early on, customising support services to meet their requirements, and allocating resources as efficiently as possible. HEIs can therefore use these technologies to make data-driven decisions that will improve teaching and learning methods and increase student achievement in sustainable ways.