Educational Data Mining Utilization to Support the Admission Process in Higher Education Institutions: A Systematic Literature Review

Abstract

The behavior of undergraduates’ withdrawing from higher education institutions before completing their degrees, and the low percentage of on-time degree completion, are considered major challenges that higher education institutions face because of the inefficient admission process. This study aims to showcase the Educational Data Mining techniques that are used in predicting the academic performance of prospective students and the factors that influence students’ academic performance prior to admission. The study conducts a systematic literature review based on a developed systematic methodology consisting of the three recommended stages: planning, conducting, and reporting the review. The study selects 11 scientific papers published during 2018 and 2022. The results showed that 13 Educational Data Mining techniques are used to support the admission process in higher education institutions and predictive models that aid in choosing successful prospective students are developed using Machine Learning Algorithms. Moreover, the study found 35 factors that predict the performance of students prior to admission, categorized into three groups: (a) Socio-demographic Attributes Data (12 factors), (b) High School or Secondary Education Data (11 factors), and (c) Admission to Higher Education Data (12 factors). The study also revealed 10 recommendations and future studies for the admission process in higher education institutions.