Predictive Analysis of Higher Education Graduation and Retention in Saudi Arabia using Multinomial Logistic Regression
Ahmed Bagabir1, Mohammad Zaino2, Ahmed Abutaleb3, Ahmed Fagehi4
1Ahmed Bagabir*, College of Engineering, Jazan University, Jazan, Saudi Arabia.
2Mohammad Zaino, College of Applied Medical Sciences, Jazan University, Jazan, Saudi Arabia.
3Ahmed Abutaleb, College of Engineering, Jazan University, Jazan, Saudi Arabia.
4Ahmed Fagehi, College of Education, Jazan University, Jazan, Saudi Arabia.
Manuscript received on November 01, 2021. | Revised Manuscript received on November 16, 2021. | Manuscript published on November 30, 2021. | PP: 1-8 | Volume-3, Issue-6, November 2021. | Retrieval Number: 100.1/ijbsac.F0466113621 | DOI: 10.35940/ijbsac.F0466.113621
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: It is suggested that this study contributes by establishing a robust methodology for analyzing the longitudinal outcomes of higher education. The current research uses multinomial logistic regression. To the knowledge of the authors, this is the first logistic regression analysis performed at Saudi higher education institutions. The study can help decision-makers take action to improve the academic performance of at-risk students. The analyses are based on enrollment and completion data of 5,203 undergraduate students in the colleges of engineering and medicine. The observation period was extended for ten academic years from 2010 to 2020. Four outcomes were identified for students: (i) degree completion on time, (ii) degree completion with delay, (iii) dropout, and (iv) still enrolled in programs. The objectives are twofold: (i) to study the present situation by measuring graduation and retention rates with benchmarking, and (ii) to determine the effect of twelve continuous and dummy predictors (covariates) on outcomes. The present results show that the pre-admission covariates slightly affect performance in higher education programs. The results indicate that the most important indicator of graduation is the student’s achievement in the first year of the program. Finally, it is highly suggested that initiatives be taken to increase graduation and retention rates and to review the admissions policy currently in place.
Keywords: Admission policy, cohort analysis, education, logistic regression, statistics, university outcome.