|
|
| |
| Anaylsis of Students' Enrollment and Graduation Rates in Gulu University in Uganda. A Case Study of Gulu University
Download PDF
|
Nabaasa Mark , , Mr., Dr Ariyo Gracious Kazaara , Tukamuhebwa Deus , Dr Arinaitwe Julius , Kamugisha Nelson , Mutesi Catherine , Friday Christopher , Isabirye Benefansi , Kimaku Alex
|
Abstract:
This study's main goal was to examine the enrollment and graduation rates of students at Uganda's Gulu University. A quantitative research methodology was used in this investigation. All students enrolled at Gulu University from 2002/2003 to 2019/2020 made up the study's population. The document from the admissions office that contained the study's secondary data served as the primary tool for data collection. Data needed to manage daily academic practices was obtained from academic management systems. These systems rely on transactional databases, which are often updated and as a result lack the ability to archive histories of data instances, making them unsuitable for study on enrollment prediction and graduation prediction. based on the probability of enrollment based on the fact that departments in charge of recruiting students are left to guess. In the majority of circumstances, this is inappropriate because it causes budget shortages and resource strain. Universities experience difficulty forecasting how many students will graduate from the school of interest in addition to speculating on those who are likely to enroll. There are several reasons why a student doesn't finish their degree of study, including their financial situation and family history. As a result, models for forecasting enrolment and graduation are put forth to help in making projections about how many admitted students will enroll and how many will graduate. The proposed system first works by converting the student data into time series data by adding a time variant to it. It then runs the data through Stata software and performs regression analysis using the appropriate econometric model. The inputs to the proposed prediction system were sourced from student data stored in a worksheet regarding student details. The enrollment and graduation of students will then be forecasted using the set of associated. With a 59% accuracy rate, the model's accuracy was remarkably high. The dataset employed, which was noisy because it was pulled from genuine student transactional databases, resulted in the proposed model having a slightly poorer accuracy in comparison to some of the papers assessed.
|
|
|
| | |