Home // IARIA Congress 2022, The 2022 IARIA Annual Congress on Frontiers in Science, Technology, Services, and Applications // View article

Early Risk Detection of Bachelor's Student Withdrawal or Long-Term Retention

Authors:
Isaac Caicedo-Castro
Oswaldo Vélez-Langs
Mario Macea-Anaya
Samir Castaño-Rivera
Rubby Castro-Púche

Keywords: machine learning; educational data mining; classification algorithm; University admission test; student withdrawal; student long-term retention.

Abstract:
In this research, we study the problem of forecasting recently admitted students at risk of withdrawing from the university or being long-term retained in a bachelor's program. We conduct research to study the case of students enrolled in courses up to the ninth semester, in the Department of Systems Engineering at the University of Córdoba in Colombia. At most universities throughout Colombia, including the University of Córdoba, the standardized and official admission test Saber 11 has been adopted for bachelor's program admissions. Therefore, we address the following research question: Might the admission test Saber 11 be used to forecast if the recently admitted student will be at either withdrawal or long-term retention risk, in the foreseeable future, before starting the first semester? We are motivated to solve the previously mentioned question because once the admitted students at risk have been identified, the University might make choices to help such students. To this end, we collected a dataset from 86 surveyed students. Although the original dataset has 86 records, after cleaning the dataset, and removing records with missing or inconsistent values, the final version of the dataset contains records of 47 students. According to the results of this research, given the student's test admission outcomes, machine learning algorithms learn regular patterns for forecasting if a recently admitted student is at withdrawal or long-term retention risk with a mean accuracy of about 72.5% (i.e., mean error of approximately 27.5%).

Pages: 76 to 84

Copyright: Copyright (c) IARIA, 2022

Publication date: July 24, 2022

Published in: conference

ISBN: 978-1-68558-017-9

Location: Nice, France

Dates: from July 24, 2022 to July 28, 2022