An Analytical Investigation of the Characteristics of the Dropout Students in Higher Education
[This Proceedings paper was revised and published in the 2018 issue of the journal Issues in Informing Science and Information Technology, Volume 15]
Student dropout in higher education institutions is a universal problem. This study identifies the characteristics of dropout. In addition, it develops a mathematical model to predict students who may dropout.
This study compared dropout rates of one and a half year of enrollment among Traditional Undergraduate Students. The sample includes 555 freshmen in a non-profit private university.
The study uses both descriptive statistics such as cross tabulation and a binary regression model to predict student dropout.
There are two major contributions for the paper, one it raises questions regarding causes of dropout thus, hopefully, it can result in better allocation of resources at higher education institutions. It also develops a predictive model that may be used in order to predict the probability of a student dropping out and take preventive actions.
Two major findings are that some of the resources designed to assist student are misallocated, and that the proposed model predicted with 66.6% accuracy students who will dropout.
The study recommends that institutions must create initiatives to assist freshmen students and have annual assessment to measure the success of the initiatives.
Secondly that analytical models be used to predicts dropout with fair accuracy.
The study should result in better allocation of resources in higher education institutions
The research will continue developing and testing the model using a wider sample and other institutions.