The Turkish Online Journal of Distance Education
A Logistic Regression Analysis of Factors Affecting Enrollment Decisions of Prospective Students of Distance Education Programs in Anadolu University
Some prospective students placed by the Assessment, Selection and Placement Center (OSYM) to associate and undergraduate programs in the Faculty of Open Education of Anadolu University enroll in these programs, while others do not. A decision to enroll may be affected by the following variables: the prospective student’s gender, educational status, plans for retaking the university entrance exam, the number of times they took the entrance exam, marital status, employment status, household income, parents’ level of education, and age. The aim of this study is to use the above-mentioned variables to develop a model that will help classify prospective students placed on the distance education programs of Anadolu University into two groups, i.e. students who are more likely to enroll and those who are not; and to discuss the potential benefits of the model in the administrative processes. Students assigned by OSYM in the 2015-2016 academic year to the distance education programs of Anadolu University comprise the population of this study, which is a finite population consisting of 178.229 people. Data were collected via an online survey. The survey included items on demographics, enrollment status, and the above-mentioned variables. A total of 1.829 students completed the survey, of which 1.117 enrolled and 712 did not. Data were analyzed using the SPSS 22.0 software package. Binary logistic regression analysis was used to develop a model to classify the students as enrollees and non-enrollees. Education status, marital status, plans to retake the university entrance exam, the number of times the entrance exam was taken, employment status, and age were found to affect the prospective students’ decision to enroll.
KEYWORDS: Distance education, decision to enrollment, enrollment in distance education programs, logistic regression, binary logistic regression.
DOI : 10.17718/tojde.522459