Joint Model for Study Programme Completion and Performance of Students in Distance Education in Sri Lanka

Abstract

The completion rates of study programmes by students under the open and distance learning (ODL) mode is important to analyze as the low completion rates or dropouts in ODL mode is an area which has taken much attention over the past years. The time to completion has a relationship with the number of pass grades in the first attempt at final examinations, as if the students cannot go through the exams by first attempt, it will take more time to complete the program. On the other hand, the number of pass grades in the first attempt indicates the performance of the students and the poorly performing students might dropout from programs. When the completion time and the number of pass grades in the first attempt are correlated, it is better to model the two variables together, creating a joint/bivariate model. Thus, this study investigates the impact of individual level factors of the students on the completion status of a study program and number of pass grades in the first attempt at final examinations through a joint/bivariate model of time and count. As the students are clustered/ grouped within the study centers they are attached to, the analysis was undertaken by adjusting for the cluster effect as well. The individual level student factors that are of interest in this study are gender, age, employment status, years of employment experience, civil status, medium of study and attached study center. The records of 1657 students who have registered for Bachelors of Management Studies study programme in 2010 and 2011 at The Open University of Sri Lanka were considered in this study. The findings revealed the significant factors that affect the two variables under consideration and it also suggests that the joint model performs better than models taking the two response variables separately.