Assessing Student’s At-Risk of Non-Completion in an Open and Distance Learning Course

Student attrition is a well-documented problem concerning open and distance learning (ODL) institutions. Evidence shows that the non-completion rate on an ODL course can be reduced if the at- risk students are followed up at an early stage. There is a problem in identifying such at-risk students as they may not be obvious at the beginning of their studies. Moreover, it would be difficult to collect at-risk evidence from students during the course presentation for personal assessment. This paper presents a Logistic Regression Model for assessing student s at-risk levels in an ODL course. The model is defined based on the findings in a previous study that ODL experience, academic background and assignment performance are three major variables relating to student attrition. Research results have shown that the model can successfully classify about 80% of students into completion or non-completion after the first assignment score is available. The simple choice of predictors and high classification rate make the model a practical instrument for an early identification of at-risk students.