Abstract:
Higher education faces the challenge of high student attrition, which is especially disconcerting if associated with low participation rates, as is the case in South Africa. Recently, the use of learning analytics has increased, enabling institutions to make data-informed decisions to improve teaching, learning, and student success. The aim of this study is to improve student performance in the first year by probing students’ learning strategies and by examining the effectiveness and efficiency of a blended course design. To date, most student success studies has focused on the at-risk students. This study takes a difference approach to student success by focusing on a group of students termed the “murky middle’ (MM).
The first part of this study used demographic and prior learning data to define three subgroups of students, those at-risk of failing without substantial intervention, the MM, and the students that are likely to pass. Subsequent to the identification of the MM students, self-report learning strategies were analyzed to examine the strategies of successful students. The second part of this study focused on evaluation of the blended course design by investigating patterns of student engagement with the learning opportunities in the course. This was followed by an analysis of which of the learning opportunities contributed most to success of the subgroups of students.
The results of this study showed that it is possible to identify the MM using data available at the start of their academic career. The analysis of learning strategies provided useful information to guide the design of interventions aimed at improving the prospect of success for all students but specifically for the MM. Results from the analysis of the course design validated the use of blended learning, as we could show that face-to-face tutorial classes and online formative assessments contributed the most to student success. We also showed that the at-risk and MM students’ engagement with compulsory learning opportunities declined during the semester. The information generated in this study is useful for course design, classroom practice and student advising and could potentially contribute to student success.