A Learning Analytics-informed Activity to Improve Student Performance in a First Year Physiology Course

Abstract

Learning Analytics (LA) can be employed to identify course-specific factors that hinder student course (outcome) performance, which can be subsequently rectified using targeted interventions. Supplementing interventions with predictive modelling also permits the identification of students who are at-risk of failing the course and encourages their participation. LA findings suggested that a targeted intervention for our course should focus on improving student short answer question (SAQ) performance, which we attempted to achieve by improving their understanding of features pertaining to various SAQ answer standards and how to achieve them using examples of varying scores. Every student was invited to the intervention via a course-wide announcement through the course learning management system. At-risk students identified using predictive models were given an additional invitation in the form of a personalised email. Results suggest that intervention improved student understanding of SAQ performance criteria. The intervention also enhanced student end-of-semester SAQ performance by 12% and 11% for at-risk and no-risk students respectively. Course failure rate was also lower by 26% and 9% among at-risk and no-risk intervention participants. Student perception of the intervention was also positive where an overwhelming majority of participants (96%) found the interventional activity to be useful for their learning and exam preparations.