Analysing Student VLE Behaviour Intensity and Performance

Abstract

Almost all higher educational institutions use Virtual Learning Environments (VLE) for the delivery of educational content to the students. Those systems collect information about student behaviour, and university can take advantage of analysing such data to model and predict student outcomes. Our work aims at discovering whether there exists a direct connection between the intensity of VLE behaviour represented as recorded student activities and their study outcomes and analyse how intense this connection is. For that purpose, we employed the clustering method to divide students into so-called VLE intensity groups and compared formed groups (clusters) with the student outcomes in the course. Our analysis has been performed using Open University Learning Analytics dataset (OULAD).