Abstract:

This paper reports on the analysis of Moodle (an open source online learning platform) data from first-year computing students (n=200) participating in a Maths module. 2020/21 data revealed about 266,000 interactions by the student group on the module. These interactions were analysed and showed a strong correlation with student performance.

This paper shows how machine learning models were explored as a means of clustering students by their level of engagement. The risk of disengagement is analysed in relation to interactions and academic performance. This presentation begins with a brief introduction of learning technologies which we currently use to evaluate student engagement.

Data is extracted from Moodle into a database where it is cleaned put into structured format. The analysis of this data identified engagement trends in the student data which was explored by day, time and activity type. By comparing the engagement analysis and current student data we are able to create classification thresholds. This allows us to identify student groups and create targeted and personalized interventions.

In this study, students are partners in the design and feedback was collected from their experiences. The students are also introduced to explainable machine learning technologies which are used to identify these student groups as part of our analysis of the data. This provides a constructive alignment between our research aims, achievement of programme learning outcomes (as computing students) and enhancement of the student experience.

Keywords:

Moodle.