Monitoring student engagement in distance learning: an action research to examine and measure the behavioral engagement of students

Educators and policymakers have widely researched student engagement, an essential yet complex construct strongly associated with a student’s academic performance. A plethora of literature is available on student engagement with considerable agreement on the multi-dimensional nature of student engagement, broadly categorized as affective, behavioral, and cognitive engagement (Appleton et al., 2008; Jimerson et al., 2003). Though affective and cognitive indicators are equally important dimensions, the focus of this paper is behavioral engagement which includes observable student actions such as participation in class and extra-curricular activities, attendance, and work habits (Fredricks et al., 2004).  

Along with exploring the phenomenon of 'behavioral engagement,' this study applies Learning Analytics to a set of student behavioral engagement and academic performance data. Literature shows that Learning Analytics (LA) data can help teachers predict learners' performance and recognize behavioral patterns (Anderson, Rourke, Garrison, & Archer, 2001). Studies on the use of LA are mostly based on deploying Virtual Learning Environment data like Moodle to observe learners and engage them in online learning, visualizing and clustering learner groups.

When observing student behavioral engagement in online learning classrooms, we have realized how some students lose interest over time. The students’ attention and participation drop as time passes. Due to the loss of motivation, they tend to miss more classes and fail to keep up with coursework. This makes it difficult for academic staff to understand/ the students' learning behaviors, whereby they can either academically challenge the more able students or provide timely support to weak students, where needed.

This manual approach of measuring students' behavioral engagement through traditional observation has shown its limits in the online learning environment. In this study, we propose the use of Insights generated by MS Teams as an automated technique for (1) monitoring digital engagement of students and (2) for applying learning analytics to guide academics to understand their students’ learning behavior, classify them into learning groups (also referred to as behavioral learner archetypes) and take proactive measures and timely action across for each group across their learning modules and extra-curricular activities.

The study aims to introduce a proof-of-concept that applies Learning Analytics to observe and measure students’ behavioral engagement classifying them into learner groups which can be further used to guide and support students in their academic journey.