Improving student engagement using learning analytics

The problem 

Student engagement is a central theme in many education, learning and teaching conferences. It is also a frequently discussed issue among the faculty in formal and informal meetings at universities. Student engagement is a necessary ingredient in a university’s mission to foster learning and growth. Achieving excellence in student engagement is a strategic objective for universities. In practice, however, achieving desired student engagement in and outside the classroom is a challenge for educators. In my academic practice, for example, student engagement gradually worsens during the semester. This could be due to my laissez-faire approach to teaching. Naturally, attendance in lectures and seminars visibly falls, email enquiries become irregular, and I could spend my office hours writing research papers. When nudged, students promise me that they would catch up on the missed content using the learning resources on the course page. That’s a reassuring response, but how would I know that students keep their promise? I do not like pestering student with emails. I could only hope that they engage with online resources. 

Potential solution 

Recently, I discovered that I could track learner engagement with online resources on course pages. I now use learner engagement markers on the course pages to track student engagement with online resources in real time. Engagement markers can be formative and summative assessments, forum posts, polls and interactive tasks among other activities. Student engagement with engagement markers feeds into activity completion reports on the course page. Combined with attendance records, activity completion reports provide a valuable source of learner analytics. Activity completion reports show not only student engagement with online resources, but also students’ learning behaviours and habits. Instructors can view the time it takes for students to complete tasks, time between attempts of formative assessments, sequence of engagement with resources and activities, participation in forum discussions, and activities students complete along with other analytics.  

Learner analytics help flag issues in student engagement and highlight where interventions are needed to help struggling students get on with the course. Last year, I used learning analytics to closely track student engagement in my first-year economics course. I identified clear patterns of student learning behaviour. Early in the term, enthusiastic students completed the tasks promptly and moved on. As we progressed into the third and fourth weeks, some students began delaying completing tasks and even stopped attending the lectures by mid-semester. I emailed and invited them back to the classroom and course page. Some students responded and a few of them were apologetic. They promised to catch up on the missed content ‘soon’. I was also able to identify more than dozen students who existed only in the list of students enrolled on the course but never actually started the course. These students failed to respond to my communications. I sought help from the student support service team who kindly reached out to the ‘phantom’ students and where necessary, helped them to resume or terminate their studies. 

There is more … 

Instructors can do more than just identifying and contacting disengaged students using learner analytics. I sometimes ‘confront’ my students with engagement data from the course page. My brief confrontations take place at the beginning of lectures. I present a summary of student engagement data from activity completion reports. This enables students assess where they stand relative to the whole class in terms of how they learn and engage with the resources. Some students may not know how to use course materials, which could be why their digital engagement trails indicate poor engagement. This provides opportunities to advise students on appropriate learning strategies they could adopt going forward and explain them how they could regulate their learning.  

Evidence from literature and conclusion 

The literature on learning analytics has reported various ways learning analytics can be used to identify at-risk students who maybe experiencing challenges (Herodotou, et al. 2020), to enhance learner performance and outcomes (Hellings and Haelermans, 2020; Foster and Francis, 2020), to improve teaching and learning (Pistilli and Heileman, 2017), to help student develop self-regulation (Siemens, 2012) and to predict student outcomes (Raj and Renumol, 2022). The evidence from the literature is overwhelmingly suggestive of many benefits and advantages of using learning analytics. Yet, there appears to be a slow adoption of learning analytics policies among higher education institutions. This is especially true for UK where Office for Students, a higher education regulator, rates universities for excellence in education in Teaching Excellence Framework (TEF). Adopting learning analytics policies and demonstrating a consistent approach to how a university monitors and responds to student engagement, experience and outcomes can be an important input into its TEF submission. 

Ravshonbek Otojanov is an Associate Professor in Economics. He is a Fellow of the Advance HE. Ravshonbek is passionate about using innovative methods to help students develop into self-regulated learners.

Reference  

Foster, C. & Francis, P. (2020). A systematic review on the deployment and effectiveness of data analytics in higher education to improve student outcomes. Assessment & Evaluation in Higher Education, 45(6), 822-841. 

Hellings, J. & Haelermans, C. (2020). The effect of providing learning analytics on student behaviour and performance in programming: a randomised controlled experiment. Higher Education, 1-18 

Herodotou, C., Naydenova, G., Boroowa, A., Gilmour, A., and Rienties, B. (2020) How Can Predictive Learning Analytics and Motivational Interventions Increase Student Retention and Enhance Administrative Support in Distance Education? Journal of Learning Analytics, Volume 7(2), 72-83.