Now that we’re four weeks into the semester, you may be wondering: how can you effectively help students who are struggling to keep up in your course? This issue can be broadly categorized into two major components: (1) identifying at-risk students (ideally early in the semester), and (2) intervening with those students to increase their academic performance. The former is a fairly established area of research [1,2,3,4], while the latter is an emerging topic. There is not much research, however, that attempts to combine the two and close the learning analytics loop [5].
Recent work conducted at the University of Notre Dame has been shown to boost student success, by using an integrated closed-loop learning analytics scheme that consists of multiple steps broken into three main phases, as follows: (1) capture gradebook data in real time, (2) analyze that data to identify a trigger for potentially non-thriving students, (3) intervene with those students to boost their performance [5]. The general idea of this technique is to close the loop by allowing later steps to inform earlier ones in real-time during a semester and iteratively year to year, thereby improving the course from data-driven insights.
This new technique effectively takes the burden of identifying and helping non-thriving students off individual instructors in larger, multi-section courses. Instead, the real-time data-driven analytics do much of the detail-oriented work at identifying non-thriving behavior, and templated bottom-up interventions are provided to students to boost their academic performance.
Even if you do not have the appropriate resources or class size to implement the above techniques, the general structure can be followed: (1) As a domain expert, do your best to decide on a gradebook trigger which you believe will identify the majority of non-thriving students (e.g., if a student has missed one or more of the first four assignments). (2) Intervene with those students early on in the semester, either directly (say, an email expressing concern and an invitation to meet with you or come to office hours) or via their academic advisor. The goal with this step is not to scare the student into doing better, but rather to get at the root of the issue (e.g., they just forgot, there was a grading mistake, they are struggling academically, they have home- or campus-related stress, etc.) This bottom-up method of intervening helps get at the root of the problem, allowing you to provide the student with the appropriate resources to boost their performance (e.g., strategies for managing their workload, scheduled meetings with the instructor or advisor, referrals to the rector, a care consultant, a peer advisor, a counselor, etc.) (3) Track those students throughout the semester and assess the effectiveness of your trigger and/or intervention, and update it as needed for future semesters.
Hopefully this strategy can help you minimize the damage for students who fall behind early in the semester this year and years to come.
References
[1] Arnold, K. E., and Pistilli M. D. (2012). “Course Signals at Purdue: Using Learning Analytics to Increase Student Success.” 2nd International Conference on Learning Analytics and Knowledge
[2] Moon, S. et al. (2013). “High-Impact Educational Practices as Promoting Student Retention and Success.” 9th Annual National Symposium C-IDEA
[3] Murtaugh, P. A., Burns, L. D., Schuster, J. (1999). “Predicting the Retention of the University Students.” Research in Higher Education, 40, 3
[4] Wolff, A. et al. (2013). “Improving Retention: Predicting At-Risk Students by Analyzing Clicking Behavior in a Virtual Learning Environment.” 3rd International Conference on Learning Analytics and Knowledge
[5] Syed, M. et al. (2019). “Integrated Closed-Loop Learning Analytics Scheme in a First Year Experience Course.” International Conference on Learning Analytics and Knowledge
Submitted by:
Carson Running
Ph.D. Candidate, Aerospace Engineering
University of Notre Dame
