Learning analytics can support student success

April 12, 2020

The rush to push in-class learning into an online experience may be leaving out one crucial piece to helping students succeed – incentivising participation.

Many colleges and universities are facing the policy question of whether to make attendance of synchronous online sessions optional.

“There is a tendency towards dropping the participation or attendance grade when moving online, and that’s understandable given the situation and it’s a policy consideration being made from a place of compassion,” says Professor Zachary Pardos, a faculty member in the Graduate School of Education, and School of Information.

“My research has suggested there are reasons, however, to do everything you can to keep all students engaged and on-track, especially those who are most tempted to tune-out,” he said.

Beyond institutions offering students support to improve their connectivity, such as laptops, data sim cards, or mobile hotspots, adaptive learning technologies can help faculty track students’ engagement.

Research Pardos conducted with an Arizona State University colleague investigated the different pathways students took through a for-degree-credit, asynchronous online course.

Using a student online course pathway visualization tool developed in the research, the visualization told a story of two orientations toward online learning:

  • a disciplined orientation, in which students followed the prescribed path, interacting with the preparation and practice materials before the quizzes;
  • a less-disciplined orientation, in which students worked backward, first going directly to the quizzes and then searching for the answers within the materials.

The latter strategy was dominant among students who failed the class. The instructor made two changes in the next offering of the course:

  1. Incentive. Students who finished the preparatory materials early earned a small amount of extra credit; and
  2. Targeted communication. The faculty member had e-mail communications with students who did not complete those materials in the first lesson, emphasizing how important completing the preparatory materials was to the academic success of past students.

By paying attention to student online participation, and then connecting with students accordingly, the ASU faculty member saw improvements in course outcomes.

Pardos acknowledged that the impetus for the current crush of online learning has put many students and faculty in uncharted territory.

“We also have to be mindful of the different access people have, whether that is access to data; computer systems; and quiet environments. With everything going on -- and there’s a lot of commotion now – many students, and faculty as well, have responsibilities you didn’t have otherwise.”

He notes that even if online learning of all classes lasts for months, these types of learning analytics can help improve student engagement and success far beyond the current shelter-in-place.

“Campuses have already been progressing towards hybrid course approaches, where lectures are placed online and in-person attendance is optional. The ability to be in-touch with learner engagement and respond with a level of personalization is a technical affordance that will help in this scenario and others long after this forced-online moment has passed," he said.

This ability to communicate with students based on their online engagement in edX, an online course platform provider, is already a research project underway.

The tools enabling this research are being contributed by the extended Learning Analytics academic community to help better understand learning in online environments and make the story of online education one that institutions, instructors, and their students want to tell in the future.