How to Boost Student Engagement in Online Courses

Engagement in online learning remains a curious paradox. On the one hand, demand for online education platforms has skyrocketed recently during the pandemic. Coursera, the leading provider of massive open online courses (MOOCs), reported that during 2020, its enrollments topped 59 million students, soaring 248 percent over the previous year.

But on the other hand, engagement in these types of courses remains abysmal, with one team of researchers having estimated completion rates of around 3 percent. These rates are highly significant to society because low persistence in completing educational programs affords a major barrier to social mobility.

Ever since the introduction of MOOCs in 2010, the online education industry has unsuccessfully struggled to hike these dreadful completion rates. But a new study by researchers at the University of Illinois Gies College of Business has, for the first time, identified an effective way to boost engagement in these kinds of classes.

It appears that nobody has tried their method before—and that’s most likely because the technique seems so counterintuitive. And although their approach appears deceptively simple, it’s poised to spur engagement in all sorts of online learning environments and could have substantial implications for the marketing and financial performance of players in the online education industry.

To fully appreciate the significance of the Gies team’s research, it’s first helpful to briefly review some of the challenges that faced massive online course providers like Coursera, Udacity, and edX during the years before the pandemic.

In short, MOOCs offered tremendous promise because “they were supposed to be free and open to everyone with online access, bringing the best possible content from the best schools and best professors to everyone,” according to Forbes higher education contributor Derek Newton.

In other words, no longer would barriers to entry like high tuition, limited enrollment capacity, or long distances away from college campuses block aspirants from access to quality higher education. Many of us remember that when this word about the potential of MOOCs first leaked in 2012, for several months, the press couldn’t stop talking about how society was about to be radically transformed by this new innovation in higher education.

In fact, much in the way that Time Magazine had named the personal computer its Man of the Year for 1983, in a famous feature article, the New York Times even proclaimed 2012 to be “The Year of the MOOC.”

What could go wrong? “The hype was exceeded only by the failure,” says Newton. It turns out that a series of reports soon revealed that the vast majority of students in these courses did not complete them. For example, one particularly disconcerting study of highly-motivated undergraduates enrolled in MOOC courses at Harvard, Stanford, and MIT demonstrated completion rates among those who intended to complete the courses eventually turned out to be less than 20 percent.

What’s more, it soon became apparent that many MOOC students worldwide weren’t taking these classes to earn any employment-related credential because they had already earned undergraduate degrees.

In 2019 the definitive paper on this topic appeared in the prestigious journal Science entitled “The MOOC Pivot: From Teaching the World to Online Professional Degrees.” In the report, MIT Media Lab professors Dr. Justin Reich and Dr. José A. Ruipérez-Valiente not only announced the 3 percent completion rate we cited above but also announced additional findings that were troubling for the industry.

Their first finding was that despite six years of investments in developing virtual courses and online learning research, completion rates did not increase but had actually plummeted by 50 percent overall. More specifically, the team also found that the completion rates even dropped by about 18 percent—from 56 percent in 2016 to 46 percent before the study’s release—among the “verified” students. Those were the students who had paid tuition to take the class to earn a credential upon completion, like a certificate that they could show employers. Experts considered this market segment critical to the growth and development of the online education industry because they were a potentially lucrative source of earnings compared with other revenue options like advertising.

Things got so bad that in mid-2020 that an all-star team of nine professors from universities that included Harvard, Stanford, Carnegie Mellon, Cornell, and the University of Toronto had revealed research into a series of behavioral “interventions” intended to prompt students to complete MOOC course assignments. The researchers had good reason to believe that three of these interventions would work because they demonstrated statistically significant effectiveness in other behavioral contexts.

For example, because plan-making interventions had previously been proven effective at encouraging behaviors like voting and keeping doctors’ appointments, one of the interventions prompted students to plan specific details about when and how they would finish assignments during their course’s term.

A variant of this intervention the researchers tried even asked students to envision some of the barriers to completing their course assignments, and then plan some ways in which they might overcome those obstacles.

Moreover, a third “social accountability” intervention asked the students to enlist the support of friends and family, who would ask the students regularly to share updates on their progress in completing course requirements.

Unfortunately, front-loading these interventions at the beginning of MOOC classes resulted in increased completion rates for some of the students’ first assignments, but failed to boost their completion rates for the entire courses. These results led the researchers to conclude that the low course completion rates posed an intractable problem.

In a June 2020 paper entitled “Scaling Up Behavioral Science Interventions in Online Education,” the results were published in PNAS, the Proceedings of the National Academy of Sciences. And almost immediately, the fallout within the industry seemed devastating. A few weeks later, EdSurge reported that the MIT author involved with both of the studies above, Dr. Justin Reich, told the outlet that the results published by PNAS felt “depressing, frustrating, and deflating.” Another faculty member involved with the PNAS study, Cornell professor, Rene Kizilcec, said the results felt “humbling and disheartening.”

Meanwhile, in Champaign, Illinois, a business administration professor had been wondering about a new idea for the first strategy that might actually boost MOOC student engagement. Dr. Unnati Narang had started working on this problem about four years earlier while a Columbia Business School graduate research assistant and PhD candidate at Texas A&M University.

But although she didn’t recognize it at the time, her tentative hypothesis would never work out. As she explained to Helix Education Vice President Eric Olsen:

I was starting to think about what can we do to improve learner engagement in these massive online courses where most people feel lost or anonymous and don’t really have a community that an in-person student has access to. So I thought about these traditional strategies that work in physical classrooms where instructors will often invite students on day one of class to introduce themselves, to share something about their life with their peers, and I wondered, “How effective will these be in online environments?”

And going in, my hypothesis was that perhaps these would increase some sense of belongingness or community. So this was going to be a paper about how identity sharing, sharing something about who you are, might be very helpful.

With help from 2,000 randomized research subjects enrolled within two online cohorts of a 2019 Geis marketing course carried by Coursera, Dr. Narang then designed an experiment to test two conditions. In both conditions, all students watched a 30 second welcome video from their professor during the first day of the course.

However, each cohort viewed a different version of that video clip—and the modification was extremely subtle. The altered welcome videos comprised the only difference in any of the course materials provided to each cohort during the entire semester, which included 41 required videos on the syllabus. However, this exercise was not required to complete the course; students could opt out if they didn’t want to participate.

Here’s how the two videos differed:

• The “identity” cohort watched a video where the professor welcomed the students to the course, and then invited them to share something with their community of peer learners about who they are.
• The “ideas” cohort watched a similar video where the professor welcomed the students to the course. However, then the professor invited them to share something with their community of peer learners about how the digital world and digital marketing are shaping firms and consumers.

Along with each video, the students received a prompt in their chat window next to the video where they could post their responses. The two video versions comprised the only difference in the course materials provided to each cohort during the four-week course term.

Then the researchers measured the students’ engagement with all the videos and assessments, such as quizzes, during the course duration. It turned out that the results were dramatically different between the two cohorts, and startling to the investigators because the outcomes weren’t what the Gies research team had expected.

In short, the students asked to chat about the effects of the digital world and digital marketing completed 30 percent more of the course’s videos and assessments than the students asked to chat about who they are.

By contrast, Dr. Narang’s team found that asking students to share their identities failed to produce any effect, contrary to the professors’ expectations. These are counterintuitive results that run completely contrary to the routine practice of many professors when they kick off new classes in person by asking students to go around the room, introduce themselves, and share a little about their personal identities. Moreover, in some disciplines like speech communication, it’s challenging to find any in-person class that doesn’t begin this way.

The researchers acknowledge that encouraging learners to disclose background information about their identities might be effective in traditional, face-to-face classrooms at building engagement. However, as they explain in their 2022 paper published by the Journal of Marketing Research entitled “The ‘Idea Advantage:’ How Content Sharing Strategies Impact Engagement in Online Learning Platforms,” they conclude that within large, impersonal online classrooms, the efficacy of this practice is limited. So why does this effect occur?

When they analyzed the responses typed into chat windows by the students asked to share their ideas, the researchers found that their language was typically more extensive, detailed, and complex.

The professors believe that when prompted to share ideas, online learners put more thought into their posts than those only invited to share information about their identities. According to Dr. Narang’s Gies collaborator and John M. Jones Professor of Marketing Dr. Aric Rindfleisch, “We discovered that learners who are asked to share ideas elaborate more and share longer responses. This elaboration played a key role in their subsequent engagement.”

Finally, what does this research mean for professionals like educators, online course designers, and edupreneurs? The Gies team suggests their research offers three principal implications.

First, professionals like these should focus without delay on encouraging idea-sharing to drive up engagement. Although this technique is easy and inexpensive to implement, its potential benefits are compelling—and potentially, quite substantial.

The professors also suggest that because elaboration appears to explain increased engagement, platform designers and course providers can apply simple design interventions to test the correlations with engagement of varying levels of elaboration. For example, they might experiment with controlling elaboration by varying the limits they set on how much prompt-response language students can type into chat boxes.

Additionally, the researchers’ findings suggest that this simple and inexpensive course intervention should result in a substantial increase in “watch time” video consumption per student, an important engagement metric on video platforms like YouTube.

Specifically, the professors emphasize that “in many marketing contexts, such as retail stores and video games, time spent engaging in an activity is positively associated with increased revenue.” Therefore—because of this project that may eventually be considered a landmark study—the “idea advantage” found by the professors at Illinois Gies could have tremendous implications for the marketing and financial performance results of players within the online education industry.