Computing Profession

Moocs Evolve; Version 2.0 Aims to Retain More Students

A representation of an electronic textbook.
The evolution of Massive Open Online Courses (MOOCs) is aimed at increasing student retention rates.

Critics of Massive Open Online Courses (MOOCs) frequently point out that student retention rates – a common method of grading the success of a MOOC — most often hover at just around 4-5% of those participating in the free Web classes.

However, one MOOC — Udacity — is now claiming that 83% of its students completed some of its courses, which is more than 16 times higher than the retention rates of its two biggest competitors, Coursera and edX. What’s going on here?

The answer, say the founders of all three MOOCs, is that some online courses have evolved to a sort of "MOOC 2.0" in an effort to retain students, and that comparing the retention rates of classic-style MOOCs to those of the updated versions just isn’t fair.

For instance, this spring, Udacity conducted a pilot program in which it altered three principals of the classic MOOC:

  • It charged a small fee for the courses.
  • It offered a certificate of value (in this case, for academic credit).
  • It gave students guidance and services, including access to a help line.

"You can argue about whether our MOOC 2.0 classes are MOOCs or not," says Udacity CEO and co-founder Sebastian Thrun. "Certainly they aren’t as ‘massive’ anymore because, when you ask for money, you immediately cut down on the number of students, but nevertheless, the evolved courses gave us a lot of data that shows that these changes made a huge difference in the success rate of the classes."

Udacity still offers "classic" MOOCs and for those, the retention rate remains at 5-7%, which is why Thrun says his "heart is with the better version, even though it costs money and is much more service-involved. We’ve done a lot of experiments to see if we could change the 5% finishing rate to a larger number, and almost nothing has worked. The only thing that has really worked is adding really strong services."

Anant Agarwal sees no shame in the average 7% successful pass rates of MOOCs at edX, where he is president.

"Some may infer from the numbers that the MOOC did not serve its purpose," he says. "However, if the student is what I call ‘a tourist’ – meaning they are only interested in coming in, watching a few videos, and leaving – then the MOOC served that student’s purpose. They never had any intention of taking an exam or completing the course; they just wanted to learn something."

Agarwal adds that it would be easy to boost those numbers by creating an admissions test for edX MOOCs, "but then suddenly we’re not democratizing education," he says. That is why he chooses to "keep the funnel open" and allow anyone to take a course.

However, Agarwal reports that he is trying four new approaches in order to improve student outcomes:

  • Self-assessment tests: Many students don’t have the background to take a particular MOOC. edX encourages but doesn’t mandate a test (that is for the student’s edification only) that can help them decide whether to take the course now or perhaps wait until they have a better understanding of the subject.
  • Buddy learning: edX suggests students form "meet-up groups" and work with a buddy. Recent research shows that students who work in groups are more successful.
  • Instructional tutorials: When professors determine in which areas their students are having difficulties, rather than leave them on their own to be frustrated, they are urged to create short videos to explain some of the more difficult concepts.
  • Blended learning: Just as Udacity has its evolved version of a MOOC, edX has developed SPOCs – or Small Private Online Courses – in which instructors teach students along with the online content. For example, when San Jose State University experimented with using a "blended learning" model in its classroom using an edX course, pass rates in that Circuits and Electronics course soared to approximately 90%. Unlike MOOC learners, SJSU students paid tuition, received campus credit for the course, and worked in groups and interacted with an instructor in the classroom.

In a recent article co-authored by Daphne Koller, the Coursera co-founder explains that most people who sign up for a MOOC never actually intend to complete it, but rather just want to explore it and perhaps watch a few videos. These so-called "browsers" or "explorers," she says, who never submit a single piece of homework, are "lumped into the ‘non-retained’ category in a way that’s just not reflective of students’ intents, which makes retention rates uninformative at best."

At Coursera, while classic MOOC rates remain at approximately 5%, its "signature track" brings in a 70% retention rate from students who pay a $50 fee for the option of earning a verified certificate at the end of the course. Of students who declare in an entry survey that they are "moderately committed" to completing the course, the retention rates rise even higher, to 78%. Furthermore, when students identify themselves as "highly committed" in the survey, retention rates rise to 88%.

"I think it’s very clear that it’s hard to do reasonable one-to-one comparisons when it comes to MOOC retention rates," Koller says.

Critics of MOOCs remain skeptical. In a Gallup poll surveying 2,251 professors, only 7% strongly agreed that online courses can achieve student learning outcomes that are at least equivalent to those of in-person courses.

Coursera’s Koller counters that professors need to "reposition their thinking around the purpose of MOOCs, and understand that MOOCs can enhance the in-classroom learning experience and give professors access to audiences around the world they would never reach otherwise."

Udacity’s Thrun agrees: "Implicit in the professors’ responses is that students have a choice between being on-campus and online. However, many have no choice. It’s a choice between online or nothing," he says.

"Someone should add a question to that survey that asks, ‘which do you think is better for a student in, say, Cambodia or Africa, who can’t get an education: an online MOOC or nothing at all?’ I’d be surprised if many professors would say, ‘yes, it’s actually better for them to get no education at all.’"

Paul Hyman is a science and technology writer based in Great Neck, NY.

Join the Discussion (0)

Become a Member or Sign In to Post a Comment

The Latest from CACM

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved

Communications of the ACM (CACM) is now a fully Open Access publication.

By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.

Learn More