The third “big question” Coursera has posed has to do with retention (though I think they’d be loath to use that term):

How can we incentivize students to diligently pursue their online courses of study?

I haven’t done a lot of research on this, but some of the statistics I’ve seen put the number of students who actually complete any given Coursera class at about 10%. Indeed, in an interview Daphne Koller gave with the Wharton School of Business back in November,  she claimed that of the students who register to take a Coursera class, only about 70% “show-up” once it opens. Of this 70%, Koller goes on to say that there’s a clear “bifurcation” between students who primarily enroll to watch the videos, and those who really want to “take the course for real.”

Among the first group (those who just seem to want to watch the videos), between 30-40% actually make it all the way through all the videos in the course. Among the second group (those who genuinely intend to complete the class), Koller claims that about 30% of students who submit the first assignment actually end up completing the last one.  For both these groups, there seems to be a steady drop-off rate week by week.

By her own estimates, then, of students who originally register for a class, only 7-9% actually complete it. (Koller also says that she believes this is the “wrong way to look at it” because many students who enroll in classes never have any intention of completing the course.) Still, this is clearly an issue that Coursera wants to address.

But I think flipping this question on its head is what’s needed.

It isn’t about incentivizing students to complete the course, it’s about designing experiences that engage learners in such ways that they need and want to do more. And I think the way to do that is to incentivize teachers to use these classes as opportunities to engage in real world research. If the course is designed as a vital space for scholarship — where vital questions are being asked, and the work of students contributes to thinking through complex issues, then the course will be great.

As the Koller interview suggests, there are also many different kinds of  students enrolling in these courses, each with their own needs, interests and concerns. So while there’s some overlap (and some blanket strategies might help, like a “Coursera concierge”)  they each deserve to be thought about differently. And of course,  developing a good engagement strategy also depends on having a clear sense of what the end-goals are.

For example, if this were just a numbers game (and I certainly don’t think it is), and we just wanted to just boost the overall percentage of, say, students who watch all the videos, then keeping these students tied to the “weekly release” schedule seems counter-productive. (They are dropping off week by week for a reason, right?!) Why not make all the videos available for download on day one of the class to this particular group of people, so they could watch them at their leisure?  Or have an easier way for them to subscribe to the series without having to visit the Coursera site every week? Coursera videos could become marketed and distributed like TED talks — in a wide-range of formats and across different platforms. (June Cohen at TED has a great talk on the evolution of the TED talk here.)

Of course, perhaps the hope is to “convert” some of these students into (for lack of a better term) completers? If so, then I can see why someone would want them to interact with the videos within the traditional frame of the course…thinking that this might somehow “tempt” them into completing one of more of the exercises/activities, and then get “hooked” into completing the class. But this is a mistaken hope. Like you’re going to “lure” someone who only wants to watch the videos into actually completing the class by saying something like, “Oh, and now here’s the unit assessment!”

Ugh. No way.

Talk about creepy treehouse!

If you want students to become actual completers (and I can think of a host of reasons why it might be just fine if they don’t). then you need to give them lots of options on how they can re-engage with the course at various points and in different ways. So, for example, if some of your users watch the class videos on YouTube, or it has been delivered to them in an email or other format, a link back to the course site to complete a specific task, quiz, or activity might work better. Especially if these tasks were fun, interesting, complex, compelling — and the videos themselves set up the task in intriguing ways.

But I wonder, too, if this is still the wrong way to think about this.

In the class I’m currently taking, “Critical Thinking in Global Challenges,” some pretty active groups that have formed on both Facebook and LinkedIn. The discussions that have emerged in response to the first couple of lessons have been wide and varied, and although I have no data to back this up, I suspect that there are many people who will continue to stay fairly actively involved in these groups even after they’ve stopped participating in the course in other ways.

By analogy, think about the ways that students enrolled in a traditional face to face course interact with each other outside of class. The course is an occasion for people who are interested in a particular topic to meet, but the social (and intellectual) interaction that occurs once they’ve connected is where the real engagement is, and probably the real learning.

I’m not saying anything new here. Most critics of Coursera — and xMOOCs more generally — contend that these systems sacrifice connectivity in favor of uniformity and consistency. Actually, the critics say a whole bunch of other things, some of them pretty silly, and I’ll address these complaints in a later post. But Coursera courses can and are already wildly connected spaces. They are occasions for people to meet, and in meeting, to explore.

How many hours do people spend taking online quizzes and surveys when they are engaged by a topic? How many sites do they visit in order to find the answers to questions that really interest them? When you look at the kind of self-directed learning that people get inspired to do when they care, then the issue of “incentivizing” actually changes. If a video or activity is interesting and its relevance seems clear, users are going to want to engage more. (By “relevance” I don’t mean relevance to the course design, I mean relevance in terms of the learning-experience.)  And this would be especially true if designers at Coursera offered quick links to course material that by-passed the typical “store-front” of their classes.

Designing activities/tasks that are so interesting and challenging that people recognize their need to know more can also be wonderful motivators. (It’s what Julie Sanford and I have been trying to work out in our own writing text-book, using “Try this!” as a replacement for the regular apparatus of “activities.”) For example, most of the quizzes in “Critical Thinking in Global Challenges” are pure skill and drill. Honestly, not one of them so far is interesting and engaging as a task in itself, and so there’s no way a casual user is going to be persuaded to become more engaged as a student by posting these activities on the inter-webs!

Again, I think it all comes back to the motives of the teachers/designers of the course itself. If the class is conceived from the start primarily as an “information transmission” experience, then the class might be okay or even good, but it’s never going to be great. And as long as a course is just okay or good, then large numbers of students are just going to drift away from it.