Swimming an Ocean: Motivation and Persistence in Massive Open Online Courses

Introduction

As I’ve discussed in a number of recent posts, MOOCs, or Massive Open Online Courses are a big phenomenon lately. A big issue of discussion with regards to MOOCs is the question of how to support motivation and persistence in Massive Open Online Courses.

This is a form of learning that works very well for certain groups of learners, but, traditionally, not so well for others. This has been an issue for distance learning in general.

In particular, MOOCs work well for for autodidacts, or self-directed learners. These are people that are able to, and enjoy learning on their own, and who either don’t need a teacher. (I count myself as one of these eccentric and annoying creatures!) Such learners are able to learn regardless of the format, and the online aspect makes the learning process very convenient. The course materials can be accessed from a home computer, any time of the day or week.

It is a more difficult environment however for marginal students. These are the students who succeed at a decent rate in traditional universities largely thanks to support systems within the university campus, both formal and informal. Whether official tutoring services, or study technique mini-courses, mental and physical health support, campus social support organizations for different minority groups, unofficial study groups of students in the same specialization, or even just campus extra-curricular and social organizations. All of these features of the brick and mortar landscape, formal and informal, help support students and keep them mentally, socially, and psychologically engaged in the university community and in their courses.  An active area of inquiry centers on how to keep these sorts of students attracted enough to persist online with courses all the way through to completion.

 

Measuring Student Engagement in Massive Open Online Courses

One metric that is talked about a lot by critics with respect to MOOCs is the low completion rates. Typically only about 5-10% of “enrolled” students “complete” the class. This looks pretty bad on the surface.

However, we need to understand that these students are not all the same, and that enrollment in a free online course doesn’t necessarily mean the same thing as it does in paid enrollment at a university. These traditional statistics are somewhat misleading. Students enrolled in MOOC courses are looking for different things. Because it’s free, online, somewhat anonymous, students don’t have to commit money or time to move somewhere or commute to study. As a result, registering for a MOOC is not the same implicit commitment as registering for a paid course at the university. Different students are bringing a wide range of different levels of commitment, different intentions or goals. Even the same student it may differ from course to course.

Take myself as an example. I am “enrolled” in a few different MOOCs right now. (I’ve been between work contracts for a few weeks, so I’ve been taking advantage of the time to take some courses and expand my skill set. This may level off a bit as I start a new job next week)

One Coursera course,  in Interactive Programming in Python from Rice University, I have been actively engaged in the course activities. I’ve spent a good 10 hours a week or more, all told, watching all the lectures, doing all the assignments and quizzes on the official schedule, engaging in the course discussion forums, posting and engaging with other students. I am also a paying participant in the course. For a number of reasons. One, just to give a little kick of external motivation. Two, because I’m interested in certification in MOOCs as a personal informal research issue and would like the verified completion with honors certificate at the end for my LinkedIn.

Other courses, though, whether because of time and / or money limitations, or because the course is interesting, but overly challenging, or because the course is finished, I’m following along with the free and open online record of lectures and assignment exercises, but not fully engaged. For example, I’m also looking at a course in Functional Programming in Scala from Lausanne Polytechnique. For a few reasons. First, because it’s a trendy language used on a lot of sites like Twitter and Quora. Second, because it’s focusing on a more exotic programming paradigm of functional programming. Third, because the course is taught by Martin Odersky, a star computer scientist who created the language. It’s fascinating stuff, and I’m roughly following along with  the lectures, (though a little behind now) and taking a look at the weekly exercises. But I’m more peripherally involved. Again, for a few reasons. First, because the teacher and the presentation are very abstract, and because the course is intended for an advanced audience. Also  there’s not much support for low level things like guidance on syntax of the language, and I don’t have enough time to properly dig around on the internet, at least for now. Also, I’m still refreshing my knowledge of computer programming, so it’s not easy to process more exotic and abstract ideas right now. So I watch lectures to absorb some of the theory, and maybe next time it’s offered I can engage more. Or I can play with the exercises later when I have time.

Another course on Coursera I’m following is in Gamification from University of Pennsylvania, put on by a very engaging Wharton Business school lecturer, Kevin Werbach. It’s not a technically demanding course; it’s theory and c0ncepts mostly, but peppered with lots of practical examples. The course is already finished, so I’m following the lecture record with interest.

Some other interesting looking courses I registered almost as a bookmarking exercise for my later reference so that I can find the courses later in my account on Coursera. Probably not how Coursera intends for you to use the enroll button, but it’s a functional hack that works for me. I similarly have a few courses bookmarked on Udacity and EdX

So just to illustrate, people have different interests and goals. The percentage of people completing / people enrolled is based on the idea that everyone has the same goals. Which is not at all the case. Some want to master a new skill and get a certification they can try to leverage professionally, maybe to expand career options. These ones tend to spend more time, and tend to complete at a higher rate. Others just want to learn some new things seriously, but casually. Others are just poking around, trying out a few courses to see which few they’ll spend more time with. So these completion rates are misleading. The more important metric is how much people are getting out of the course in relation to what they hoped to get out of it going in. And on this metric, many or most are getting as much or more out of it than they intended.

That said, there is doubtless room for improvement in making courses so that more people will engage in them to a deeper, more persistent manner. And this is where I’d like to turn to now.

Again, I’d like to illustrate from my own experiences, particularly with the interactive programming in Python course. The course had lectures, weekly graded multipler choice and short answer exercises, and  weekly mini-project programming challenges. The programming challenges were typically games. An early project was a textr based game of Rock-Paper-Scissors-Lizard-Spock (a nerd-variant of the popular game Rock Paper Scissors). Each week, this stepped up in difficulty, through programming a version of Pong, a version of memory, a visual game of Blackjack, and finally, a two week project building a space shooter game based on Atari Asteroids. Here is my final submission, by the way: http://www.codeskulptor.org/#user32_yFpWoboCQZ_80.py

People started having trouble when it got to the Blackjack game. This was the week when the programming paradigm of “Object Oriented Programming” was introduced. Coming from an abstract mathematical sort of background, I was able to grok the basics of this pretty easily, and managed to finish the project early. As someone who is also from an education background, I decided to go onto the course discussion forums to try to help out anyone who was having difficulties. There was a  lot of chatter during that particular week from people thinking about quitting. I spent a lot of time on the weekend that project was due on the forum trying to talk people down from the ledge and  trying to help different strangers who were stuck. It was a rewarding experience. But it illustrates the fragility of the online environment for those who are borderline.

Supporting Engagement in Massive Open Online Courses

So this is a big topic, how to support learners who are less auto-didactic, less internally motivated than someone like me. I’ve always been someone who benefitted from a good teacher, but didn’t really need one. Online works  naturally for people like me. But what about for others? There are a few different ideas.

More development of discussion forums. This is a great option currently used by different courses. It helps to build a course community and gives students a forum to discuss, exchange ideas, ask questions, get answers. People like online discussion forums, and they play a useful role in the educational experience. The key is to motivate possibly borderline students to engage on the forums and to make sure that the forums are a welcoming, friendly, and safe place to try out ideas and learn. This takes promotional efforts from the instructors to spend time encouraging the use of the forums, and periodically reminding about the existence and usefulness of the forums. It also requires volunteer or paid TAs to help monitor and moderate the discussions.

Another option currently enabled by some MOOC providers is local meetups and other similar projects. Coursera for example enables the ability to try to organize face to face group meetings of people in the same city who are taking the same course. In this way, people could theoretically get together to form real world study groups.

Another idea would be assigned buddies or groups made up of other students. Whether for all or for those who ask for it as an optional support. One possible way to incentivize this into a voluntary effort would be to have a survey at the beginning of the course where people can identify their level of familiarity with the subject going in and their level of confidence in their ability to succeed. People with little experience and low confidence could be directed to a page where they can opt in to being paired with a volunteer student with more experience and confidence. Conversely, those with more experience and confidence could be directed to a page where they can opt to be put into a pool of people who will be randomly paired with a few students who anticipate needing help. It could be entirely voluntary, but incentivized for example with marks. If you opt in to this and actually help people out, then you can get a few bonus marks that can make up for a bad performance on an assignment or quiz.

Another option to make things more lively and interactive than discussion forums would be something like Google Hangouts. People could talk face to face about ideas in the course. This would work best probably with arts and social sciences content, but could find application more generally too.

Finally, another option is to have some minor collaborative exercises in the course where you need to work together with other enrolled students. There would be logistical challenges to dealing with this, for example, if the random forming of teams included a number of more peripheral students not taking active part in the assignments. But if you can find solutions to these challenges, the social, collaborative aspect could help to engage many in the course.