MOOC motivations and magnitudes: Reflections on the MOOC experience vs the MOOC drop out #moocmooc


This is a long post. What can I say? The MOOC made me do it!

When 90% drop-out rate is less than a 10% drop-out rate

English: A diagram of the geological time scal...

English: A diagram of the geological time scale (Photo credit: Wikipedia)

One of the foundations of the, I think largely unfounded, criticisms of the MOOC is the high drop out rate.

But I think this is a complete misunderstanding of how proportions work. As in so many other cases, people are basically taking two similar looking different domains of experience and translating meaningless percentages between them without consideration of scale.

A brief diversion: The thing we know from complexity science (or chaos theory to be a bit inaccurate) is that magnitudes matter. Complex systems are sensitive to initial conditions, so that a small variation in input can result in an unpredictably large variation in output (to simplify the idea out of recognition). However, the variation is not completely random. It happens in patterns (attractors) that bear a resemblance to each other but are not the same. Weather, for instance is largely unpredictable but whatever the outcome, it still looks like weather (so we may say it’s boiling hot, but it will never actually be boiling). But the difference between 32 degrees or 33 degrees today may be the difference between 12 and 42 degrees tomorrow. But the sensitivity to initial conditions only happens at a certain scale. Whether it snows in a particular July or boils makes no difference to the climate. But climate (on its own scale not experienced by humans) also has its sensitivities. It seems predictable to us in human time but not in geological time.

This is a roundabout way of saying that even if it looks like a drop out rate, it may not be the same kind of drop out rate. When you take scales into account, you may come to the conclusion that the drop out rate of a MOOC is actually smaller than that of a university.

So let’s consider a few numbers. The largest MOOCs have enrollments around 100,000 and graduations around 10,000. So that’s a 90% drop out rate. Horror of horrors. This is awful education with old style pedagogy moans complain the pedagogues.

But what the complainers don’t take into account is the scale. Sure, if their classes experienced a drop out rate of 90%, they would end up with what would essentially look like no students (and no money). Imagine 90% if a class of 20 would drop out. Which is why universities take extreme measures to prevent students from dropping courses. But students still drop courses and would drop them more if they were allowed to. The only reason university students can’t drop out of courses is to fit in with financial logistics of the overpriced bureaucratic institutions that gauge them for the promise of a better future. Where’s the pedagogy in that?

The Stanford AI course reportedly had 160,000 enrollments with roughly 15% completion rate. That’s 23,000 students. Which is 10,000 more than are enrolled at Stanford at any one time (which by the way has 22% drop out rate). So to complain about the rate of drop out is to ignore the magnitude. Even if 95% of the students drop out, it is still having more impact than a course at an institution on a normal scale where only 5% of the students drop out.

Let’s look for some parallels. If I make 1 million dollars a year and my salary drops by 90% the next year, I won’t experience the same hardship as if I’m making 100 thousand and year and get the same drop in income. If the gun crime doubled in Switzerland from year to year, nothing would essentially change because there are negligible amounts of it to start with. If it doubled in some parts of the US, there’d be bodies lying in the street. Magnitude’s also why a tiny organism can carry heavier objects in proportion to their body weight or fall from greater heights. This strength does not scale. A human who can only carry several times her body weight will still win with a fight with an ant. Magnitudes matter and Spiderman is impossible.

But there are other reasons why the enrollments and dropouts in xMOOCs and traditional classes are not really comparable even at the same scales. Let’s pick some of them apart.

Why do people enroll in MOOCs and universities

Motivations play a big role in enrollment and retention.

The range motivation from enrolling in a MOOC is great and no two motivations are the same. However, we can probably describe some of the basic motivation profiles and cover enough points on the spectrum to give an accurate portrayal. The motivation profiles are something like this:

  • Desire to learn something from an expert to whom you would otherwise not get access
  • Desire to learn something on a subject with which you are not familiar
  • Desire to get some sort of recognition for learning done (even though it is not a formal credit)
  • Desire to complete what was started (see below)
  • Desire to connect with other learners from a wide range of backgrounds (age, region, language, level of expertise)
  • Idle or professional curiosity in a new thing (be it methodological, technological or simply because it’s there)

Every single person’s motivation will combine some characteristics of one of these profiles making everyone completely unique in one sense but not enough to really make a difference.

For comparison, these are the motivation profiles for enrollment for a university degree:

  • Get a university degree or another qualification
  • Fullfil an expectation imposed by the environment (parents, peers, etc.)
  • Improve employment prospects (related to but distinct form the above)
  • Acquire specific knowledge or skills
  • Enjoy a growing experience with your peer group
  • Become a more rounded / educated person (often more conversational requirement than real motivation)

So we see the profiles are different, but I’d also suggest that their priorities are different. But we also need to look at why students enroll in individual courses at a university:

  • It is a required course for a degree (or one of a set of required courses)
  • It is the only course available due to scheduling restrictions
  • It is a course reputed to be easy or passable
  • It is a course that they heard something good about
  • The course is taught by a star professor
  • The student wants to learn what the course has to offer

So this shows us that MOOCs and university course are not comparable in any meaningful sense when it comes to student motivation at the moment. MOOCs are somewhere between classes and degrees. The time commitment and level of achievement is comparable to the idea of a single class. But the motivation and student engagement with the institution is more similar to a whole course of study.

This will change if MOOCs will become credit bearning or even compulsory as part of a degree. Long time ago, I wrote about the Becker attractor, suggesting that over time students will always organize around meeting the institutional requirements rather than an ideal. So as they become embedded in these structures (or obstacle courses), MOOCs will become more like any other course when it comes to motivation for dealing with their requirements.

Why do people drop out from MOOCs, university courses and universities?

2005-12-25 Magnifying drop

2005-12-25 Magnifying drop (Photo credit: Wikipedia)

So what are the reasons people drop out of MOOCs? This is not hard to imagine:

  • Lack of time or conflict of commitments
  • Lack of continued motivation due to no recognition for hard work
  • Lack of ability to perform the course assignments (language barrier, pre-req knowledge, etc.)
  • Dissatisfaction with an aspect of the course (support, peer group, course tools, etc.)

This is not that different from when people drop a university class:

  • Scheduling conflict or lack of time or other resource
  • Dissatisfaction with an aspect of the class
  • Inability to complete assignments

But why do people drop out of university education completely?

  • Run out of money (my case with my PhD)
  • Conflict with other commitments (job, sport, etc.)
  • Find it too hard
  • Other personal reasons (illness, family situation, move, etc.)

So the range of reasons for dropping out does not appear as great as the range of motivations for joining. Other than cost, they are similar between MOOCs and universities. But before we go any further, we also need to compare the level of investment.

The effort to enroll in a university is incomensurable with that of a MOOC which only requires you to fill out a 10 second form. Days and days spent on application forms, research about the university. And then there is the cost. Tuition in thousands or tens of thousands and living expenses. Why would anyone waste all of that and drop out?

Yet, students drop out of British universities as a whole at a rate of about 20%. And the drop out rates for US universities can be as high as 40% (and as low as 10% for many – the Webb institute has 96% completion rate but their entire student body is 80 – so we’re at magnitudes again). And this is after teh students invested a year or more of their time and thousands of pounds in fees and other expenses. Not even counting how much is potentially at stake in terms of their future job prospects and financial stability. And the impact on their standing in their peer group and the pressure from their family. And that’s not even considering  all the effort and resource, the universities expend to keep the students.

What is the investment in a MOOC? Some time (measured in hours) and some effort. What are the consequences of dropping a MOOC? No practical ones. Some mild sense disappointment, perhaps.

So when we compare the two situations, my conclusion is that the university drop out rate is higher in relative terms than the MOOC drop out rate.

But that’s not the point. The point is that the level of MOOC drop out has little to do with the MOOC pedagogy (individual dropouts do but not the scale of them). It has mostly to do with investment and motivation. So it is a bit strange that the people who essentially keep their students in their classes by force find that in any way alarming. They should be alarmed about the symbolic violence and exploitation to which they signed up!

So what is it like to take a MOOC?

drop out - cracker [portrait]

drop out – cracker [portrait] (Photo credit: the|G|™)

But it’s not just what brings us in and how we leave. What it’s like, is also important.

We all sort of know what it is like to take a university class. Lots of sitting, listening, reading and writing is involved. At that level, it is not so different from taking a MOOC.

What is different is the engagement. And here the type of MOOC you’re taking matters.

I’ve tried and dropped (or lurked on and dropped) several MOOCs but my motivation was always professional curiosity rather than a desire to learn. Conflicting time commitments then got in the way. But it was enough to get a sense, I believe, for what it’s like.

I’ve peeked in on 2 xMOOCs (generally larger extension MOOCs) and 2 cMOOCs (generally smaller connectivist ones). The day to day experience of the xMOOCs and the cMOOCs was very different for me. The xMOOCs were very much focused on knowledge and skills and peer interaction was always in support of knowledge and skill acquisition. All the interaction was done inside the course’s online platform (forums, etc.). This would have been perfectly suitable if I’d been motivated enough. I really do want to learn Python and I have no doubt I would have learned enough had I followed the Udacity course faithfully. But I also really want to learn Chinese. Both of these motivations are real and sincere but they are aspirational and any steps I may be taking towards their fullfilment will be derailled by the slightest of obstacles (including missing one class or just not feeling like it).

The cMOOCs, on the other hand, felt much more interactive and exploratory. To the point I didn’t actually feel I was learning all that much (sense, I know, shared by others). The learning was all coming from inside me. Which would be great, if I truly felt I wanted to learn about what was on offer. These MOOCs were much more distributed. There was a central hub (with some) where discussions would take place and assignments would be shared but most of the work happened on blogs, YouTube and Twitter. The motivation to stay involved would stem from establishing peer relationships. But if you’re only lurking, as I was, then the ties are much weaker and chance of drop out (or rather fade out) is much higher. And indeed that’s what happened. I started with the best of intentions but the smallest stumble on the way was enough to derail me completely.

But it need not necessarily be like that. The one xMOOC where I’ve done all the work for at least a part of the time is the still running Coursera course on SciFi and Fantasy. That one is less about knowledge and more about experience. The peer discussions were quite supportive (if a bit scattered and hampered by a limited software platform) and the assignments and the feedback themselves were very useful. But I’m not going to continue because I don’t actually feel, I need that course for anything. I joined to see what teaching literature is like having taught it briefly myself and having on occasion engaged in literary criticism. What I’ve seen convinced me that (despite the many mostly valid quibbles some people post about the course) it would have given me a very good foundation in reading and writing about reading if I felt like I needed it.

The first MOOC where I’ve completed every assignment (though one rather skimpily) is the MOOC MOOC – a MOOC about MOOCs. I enrolled fully intending to lurk. But I found that I was learning more than I’d expected. I thought that I was prety well informed MOOCwise (enough to run my own MOOC) but found out that I had a lot of room for more MOOC learning. Plus I really appreciated the opportunity to exchange views with others.

But the MOOC MOOC only lasted one week (six days to be exact). I may have learned as much without completing the assignments but I found my motivation to be self-perpetuating. I started out doing my own thing but found myself converging on the assingments for the day and wanting to complete the set (thus this post). I also like the aspect of light touch but quite intensive interaction via Twitter (I posted close to 200 tweets in 5 days). But if it had been much longer or more spread out, I may have dropped out anyway. I will of course continue my thinking and learning about MOOCs and writing about them on this blog (though most likely with much lesser frequency – I average about a post a month).

What could have helped me complete the other MOOCs, as well, would have been some expectation of a recognition. I would have certainly appreciated some badges to collect as I went along. An expectation of some certificate at the end was not enough.

But a more immediate need for learning is probably even a bigger factor. Since I’m in the middle of working on a MOOC of my own, the MOOC MOOC fit perfectly with my needs in a way the other connectivist courses I lurked on did not.

And finally, having some time set away without any other commitments would certainly have helped. That is the one advantage of going to a place to a course. You’ve carved out some space where nothing intrudes. As I’ve argued in “Space, the final frontier of online education“.

Now, could I have met the course objectives of all the MOOCs I’ve tried given the resources and forms of “delivery” offered? Absolutely, I could have been a contender! And I’m sure there would have been frustrations, dead ends and false starts along the way and some of the learning would be incomplete. But that’s no different from traditional courses or any learning.

Bringing it all together

So what do magnitudes (which are number like) and experiences (which are decidedly analog) have in common? Magnitudes are about the human experience of numbers. A person earning 10,000/year won’t be able to differentiate between someone earning 1 or 2 million. The doubling of a small salary is more impactful than the doubling of a large salary. If a balcony falls on you from great height, it won’t be that much different from a whole space ship falling on you. And you won’t feel the difference between one feather and two feathers. But there is a point where the doubling of the weight will matter.

We need to consider the impact dropping out of a course or education has one the individual. And the impact the dropping out of a large numbers of individuals will have on the a whole group. So we need to explore those experience to understand the magnitudes. Otherwise the numbers we’re throwing around won’t have much more value than those called out at a Bingo game.

Thus endeth the MOOC. MOOC.

  • Pingback: MOOC motivations and magnitudes: Reflections on the MOOC experience vs the MOOC drop out #moocmooc | Ideias |

  • Wise After the Event

    Great stuff! As a former scientist in chaos theory, I want you to know that you actually get it! But people have a very difficult time with both the nonlinear and scaling paradigm.

    There needs to be a larger discussion on differences between types of cognition so that people can understand that learning different types of things is really different. FWIW, I have tried, but have failed pretty profoundly with my peers, though I find my students get it. MOOCs are going to be good at certain types of learning, and not-so-good at others, or must be structured differently. But we have a one-size-fits-all paradigm for most knowledge in the university, and so I think the prospect is relatively hopeless.

    Regardless, thanks for writing.


Get every new post delivered to your Inbox

Join other followers:

%d bloggers like this: