The MOOC Experiment

One of the benefits of this project getting so much traction is that it gives me the opportunity to engage in conversations with people working in the independent learning space on a daily basis.

It’s hard to believe it’s been less than two years since Stanford’s AI course got overenrolled by a hundred thousand students, a phenom which begat huge media attention, which begat the creation of companies like Udacity and Coursera as well as the non-profit edX, which begat the whole big, wide, deep conversation over what it means to be a school, a teacher and a student in an age of anytime, anywhere learning.

While some of the conversations I’ve been having still start with questions over whether MOOCs mean the end of the traditional university (or at least the tiers of universities that don’t qualify as “elite”), arguments over such “all-or-nothing” alternatives are becoming fewer and further between.

I suspect this is because belief that MOOCs represent harbingers of a wall-less educational utopia or doom for the student-teacher relationship (if not education as a whole) are giving way to a recognition of the complexities created by the entry of the MOOC into the educational ecosystem.

This complexity is encapsulated in a word we use a lot when talking about massive learning: experiment.  For experiments are not something one engages in because you know the answer to a question.  Rather, they are the means by which you put potential answers to different questions to the test (which inevitably leads to the collection of data and observations of new phenomena that can become the basis for the next set of experiments).

For example, when edX limited enrollment to its Copyright course to “only” 500 students, it was engaged in an experiment to see if some level of up-front commitment (in the form of asking students to apply to participate in the course) would lead to completion rates in the 80-90%+ range (vs. retention rates of <5%-10% for most courses that have enrollments in the tens of thousands).

But these huge enrollment classes are also experiments, ones which generate reams of data that are only now being collected and analyzed.  And one of the questions this analysis should be able to answer is how much we should be paying attention to stats like completion rates (which only look at two numbers: course completers and enrollees, respectively placed at the top and bottom of a fraction) vs. statistics that should tell us more about the varying patterns of behavior students engage in after they register for a course.

The broader MOOC experiment has also triggered additional independent experiments.  This blog is one of them (an attempt to answer the question of whether one can get as much from free learning as you can from a traditional college education).  And the many startups springing up to fill niches in the rapidly changing MOOC ecosystem (such as review sites to help students understand what they’re getting into, or portfolio sites that help students document their independent learning) are themselves experiments which may succeed and prosper, flop, or transform based on what they learn from their own experimental results.

In some environments (such as corporate product development), characterizing a venture as an “experiment” has become an excuse for failure (often associated with too little up-front thinking before launching in a particular direction).  But in a field that’s changing as rapidly as education, I think it’s a sign of health that we don’t get all worked up because this MOOC goes off the rails (or herald a new age because that one went particularly well).

For this project needs to grow if we are going to reach a point where free learning becomes a reasonable supplement or alternative for students exploring where they go next with their own, increasingly individualized education.  After all, if you put together a list of all offerings currently traveling under the name of “MOOCs,” you’ll find it to consist of less than 500 courses (about the number you’ll find on offer any year at a reasonably sized university).

I already know the MOOC universe needs more advanced philosophy classes in it if I’m going to be able to finish my own project before the end of the year.  And it’s only by treating successes and failures as part of the overall experiment in online learning (rather than sources of uninformed optimism or pessimism) that we’ll reach the level of breadth and depth that will allow the real MOOC experiment to begin.