Moocs in transition: Not so massive, open or even online

Moocs in transition: Not so massive, open or even online

February 2, 2014 11:32 pmby Richard Waters

Like any widely hyped online phenomenon, Moocs – massive open online courses – are facing a reality-test.

Hardly anyone who starts one of the online courses actually finishes it. But even if the formula is being rethought, the potential impact on education hasn’t diminished.

The case for a rethink is made by John Hennessy, president of Stanford University, where much of the pioneering work on Moocs has taken place. In an interview with the FT, he questioned whether any university-level course could be opened up successfully to a huge online audience:

The range of student capability within the course simply becomes so large that you’ve got 10 students that are way ahead of the class – or 1,000 students that are way ahead – you’ve got 5,000 that are lost and you have some in the middle – and it just doesn’t work … [When] you give an exam, that exam is going to be a cakewalk for some students and it’s going to be a disaster for others – and you could see that in some of the early experiments we’ve done with the really massive activities.

Stanford has plenty of data to assess what’s been going on. Since 2011, more than 1.9m people have registered for one or more free public online courses taught by the university’s faculty and offered through four different platforms (including one of the best-known, Coursera, a social enterprise in which Stanford, Princeton, Yale and other universities are partners).

But drop-out rates start are high. Only 47 per cent of those who enroll get as far as watching at least one course video, according to Stanford, with 8 per cent watching more than half the course.

Does this mean Moocs have failed?

Andrew Ng, CEO of Coursera, says that this is to judge them by the wrong yardstick. Many people who sign up are engaged in a form of risk-free sampling, he says, so it’s only to be expected that many will drop out before they have to make a real commitment of time and effort. Even the ones who don’t stay may develop some interest and knowledge in a subject along the way.

Mr Ng (who teaches a computer science course that has attracted 400,000 enrolments over the past three years) is more concerned about the 53 per cent of people who get as far as submitting a major homework assignment but then never make it to the end.

He says the answer is to make the courses more useful in helping people land jobs. The median age of Coursera’s users is 35, and three-quarters of them have at least a bachelor’s degree: these are people who are looking for extra skills to make them more employable.

Since last year, Coursera has made it possible for users to print out verified certificates of their course completions. Three-quarters who do this list the achievements on their résumé’s, says Mr Ng.

Coursera is also starting to package courses into “curricula” geared to particular job descriptions. Anyone completing the data scientist curriculum, for instance, would be able to point to a level of achievement in the key skills needed in that job.

While this approach still applies to the wide reach and open access of the internet, others are adapting the Mooc model to make it a more useful tool in university-level education.

A good example is the copyright law course being taught by Professor Terry Fisher at Harvard University.

Although free, Fisher limited his first online course last year to 500 people chosen after extensive applications. Subjects like law are best taught in seminars of no more than 15 led by an expert, he argues. So he split his online audience into groups of 25 and allocated a teaching assistant to each, assuming a 40 per cent drop-out rate.

The online course was identical to one being held at the same time for students at Harvard. Lectures were put online, along with reading material. Discussions were both “live” (which Fisher says worked well, once teething problems with using Adobe Connect software had been worked out) and asynchronous (these were less successful, either because of the technology platform or course design and support). One group, in Jamaica, met in person for weekly discussions.

The outcome was very close to what Fisher had targeted, with 307 people completing the course – though 60 of these didn’t take the final exam and 52 who did take it failed.

Courses like this spread the cost of the content across a much bigger group of students, while concentrating more faculty time into actually providing the teaching that students value most, Fisher says.

This is very similar to what Hennessy at Stanford hopes for from online education. Universities face a “financial mismatch” as students’ families and governments struggle to meet their fees, he says.

So raising the productivity of teachers by distributing the content they produce to more students online would free them up: “What people are really willing to pay for is more personal interaction, as well as the credentialing process,” says the Stanford president.

Fisher is also adding a greater offline component to his online copyright course. For this year’s version there are 10 satellite discussion groups, including at universities in Delhi and Sao Paolo. Six are in sub-Saharan Africa, where bandwidth limitations have made it hard for students to participate.

Getting students to meet face-to-face is about more than honing their intellect, says Fisher:

Online education is alienating. Community is an important supplement to the educational experience, it tends to keep people involved.

As the Mooc revolution moves forward, it is evolving – but its potential impact on education does not look to be in any way diminished.