MOOCs, and the man leading the UK's rush to charge

2 min read

Someone has written the MOOC article again - this time Peter Wilby in the Guardian presents a gee-whizz gift of an interview to Simon Nelson of FutureLearn.

MOOC "scepticism" (an odd term for "the findings of actual research", but there you are...) merits an entire paragraph. All of which is brushed away, via a series of unchallenged counter-positions from Nelson.

I just want to dwell on "There are huge differences between the providers. Learning is not something you can commodify." for a moment. In my admittedly limited understanding the whole point of online massification is to "commodify" education - the unit cost drops, after all, to the point where you can offer it for free without losing "too much" money, and you do this by substituting an allegedly personalised (algorithmic) student experience for a truly personal (human) one.

All of the major providers (without exception) have done this in the same way, videos, readings, multiple choice quizzes and a few places for students to chat about it. Nelson (who is a nice chap, having met him at a conference in Stockholm) for me overplays the effect of offering a place to comment after the video (youtube style) rather than in a dedicated forum (BlackBoard style). Coursera forums tend to be full of students asking each other for advice, F/L comment streams tend to be "gosh, wow I never knew that" type things.

But the big warning klaxon sounds with the passage on F/L drop out rates. It's the classic pop-sci journalism trope of not properly describing data-sets, and comparing like with like. You can't just "[discount] those who sign up but never start", if you are citing data from other providers which includes all students who sign up. And Simon Nelson definitely cannot go on to say that participation rates are "two to three times better than other providers".

Katy Jordan's canonical dataset puts F/L MOOC completion almost exactly were you would expect it to be given the (smallish) size of the cohort, neatly on the Pareto distribution that appears to be developing when you plot % completion rate against enrolment numbers.

Oh, and Kyloe is not Scottish Gaelic for cow. You are thinking of Bò Ghàidhealach.

source: http://www.theguardian.com/education/2014/aug/19/moocs-man-leading-uk-foray-simon-nelson-futurelearn
source: http://www.katyjordan.com/MOOCproject.html
source: http://en.wikipedia.org/wiki/Highland_cattle