In case you haven't seen it, Ulf-Daniel Ehlers & Ebba Ossiannilsson & Alastair Creelman are running a project on Quality and MOOCs. It has a different blog post from an invited author every week for 12 weeks. They've all been of a very high standard (until this week, see below). I'd recommend Stephen Downes and Grainne Conole's posts in particular, but they're all worth reading.
So last week was my turn, and you can read my post here. The point I wanted to make was that we have created a set of quality criteria for education that is, obviously, matched to the context of education. So we are interested in things like student satisfaction ratings, student completion rates, benchmarking against agreed criteria and external review. Some of these are applicable to MOOCs, for instance you might still want external review of content before releasing it, and some courses may need to be linked to official criteria, for instance if they are related to specific vocations.
But MOOC audiences are very different, and the motivations for people studying MOOCs are different for those in higher education, or are at least of a different priority. So the same quality measures are not applicable. The most obvious of these is the much discussed drop-out rate. Course completion is just not as important for MOOC learners as it is for formal education students. And, as the (excellent) report on Edinburgh MOOCs highlights, the numbers of MOOC learners who complete surveys is a small fraction of the small fraction who complete, so student satisfaction is difficult to compare also.
My blog post ends with a plea though – and that is not to constrain MOOCs with the same quality demands we have placed on higher education. Those constraints are understandable when people are parting with lots of money, but the point of MOOCs is that they are liberated from these considerations. They allow experimentation, because the relationship between learner and educators is fundamentally different when money hasn't changed hands. We don't want to make MOOC providers fearful of experimentation because they'll fall short of quality measures that have been imposed from a different context.