This is a post whose original core idea has aged badly since its stubbing nearly four years ago, but looking at it, I thought the total rewrite it now needed was actually indicative of something worth saying. So I hope you’ll forgive me if I lead you through some old thinking before I take stock of where we now seem to be. The subject is the digital transformation of university teaching, and specifically that demon of internet commentary circa 2015, the Massive Open Online Course or MOOC.
It is one of my contracted professional obligations in my current job to keep abreast of trends in the scholarship of education itself, as well as my mainline academic field or fields. This makes sense in as much as teaching is most of what, professionally, I do, but it is not without its frustrations, as that scholarship is prolific and I don’t think it’s controversial to say that it has yet really to establish clear standards of quality control. There certainly are serious, large-cohort or long-running studies of techniques or teaching environments and so on that make full use of the potential of the classroom (or rather, lots of classrooms in collaboration) as a laboratory for educational strategy and technique.1 The trouble for the people who put that kind of experimental planning and effort into their pedagogical scholarship is that they seem to stand as much chance of getting published as, and to share more or less equal standing in publication and citation when they do with, studies whose experimental basis is more like ‘I’ve been teaching small groups in the same small instititution to the same general demographic for thirty years now and though I don’t have records or anything, here’s what seems to work.’2 And I have to say, I have probably learnt at least as much about teaching from some of the latter as I have from the former, which might unkindly be seen mainly to provide data that shows what we mostly already suspected.3
But there is another trend in this scholarship, anyway, which is prognostication. I could list you scads of articles that exist only to say, “we predict that everything is going to change in the next few years in such-and-such a direction”, and they would probably almost all fall into three groups, being about either the ‘flipped classroom‘, ‘blended learning‘ or MOOCs. Of these, the most accurate looks like the ‘flipped classroom’ group, arguing with often-sound data for the improvement in learning and retention that happens if you use your classroom time for interaction and discussion rather than for lecturing, and instead have the information acquisition set as preparation and use the classroom time to work with that information and make sure it’s understood at the level where the students can do stuff with it.4 The only thing that annoys me about that scholarship is how they manage to keep selling as new methods that represent how the humanities, at least, has largely been taught for decades. Asking students to read in advance and come to class prepared to discuss, or even already having written something about, it is how I was taught and, I think I could safely say, how my father was taught either side of the Second World War, so how this keeps being published as a new idea beats me.5 Of course, it blurs into the second category, blended learning, when the prep work is digital content rather than just reading or a physical-space lecture, but the actual structure of the learning is not different. Anyway, this is not actually my subject for the day, just a frustration.
So instead my target is the scholarship or commentary which, in 2017, was saying that the future was either partly-online or partly-digital education, i. e. blended learning, which has been going on for a while, or else fully-online and centralised in the form of the MOOC. I was actually signed up to a MOOC for 2015-16—is it ironic that it was about blended learning? I’ll let you decide—as training for an administrative role I then held and so I got some idea not just of how they were supposed to work but how they actually do, or don’t. (Of course, this is the same sample-of-one no-control anecdotal standard of scholarship I was just separating from stuff with any wider basis, but this is why it’s a blog post not a journal article, isn’t it?) But even as I was doing this, the generally hyperbolic level of excitement about them had struck me as weird. I think it was a recurrent thread on a blog I wish I still had time to read, Not of General Interest, that alerted me to this, but then I started collecting references, and some example titles would be, ‘MOOCs: another weapon in the outreach armoury’, ‘MOOCs are Coming’, and my two favourites, ‘This Could Be Huge’ and ‘An Avalanche is Coming: Higher Education and the Revolution Ahead’.6 It’s probably worth noting that none of these were peer-reviewed work, although the third is good journalism and I cite it often below; but it’s almost as if genuine educators didn’t find all this plausible… And indeed, as Undine’s posts already linked show, it wasn’t hard at the time to find push-back and criticism as well, but we seemed to be pushing against a technology-evangelical wall.
I say ‘we’ because I had definitely settled against the whole MOOC idea by the time I originally stubbed this post. This was not because I thought they were dreadful ways to learn, as such, but I did think that they were somewhat missold, and most of all that there was a bait-and-switch going on from their evangelists about the economics involved. The rhetoric that all the literature I could find, and the teaching I received, was to the effect that you, the educator, can now access this mass of pre-prepared digital content and incorporate it into your own courses/modules/whatever, thus saving you valuable time and enlivening your material for your students!7 And call me a cynic, but my reaction was, “that sounds a lot like, ‘your content is boring because it’s not modern and digital; someone else is already doing what we want from you better than you are; if we can just replace all your stuff with content we bodge in from elsewhere, or even just record your content, we can get rid of you, save your wage and just pay teaching assistants to do seminars on the recorded content, because hey man, it’s history, it’s literally all in the past, not like it’s gonna change amirite?'” Using MOOCs was pretty clearly out-sourcing, and one only ever does that to save labour costs. In other words, I saw MOOCs pretty squarely as a resource to which cash-poor universities could resort to cut teachers, and which cash-rich universities who didn’t need to cut teachers could create to sell to the cash-poor ones. That was more or less explicitly Harvard’s rationale for starting to generate them, and I’m sure there were other universities whose digital learning teams had similar glinty-eyed aspirations.8 But to me it seemed obvious that we should neither use nor make these things because by doing do we would, somewhere or other, be making a colleague and maybe eventually ourselves redundant.
Now, that was the point of the original post, but the thing is, from four years on, it’s obvious that the promised revolution hasn’t happened, isn’t it? This was already becoming evident even as I stubbed the post; only a few days before major MOOC provider Udacity had decided to stop generating new courses and declared MOOCs ‘dead’, and by 2019 the subsidence of the phenomenon had been noted even in Science.9 But then came the Great Digital Pivot of 2020, which you might have thought was the tidal wave that should reverse the fortune of this dying tool, the point where everyone had to go not even blended but full-online. But it seems to have made no difference. Coursera, FutureLearn, EdX and a few other firms continue to offer such courses, but we have seen neither of the outcomes once predicted, where cheap online learning replaces universities or where universities start using extensive external content instead of their own staff (though if some of the staff cuts currently being fought in the UK go ahead, I guess that could still change).
So why not? Why didn’t the pandemic save the MOOC and why was it so ill in the first place? I think there are two big reasons. Firstly, as was being observed even in 2012, no-one was quite sure how to make them pay. The point of them was to be open access, after all, so you couldn’t charge up-front. The way Coursera used to work, and may still, is that you could do the whole course for free but had to pay a small amount to get certification that you had done so; I guess that the logic was that those who had actually completed the course would want to be able to prove it, and with literally millions of learners, even if conversion rates were terrible, you should still see enough of those payments to cover creation or licensing of content and running costs of the IT infrastructure, and even minimal ongoing staff time; a teacher-pupil ratio of 1:21,000 (as instanced below) should make that possible… But it seems that drop-out rates were even more terrible than that, and that actually you can just about keep going on that model but not make the forecasted mint. So only the biggest offerers have survived and very few universities have built MOOCs to try and make money; a few embraced them for a while as advertising for full degree courses, but I’ve seen nothing to suggest that that seems to work either.10 So reason one: a massive misjudgement of the world population’s willingess to pay for this kind of product.
I would like to think that reason two was that the turkeys who would have had to generate this content saw the wisdom in not voting for Christmas, but I’m pretty sure that in this employment climate you can pay would-be academics to teach pretty much anything, however much it may be against their long-term interests. So reason two might instead be a thing which the pandemic has very clearly exposed, which is that whether it’s because they think it results in better learning or because it forms part of the much-championed but little-specified ‘university experience’, people doing degree-level education actually want to receive it direct from people they think are experts, and have the chance to interact with those strange beasts in real time. In theory, that was possible in MOOCs, if you signed up to them when they were new and kept up with their schedules; the designers and instructors would be around and responding to comments as far as they could. But with 38,000 people signed up to a course, many contributing several times weekly, and probably two staff members running it in only some of their time, you can see how much chance there is of reaction from the instructor really happening for most students.11 If you were well behind on the content, then you probably didn’t even have the benefit of peer discussion and learning; there just wouldn’t be enough people on the same unit as you at the same time to sustain a conversation. This was certainly my experience, and I suppose illustratively, I never actually finished the course myself. An Edinburgh study in 2013 made the problems even more clear, however: they found that a MOOC took eight hours a week to set up and sixteen hours a week to run, for only one of the two staff involved; only 2,000 of 42,000 students enrolled actually completed it; the students expected instructors to be more present than was actually possible; those who fell behind didn’t find it desirable to catch back up (as I also found); and, as one of their students (identified as “Bertin”) put it on their blog, “the overall effect for me was knowing that I don’t want to do an e-learning course they run that I had previously been interested in taking”.12 Oops!
What this means is that making these things work is actually very hard, but even when they do there’s really very little difference, other then the endurance required to finish a MOOC, between it and any other online training course like the ones they give university staff on heavy lifting or fire safety or gender and race equality, all useful within limits, but basically canned content with zero interaction with the supposed teacher. And it seems clear that, even though often enough when you have students in a classroom with you it seems like interacting with you is absolutely the last thing they want to do, when the alternative is no interaction at all, it’s worse, and they’ll pay to avoid it, or at least accumulate fairly abstract debt (in England and Wales, anyway; I realise that student debt has more serious implications where the Student Loans Company doesn’t periodically go bankrupt from shortage of repayments). And MOOCs, it seems, were and are that alternative. Perhaps they could be made to run better, with more student-teacher interaction and more live content; but it would send their costs up, reduce their universal accessibility (because live means in a fixed time and place) and probably therefore make their margins worse not better. So I cautiously think that university teachers might now be safe from this revolution, at least. The blended learning evangelists look like being a lot closer to the future, and indeed the present really, but that would be another post. Let me mature that one a few years too before trying it, eh?
1. I hope I can be forgiven no more than one example per generalisation, even though by so doing I am myself normalising bad scientific practice; sorry. But: the biggest large-cohort highly-designed meta-study I know is Elif Kara, Mirco Tonin and Michael Vlassopoulos, ‘Class size effects in higher education: Differences across STEM and non-STEM fields’ in Economics of Education Review Vol. 82 (Amsterdam 2021), 102104, DOI: 10.1016/j.econedurev.2021.102104.
2. For example, the perfectly worthy but not really scientific Anne Firor Scott, “Why I Teach By Discussion” in A. Leigh Deneef and Craufurd D. Goodwin (edd.), The Academic’s Handbook, 3rd ed. (Durham NC 2007), pp. 212–216, on JSTOR here. A charming example I also have to cite is Harry Brighouse, “Becoming a Better College Teacher (If You’re Lucky)” in Daedalus Vol. 148 (2019), pp. 14–28, DOI: 10.1162/daed_a_01758.
3. A really good anecdotal practice paper is Brett Lunceford, “There Are No Girls in My Classroom: A Pedagogical Note” in ETC: A Review of General Semantics Vol. 68 (New York City NY 2001), pp. 63–67, which I don’t know how you’d find without being told. A somewhat unsurprising large-scale study is Louis Deslauriers, Logan S. McCarty, Kelly Miller, Kristina Callaghan and Greg Kestin, “Measuring actual learning versus feeling of learning in response to being actively engaged in the classroom” in Proceedings of the National Academy of Sciences Vol. 116 (Washington DC 2019), 19251, DOI: 10.1073/pnas.1821936116, which laboriously shows that on the whole students prefer it when they don’t have to work as hard to grasp what’s being taught even if it teaches them better to do so.
4. There is an incredible amount of work trialling flipped-classroom approaches or resisting them. A very recent and huge meta-study, which ought to lay the matter to rest but probably won’t, is Khe Foon Hew, Shurui Bai, Weijao Huang, Phillip Dawson, Jiahui Du, Guoyuhui Huang, Chengyuan Jia & Khongjan Thankrit, “On the use of flipped classroom across various disciplines: Insights from a second-order meta-analysis” in Australasian Journal of Educational Technology Vol. 37 (Tugun 2021), pp. 132–151, DOI: 10.14742/ajet.6475. Some suggestions that the effects might be socially variable in Elizabeth Setren, Kyle Greenberg, Oliver Moore & Michael Yankovich, Effects of the Flipped Classroom: Evidence from a Randomized Trial, discussion paper #2019.07 (Cambridge MA 2019), online here.
5. Evangelism: Dan Berrett, ‘How “Flipping” the Classroom Can Improve the Traditional Lecture’ in Chronicle of Higher Education, 19th February 2012, online here; Jonathan Bergmann and Aaron Sams, Flip your classroom: reach every student in every class every day (Eugene OR 2012); Jennifer Gavriel, “The flipped classroom” in Education for Primary Care Vol. 26 (Abingdon 2015), pp. 424–425, DOI: 10.1080/14739879.2015.1109809; or Betty Love, Angie Hodge, Cynthia Corritore and Dana C. Ernst, “Inquiry-Based Learning and the Flipped Classroom Model” in Problems, Resources, and Issues in Mathematics Undergraduate Studies Vol. 25 (Abingdon 2015), pp. 745–762, DOI: 10.1080/10511970.2015.1046005.
6. Respectively Chris Parr, “MOOCs: another weapon in the outreach armoury” in Times Higher Education, 11th July 2013, p. 11; David Williams, “MOOCs are coming”, AdvanceHE, n.d., online here; Zoë Corbyn, “This could be huge…” in Times Higher Education, 6th December 2012, pp. 34–39; and Michael Barber, Katelyn Donnelly & Saad Rizvi, An Avalanche is Coming: Higher Education and the Revolution Ahead (London 2013), online here. Corbyn says some academics were then forecasting a ‘Napster moment’ (p. 36), which seems unintentionally accurate given that in the end Napster perished and traditional recording labels somehow survive…
7. See even now Peter G. M. de Jong, James D. Pickering, Renée A. Hendriks, Bronwen J. Swinnerton, Fereshte Goshtasbpour and Marlies E. J. Reinders, “Twelve tips for integrating massive open online course content into classroom teaching” in Medical Teacher Vol. 42 (Abingdon 2020), pp. 393–397, DOI: 10.1080/0142159X.2019.1571569.
8. Harvard’s offering discussed in Corbyn, “This could be huge”, pp. 36 & 38.
9. Justin Reich and José A. Ruipérez-Valiente, “The MOOC pivot” in Science Vol. 363 (Washington DC 2019), pp. 130–131.
10. Corbyn, “This could be huge”, p. 38, where the example is the University of London, in one of its very rare corporate actions.
11. Numbers from the Stanford course around which Corbyn, “This could be huge”, is centred; see esp. pp. 36 & 38.
12. Reportage from Chris Parr, “Wisdom and Crowds” in Times Higher Education, 18th April 2013, pp. 24–25.