(De?) Personalised Learning

[NB this was written pre-GenAI which may open up many new oppportunities]

There’s a lot of talk, in the educational press, about personalised learning, which I have seen described as a diverse variety of educational programs, learning experiences, instructional approaches, and academic-support strategies that are intended to address the distinct learning needs, interests, aspirations, or cultural backgrounds of individual students.

Framed to support diverse individual needs like that, it sounds like a great idea, and has been getting airtime and support from major philanthropists and investors such as Bill Gates and Mark Zuckerberg.   However, the two big studies that have been undertaken – the RAND evaluation of 40 personalized-learning schools and the Carnegie Corporation evaluation of “Opportunity by Design” high schools – both indicate that the evidence of any improvement as a result of personalised learning is underwhelming.

My initial reaction was one of puzzlement; of course all learning is personal, all the time, everywhere.  It happens in peoples minds, so what else could it be?  How can personalised learning not work?  But then I realised that the very term personalised learning is deeply ambiguous, potentially troubled, and perhaps even dangerous.  At its worst, there is doublespeak here; like the friendly fire that is anything but friendly, the personal in personalised learning may be highly dubious.  In fact, as Dylan Wiliam has tweeted, “…what people generally mean [by personalised learning] is personalised instruction, and it’s generally not well defined.”

But let’s give it the benefit of the doubt to start with, and examine what a move towards personalised learning might mean.

Firstly, it could might mean that we seek personalised pedagogy. This would require that we can shape pedagogy in light of reliably categorised and identified  learning styles.  Alas, as the robustly-debunked Learning Styles myth shows, there are no such identified individual needs; and in fact, children learn in largely the same ways (this is the reason that we can even talk about learning principles, for example).  So this is not promising.

Secondly, we can explore personalized pace.  This seems to make a lot of sense to me – we can find ways to support those who take a little longer and extend those to whom the ideas come easily.  But this happens in good classrooms already; it is one form of what educators have for decades called differentiation and I see it every week.  Of course there is always room for improvement, but many learning tasks have it systemically built in already.  No news here!

Thirdly, and here’s where it sounds so appealing – we can look at personalized content whereby we allow students a greater voice in selecting the content that makes up their education.  The idea is that when students can follow their passions, they will be more motivated.  I am sure that is true.  However, even if we can overcome the formidable practical challenges of running classes where students choose the content (and a real choice would mean that there could be vastly different ideas being covered), there is a more fundamental worry.  Students do not know what they do not know, by definition, and so given a free choice they will be unable to make a meaningful choice.  They will likely shy away from many of the things that take years to master, and may end up being expert in very narrow and perhaps marginal areas.  Perhaps no-one would chose to learn how to calculate with fraction, ratios and percentages; or to learn a new language; or to read challenging literature; or how chemical react.  I asked one grade 6 student what he would study given a choice, and he told me he would love to devote himself to uni-cycling.  This is a genuine problem (the choice, not the unicycling which is great).

There are ways to address this third point.  One way is to make specific content so relevant that students are motivated and engaging.  That is a real and familiar challenge for educators – and we can all recall those teachers who succeed here.  Another is to offer choice within a prescribed content area – which is, of course, familiar territory to anyone who runs projects or allows students to choose their own ways and examples to illustrate key ideas.  Where possible, choice, understood properly, should be built into any learning programme – thought not an entirely free choice.

So these three areas are difficult.  That is not to say, of course, that there is nothing to pursue here.  There is, and I know that good teachers connect with individuals, not classes, and inspire people in personalised ways.

George Wood’s quote opposite suggest that this idea is a modern manifestation of what good teachers have always known, and there is wisdom here.  But if you were less inclined to be so generous, you might see some far more dangerous trends here.

One proponent of personalised learning writes that the key with personalised learning [lies in] monitoring progress in real time, or near-real time…that gets really hard to do on paper.  What that means is that we need to buy big expensive data-systems, and I currently get weekly requests to meet with sales-folk.  These systems seem to be about an education which is about the transmission of bits of information, ideally suited to multiple-choice testing, that can be monitored, graphed and used to generate additional tasks.  The students need to work individually, in what amounts to a world of perpetual testing.  The construction of deep, rich, ambiguous and problematic meaning is entirely lost. Take it to its logical extreme and we have traded education for a world of stimulus-response behaviourist training.  Like Facebook, which should in theory represent what is unique about my life but which in practice tends to reduce me to superficiality, the personalisation is in fact utterly homogenising.

Taking a slightly wider view, we have seen this sort of thing before, as the graphic illustrates.  My hope is that this technology, like others, will not live up to the hype; it will, however, generate some excellent tools that talented teachers will use well; and that weaker teachers will overuse.  I am therefore looking  forward to approaching with a skeptical eye.

Taking an even wider view, we have every reason to be wary of anyone selling us things designed precisely for us; because we know that our individual interests are frequently not aligned with corporate interests or the broader social good.  This may seem extreme, but Valerie Strauss quotes the inimitable Alfie Kohn on the broader theme of faux individualism:  For some time, corporations have sold mass-produced commodities of questionable value and then permitted us to customize peripheral details to suit our ‘preferences.’ In the 1970s, Burger King rolled out its ‘Have it your way!’ campaign, announcing that we were now empowered to request a recently thawed slab of factory-produced ground meat without the usual pickle — or even with extra lettuce! In America, I can be me!

The danger is less that these products are on the market; more that they corrupt our way of thinking about the market itself.  I close by noting the old joke; no country with the word ‘democratic’ or ‘united’ in its name is.  The word is there to persuade, not to describe.  Perhaps personalised is the same.