Personalization is one of the key leitmotifs in current educational discourse. The message is clear: personalization is good, one-size-fits-all is bad. ‘How to personalize learning and how to differentiate instruction for diverse classrooms are two of the great educational challenges of the 21st century,’ write Trilling and Fadel, leading lights in the Partnership for 21st Century Skills (P21)[1]. Barack Obama has repeatedly sung the praises of, and the need for, personalized learning and his policies are fleshed out by his Secretary of State, Arne Duncan, in speeches and on the White House blog: ‘President Obama described the promise of personalized learning when he launched the ConnectED initiative last June. Technology is a powerful tool that helps create robust personalized learning environments.’ In the UK, personalized learning has been government mantra for over 10 years. The EU, UNESCO, OECD, the Gates Foundation – everyone, it seems, is singing the same tune.
Personalization, we might all agree, is a good thing. How could it be otherwise? No one these days is going to promote depersonalization or impersonalization in education. What exactly it means, however, is less clear. According to a UNESCO Policy Brief[2], the term was first used in the context of education in the 1970s by Victor Garcìa Hoz, a senior Spanish educationalist and member of Opus Dei at the University of Madrid. This UNESCO document then points out that ‘unfortunately, up to this date there is no single definition of this concept’.
In ELT, the term has been used in a very wide variety of ways. These range from the far-reaching ideas of people like Gertrude Moskowitz, who advocated a fundamentally learner-centred form of instruction, to the much more banal practice of getting students to produce a few personalized examples of an item of grammar they have just studied. See Scott Thornbury’s A-Z blog for an interesting discussion of personalization in ELT.
As with education in general, and ELT in particular, ‘personalization’ is also bandied around the adaptive learning table. Duolingo advertises itself as the opposite of one-size-fits-all, and as an online equivalent of the ‘personalized education you can get from a small classroom teacher or private tutor’. Babbel offers a ‘personalized review manager’ and Rosetta Stone’s Classroom online solution allows educational institutions ‘to shift their language program away from a ‘one-size-fits-all-curriculum’ to a more individualized approach’. As far as I can tell, the personalization in these examples is extremely restricted. The language syllabus is fixed and although users can take different routes up the ‘skills tree’ or ‘knowledge graph’, they are totally confined by the pre-determination of those trees and graphs. This is no more personalized learning than asking students to make five true sentences using the present perfect. Arguably, it is even less!
This is not, in any case, the kind of personalization that Obama, the Gates Foundation, Knewton, et al have in mind when they conflate adaptive learning with personalization. Their definition is much broader and summarised in the US National Education Technology Plan of 2010: ‘Personalized learning means instruction is paced to learning needs, tailored to learning preferences, and tailored to the specific interests of different learners. In an environment that is fully personalized, the learning objectives and content as well as the method and pace may all vary (so personalization encompasses differentiation and individualization).’ What drives this is the big data generated by the students’ interactions with the technology (see ‘Part 4: big data and analytics’ of ‘The Guide’ on this blog).
What remains unclear is exactly how this might work in English language learning. Adaptive software can only personalize to the extent that the content of an English language learning programme allows it to do so. It may be true that each student using adaptive software ‘gets a more personalised experience no matter whose content the student is consuming’, as Knewton’s David Liu puts it. But the potential for any really meaningful personalization depends crucially on the nature and extent of this content, along with the possibility of variable learning outcomes. For this reason, we are not likely to see any truly personalized large-scale adaptive learning programs for English any time soon.
Nevertheless, technology is now central to personalized language learning. A good learning platform, which allows learners to connect to ‘social networking systems, podcasts, wikis, blogs, encyclopedias, online dictionaries, webinars, online English courses, various apps’, etc (see Alexandra Chistyakova’s eltdiary), means that personalization could be more easily achieved.
For the time being, at least, adaptive learning systems would seem to work best for ‘those things that can be easily digitized and tested like math problems and reading passages’ writes Barbara Bray . Or low level vocabulary and grammar McNuggets, we might add. Ideal for, say, ‘English Grammar in Use’. But meaningfully personalized language learning?
‘Personalized learning’ sounds very progressive, a utopian educational horizon, and it sounds like it ought to be the future of ELT (as Cleve Miller argues). It also sounds like a pretty good slogan on which to hitch the adaptive bandwagon. But somehow, just somehow, I suspect that when it comes to adaptive learning we’re more likely to see more testing, more data collection and more depersonalization.
[1] Trilling, B. & Fadel, C. 2009 21st Century Skills (San Francisco: Wiley) p.33
[2] Personalized learning: a new ICTenabled education approach, UNESCO Institute for Information Technologies in Education, Policy Brief March 2012 iite.unesco.org/pics/publications/en/files/3214716.pdf
Adaptive Learning in ELT 
“But the potential for any really meaningful personalization depends crucially on the nature and extent of this content, along with the possibility of variable learning outcomes.”
This last point (variable learning outcomes) would seem to be key. A program of instruction is only personalized (or individualized) to the extent that it adapts to my specific goals. If my goal is to achieve communicative competence at around B2 level in specific domains (e.g. rocket science) in specific registers (e.g. giving conference presentations) then the program will need to know how to differentiate me from the learner who wants to achieve C1 level in the domain of economics with the view to writing research papers. Or the user who needs only A2 to work in a call centre. Does such a program exist? Will it?
I think the short answer to your questions, Scott, are ‘no’ and ‘no’. Truly personalized learning outcomes would require an unimaginably colossal amount of content (relevant, useful content) (1) that the learner could access and (2) that has been tagged for the adaptive software. Given the number of variables involved (you want B2 rocket science for a conference, I want C2+ Spinozan Ethics to translate something from Latin into English), and the fact that these variables continually expand (at the same time that the English language itself continues to change), it will never be possible to tag all of it. It has often been dreamt of: the Mundaneum of the early 20th century, Google’s own Mundaneum ambitions, or the Wolfram Alpha project, whose goal is ‘to make all systematic knowledge immediately computable and accessible to everyone’. It’s a dream straight out of Borges.
Tagged content has to be finite. Individual learner differences are not. Analytics could certainly help in finding connections between these two (and therefore in guiding a learner and a teacher), but analytics can’t recommend content that isn’t there. Adaptive learning could take us away from ‘one-size-fits-all’ towards ‘quite-a-lot-of-sizes-fit-all’ … but no further. Pearson and Macmillan are trying to aggregate as much of their content as possible onto their platforms – and there is / will be a lot of it – but I doubt they’ll ever be able to address the needs of the Spinoza translator.
As Cleve says, a human teacher, possibly helped by insights from the analytics, will be the most effective personalizer.
Scott, I agree that variable outcomes is a key goal. Could we also say that variable paths to the same goal is also an aspect of personalization? I.e. an exam prep class (same goal) but the online component adjusts to each student’s strengths and weaknesses, adjusting to recycle more heavily on the areas the student needs more work on, and recycling less on what they seem to have acquired. It’d be a case where the destination is the same but the routes may end up being different.
Also I think the question you ask about the program that automagically adjusts to different outcomes could be a reality within 5 years, although in many respects still won’t be as effective as a teacher in personalizing to these goals. In other respects it could be superior. The ideal would be to use both the teacher and this adaptive program, but then you knew I’d say that 😉
Philip I agree that we’re “likely to see more testing, more data collection”, but am not sure that this necessarily means “depersonalized”, especially given the variable paths point above. But then I’m leaning towards the position of not blaming the tool but instead blaming those who misuse it. So I’m viewing adaptive learning as neutral, and each “use case” as establishing whether things are better or worse. And the use cases are defined by decision makers in education, many of whom are former teachers.
Interestingly there is some media attention now in the US to parents that are pulling students out of standardized tests – it’d be interesting if this became a thing.
Cleve, I think we could say that ‘variable paths to the same goal is also an aspect of personalization’, but the degree to which we could call it personalized in any meaningful sense will depend on the goal we’re referring to. A variable path towards mastering, say, affirmative present continuous for now-actions with a limited set of verbs would not be what I would call personalized. In other words, I see nothing personalized in the kind of stuff that duolingo and babble are doing. When Knewton talk about this issue, they talk about granular learning outcomes, but we’ve argued about that one before, so I won’t say more. Granularity is at the heart of Knewton’s model: ‘how students learn isosceles triangles can predict how they learn scalene triangles’ is an example that Jose Ferreira gives. Ferreira also gives examples of the way that adaptive software could personalize the learning experience by recommending that one learner studies in shorter (rather than longer) blocks or that one learner studies at a particular time of the day. I would agree that these could be useful insights, but it remains a very poor kind of personalization as these learners are all following the same pre-ordained granularized knowledge graph.
Your penultimate paragraph raises another big difference between our viewpoints. You say that you view adaptive learning (technology) as neutral. Wikipedia has an interesting little paragraph on this topic:
‘Individuals who consider technology as neutral see technology as neither good nor bad and what matters are the ways in which we use technology. An example of a neutral viewpoint is, “guns are neutral and its up to how we use them whether it would be ‘good or bad'” (Green, 2001). Mackenzie and Wajcman believe that technology is neutral only if it’s never been used before, or if no one knows what it is going to be used for (Green, 2001). In effect, guns would be classified as neutral if and only if society were none the wiser of their existence and functionality (Green, 2001). Obviously, such a society is non-existent and once becoming knowledgeable about technology, the society is drawn into a social progression where nothing is ‘neutral about society’ (Green). According to Lelia Green, if one believes technology is neutral, one would disregard the cultural and social conditions that technology has produced (Green, 2001).’
In the case of adaptive technology, it is uncontroversial to state that it has been used before, and we can list a number of reasons why it has been used. We can identify the advocacy networks that promote this technology, and we can map out the nexus of commercial and political interests that fund both the technology and its advocacy. It can hardly be called neutral any more than, say, nuclear technology.
Finally, I think that there is a problem in restricting our evaluation of this technology to looking only at use cases. Yes, use cases are defined by decision makers in education. You suggest that many of these are former teachers. But many are not. The most significant ones are not. Arne Duncan, Joel Klein, Jeb Bush, Bill Gates … Others, like Michael Barber and James Tooley, were teachers for a relatively short time. Practically every significant voice in the global world of EdTech has close associations with neo-liberal foundations and think-tanks. Not many teachers there.
However, if we were to take a use case, Arizona State University would be a good place to start, not least because of the scale of its use of adaptive software. It’s an interesting case because when Knewton’s software was introduced to the maths programme, the university’s director of mathematics wasn’t even consulted. The university administration’s decision to buy in Knewton expertise was ‘swift and unilateral’. Teachers were not consulted or informed.
Yikes Philip that was a great response and I’ll need to summons all my powers to provide an at least adequate response. Seems as though I’ll need to rebut Green, McKenzie, and Wajcman as well. Unfortunately I’m in Heathrow Terminal 1 on the way to Riyadh, so it’ll be a couple days before I can get back to you. But as a preview it appears that there are at least 2 logical fallacies in the first paragraph…not to be snarky – I enjoy this quite a bit and am learning a lot 🙂
I look forward to hearing from you when you’re back from Riyadh, Cleve.
One thing I forgot to mention in my other reply – about the growing resistance in the US to standardized testing. It will, indeed, be very interesting to see how this develops. The big testing corporations (e.g. Pearson) are all into adaptive software in a big way, and they have repeatedly screwed up test administration and marking. Will they be able to sort it out? Probably not, because standardized (and cheap) testing has so many internal problems. But standardized testing won’t be going away in the near future because it’s so central to the justification of charter schools, vouchers and for-profit interventions in education.
Interestingly, though, standardized testing is also coming under attack from the 21st century skills lobby who realize that more complex assessment methods are needed if we really want to educate our children for the ’21st century Knowledge Economy’. And, even more interesting (!), is the fact this lobby includes precisely the same organisations, under the P21 banner (Pearson, McGraw-Hill, EF Education, Microsoft, etc.), as those who have been so instrumental in promoting and selling the standardized tests.
Barbara Means, Marianne Bakia and Robert Murphy in their recent book, ‘Learning Online’ (Routledge, 2014), discuss the various definitions of ‘personalization’ in a section entitled ‘Deconstructing the Rhetoric Around the Advantages of Online learning’ (pp. 14-16). They also point out that the meta-analyses (Aiello & Wolfle, 1980; Hattie, 2009; Slemmer, 2002) of the matching of ‘learning styles’ to instructional modes show that this kind of matching indicates ‘a weak intervention at best’ (p. 32-33)