Disrupting Education

Education is a strange, looking-glass world where conservatives pose as radicals. It is not a progressive education utopia, even if the ideas of the progressive education movement are drawn upon heavily by educational academics. Schools are intensely practical places and children pose myriad problems which educators have to muddle-through on a daily basis.

Instead of providing solutions to such problems, education academics are generally committed to some sort of an ‘approach’. Typically, academics construct complicated-sounding nouns and then try to rationalise why these ‘things’ are important. ‘Reification’ is the complicated-sounding noun that describes this process.

Educationalists would do better to adopt part of what is recognised as the scientific method. They should ask, ‘What would be the case if I were wrong?’ and then seek that evidence. This is what scientists do. This is why science is the best magic available at present and why we haven’t yet benefited from The Sociological Revolution. A good theory echoes Pat Benatar. “Hit me with your best shot,” it says, “fire away!”.

And so we must welcome the disruptors. And I’m not on about those who are giving kids iPads or anything as daft as that. I am talking about those who are now holding received educational wisdom up to the light and seeing right through it.

They ask whether learning styles or multiple intelligences exist. They wonder whether ‘skills’ such as ‘collaboration’ or ‘creativity’ are really skills and whether they look the same in different contexts. Perhaps these skills have just been made-up – ‘socially constructed’ if you will – and are not really ‘things’ at all.

The disruptors wonder whether teachers really are to blame for children’s deep-seated behavioural problems. They challenge the narrative that these are caused because teachers don’t listen or don’t plan engaging enough lessons. Instead, they question whether there are other causes and other solutions.

Indeed, they wonder about the notion of engagement, sanctified as it may be. Perhaps there are many kinds of engagement of which relatively few lead to learning. Perhaps hard work can in fact cause the right kind of engagement. Perhaps defining engagement in ‘active’ behavioural terms risks circular reasoning, particularly if we wish to promote behaviourally active teaching methods. Other kinds of engagement are possible, after all. 

The disruptors ask what is wrong with telling kids things. They challenge the idea that inquiry/discovery/constructivist/insert-new-name-for-it-here methods are better and point to the empirical evidence that they are not.

Of course, those who inhabit the current landscape are not going to welcome the disruptors. They are the conservatives who wish to preserve the way of things. They are likely to dismiss, patronise and even openly insult the new punks who are insisting on thinking for themselves.

But they can’t stop it. These are revolutionary times.


26 thoughts on “Disrupting Education

  1. Could I put in another plea for ‘the disruptors’ not to lump things together that shouldn’t be lumped together?

    I’m aware that learning styles and multiple intelligences have been rolled out unquestioningly in schools. However, few critics of learning styles appear to have read the Coffield review or cite it and are happy to simply refute the concept completely because someone said on their blog that there was ‘no’ evidence.

    Howard Gardner’s multiple intelligences framework was derived from his work with neurological patients. You couldn’t get much more scientific than his earlier book ‘The Shattered Mind’, but none of the disruptors are citing that; they are blithely junking wholesale the idea behind multiple intelligences based largely on the way the MI model has been applied in schools.

    Then there are education academics. Some do what you are complaining about. Others devise assessment methods for children with hearing or speech difficulties; there is no justification for lumping them all together either.

    If different things are lumped together there’s a big risk of the disruptors replacing one inadequate over-simplified model with another.

    • I have read a number of reviews of learning styles and also multiple intelligences. I am aware of no convincing evidence that they exist. Are you saying that there is such evidence? If so, please link.

      Yes – my comments on educational academics are a generalisation.

      • A broader and stronger concept is that of ‘cognitive styles’, see, f.e.,

        Click to access cognitive_styles2007.pdf

        That differences in intelligence are multiple in kind, also granted broad factors fluid and crystallized intelligence explain the bulk of those differences), is no secret. The main work here is by John Carroll, 1993, its introductory chapter:

        Click to access Carroll_1.pdf

        and yes indeed, the second chapter

        Click to access Carroll_2.pdf

        The problem for the teacher then is; how is this material relevant for my work? Well, not in the sense of diagnosing the specific abilities of every individual student (lots of ethical issues here, also, quite apart from the inefficiencies entailed by this kind of ‘personalizing’), but by taking care that teaching methods and assessment forms do not get narrowly defined. So there is no need to study John Carroll, or research on cognitive styles 😉

      • Comment in two parts. First about reification of constructs.

        I completely agree with your point about reification. I’ve found it useful to make a distinction between descriptive terms and the reifications that emerge from them.

        We usually know what people mean when they say person A is, say, patient, loving or intelligent, by the way they use those adjectives even if different people would define the words slightly differently. We also know what they mean if they describe person A as having a lot of patience, much love or great intelligence. The nouns are still being used descriptively, even though they’re effectively adjectives that have been reified. It doesn’t follow that patience, love or intelligence are free-floating things that we have in different quantities.

        We could operationalise the constructs ‘patience’, ‘love’ or ‘intelligence’ and then devise instruments to measure those constructs – we’ve done that in the case of ‘intelligence’. The fact that we’ve operationalised and can measure a construct doesn’t mean that patience, love or intelligence exist as free-floating qualities independent of human beings either. But the fact that they don’t exist independently doesn’t mean it’s invalid to use them descriptively.

      • Part 2

        ‘Learning styles’ is a valid descriptive term for the various ways in which people learn – or prefer to learn – stuff. It could refer to a whole host of things. I agree that it’s term that’s been reified to mean a thing that people have, when there’s little evidence that a learning style is a thing that people have.

        I don’t know what reviews you’ve read, but the Coffield one is pretty comprehensive and rigorous. Coffield et al reviewed a dozen or so different learning styles models widely used in post-16 education and training and assessed them for reliability and validity. Several of the models met some of their reliability/validity criteria. Allinson and Hayes’ Cognitive Style Index met both.

        My guess is that what you mean by ‘learning styles’ is the Dunn and Dunn Learning Styles Inventory (VAKT model), which was criticised in the Coffield review. If that’s the case, it might be worth saying so, because it’s clear from Coffield that not all learning styles models are the same.

        The same descriptive/reification framework applies to the Multiple Intelligences idea. It’s clear from Gardner’s work that he doesn’t buy into reifications and in part his multiple intelligences idea was developed to make exactly that point; that a wide range of competencies make up the construct ‘intelligence’ and what our view of intelligence is depends on the competencies we choose to include. What has happened is that a lot of people who have come across his idea have bought into the reified construct ‘intelligence’ and have simply assumed that there must be multiple intelligences instead of just one. Misinterpretations of Gardner don’t mean that people aren’t ‘intelligent’ in different ways.

        I don’t object to ‘disruptors’ at all – the more the merrier. But if they are going to go on about the scientific method and empirical evidence then perhaps they should use both, rather than wave their arms dismissively in the general direction of a select few constructs.

      • I don’t think it is a waving of the arms. It’s an objection to people just making stuff up and then expecting everyone to take it seriously. There is a lot of that in sociology and education research.

    • The objection to people just making stuff up and then expecting everyone to take it seriously is one side of the coin. The other side of the coin is the disruptors saying there’s no evidence for this, that or the other and expecting everyone to take that seriously too.

      If the disruptors are serious about using the scientific method and evidence, then why not use the scientific method and evidence rather than just claiming there’s no evidence for this, that or the other concept and concluding that *nothing* about it can possibly be valid, and expecting people to believe them?

    • Your post wasn’t citing phenomena that are irrefutable. You referred to two constructs that can be operationalised and can have their reliability and validity assessed, but you haven’t cited the evidence that has assessed their reliability and validity. Just saying ‘there’s no evidence so they can’t exist’ is the exact equivalent of saying ‘there’s no evidence they don’t exist so they must exist’.

      Neither assertion uses the scientific method or evidence.

      • No. Those statements are not equivalent. It is a different burden of proof. Those who make-up the concepts are the ones with the burden to demonstrate that they exist. These folks cannot insist that others prove a negative. Nevertheless, although I have not referred to evidence of their absence here, I have in other posts.

        This is helpful: https://yourlogicalfallacyis.com/burden-of-proof

    • You’re not talking about logic, though, you’re talking about the scientific method and empirical evidence. The scientific method doesn’t work like a mathematical/philosophical proof. It’s not about demonstrating that phenomena exist, it’s about testing hypotheses.

      1. Someone comes up with a hypothesis. They test it. They refute it. They tell everyone they’ve refuted it.

      2. They can’t refute it, ergo it could be valid.

      3. A bunch of other people test it. They either refute it, in which case it’s not valid. Or they can’t refute it either, ergo it could still be valid.

      4. We can never say a hypothesis is true or a phenomenon exists, just that they haven’t been disproven yet. We only ever have working models, though some hypotheses/phenomena have a great deal more validity than others.

      Could you give a link to the other posts? Thanks.

      • I think you will find that logic is important in science. There is no need to accept the existence of things for which there is no evidence.

    • I’m well aware that logic is important in science.

      Russell’s teapot is an example of an irrefutable or unfalsifiable hypothesis. There’s no way anyone can currently demonstrate, either way, whether or not there is a teapot orbiting the sun. In the case of an unfalsifiable hypothesis the burden of proof must lie with the person proposing it because no one would ever be able to test it.

      But learning styles and multiple intelligences are not unfalsifiable hypotheses. We could test them. Coffield et al did test the VAKT Learning Styles Inventory and found it wanting. To make sure, someone else could also test it, but it looks as if the VAK model doesn’t hold water. Coffield et al also tested Allinson and Hayes Cognitive Styles Index and found it was reliable and valid. That doesn’t mean Allinson and Hayes’ model has been proven to be right – it means it might be right. That’s as much as you can say about any hypothesis that hasn’t yet been falsified.

      What you seem to be looking right past is that there is evidence for learning styles models and for multiple intelligences, which is why both concepts were mooted in the first place. Whether that evidence has produced hypotheses that stand up to testing is another matter, but the burden of proof in the case of a falsifiable hypothesis most certainly doesn’t rest with the person proposing it, nor is it in terms of proof positive.

      • I cannot prove that learning styles or multiple intelligences don’t exist any more than I can prove Russell’s teapot does not exist. The point of interest to educators is whether any of these suppositions have implications for what we do. It seems that they do not (although I can see that you hold out hope for one particular version of one of them). If evidence is presented that suggests that I should take notice then I will reconsider.

      • It isn’t about whether the construct exists or not, it’s about whether a hypothesis is testable or not. And you might not be able to test the hypotheses, but other people might.

        The VAKT model, for example, proposes that people in general prefer to learn using a particular sensory modality. All that’s needed to refute it is to demonstrate in a large group of people that they don’t.

        If educators are gullible enough to accept as true a hypothesis that hasn’t been tested, then they have only themselves to blame if they find out the hypothesis is wrong.

        I don’t hold out ‘hope’ for any of these suppositions. I’m just pointing out you can’t say they are all either true or false – and I’m basing that claim on published evaluations of the evidence. The Coffield analysis suggests some learning styles models might be worth looking at. I’d be interested to see what your claim is based on.

      • Again, the burden of proof does not lie with me, it lies with those who put forward these things.

        Interestingly, I am quite prepared to believe that people prefer different modalities. This is something quite different to suggesting that teaching people accordingly leads to greater learning.

      • I’m not clear whether you think constructs like learning styles constitute unfalsifiable hypotheses and therefore are a form of Russell’s teapot, or if you think that scientists are required to come up with proof positive for all their hypotheses.

        Yes, people preferring different modalities is different to proposing reified Learning Styles. Both are different to claiming that teaching according to a preferred Learning Style improves learning. There are three different, testable hypotheses in there. They need to be teased apart and tested separately. If you don’t want to test them, that’s fine, but you then can’t draw any conclusions about their validity.

        Science involves precision as well as logic.

      • You are right that learning styles theories make testable claims and it is clear that they largely fail such tests, leading to conclusions such as those made by Paschler et al. However, I cannot prove that a learning styles theory may not appear at some point in the future that does appear to work.

        Indeed, you point to Allinson and Hayes’ CSI inventory as showing promise. It does, at least, seem to be a reliable instrument. However, it is based upon the largely debunked idea of left versus right brain thinking and this should at give us pause. As Coffield points out, the fact that British managers tend to be more intuitive than their subordinates may not indicate a stable feature of their brains but may be related to hierarchy and culture: They have experience of making more decisions at pace or perhaps they tend to delegate analysis. Quite what any of this might mean for the K-12 classroom is unclear.

        I am quite happy to conclude that learning styles (as commonly interpreted – cue another semantic argument) clearly do not exist on any reasonable interpretation of the evidence that exists to date. However, given their multiple forms and the fact that people keep making up new versions, I cannot say that I have evidence that no learning styles theory will ever work or provide something useful in the classroom. However, like with the existence of Russell’s teapot, I will be proceeding on the basis that they won’t.

      • I have to add that it is odd that you are so keen to defend entirely socially constructed notions for which there is so little evidence. Perhaps we have identified a new cognitive style continuum with me on the ‘minimalist’ end and you on the ‘thingification’ end?

      • Thanks for unpacking the learning styles evidence.

        What I’m still struggling with is if, as you claim in your post, the scientific method and the evidence are what educationalists should be using, why you’re happy to go beyond the evidence with the conclusions you draw from it.

        Evaluation of learning styles models shows that some elements of the models have some reliability and validity. That’s *all we can say*, based on the evidence. If sufficiently interested we could look at the elements more closely to see what’s going on. But you’re right, that doesn’t necessarily have any bearing on what happens in K-12 classrooms. It certainly doesn’t imply that there’s any justification for using learning styles models in K-12 classrooms.

        However, you then go on to lump together ‘learning styles (as commonly interpreted)’ – presumably the reified construct – with ‘learning styles theories’ that ‘make testable claims’ and dismiss them both completely. The distinction between a theory that makes a testable claim and a reified construct isn’t a ‘semantic argument’, it’s a distinction that would have to be made by anyone testing a hypothesis.

        You might be happy to lump the two together, but that’s not using the scientific method or relying on the evidence, that’s you making a personal judgement. Someone using a learning styles model in the classroom on the basis of a few reliable/valid elements wouldn’t be using the scientific method or relying on evidence either. Again, that would be a personal decision. Personal decisions are fine, but they are not the scientific method.

        You then go on to conclude that learning styles theories are like Russell’s teapot even though you’ve said they make testable claims. The whole point of Russell’s teapot is that it isn’t a testable claim.

        You’re right to be sceptical about learning styles in the light of the evidence, but dismissing them entirely is you making a personal decision, not you applying the scientific method and relying on the evidence.

        I am not defending any ‘notions’. I’m trying to defend the scientific method and the way evidence is used.

      • I predicted that this would turn into a semantic argument and so it goes. You put too much weight on the reliability of an instrument that has no potential impact on the classroom. It’s rather pointless and I will be taking no further comments.

        I make personal judgements all the time when I make decisions about what to do in the classroom. I do not restrict myself entirely to the scientific method not have I ever claimed to. In this post, I have suggested that we pay more attention to the scientific method when people make up new constructs.

        Specific learning styles theories do make testable claims, unlike Russell’s teapot; claims that they generally fail to meet. However, as you have demonstrated, there are myriad different learning styles theories and I cannot a priori falsify all of them. That is the similarity.

  2. Nick Hitchen says:

    ‘And I’m not on about those who are giving kids iPads or anything as daft as that’

    Buy your argument, but not the line above: why would giving kids a tool – a tool you can use to write blog posts and leave comments, for example – be daft?

  3. Pingback: A response to Dan Meyer | Filling the pail

Comments are closed.