A response to Doug Holton

Last year, Doug Holton wrote a blog post about ‘The evidence for various research-based instructional strategies‘. I went to the trouble of commenting on this blog. It appears that my comment has now been deleted. That’s fine: Doug has every right to delete comments from his blog. 

I never had a response to my comment – I suspect that Doug is unable to respond – and Doug has recently been linking people to the original blog post on Twitter. I kept my own copy of my comment and so I reproduce it below:

Hi Doug.

Thanks for this – very interesting.

I am someone who thinks critically about ‘constructivist’ approaches to teaching. I use this term with caution as I accept many of the premises of constructivism as a theory of learning. Conflating a theory of learning with a theory of teaching is what Mayer calls ‘the constructivist teaching fallacy’ (see http://psycnet.apa.org/psycinfo/2009-09809-010) and so I try to separate these ideas.

I would like to make three points.

Firstly, I did not get my views from Wikipedia. In fact, I first found out about Kirschner, Sweller and Clark’s 2006 paper from a reference in John Hattie’s ‘Visible Learning’. The paper is quite significant, in my opinion, having sparked numerous responses, a conference and a book. I am certain that you did not intend this but stating that, “many bloggers today get their opinions about education and psychology from Wikipedia,” gives the impression that those who disagree with you are rather simple-minded and amateurish. You might want to correct this impression.

Secondly, I would suggest that it is wrong to conflate lecturing with explicit instruction. If we were to run a trial, for instance, where a non-interactive lecture was compared to a lecture where clickers were used to gain live audience feedback then I would predict that the latter would be more effective. It is also closer to explicit teaching. Any lecturing within an explicit teaching approach is highly interactive and explicit teaching will also involve subsequent periods of individual practice and teacher feedback (See Rosenshine for a full definition: http://psycnet.apa.org/psycinfo/2009-09809-011). The key feature of explicit teaching is that new concepts are fully explained. Students are not required to work any of this out for themselves.

The effectiveness of EI for novices is in accordance with the predictions of cognitive load theory (CLT). However, CLT also predicts that students who are more expert in a domain will benefit from less guided approaches; something known as the expertise reversal effect (See Kalyuga et. al. http://www.tandfonline.com/doi/abs/10.1207/S15326985EP3801_4#.VR8p_fmUd7E). This would mean that experiments conducted with students who already have considerable domain knowledge will not invalidate the value of explicit instruction for novice learners. 

Finally, it is often asserted that constructivist teaching approaches are superior for transfer of learning whereas explicit instruction is only any good for “rote” recall. “Rote” implies recall without meaning. It’s actually very difficult to remember things that have no meaning to you and so I suspect this term is often misused, although I accept that it has rhetorical power. Klahr and Nigam investigated whether students finding things out for themselves leads to superior transfer than explicit instruction (see http://lexiconic.net/pedagogy/KlahrNigam.PsychSci.pdf). It did not.

Thanks for the discussion.

What do the public need to know about science?

In my last post, I shot a few weary arrows towards adherents of the fashion for describing school subjects as ‘literacies’. My objection is that this takes a productive word and renders it vague. My suspicion is that people tend to do this in order to lend substance to the nebulous, significance to the trivial and prestige to a set of skills that they’ve just made-up.

There were quite a few comments on the blog itself and on Twitter and an issue arose that I would like to address in more depth. The issue is that of supposed ‘scientific literacy’.

One of my most vivid recollections of training to be a teacher was reading a piece by Robin Millar; “Towards a Science Curriculum for Public Understanding.” It was a dispiriting paper about which I was required to write my own response, now lost to a nondescript landfill somewhere in West London. I can no longer access Millar’s text, even as a PhD student. It seems that my university does not have electronic copies of the School Science Review. However, Millar has made similar arguments elsewhere and is by no means alone in making them. Indeed, Alex Weatherall pointed me towards a GCSE course that aims to teach scientific literacy.

The basic idea sets up a false choice between the kind of science that students need to learn in order to prepare them for further scientific study and the science that ordinary members of the public need to know. The contention is that much of school science is about arcane, technical facts and that many students see these as pointless and irrelevant, switching them off science entirely. Therefore, they miss out on learning about the processes of science, something that would serve them well as citizens and consumers.

It is a superficially appealing idea. It’s a classic educational shortcut of the kind often dreamt of by educationalists. ‘What if we could deliver the benefits of a scientific education without all the dreary bits?’ sits up there with ‘What if we could teach reading by just exposing kids to books?’ and ‘What if we could just get children thinking like historians without having to learn all that damnable history?’

Once you expose the hairy backside of this idea it seems less appealing. Advocates will talk of teaching a few ‘big ideas’ such as evolution and climate change alongside a whole lot of process; the scientific method, peer review, correlation versus causation. This fuses seamlessly with the vogue for ‘inquiry’. Soon we will have students who can parrot a formula for constructing a hypothesis, yet who don’t know the difference between a series and a parallel circuit.

And, of course, it doesn’t work. It is an attempt to teach critical thinking skills directly without teaching the content that students need to think critically about. One of the most ridiculous pieces of pseudoscience that our students are likely to meet in their adult lives is homeopathy. It is obviously nonsense. I know this because I know the science. I know what water is made of. I know about molecules. I understand what repeated dilution will do. I understand this because of my content knowledge.

How would knowledge of the scientific method help me evaluate homeopathy? Perhaps I could run my own trial or ask homeopaths about their evidence? What about looking for peer-reviewed sources? Well, they exist.

As ever, there are no shortcuts in education. Well-educated adults still need to know quite a lot of basic science. Scientific ‘literacy’ is not required.

Bonfire of the literacies

This post has one message: If it’s not reading or writing then be suspicious if it’s described as ‘literacy’.

I am well aware that ‘literacy’ has a number of meanings. Google gives me two:

1. The ability to read and write

2. Competence or knowledge in a specified area

It is the second usage that I object to because it is one of the many manoeuvres that make the process of education less easy to understand. It acts against clarity and allows vague and ill-defined skills and competencies to feed off the credibility and self-evident value of that first definition; the ability to read and write. Everyone understands that learning to read and write are essential. These are probably the two objectives that all those involved in the education debate can agree upon, even if some promote dubious methods to achieve them.

Firstly, it is not clear that even reading, writing, speaking and listening should be grouped together as one thing that we can call ‘literacy’. Speaking and listening are skills that humans have likely been utilising for many millions of years. We have therefore evolved mechanisms for picking them up implicitly from our social surroundings, hence the seemingly effortless way that most children learn to talk. Yet reading and writing are inventions of the last few thousand years and for much of this time, they were the preserve of a tiny elite. This is why we invented schools in order to efficiently pass on this unnatural, artificial skill to large numbers of people.

This doesn’t mean that we cannot teach children anything that will improve their speaking abilities: we might stress enunciation or pass on knowledge of rhetorical devices, knowledge that would also apply to writing. But we don’t have to start from the very beginning like we do with reading and writing so the processes are qualitatively different.

It becomes even more confusing when we start to refer to any specific body of knowledge as ‘literacy’. Why do people talk of ’emotional literacy’ or ‘visual literacy’? It’s instructive that with well-worn school subjects we don’t tend to add this redundant term. We don’t generally talk of learning ‘scientific literacy’ or ‘historical literacy’; we talk of learning ‘science’ or ‘history’. Yes, the term ‘scientific literacy’ does crop-up but it is often when someone is trying to sell a dumbed-down version of a science course with all the hard bits taken out. I think this is telling.

Let’s reverse the perspective. Why would we talk of learning about ’emotional literacy’ rather than learning about ’emotions’ which is, presumably, what we mean? Well, it seems a bit lightweight and insubstantial to talk of ‘learning about emotions’. Why not talk of learning about ‘images’ instead of ‘visual literacy’? The same problem arises – does that really warrant extended classroom time? Is it really as important as learning to read?

No. The ‘literacy’ bit has been stuck on the end in order to give the subject matter more gravitas. It is there to hijack the esteem we have for genuine literacy. It is a bait-and-switch approach that diverts attention from the fact that we are asking children to engage in fluffy, insubstantial and largely pointless activities instead of actually teaching them something worthwhile.


Ed Schools versus Consultants

Which group has been most damaging to the teaching profession; schools of education or education consultants? Here’s what I think.


The most bizarre ideas in education are definitely propagated by consultants. These are the folks who gave us Brain Gym, for instance. And the worst training that I ever received was on Building Learning Power. Imagine sitting around a table with a group of science teachers being forced to discuss ways in which we could better exercise our students’ ‘learning muscles’. The upshot of that was a compulsory routine that we had to go through at the end of each lesson, complete with its own box on the school lesson plan proforma. 

Horror stories about consultant-led professional development sessions abound, from the dreaded role-play to someone trying to plug software that is both pointless and unreliable. Feel free to share your own examples in the comments.

Yet consultants have been responsible for some of the best professional development that I have had. There are one-or-two behaviour experts out there that are worth listening to. And I quite fondly remember the start of the “Key Stage Three Strategy” in England: A consultant from the local authority came to visit and we had training on formative assessment strategies, some of which I still use. It was the first time that I was asked to read “Inside the Black Box” and that opened my eyes to some genuinely useful education research.

Education Schools

If consultants are responsible for the extremes of professional development then the sins of education schools are mainly sins of omission. I thoroughly enjoyed my Postgraduate Certificate of Education, taught by urbane and interesting researchers. But there were great swathes of relevant content knowledge and pedagogy that were not addressed. There was little on how to manage behaviour, for instance. The only pedagogical approach going was essentially constructivist teaching, although it wasn’t described that way at the time. Instead, it was generally referred to as ‘teaching’. When I recently leafed through my teacher education text, I noticed that there was nothing on direct instruction, DISTAR or explicit instruction, despite there being a wealth of research evidence on these approaches. Instead, where explicit teaching was referred to at all, the pejorative label of ‘didactic‘ was used. Even when we were guided towards using group work – a preferred approach – we weren’t told about the evidence on how to ensure that it was effective.

Apart from some basic psychology, the content knowledge of most use to emergent science teachers would be that of students’ misconceptions about science and ways to address these. I wasn’t taught this stuff although we were encouraged to investigate misconceptions through our project work. Perhaps it is too much to ask for in a single year course. Perhaps there’s not enough time. But this wouldn’t explain why primary school teachers who have generally studied for four years have not been taught about graphemes and phonemes.

The result

On balance, I think consultants do the most damage. Despite there being good ones around, the propensity to focus on gimmicks and the very silliest ideas whilst also setting the agenda in many poorly-led schools means that they do the most harm. At least schools of education are generally more sober places where reflection is encouraged and where we can be assured that the most foolish notions are unlikely to ever be put into practice.

Fishing for termites

Chimps have a culture. Watch them fish for termites with twigs that they have fashioned for the purpose:

This behaviour was first scientifically observed in the chimps of Gombe National Park in Tanzania by Jane Goodall. At the time, this was a remarkable discovery because tool use was considered a defining characteristic of humans. We now know that chimps use a variety of tools in different ways. However, of most interest to me as a teacher is the fact that this tool use is cultural; different groups of chimps use tools in different ways and they pass these uses on through the generations, seemingly via imitation. This is not a genetically programmed behaviour.

So a nascent form of education is at work here.

How would you describe the skill that the chimps are learning through this process? I’ll give you two options and you decide which you think fits best:

  1. The skill of stripping a twig of leaves and then using it to ‘fish’ for termites
  2. The skill of acquiring food

Clearly, the first description is more specific. But what of the second one? Does it have any value? It implies that ‘acquiring food’ is something that can be learnt in a general sense. It implies that learning how to collect fruit might, in some way, help you learn how to fish for termites. And yet it is clear that these are two quite distinct activities.

This is what I mean when I suggest that there is no such thing as a skill of ‘analysis’. It’s too general an idea. In order for chimps to learn to fish for termites, they need to be shown how to fish for termites. If we want students to learn how to analyse a poem then we need to show them how to analyse a poem. And that’s it.

Bloom’s Taxonomy

Bloom’s taxonomy was developed in the early 1950s by a committee led by the educational psychologist, Benjamin Bloom. The cognitive domain of this taxonomy – which is usually what people mean when they refer to “Bloom’s taxonomy” – is ubiquitous in education and is probably one of the few concepts that most teachers have heard about. It was revised in 2000 and the original and revised versions have a slightly different emphasis, with nouns having turned into verbs and an equalisation of the top three layers:

Ashman Version

The taxonomy is a method for classifying educational objectives. Does it fit with what we know about research? Well, perhaps this question is moot. The only way that we could base a classification of learning objectives on anything scientific is if we had firm evidence that there were different brain processes involved. We certainly did not have such evidence in 1956 and we don’t have much now. However, I don’t think Bloom and his committee made or implied such a claim.

There are aspects that I like about the taxonomy. One way of reading the hierarchy is to suggest that we clearly need to know something before we can progress to applying it and so on. However, there are other forms of exegesis that are deeply troubling.

Higher order thinking skills

The top end of Bloom’s taxonomy – the parts that I’ve coloured yellow – are often known as ‘higher order thinking skills’. Again, the ‘higher order’ part is just another classification and, as such, is pretty neutral, although I would disagree with the idea of calling these objectives ‘skills’ and I will explain why later.

The main problem arises when educators start to infer that these are superior kinds of objectives. This can lead to the suggestion that these objectives should be emphasised over the ‘lower order’ objectives of merely remembering rote, disconnected facts. Most teachers who have worked in schools over the last 20 years have probably been exposed to some kind of process where observers come and look at classrooms and then pronounce that not enough higher order questions are being asked or tasks set.

I don’t think that this is what the authors of the taxonomy would have wanted. I think the intention is to emphasise the crucial, foundational importance of knowledge; knowledge first and then everything else. But this doesn’t seem to be how it is interpreted in the wild. As Karen Tankersley suggests in this extract from her book on higher order thinking, “There is simply too much information in the world for us to waste students’ time with regurgitations of basic facts.”

The conjuring into being of a class of thing

The other major problem with Bloom’s taxonomy is that it encourages us to imagine that there is a stable thing that we can label ‘analysis’, ‘evaluation’ or whatever. We assume that ‘analysis’ involves a certain set of mental processes and that, by practising analysis, we may become better at analysis. The trouble is that it’s not at all clear that the mental processes involved in analysing a bar chart have anything in common with those involved in analysing a TV commercial. The domains are completely different. To become proficient at analysing bar charts we must first learn about bar charts before seeing different examples and different contexts. After practice, we may reach our objective. However, once we move on to TV commercials we must start again from the bottom of the hierarchy.

This, at least, is what the research tends to show: First you need a solid base of knowledge. Analysis, evaluation or creativity are not skills that can be trained or exercised like a muscle and then applied flexibly, as required. If we think that they are then we might miss out the vital steps of teaching the domain knowledge that true expertise requires.


I think it is time to put Bloom’s taxonomy out to pasture. I know less about other generic learning taxonomies such as SOLO but they do, superficially at least, seem to suffer from the same issues of interpretation. Instead, we should invest our efforts in attempting to sketch-out out the pathways of progress that are followed in developing an understanding of specific concepts. These would be of real, practical value to teachers, and far more use than a set of cliches to be trotted-out at the next whole-staff meeting.

Update: @penpln has sent me a link to this paper that well worth reading.

The urgent need to improve teacher education

Imagine a medical school where the graduating students demonstrated little knowledge of anatomy. Now imagine that all medical schools were like this and the failure to adequately teach anatomy was an open secret. Picture a medical school professor who, when challenged, wrote largely empty tracts about how anatomy was an important part of medicine but that a key focus also had to be on ‘making healthy’.

Such medical schools would probably not produce the doctors that we need. We would see a decline in population health and particularly so amongst the poorest and most disadvantaged because they would lack the resources to make healthy choices for themselves or to seek paid help from alternative practitioners.

This is the situation as it stands in teacher education. The equivalent of doctors not learning about anatomy is teachers of early reading and writing not learning about basic linguistic constructs such as graphemes and phonemes. For instance, could you identify how many speech sounds there are in the word ‘box’? This is key to being able to teach phonemic awareness. Could you identify which of the following words has a soft ‘c’; Chicago, cat, chair, city? Do you know what a voiced consonant digraph is? Indeed, do you know the difference between phonemic awareness and phonics?

These were amongst a set of questions put to seventy-eight Prep teachers from schools across the Australian state of Victoria (Prep is the first year of formal schooling). The answers showed that knowledge of these ideas was ‘limited and highly variable’. Worse still, the teachers were asked to rate their own knowledge and ability to teach these concepts and yet these ratings had no relationship to the teachers’ performance on the knowledge questions. Interestingly, the longer a teacher had been teaching, the more likely he was to highly rate his ability to teach phonemic awareness and phonics.

The Dunning-Kruger effect is relevant here. We don’t know what we don’t know and so we may overestimate our abilities. Also, cognitive dissonance plays a part; it is hard to accept the idea that we might have been teaching something poorly for many years and so the better alternative is to assume that we have been teaching it well. The more years we have invested, the stronger the effect.

I do not blame teachers. Dunning-Kruger actually absolves them of much of the responsibility.  You train to be a teacher and you assume that you are told the stuff that you need to know. If you’ve never heard of a diphthong then you won’t know that you haven’t heard of it.

And I don’t blame teacher educators. They probably know as little about these concepts as their students. Why would they know more given the culture in which they have built their careers? Once inducted into the guild, you start to see things through the guild’s eyes. Phonics is important blah, blah, blah, BUT there are other things that we need to focus on such as ‘making meaning’ – whatever that means – and other vague and general, good-sounding things.

The guild has its own answer for educational under-performance: You cannot make-up for poverty through education. Except that nobody is claiming that you can. It’s like the person whose job it is to install your new boiler claiming that she can’t make up for the fact that it’s winter. Nobody is asking teacher educators to end poverty but we are asking them to better prepare new teachers.

This idea is met with incredulity. The guild has its ritual enemy. Sporadically, it erupts into a two-minutes hate against neoliberalism or politicians who won’t commit more funding to schools. Education academics know that these demands are not politically possible and so, like a minor political party promising both tax breaks and extra spending, they can keep on advocating for Utopia whilst blaming the evil overlords for thwarting it. And all this without ever having to examine their own practices; their own part in the current state of affairs. The problem is caused by someone else.

I have argued previously for the disruption of teacher education. This is not because I believe that school-based training or the other alternatives will be any better. I am looking for a circuit-breaker; something that will jolt teacher educators out of their group-think of philosophical blather, make them take a hard look at what they are doing and start teaching the educational equivalent of basic anatomy. This is urgent.

But let us imagine for a moment that we simply decided to do this for ourselves, disruption or not. Imagine that we could improve teacher preparation from within the world of education; no government imposed targets, no tests. We could do that and we could claim those successes. Perhaps, in time, we would demonstrate that we don’t need harsh scrutiny or accountability measures. Education academics, having achieved real successes, might start to be listened to by the government and be given some rope to run policy without the mediating influence of think-tanks, columnists and irritating bloggers.