Beyond ‘Constructivism versus Direct Instruction’

In Debates and Polemics by Alex Quigley18 Comments

This morning I read an excellent blog post by Martin Robinson (@SurrealAnarchy), writer of ‘Trivium 21c‘ – see here. I was particularly struck by this sage statement:

“Our need is not to fashion an easy answer but to hold the competing ideologies together in an awkward contradictory balance, as we do in liberal democracies, for education is human and it belongs to all, whether conservative or radical, and it is frail, flawed and fantastic because of it.”

It was an excellent counterpoint to the debates over teaching methods that characterise much of the discourse on Twitter, in universities and beyond. I find this battle of ideas and methods useful because at least it indicates that a great deal of thought and interest is being invested in how we instruct our students so that may learn best and succeed. I have written before about how I am averse to binary thinking that refuses to countenance all the complexities of the minefield that is the real classroom. Ultimately, each school and each classroom has its own unique context and therefore any pronouncements about research into the effectiveness of methods of instruction should be treated as useful guidance and not didactic commands.

Ideology and theory usually doesn’t translate easily into ‘real’ practice.

A really useful summary of the research into teacher effectiveness, conducted by the CfBT Education Trust, provides a concise and excellent summary of the debate between constructivist and direct instruction methods. These are the two warring faction of teaching and theories of instruction (put simply: direct instruction puts the teacher at the centre of instruction, whereas constructivism puts the student at the centre). The report can be found here. Once more, specific elements of the report struck home, reiterating the message of Martin Robinson, about the often contradictory, but not mutually exclusive, balance between different methods of teaching and learning. I have selected the following quotation from the report as it best articulates my position in the debate:

“The relative utility of direct instruction and constructivist approaches to teaching and learning are neither mutually exclusive nor independent. Both approaches have merit in their own right, provided that students have the basic knowledge and skills (best provided initially by direct instruction) before engagement in ‘rich’ constructivist learning activities. The problem arises when constructivist learning activities precede explicit teaching, or replace it, with the assumption that students have adequate knowledge and skills to efficiently and effectively engage with constructivist learning activities designed to generate new learning.” (Rowe 2006)

from page 24 of the CfBT report.

The report highlights the importance of teacher led approaches, particularly for younger students, who are effectively novices, and particularly disadvantaged groups of students. Dr Ken Rowe, an Australian educational researcher (who tragically died in the bush fires of 2009), explored this issue with interesting findings.The following research in this area is suggested by the report: Galton et al. (1980); Mortimore et al. (1988); Muijs & Reynolds (2000); Rowe (2006). This appeals to my educated intuition that students who are novices require much more direct instruction, before reaching a stage of some expertise, at which point learner centred activities (such as assessment for learning strategies or constructivist methods, like co-construction) can be integrated into our teaching methods with success.

My personal example would be the difference between my current year 8 SEN group and my year 13 English Language A level group. I teach my year 8 students with a great deal of direct instruction. They are consistently reliant on my guidance. There is lots of teacher led writing- modelled and scaffolded appropriately. That being said, I have them working in pairs regularly, such as finding word definitions for Beowulf, and their combined efforts have proven constructive and successful (with carefully chosen pairings!). Individual students work with the TA in being ‘meaning moles’ finding tricky new words as we read, before ‘teaching’ the class their meanings.

With my year 12 group we have recently started Child Language Acquisition. I am very much taking a predominantly direct instruction approach in the early stages to develop their knowledge; however, I will encourage constructivist independent research and presentations on aspects of CLA to ensure students have the capacity to undertake independent study (of course, this is essential for any further education they undertake). We will also move to some reciprocal teaching strategies, with students leading on teaching the revision of aspects of CLA. It works well and sees successful results.

For each lesson there is an appropriate strategy, for every scheme of learning there is a combination of methods required. The most effective teachers have a strong grasp of a range of teaching methods and they have the agility to negotiable between different strategies for different groups and individuals.

Every group is, of course, different. I don’t think any pure method of instruction wholly translates to the complicated and nuanced range teaching and learning that characterises the typical classroom. Englemann’s pure Direct Instruction, with lesson scripts and quick fire questioning etc., does not translate to the needs of my A level class; whereas a project learned based approach for my year 8s would still require a serious amount of direct instruction and would have to be teacher directed to succeed. Pure theorising about teaching approaches becomes a limitation on teachers trying to do the best for their students.

We should continue the debate, we should undertake research, conduct research and synthesise the best of evidence, but we should be wary of every finding a ‘answer‘ or ‘solution‘ to the most effective method of teaching. Teaching and learning is far too “frail, flawed and fantastic.”


  1. Hi Alex

    Great post. A few points.

    The evidence you cite does not really support an equal balance between the two approaches. Ken Rowe is extremely hostile to constructivism, questioning whether it is even a legitimate theory of teaching.

    Direct Instruction has a very specific meaning – it is the system of programmed instruction developed by Seigfried Englemann. In this system, teachers do not plan their own lessons. Instead, they defer to a script produced by an expert planner. I think that you are really describing what I would term ‘explicit instruction’ where a teacher tells / explain the key concepts. DI would be a subset of explicit instruction more generally.

    I agree that the expert reversal effect means that students need less guidance as they move from novice to expert. However, the centre of gravity in much educational practice means that this often happens too early. Whenever new content is being delivered, students are effectively novices and so the evidence around direct / explicit instruction applies.

    Indeed, the followers of Englemann in the US would take issue with the idea that direct instruction is less important at higher levels of knowledge. The DI schools there teach advanced algebra in the same way that they teach remedial grade 6 maths. They would argue that it is much more efficient because the scripted process anticipates and mitigates common errors and problems.

    I don’t advocate Englmann’s DI but I would support explicit instruction for all new content.

    Constructivism also has damaging effects other than those associated with minimal guidance. It delegitimises the teacher as an expert. Teachers see themselves as needing to ‘engage’ students with ‘relevant’ and ‘authentic’ tasks; a sort of ‘customer knows best’ mentality. This pulls teachers away from focusing on cognitively taxing problems and forces them to introduce potentially distracting contexts. It also creates debilitating levels of guilt around a failure to deliver on nebulous concepts such as differentiation.

    A well researched post.


    1. Author

      Thanks for the reply Harry. I agree, Rowe favours direct instruction (I aimed for lower case to try to differentiate from Engelmann’s pure version), but he does temper that view in some quotes, whilst strongly questioning constructivism. His quote obviously priorities direct instruction, therefore it isn’t some even balance. In many cases I would prioritise direct instruction too, as I conveyed in my examples, and I would likely spend more curriculum time using direct instruction methods, but I haven’t analysed this. That being said, I use methods from both perspectives on a daily basis, so I don’t argue for one or anger when in debate.

      I wrote a whole paragraph of Englemann’s DI, but I cut it out because the post was becoming too cumbersome and was losing the message. I used direct instruction, lower case, because like the report I cite uses it. It has become a generalised term beyond just Engelmann, particularly outside of America, where Direct Instruction has stronger roots. I agree, I am taking about what would easily be termed explicit instruction. But, having quoted the report, using ‘direct instruction’ I think I would have confused people and required another explanatory paragraph! Whilst not thinking I have some vast readership, I did want some readers who haven’t come across the debate to find an accessible way in.

      With regards to your last point, I do carefully weigh up the points you make. I have issues with the associations pinned on ‘engaging’ learning, but I think that ‘authentic’ can add value. I reckon I’d need a post getting to the root of hose definitions! I loath the ‘customer knows best’ drive in education, without a doubt. For me, it isn’t an issue just related to teaching methods and I don’t attach it to one method or another really. It is more about parental attitudes and responses to school and the behaviour (irregular from personal my experience) patterns exhibited by some students.

      1. Sorry. I didn’t mean to be a pedant but I understand why you might think that. You are quite right to use direct instruction in the way that you did. However, I prefer explicit instruction as I think it makes things… well… more explicit. There is a key point about the scripting of lessons that otherwise gets lots. I wasn’t complaining, just trying to point that out.

  2. Great post! I agree that constructivist approaches and direct/explicit instruction are not mutually exclusive and the dichotomy is damaging in educational discourse. I believe that explicit instruction is very important, especially to ensure acquisition of knowledge and fluency with skills but only a well designed question/inquiry enables learners to construct deep understanding, which they can then apply with independence and insight in original situations. I would also argue that novice vs. expert has little to do with age and more to do with mastery in that field. Early childhood experts who work to build students’ capacity as learners, inquirers and wonderers, would disagree that only teenagers are ready for rich, engaging, student-led activities. Very young children, led by their natural curiosity and passions, construct meaning every day. With the addition of an expert teacher to guide, instruct, question and provoke, the level of thinking, processing and understanding can be increased.

  3. An brilliant post bridging the evidence from research with common sense with common sense and judgement.

    I think you have supported very well the idea that Direct Instruction and Constructivism are not mutually exclusive. For me the need to define the terms is very important. The Direct Instruction/direct instruction/explicit instruction confusion thing is very important for me as is the Constructivist/ constructivist ideas/constructive learning/constructive teaching/constructionist confusion. These for me make a big difference to implementing the ideas effectively.

    The best post that I have seen on the topicand with Harry’s reply even more illuminating. Hopefully other will join the debate here.


  4. I’m curious about the resonance between the point made by Martin Robinson and the notion, initially propagated in models of ‘giftedness’, of “Tolerance for Ambiguity”, which is an underpinning of a great deal of my work with my own students when asking them to extend and develop their work in English.

    I feel strongly that the tendency towards the perpetration of all of these polar opposites in educational discourse in the UK comes from some (alien to me) need for there to be a distinct right and wrong. This is probably the deeper cause for my concern over the absurd distortion of education caused by the slavish adherence to GCSE grades as being the sole ‘outcome of education’ and also gives me cause for concern about the current drive towards an instrumental view of education as a whole.

    It heartens me to hear you, with the significant influence you have, expressing a nuanced view. I want my students – and my colleagues – to become a lot better at encompassing paradox and contradiction, as to me, that competency is going to become one of the keys to success in the future.

    Thanks – as always – for your thought-provoking (and great looking) blog.

    Chris Waugh

    1. Author

      Cheers Chris. I don’t know about having any influence really, but I get tired of the binary debates that fail to recognise the reality of the classroom is far more nuanced than singular methods or ideologies. From what I can see, the vast mass of ordinary teachers are outside of much of these polarising debates, busy getting on with teaching and fending off their many external pressures. The binary thinking largely stems from political thinking. It is why governments and many union positions annoy me and they usually don’t further the profession or improve anything in the classroom. I note the ‘Royal College of Teaching’ idea went very quiet. Too sensible perhaps?!

      1. In my mind Willingham’s ‘factual knowledge must precede skill’ principle is correct. It depends, however, on ‘natural curiosity’. Occasionally (when nature doesn’t provide) that curiosity needs a kick start with something interesting. Otherwise the teenage brain just carries on thinking about peripheral matters in their life. They may not see teachers as who we think we are to them.

        It’s a bit like the whole part whole method of coaching skills. Let ’em have a go. Break it down and build it back up.

        I agree that the plethora of variables make any one method being ‘right’ impossible to argue or believe in. If it works well, it works well but it might not elsewhere.

  5. Pingback: Beyond ‘Constructivism versus Direct Instruction’ | HuntingEnglishHuntingEnglish | The Echo Chamber

  6. Pingback: Drills, skills and being match-fit. | headguruteacher

  7. Alex, what do you make of Hattie’s summary of 900 meta-analyses over 18 years in Visible Learning? ‘Too often direct teaching is portrayed as bad, while constructivist teaching is considered to be good… This is almost directly opposite to the successful recipe for teaching and learning.’ … ‘Every year I present lectures to teacher education students and find that they are already indoctrinated with the mantra “constructivism good, direct instruction bad”. When I show them the results of these meta-analyses, they are stunned, and they often become angry at having been given an agreed set of truths and commandments against direct instruction.’

    1. Author

      Although there are valid questions about effect sizes and how the meta-analysis averages are calculated, I think there is much to be gleaned from ‘Visible Learning’. I would credit it as being a book that was a turning point for me. Less for some Damascene conversion, but more for engaging me inr research and the pursuit of evidence. Ironically, a year later, I found out a great deal about Hattie’s meods that made me pause. Such is the nature of research in our adaptive, complex context.

      I think his quote here is quite compatible with my post, although to state there is a ‘successful recipe’ is rather overstating it as his many effect sizes hardly compose a synthesized and harmonious whole. I think many of the positive impact effect sizes show some success, in some contexts, for many direct instruction and some constructivist approaches. I think my exemplars and my focus on direct instruction as the basis for new knowledge stands in partial agreement with Hattie! as I reject he rejection of ‘direct instruction’.

      I would say my repertoire includes ‘direct instruction’, or the less pre-packaged ‘explicit instruction’; reciprocal teaching; student self-assessment and self-grading; spaced practice; feedback; concept mapping; mastery learning; worked examples; some student centred teaching; questioning; the Read/Write writing programme and many more I’m sure are buried in the specific evidence of individual studies. I would find it hard to cohere my approaches in any one ‘recipe’. Hattie, whose work I value (I have just purchased his latest book on the brain), does simplify here by using those two terms and saying either “recipe” is an exact science. I agree, any mantra that holds “constructivism good, direct instruction bad’ is a nonsense. Many of my methods would be characterised under the broad church Hattie uses with his label! ‘direct instruction’. I also value the work of Vygotsky and I think that learning does have a strongly social context, whether we want it to or not! I think Nutthall’s ‘The Hidden Lives of Learners’ reveals a fascinating really of covert feedback and social communication that is integral to the classroom. I will post on that one day I am sure.

      I genuinely remember very little of my teacher training. I wasn’t engaged in research like I am now. In my work with trainees now I would have little time for such mantras like “constructivism good, direct instruction bad’. The reality is much more complex, nuanced and relative than that. I would encourage the ‘best’ strategies through reflective practice and engaged research. The important thing is to be engaged in research, critically and with as much educated intuition as we can muster.

  8. Pingback: 365 days in my shoes Day 283 | high heels and high notes

  9. Pingback: Denying the Debate | Scenes From The Battleground

  10. Thanks for a well written and considered post.

    I’ve been following the direct instruction/unguided constructivism debate for a while now, and the actual debate mirrors aspects of your own teaching approach – novices need guidance. Possibly less as they achieve expertise, and experts may benefit from a lot of freedom.

    Kirschner, Sweller and Clarks papers (there are two, the initial one, and a repsonse to criticisms) seem to be the key and kernel, and specifically make their points against unguided constructivism, and for direct instruction, specifically where novices are concerned – which would seem to tally with your intuition. The main paper includes this (the unguided and novice provisos) in the byline, but it often gets skipped over when people describe the debate. Which is unfortunate.

    Admittedly their first paper might read as lacking in some regards. They do seem to wander off target a little, and make some claims that are above and beyond their evidence, but the second paper does a good job of clarifying the position, and overall they are clear. Their case is with regard to unguided constructivism and novices, though they do suggest the question may be asked about higher levels too.

    Some of the Constructivist replies have seemed to eke out a common ground – Schmidt, I think, argues for scaffolding of activities within Constructivist practices, and says it commonly occurs, which the Cognitivists take as a form of direct instruction embedded in Constructivist practice. They seem to be saying that scaffolding, as put forth by constructivists, is pretty much what they are arguing for, and an admission that they have a point.

    The constructivist evidenve for unguided constructivism for novices seems weak, at best, and at times inept, or, frankly, bizarre and the congitivist evidence for guided instruction of novices seems reasonably strong, and spans several decades of data, and some seemingly good RCTs.

    The dichotomy in the argument is both less than is supposed (there is some common ground, with some constructivist commentators, and probably a lot of actual constructivist practioce. Constructivists do directly instruct in practice, and cognitivsts are aware of this, and point to it as an example of what they might mean) and more specific – The Cognitivists/Direct Instructionists argue that they have good evidence specifically relating to novice and near novice contexts, and admit there seems to be some evidence that the less guided pedagogies may result in good or better outcomes at high levels of expertise.

    My own experience of lecturers describing the debate are in terms of binary opposition and dichotomy (“there’s been an attempt to do down constructivism” most memorably), and typically take the shape of statements suggestiing it’s a general attack on Constructivism in all contexts, and of all stripes. My reading of the papers makes the debate seem, though divided, less binary, and more specific and nuanced.

    As a side note, Bandura, who is a fairly big influence on some of the pro direct instruction cognitivists, punted pretty much this compromise structure in some of his seminal self-efficacy work in the seventies, arguing that novices need structured guidance with challenge level approrpriate tasks and a lot of corrective feedback from experts, but once they achieve mastery need to be allowed to experiment relatively freely in self determining ways with their skills in order to ensure retention.

    There’s links to pdfs of the main papers at the bottom of this page

  11. Pingback: Why we shouldn’t close down the skills-knowledge debate | Pragmatic Education

Leave a Reply