Whatever happened to constructivism?

When I started blogging eight years ago, constructivism was a big deal. Whenever I wrote about it, which was often, plenty of people would appear and tell me that I didn’t understand it. So I even went to the lengths of creating a FAQ post to address the points they made. Yet now, constructivism seems to have faded. It seems to be mentioned less in academic papers and far less in the blogosphere. So what happened, which side, if any, won and what lessons can he take from this?

Constructivism is a theory of learning that comes in a variety of flavours, from radical constructivism to cognitive constructivism. What they all have in common is the idea that learners are not blank slates – that new knowledge does not simply accumulate in the mind but has to slot in to the old knowledge that is already there in some way. One way we may look at this is through schema theory. Knowledge is organised in the mind in the form of schemas – think concept maps – and new knowledge is either added to the relevant schema or, in some circumstances, changes the schema. Although there is debate about exactly what takes place and how, schema theory remains a well-accepted concept within cognitive science and educational psychology. If true, constructivism is the way we learn, whatever teaching method is used.

However, constructivism doesn’t end there, in what Richard Mayer has described as the ‘constructivist teaching fallacy’, many have drawn implications for how we should teach from the theory of constructivism, designing constructivist teaching methods. These are thought to go with the grain of how we learn and therefore be more effective. The idea that knowledge is constructed in the mind is interpreted as a need for learners to construct their own knowledge in the classroom through approaches such as inquiry learning that involve specific kinds of behavioural activity.

If you see a similarity to the methods proposed by late nineteenth and early twentieth century progressivists such as Herbert Spencer, William Heard Kilpatrick and the impenetrable John Dewey, this is, I believe, no coincidence. Those people who are drawn to the older ideas are drawn to constructivism. It’s less a case of people being persuaded by the logic of constructivism and more that constructivism provides a scientific framework and therefore justification for pre-existing preferences.

And that’s its weakness. Ideas are not scientific because they are true – that’s a common misconception. A idea or model is scientific because it is testable. And we can test constructivism, more or less.

Probably the biggest test was the late twentieth century experiment with whole language reading instruction. An analysis of this is beyond the scope of this post, suffice it to say that asking small children to construct their own knowledge of letter-sound relationships has been found to be less effective than explicitly teaching them. See, for example, the 2000 U.S. reading panel report.

Nevertheless, like a teenager at a party whose date is kissing someone else, constructivism hung around in the kitchen for far too long. It’s usefulness, it seems, had not yet been spent. So, what happened next?

There was a slow move away from constructivism among cutting-edge educational psychology researchers. I would say this started with the 2006 publication of Kirschner, Sweller and Clark’s paper that outlined theoretical problems with, and empirical failings of, the model. This was met with a lot of sound and fury at the time, the echoes of which stretched well into the 2010s. There were three rebuttals published in the same journal followed by a response from the original authors. There was a debate organised at the American Educational Research Association conference in 2007. This was followed by a 2009 book in which both sides of the debate were given the opportunity to air their views. In his concluding comments, Sigmund Tobias, one of the editors, summarised his own thoughts on constructivism:

A careful reading and re-reading of all the chapters in this book, and the related literature, has indicated to me that there is stimulating rhetoric for the constructivist position, but relatively little research supporting it.

From then on, constructivism was living on borrowed time as the gradual shift away from it in the research community worked its way out into the wider educational ecosystem where constructivism probably still survives in some dark corners.

But remember that the teaching practices associated with constructivism preceded their constructivist justification and they will undoubtedly outlive it. At the moment, they are looking for a new host. For instance, the identity politics movement will perhaps be used to label explicit teaching as upholding white supremacy or something. However, the inherent absurdity and deep unpopularity of identity politics, coupled with the damage it does to genuine attempts to tackle intolerance and injustice, lead me to think its days are numbered.

Researchers seem to be rallying around self-regulated learning theory. This is a neat trick. If you define the issue as one about the most effective ways for learners to teach themselves new knowledge then you can draw on a whole lot of valid research. We can teach learners about retrieval practice, spaced practice and so on. The question of why we want learners to teach themselves new knowledge, when we have schools and universities and lots of people within them who are employed to teach knowledge to them, remains unaddressed.

Research interest in self-regulated learning theory is probably what’s behind the UK Education Endowment Foundation’s drive to promote ‘meta-cognition and self-regulated learning’. In that case, we have another motte-and-bailey argument where, under the banner of meta-cognition and self-regulation, all manner of curious agendas are advanced such as Philosophy for Children but, when questioned on this, proponents can retreat into arguments about teaching students how to use retrieval practice.

So, we are not done. Constructivism may have fallen for now and the battle may be won. But the war goes on.


14 thoughts on “Whatever happened to constructivism?

  1. Philosophy for children seems an utter waste of time. It does seem mostly to be about relationships and the odd ‘moral’ decision, but bears little relationship to actual philosophy – most of which is not accessible to children btw.

  2. Unfortunately, constructivism’s evil, fraternal twin, “discovery learning” is the focal point of the next Generation Science Standards adopted by 20 states including NY and CA. Students are required to become pretend mini-scientists as they try to discover the facts, laws, and principles of science. This is a recipe for misunderstanding, inefficiency, frustration, rejection, and little learning.

  3. Here are 2 recent meta-analysis with effect sizes close to 1 in favor of PBL-

    concluding – “The results showed that the overall mean weighted effect size (d+) was
    0.71, indicating that project-based learning has a medium to large positive effect on students’ academic achievement compared with traditional instruction.”
    Concluding – “The result of the meta-analysis shows that the project-based learning approach is more effective than traditional teaching methods about the academic achievements of the students in science classes. The effect size of the Project-based learning approach, between 0.777-1.218”

    • Hi George. I will have a look at these. However, I am very suspicious of any meta-analysis with an effect size close to 1. The problem with meta-analyses is similar to the problem with Hattie’s method. We are taking different studies, carried out using different methods, on different students and with different control conditions and trying to compute a metric such as an effect size. Robert Slavin explains these issues well here:


    • Chester Draws says:

      The result of the meta-analysis shows that the project-based learning approach is more effective than traditional teaching methods about the academic achievements of the students in science classes.

      This always raises my eyebrows. In the Turkish setting, what is “traditional” learning?

      I teach by direct instruction, but my classrooms are not in any sense traditional.

      If we set up Project based learning against everyone seated in rows, with dull textbooks and uninspired teaching, then I’m willing to believe it is better. Turkey, judging by PISA does not have great teaching as a norm. But I have my doubts when compared to the sort of teaching most direct instruction fans in a less rigid system would have.

      (My biggest objection to project based learning isn’t that it isn’t successful. It’s that it is massively time-inefficient. It is not helpful if I can teach one topic really well using it, only to have to cram the rest of the syllabus into less time for the rest of the year. Studies tend not to examine that aspect.)

      • They defined Traditional as lecturing+question + answer. PISA scores are not relevant to the study, also the Chen study included studies from all over the world. The problem of variations in traditional, problem or any other method, is an issue with the Sweller paper too. Sweller et al define PBL as minimal guidance. When you read some of the studies, many of them are NO guidance whatsoever. So i could argue, that is not the way I use PBL. Regarding efficiency, I ‘m not sure that is a major criterion for a teaching method. For example, some of the key skills that kids have to learn in maths, like division/fractions, they have 10 years to master.

      • Chester Draws says:

        It is true that primary school has more time for projects. But then most modern Western schools have lots of projects already. There you are preaching to the converted.

        Secondary school has no time for learning that slowly. (Especially since many of the project taught kids end up in my classrooms at the start of high school with basically nothing but arithmetic, and we have to almost teach from scratch. Another reason I tend not to believe much in PBL.)

        We don’t get to ignore data just because it is deemed “irrelevant”. If the quality of teaching in Turkey is low, then any halfway decent method will be better. I see very little evidence that Turkey teaches well and that PBL is improving on an already good system.

        What I want to see is PBL doing better than the norm in Singapore or Japan.

      • Chester Draws says:

        Glad we agree.

        There’s been a lot of studies out of the US that show direct teaching techniques are clearly superior to Constructivist ones — from Direct Instruction onwards.

        The powers that be across the West think that this means nothing, because they continue to move in the other direction. So clearly they believe studies out of the US can be ignored. Who are we to doubt them?

        To believe ones out of the US that show PBL is superior — while ignoring ones that show direct methods are superior — would be hypocritical at best.

      • Don’t think we agree. This blog argued Constructivism has disappeared. I presented 2 recent meta-analyses and also the major change in Hattie’s position, which show it has not disappeared. You focused on the Turkish study and rejected it because Turkey has mediocre PISA scores, my point was USA has mediocre PISA scores too. You’ve not addressed the Chen study nor Hattie’s new position. I have not totally dismissed the direct or explicit strategy. I’ve said the studies that support those approaches over PBL have a definition of PBL that most teachers don’t use – i.e. minimal guidance. The PBL studies do suffer similar problems in the definition of PBL Until all this is resolved, i have not dismissed either approaches. I’ve just watched Adam Boxer’s presentation on dual coding, and i found it very helpful, but i will still use a PBL approach to START some topics i teach in maths, then move to more explicit strategies after. I am yet to see a study that compares that approach with anything else.

    • thanks Greg, excellent research. It has caused me to consider changing my approach. But, it’s also good to read you say , “Nevertheless, there may be sufficient evidence in the literature to indicate that problem solving first is effective under some circumstances.”

  4. Yes agreed, the effect size comparisons across different studies is dodgy. But, some of these meta-analyses are an interesting read. Hattie has been promoting PBL/Inquiry based learning for last couple years which is at odds with what he said years ago that there is zero evidence.He is promoting his theory of learning – surface to deep to transfer and claims the ‘jigsaw method from PBL is best method for Deeper learning. I suspect this will sell a lot of books.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.