Hattie on inquiry learning

Penny Bentley (@penpln) recently tweeted a link to the following clip where John Hattie discusses inquiry learning:

I have a few comments that I would like to make.

Hattie’s view is consistent with cognitive load theory

It is reasonable to propose the use of inquiry learning once sufficient domain knowledge has been developed (even if Hattie’s choice of the pejorative term ‘shallow’ is unfortunate). In fact, this is a structural feature of our education system where the culmination of study in a particular area might be a research-based masters degree or a doctorate. However, it is also possible to conceive of ways of employing inquiry more immediately.

Imagine a science unit of work where, instead of a test at the end of the unit, there is a test two-thirds of the way through. Students may then move on to conducting some form of inquiry within this domain. Perhaps it might be contingent upon them demonstrating sufficient knowledge and understanding in the test and this would therefore act as a hinge-point, with some students being retaught the concepts rather than continuing to the inquiry.

We then might derive some significant gains from the inquiry component. Students will maybe deepen their understanding through application. For instance, if students had sufficient grounding in projectiles then a projectiles-based inquiry would further develop ideas around experimental methods and uncertainties, the effect of air resistance and so on.

It ain’t going to happen

The problem with such a model is that the reason teachers choose inquiry is linked to students having little domain knowledge of what they are inquiring into. Teachers think that students finding things out for themselves is motivating. They confuse the way that experts gain knew knowledge in a domain – epistemology – with the best way of teaching well-established knowledge to novices – pedagogy. This fundamental confusion between experts and novices is both widespread and misguided.

A student who already knows the theory behind projectiles is going to have a good idea about what is likely to happen and is not going to develop their own idiosyncratic hypotheses. Yet proponents of inquiry tend to value the ‘skill’ of inventing such hypotheses.

At best, inquiry enthusiasts will promote ‘just in time‘ teaching of concepts i.e. the focus is on the inquiry element, with teaching of domain knowledge minimised to as great an extent as possible.

The precise opposite has been proposed

The book “How People Learn” by Bransford et. al. for the National Academies Press in the US is highly influential and certainly good in parts. If anything, it can be be given credit for being freely available and for helping raise the profile of cognitive science in education, but I have two main problems with it. The exemplars of good teaching do not reflect what we know from research, and the following passage from Chapter 1 seems to conflate explicit instruction with ‘simply providing lectures’ and is in direct opposition to the position described by Hattie [my emphasis]:

“Fish Is Fish (Lionni, 1970) and attempts to teach children that the earth is round (Vosniadou and Brewer, 1989) show why simply providing lectures frequently does not work. Nevertheless, there are times, usually after people have first grappled with issues on their own, that “teaching by telling” can work extremely well (e.g., Schwartz and Bransford, 1998).”


12 thoughts on “Hattie on inquiry learning

  1. Yes, I agree, I object to the term “surface” knowledge. As for inquiry learning, it could also be argued that the reason it fails is that it is asking to much of teachers who struggle to design these lessons well. It is time consuming for both teacher & student and the student is usually better off simply being given the explanation. In primary, I see it largely as a waste of valuable class time. Primary students really need to concentrate on “accumulating” content knowledge. Also, is it not possible that inquiry/problem based learning doesn’t work well because it is just not a very good way to teach? Hattie thinks it’s introduced to early, but perhaps direct instruction is still superior to inquiry regardless of when it’s introduced. It may well just be a sub-standard way of teaching…

  2. I’m totally in agreement with your central idea here that it is usually best to teach the underpinning knowledge first and only then let learners loose.

    For me the classic example is circuits where I would always tend to teach the theory first, illustrated with appropriate models, and then use practical work (quite tightly guided) to deepen understanding. It’s hard enough to understand without having to extract the rules from practical results that are complicated by internal resistance, bits of contact resistance, and incorrect circuits. Essentially start simple and then gradually add context and complexity.

    I would disagree with one bit of what you say here, though. You say “For instance, if students had sufficient grounding in projectiles then a projectiles-based inquiry would further develop ideas around experimental methods and uncertainties, the effect of air resistance and so on” but I would argue that the LO should either be about the physics of projectiles (which might include finding out about air resistance through a carefully scaffolded investigation), or about methods and uncertainties. Millar (2009) is very good on this distinction.

    Millar, R. (2009) Analysing practical activities to assess and improve effectiveness: The Practical Activity Analysis Inventory (PAAI). York: Centre for Innovation and Research in Science Education, Department of Educational Studies, University of York

    To my mind, you are conflating inquiry-based learning here with learning investigation skills (one being a pedagogical approach and the other an area of science subject knowledge). It’s maybe a minor point but by separating these LOs it is possible to have practical work that is planned purely to reinforce conceptual understanding, or carefully scaffolded investigations that use an inquiry-based approach (e.g. to learn about something like the effect of air resistance on projectiles), or activities that develop children’s knowledge of investigation skills (like methods and uncertainties) but I think it’s a mistake to try to combine these.

    The strength of investigations that use an inquiry-based approach is partly deepening of understanding, as you suggest, but also they help demonstrate that science is not just a fixed and dusty body of knowledge determined by people with special powers, and it helps children to understand what science is about. Of course, open-ended investigations are complex so teachers need to explicitly teach some things (e.g. CVS, how to measure and record data) and often strongly scaffold others (e.g. range of independent variable, number of repeats). The critical thing is that they are not used to try and teach investigation skills and conceptual knowledge at the same time. This is an error I see a lot.

    To summarise a rather long comment: knowledge first; then adding context and complexity, which might include inquiry-based work; knowledge of investigation skills should be taught separately.

    Best wishes

  3. You might want to investigate ‘A Private Universe’, a report and Video documentary, produced by the Harvard-Smithsonian Center for Astrophysics, under a grant from the Annenberg Foundation .

    The webpage is: https://www.learner.org/resources/series28.html

    An introductory video can be found here:

    For some reason, the idea is proposed that we teach using direct, didactic instruction, because it works so well. In fact, this idea that things were better in ‘the good old days’ and that ‘today’s students just don’t measure up’ has been around for decades, in all subjects, in all countries.

    The simple fact is, the old one-size-fits-all methods just didn’t work all that well, and newer methods are seldom fully adopted…

    … So they may not work that well, either.

    The simple fact here is that knowledge evolves. It wasn’t all that long ago when respected physicians treated patients with leeches, and bled them to ‘let the bad vapours out’. Newer techniques, such as antibiotics and even sterilization of tools or handwashing, were met with scorn by the traditionalists.

    The fact that you don’t understand it doesn’t make it wrong. Just makes it new.

      1. Direct Instruction done poorly will produce poor results. Many who are striving to use an inquiry lesson format are not doing it well, and so the evidence tends toward poor results.

        Without a concurrent measure of fidelity, to ensure the instructional model is being implemented well (or at all, as happens when novices mis-label a pedagogical technique), the conclusion ‘inquiry learning is not effective’ is without basis.

        In fact, Hattie agrees that Guided Instruction (a version of inquiry learning in which the instructor matches lesson design with appropriate scaffolding) is of high effectiveness.

        Again though, it is difficult to do guided instruction well, and so a fidelity measure is required.

      2. That’s not my reading of Hattie. His view is that the place for inquiry is after students have learnt the content, as far as I remember. You seem to be dismissing the large body evidence against inquiry learning on the grounds that, in these studies, it hasn’t been done very well. What I am interested in is the evidence that convinces you that inquiry learning is effective.

  4. Inquiry and the National Science Education Standards: A Guide for Teaching and Learning (2000)


    Canadian Education Association:


    Concept to Classroom: Inquiry-Based Learning


    It’s important to note that I’m not simply claiming that “…the large body evidence against inquiry learning (should be dismissed) on the grounds that, in these studies, it hasn’t been done very well.” I’m claiming that, but also that lots of poor practices have been generalized as ‘Inquiry Learning’ just because it’s easier to demonize them in a lump.

    ANY practice that is under study needs to be narrowly defined, tested for fidelity, and controlled for student variation. If you teach Rich Kids the same way as you teach Poor Kids, they will still learn more, just because of other advantages. (like, they have more social experience with learning). If you teach average kids with an effective technique REALLY WELL, and use the same technique BADLY on other average kids, the ones you taught WELL will do better.

    The purpose of metanalysis is to discover patterns – deeper research is needed to find out WHY. if you don’t control for things, your metanalysis is useless.

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 )

Google+ photo

You are commenting using your Google+ 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.