New South Wales government promotes unproven educational fad

The paradox of education is that you can increase spending on it without actually improving outcomes. I will always be in favour of spending more money on education but I also believe that it is immoral to not ensure that this money is put to good use. The paradox of education is therefore a pressing issues for educators to deal with.

For instance, I lived through the Blair government’s massive injection of cash into schools. Tony Blair became Prime Minister in 1997, the year I started teaching, and over roughly the next ten years, education spending grew at the largest rate since the 1970s; 5.1 % or so in real terms. Yet I felt that, in 2010, I was teaching a senior physics course that was not as demanding as the one that I had been teaching in 1997 and that students were starting the course less well prepared. This is my subjective experience and the objective data on what happened over this time is strange. Whilst GCSE and Advanced level exam grades soared, performance on international tests flat-lined (see Chart 1 here). I can only draw the conclusion that the improvements in exam grades were illusory; the result of a gradual reduction of standards.

You might well argue that exam grades or international tests are a bad measure. If so, I think the onus is on you to suggest an alternative. I don’t think it is acceptable any longer to ask the taxpayer for ever larger mounts of spending on trust. We need to be able to demonstrate what we can do with it. We need to be accountable.

One reason for the paradox of education might be that we spend the money on the wrong things. John Hattie made a splash a few years ago by arguing that reducing class sizes is not only expensive, it is ineffective. If he is right and if we target money in this way then this could be a cause of the problem.

I also think that another factor is a tendency to invest resources into unproven programs that make educators feel good but that have little to no empirical evidence that they improve any student outcomes. A prime example is the idea of Project Based Learning and that’s why I wanted to highlight that the New South Wales government is currently promoting the use of Project Based Learning (PBL).

As the government make clear, this is an idea that is deeply indebted to the American philosopher, John Dewey. He held a naturalistic view of learning: we learn by doing. Just as we effortlessly learn to speak and to walk by engaging in the activity, school learning should follow the same path. This view now seems at odds with cognitive science. Academic subjects are relatively recent inventions whereas speaking and walking have been around long enough – and are so critical for human survival – for us to have evolved how to pick them up without explicit training.

Moreover, Dewey’s approach to learning through inquiry has been around for over a century. If it was going to revolutionise education then it probably should have done so by now. It is hardly cutting-edge. Naturalistic learning tends to mutate over time and gain different monikers. And yet William Heard Kilpatrick wrote about ‘The Project Method” in 1918. The central idea to PBL, inquiry learning and all the variations is that students are guided to find things out for themselves whilst addressing a pressing question that, ideally, has some clear relevance to their everyday lives (this, incidentally, was why Dewey was sceptical of students learning the traditional subject of history and preferred them to start the social studies curriculum by considering their own family – the ‘expanding horizons’ model adopted by the new version of the Australian Curriculum).

There is very little empirical evidence to support such a Deweyan approach. The evidence that does exist tends to come from poorly controlled studies. When Hattie examined such studies he found some small evidence of an effect – he also commented that very few educational interventions result in a negative effect because the odds are stacked in their favour. Yet he also found that, even so, inquiry learning, problem-based learning and the like were less effective that more teacher-led, explicit methods. This paper explains why.

I suspect that Hattie copped some heat for this from the educational community because it seems that he has been trying to play down these findings ever since. Recently, he even made a video about when inquiry learning might be appropriate i.e. at the end of a period of study when students have learn the subject-specific knowledge that will make it more effective.

Yet the New South Wales government explicitly rules this out: It’s not proper PBL:

“Unlike projects that are tacked on at the end of a unit (this is called Project-Oriented Learning), the projects in true Project-Based Learning are central to the learning. Projects are typically framed with a Driving Question that is open-ended and promotes students to investigate, research, collaborate and present their conclusions to an authentic audience.”

We need to mature as a profession, stop promoting stuff because we like the ideology and start choosing strategies because they are effective. You know, like doctors and engineers do.


10 thoughts on “New South Wales government promotes unproven educational fad

  1. Across Canada the relative performance of the provinces in Math, Science and Reading in the PISA and PCAP assessments is almost perfectly INVERSELY correlated with per-pupil spending on education. When one maps the amount of CHANGE in per-pupil spending over the last 15 years against the CHANGE in scores on those assessments you see the same picture. Provinces that hugely increased their funding crashed the furthest in educational outcomes, as a general rule.

    (Source: StatsCan, PISA 2003 and 2012 reports, PCAP and SAIP reports 2000-2011)

    R. Craigen, WISE Math

  2. John says:

    An interesting case study on the failures of PBL involved the University of Liverpool’s medical degree. In 1997 they switched to a PBL curriculum, which was abandoned a couple of years ago. The University of Lancaster still runs a PBL course; whether they will continue to do so long term is another matter entirely.

  3. You’ve often responded to the ‘jobs which don’t even exist’ perspective by saying we should teach kids the knowledge which has stood the test of time. I’m inclined to agree with this, though I have trouble justifying it. At the moment I am teaching linear diophantine equations to high achieving mathematicians who will almost certainly never use them, though I hope will appreciate them nonetheless. I find this more satisfying than teaching them e.g. statistics, which might be more applicable to their futures. But how do you convince someone who genuinely doesn’t see the value in number theory, poetry or the like? Also there is a somewhat circular aspect to the justification – when we say we teach what has stood the test of time, that means we’re teaching it because we taught it before.

    • This intersects with another argument of mine about whether education is purely preparation for work. Partly, it is about passing on culture and educating youngsters in that culture. The equations you describe may not be useful to specific individuals but they also might be. And they might have value to society more generally, otherwise we epic probably have forgotten about them.

      • Stan says:

        You could ask what would you teach if it was only that which almost everyone uses after completing their education. It would not be a lot of math. No trig, no calculus, even probability is not actually used by most people. High school math would be optional.

        If it was just for those going into science and engineering careers you could specialize earlier.
        Yes I know that might exclude some coming late to interest in math but more focus on those that really have an interest might be a better use of time.

        My guess is that a good reason to teach everyone about high school math is to remove the mystery. If everyone knows math is about ideas that work independently of the person using them: assume these ideas, then this algorithm always works or this other idea is always true. Then showing this over a range of mathematical domains is reinforcement – it works in geometry, it works in algebra and it works when we put these together in trig or calculus.

        A good read on the topic is Underwood Dudley’s What is Mathematics For?

        Click to access rtx100500608p.pdf

        Of course there is also that math is a really good way to measure if someone can understand complicated ideas. So if you want a way to decide who to give scarce university positions to or who to hire its a very useful metric.

  4. There have been some notable successes with the use of discovery learning and perhaps problem based learning (depending on what you define as PBL). Modeling Instruction, as exemplified with physics modeling, is a program based around students creating and developing a model that is then deployed for explanatory and predictive purposes. I’ve never come across any significant criticism of modeling instruction, and I’m wondering if you have. It deserves to be said that modeling instruction is definitely not inquiry, as the situations, topics and line of inquiry are heavily guided/scripted by the teacher. However, there is no question that the modeling cycle is student centered, constructivist and discovery based.

    • I am not familiar with Modelling Instruction. The evidence is clear that pure discovery doesn’t work – – because a lack of guidance ignores cognitive principles of how we learn –

      Obviously, the more guidance you add to your discovery learning approach, the more you will mitigate the problem. However, the logical conclusion is to keep going and keep adding guidance until you are using explicit instruction. The evidence for the effectiveness of explicit instruction – for novice learners – is quite overwhelming, backed by the process-product research of the 1950s-1970s, the strategy instruction research of the 1980s and the problem solving research that led to the development of cognitive load theory, amongst others.

      In contrast, I am aware of no rigorous, properly controlled research that shows a benefit of using any element of discovery with novice learners when compared with explicit instruction. The closest you get is probably the productive failure work of Manu Kapur where a short period of open ended problem solving is used prior to explicit instruction. This shows some benefits but I am still unsure about the controls used and it needs to be replicated.

  5. Pingback: Education Nation | An Interview with Professor Masters – Century 21 Teaching

  6. Pingback: Useful bits and pieces – A Chemical Orthodoxy

Leave a Reply

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

You are commenting using your 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.