I was wrong. Here’s what I’d change about education research

What happens at BERA, stays at BERA. Why the trad/prog fight is weakening the field – and what we can do about it.

There’s a big education research conference happening at the start of September. It’s big. Maybe the biggest, most important one in the UK. The latest research gets shared. Boundary-pushing discussion takes place.

Another big education research conference happens at the start of September, too. A lot of people go, but when they talk about ‘evidence’ and ‘methods’, some say they’ve got the wrong idea.

Some say they ask the wrong questions.

Some say they’re closeted and partisan and do not listen to outside views.

One is called BERA. It’s for academic researchers in education.

The other is researchED; it’s practitioner-facing, for teachers who want to understand the science of learning.

But which is which?

The problem I didn’t see

We’ve got a problem, but it took me a while to see it.

I spoke at BERA a few years ago. I overheard a couple of colleagues sharing a light joke about researchED. Standard stuff at academic conferences. You’re with your tribe. The other tribe is over there. Everyone agrees they don’t quite get it. Nobody loses any sleep.

I’d put money on versions of that conversation happening at researchED too, in reverse.

Two conferences full of people giving up their time to improve education. And what do we end up talking about? The failures of the other side.

It took a comment from a reader to spell it out for me.

Taking the Argument to the other side

Earlier this month I wrote a Substack post pushing back on Kelsey Piper’s piece in The Argument, which argued that education research is broken. I thought I was defending the field. I was – kind of.

But Marc, a reader, commented underneath and put me right.

Marc, who comes at education from the effective-teaching side, took the time to acknowledge what qualitative research can do that quantitative research can’t – and vice versa. But the point that really got to me was how research gets used:

This is what causes me to think that mainstream education academics are often ideologically-driven: research is good if it supports ‘good’ ideas.

He was applying the line to one side. I think it applies to both, and I’m going to tell you why.

I’d written 2000 words defending the plurality of educational research methods, and underneath I’d taken a swipe at the people I’d most like to convince.

So this is me trying to think about it more carefully.

Trads vs Progs. No, sorry, Reformers and Pragmatists

Let me try a reframe. Imagine there are two groups of people researching and spreading ‘evidence’ about education. (Evidence in air quotes, because they don’t always agree the other side really has any worth talking about.)

Let’s call one group reformers and the other pragmatists.

If you’ve spent time in these circles, this will start to sound familiar. Bear with me. The framing matters.

Reformers tend to think the system is broken and needs to change. They’re interested in students’ experiences, teachers’ experiences, the role of schools in reproducing inequality. They use qualitative methods alongside quantitative ones. They publish in academic journals read mostly by other reformers.

Pragmatists want to get better outcomes within the system as it currently is. They speak directly to teachers – often via blogs and books. They draw on cognitive science and the science of learning. They emphasise classroom experience and what works.

These are ideal types. Most researchers sit somewhere on a spectrum. But the public-facing versions of each camp tend to be sharper. Both contain serious work. Both have produced research that has improved schools – and children’s experiences within them. But both also produce caricatures that flatten the other side.

This makes the debate sound theoretical, but their disagreements shape what happens in schools: Should children be taught through direct instruction, or through more exploratory approaches? Should they learn a knowledge-rich curriculum, or one built around skills? How much should cognitive science shape day-to-day teaching? How should classrooms be run? How should teachers be trained?

Reformers and pragmatists don’t always sit on the opposite side of the fence on these debates – the issue is that they rarely debate at all. At least, not with the other camp. When the reformers acknowledge the pragmatists, they tend to dismiss them as reductive. When the pragmatists acknowledge the reformers, they tend to dismiss them as ideological.

Outside both camps – in the press, in policy, on social media – the dismissals get cruder still.

Why we should care

In February 2025, the US Department of Education’s Institute of Education Sciences was effectively dismantled – around 90% of staff cut, more than 100 active research contracts cancelled, the Condition of Education report missing its publication deadline for the first time in its history. The American Educational Research Association filed a lawsuit; its Executive Director called the cuts “nothing short of a fundamental harm to the education research community.”

The UK hasn’t seen anything like this. But the political appetite for ‘evidence-based’ rhetoric in service of preferred conclusions is not a uniquely American problem. As far back as 2013, Michael Gove was distinguishing between ‘good academia and bad academia’ – bad academia being, conveniently, the academics who disagreed with his curriculum reforms. Former Schools Minister Nick Gibb gave a speech in which he approvingly cited the view that university lecturers ‘have to justify their existence with all that pointless theory.’ Terry Wrigley wrote a BERA blog in 2018 setting out the same dynamic at greater length.

None of this is new. But the more we caricature each other, the weaker the public case for the field becomes – and the easier it is to dismiss. If you want to discredit education research, you don’t need to do the work yourself. You just point at the most heated bit of the trad/prog debate and say look, they can’t even agree among themselves.

At the moment, the two sides are like parents: currently estranged; each thinking they’re the one doing best for the child; fighting for custody.

It doesn’t have to be this way. Instead, we could look at what they have in common: a desire to improve the education system for those in it.

We could see pragmatists differently: as people serious about evidence; willing to translate research for the people who use it; unafraid to argue in public.

We could see reformers differently: as people who hold the structural questions in view and stop us treating each cohort of children as if the system around them didn’t exist.

How can we pontificate about learning when we close our own minds at the first sign of dissent?

The personal touch

After Covid, I quit teaching. It took me a while to work out why – in fact, I had to write a whole book proposal (cheaper than therapy, at least). It was a rollercoaster ride: half autobiography, half blistering critique of the educational system.

It ended with a conclusion. The system was broken. Broken because society’s incentive structure was all wrong. Exams matter because there’s limited room at the top. Outsized benefits for those who do best. An elite who want to protect what they have.

The solution? Rebuild society. Scrap the outsized incentive structures. A few simple changes. First, universal basic income. Second, huge taxation to reduce the ‘winner takes all’ attitude. Third, election by lot, so we don’t have such a huge ratio of Old Etonians on the ballot paper.

I thought it sounded reasonable enough and sent it off to a couple of places. A few polite rejections. And those were the positive responses.

Then the days arrived where you can’t draft an email without getting an LLM to redraft it (more accurately, politely, concisely and closer to your true voice than you could do yourself – although admittedly it’s probably not hungover and two coffees short of optimal functioning). I ran my draft through Claude, hoping it would tell me I’m a genius and to forget about the literary world (which is even more dysfunctional than the world of educational research).

Sadly, it didn’t. All that talk about sycophantic AI? Doesn’t seem like it when I run my ideas through it.

Claude told me the writing was ok, but who was actually going to read it? A teacher might agree with parts, then get to the end and think, ‘right, yeah. Cheers for that.’

Had I been a bit less reformer and a bit more pragmatist, I might have got further. When we have one side without the other, we often fail to get any message across.

Finding problems is the easy bit

I am – in many ways – an extreme example. But I’m reminded of my own ‘if this doesn’t work, nothing will’ conclusions when reading some academic papers and books. Academics are great at posing problems. Great at rigorous methodology. Even better at poking holes in other people’s rigorous methodology. Not so good at the ‘what do we do about this in the real world?’ question.

There’s something comforting about the change-society-or-nothing position. The chances of society being restructured as a result of this post are fairly small. Even if Sir Keir does read it and take drastic action, I can claim he didn’t quite do what I suggested when things don’t work out.

Finding problems is the easy bit. Suggesting practical changes is harder and more dangerous. You put your reputation on the line. Try this, you say. And you’re judged by the results.

So I’m changing my approach. Moaning about the exam system while doing nothing for the actual kids I teach isn’t satisfying. I’ve spent years researching factors that promote and suppress curiosity in the classroom. I’m going to use that knowledge – in my own classroom, and in my writing, in the hope that others can benefit.

The motive is partly selfish: it will give me agency. But there’s also a genuinely more hopeful form of curiosity to focus on – one that teachers have much more control over. I want to spread the message.

I’ll carry on poking holes in the system, because it’s what I do.

I just don’t want it to be all I do.

As a reformer at heart, I specialise in seeing problems. So here’s me exercising my pragmatic muscles. A few ideas for next steps:

Recommendations

1. Don’t weaponise the term ‘evidence’.

Everyone wants teaching to be evidence-based, but when we disagree about what the term ‘evidence’ means, the arguments will rumble on. We need to use the term more broadly. We can and should dispute methods and findings, but we should apply the same scrutiny whether the conclusions slot into our worldview or not. All evidence tells us something about the world; forgetting it shrinks our potential to improve children’s lives.

2. Become comfortable with uncertainty

It’s ok not to have all the answers. It’s ok to have questions. Especially when we speak to practitioners, we shouldn’t overstate the evidence. Data from the lab might be promising. Will it translate to the classroom? Who knows, we need to try it out. Equally, just because we have a small-scale study, we should not try to hide this. Be clear about the limitations and let someone else decide how they want to use your evidence.

3. Get curious about the other side

Attend a conference you would normally stay away from. Ask questions, even when you might get pushback. Don’t be afraid to show your ignorance. This is how we learn. Don’t attend to push an agenda – on either side. That closes doors. The flipside is to ensure transparency in our own methods and findings. Where possible, ensure others can scrutinise the work we’ve done, and prepare to enter into open debate.

4. Share what you’ve found

Education research isn’t blue-skies physics. The whole point is the practical benefit for society – broadly conceived. Researchers have an obligation to share what we’ve found, not as an afterthought, but as a central aim. Especially when that research is publicly funded.

5. Work together to draw up questions

We should discuss the questions, even when we disagree about how to reach the answers. We’re closer than we might think: how do children learn? What should they learn? What are the optimal conditions for this to take place? What other outcomes do we want for them? How do we assess these – including but going beyond exam results. Who is held accountable for all of this? Once we stop talking at cross-purposes, it becomes easier to have a constructive debate.

Moving beyond the old battle lines

Those are my suggestions for the field. Here’s how I’m going to start putting them into practice – on a personal scale.

I’m going to BERA in September to talk about the architecture of the school day, and how it suppresses curiosity in ways most teachers can see but feel powerless to change – hands-up questioning as the only sanctioned form of inquiry, practical work reduced to a tickbox protocol, social pressure that makes asking a question a confession of ignorance. That’s a critical argument. It’s drawn from my doctoral work. It belongs at BERA.

I’m going to researchED to talk about how teachers can engage students’ curiosity in lessons. Make students curious and they will be thinking about the things we’re teaching – and there’s good neuroscientific and psychological evidence underpinning the links between curiosity, learning and achievement. That belongs at researchED.

I think I can do both. More posts coming on this Substack. Some about the system. Some about the classroom. Sometimes both at once.

To me, this is what it looks like to advocate for the field. Disagreement welcome – comments section is open. The version of education research that survives is the one that’s argued for in public, not just in journals.

Thumbnail photo by Arthur Tseng on Unsplash