Now Suddenly Nuance Matters

For years, the debate around mixed attainment teaching in mathematics has often felt less like a debate and more like a declaration. A consensus emerged, or at least the appearance of one, and with it came a curious certainty. Schools were told that the evidence was clear. Setting was harmful. Mixed attainment teaching was morally and educationally superior. Those who questioned the strength of the evidence base, especially in relation to mathematics specifically, were frequently dismissed as reactionary, elitist, or worse.

I know this because I was one of them.

For a long time, I have argued that the evidence base in this area was much less settled than many claimed. Not because I have some ideological attachment to setting, nor because I believe there is a single classroom structure that magically solves the profound challenge of teaching mathematics well, but because I could see that many of the claims being made stretched well beyond what the evidence could reasonably support.

Most of the studies so confidently cited in support of mixed attainment teaching were not mathematics-specific. Some relied on weak proxies for attainment grouping. Others bundled together very different forms of grouping into broad categories that obscured more than they revealed. The complexity of curriculum sequencing in mathematics, the cumulative nature of mathematical knowledge, and the enormous variation in prior understanding between pupils were often treated as secondary concerns, if they were acknowledged at all.

When these limitations were pointed out, the response was rarely a humble engagement with the argument or a closer look at methodology. More often, it was moral condemnation. The debate was framed as settled science. To question it was portrayed not merely as disagreement, but as a kind of ethical failure.

This is why the recent Education Endowment Foundation paper has been so interesting, not simply because of its findings, but because of the reaction to them.

The study suggests that mixed attainment teaching in Year 7 and Year 8 mathematics may harm the attainment of higher attaining pupils, while setting does not appear to harm lower attaining pupils. Although I seek to avoid confirmation bias, I admit it is pleasing to read evidence of what I have long argued.

Unsurprisingly, this has caused considerable discussion.

And here is the important thing: many of the responses questioning the study are entirely reasonable.

People are correct to point out that this is simply one study. They are correct to note that it focuses specifically on mathematics. They are correct to observe that it is confined to a particular age range. They are correct to interrogate the methodology, the limitations, the possible confounding variables, and the degree to which the findings can be generalised.

This is precisely what educators should do with research.

Indeed, one of the great problems in education over the last two decades has been the tendency to turn pithy summaries of research findings into education slogans for immediate adoption, rather than contributions to an evolving body of knowledge. Studies become banners under which people gather, rather than imperfect attempts to better understand extraordinarily complex phenomena. The language of evidence becomes weaponised. Nuance disappears. Certainty expands to fill the gaps where humility ought to be.

Research should always be approached with skepticism, care, and perspective. Educators should maintain enough psychic distance from any study, however compelling its findings may appear, to avoid becoming ideologically captured by it. A single paper rarely closes a debate. In truth, even a large body of research seldom does.

But this is also why the current reaction from some quarters feels difficult to take entirely seriously.

For years, many of the very people now urging caution and nuance insisted that caution and nuance were unnecessary. When others questioned the limitations of earlier evidence against setting, they were often met not with thoughtful discussion, but with ridicule and hostility. The message was clear: the science had spoken, the debate was over, and dissent itself was suspect.

Now, suddenly, methodology matters again.

Now, suddenly, context matters again.

Now, suddenly, we are reminded that evidence must be interpreted carefully, that educational structures are complex, and that broad claims should not be drawn from narrow findings.

All of which is true. All of which was always true.

What the social media response to this new EEF study reveals is not simply disagreement about attainment grouping, but something deeper about the culture of educational debate itself. Too often, we do not merely disagree with one another; we delegitimise one another. We confuse confidence with wisdom. We treat provisional conclusions as moral truths. We use research not as a tool for inquiry, but as a means of social sorting, dividing people into the enlightened and the unenlightened.

One would hope that people working in education, the very business of learning, would remain open to learning from those with whom they disagree.

The reality is that there are excellent schools using mixed attainment teaching and excellent schools using setting. There are schools failing with both approaches. The quality of curriculum, teaching, behaviour, assessment, intervention, and leadership matter enormously. Above all, the subject matters. Classroom structures are not irrelevant, but neither are they silver bullets that will bring solutions across all subjects. Hierarchical subjects are different to non-hierarchical ones and require different approaches. Subject specificity matters.

The task, surely, is to remain intellectually honest enough to follow the evidence wherever it leads, even when it unsettles our prior assumptions, and humble enough to recognise that educational reality is usually more complicated than our preferred narratives allow.

In those moments when new evidence confirms our views, and in those moments when it unsettles them, humility is usually more valuable than the instinct to defend our prior certainties at all costs.

Research should illuminate thought, not close it down.