Does homework actually work?

It is 5:30pm on a Tuesday. Your child has been at school since 8:45am. They are tired, probably hungry, and have been concentrating in a structured environment for six hours. And now, before dinner, before play, before any kind of decompression, there is a worksheet about fronted adverbials that needs to be completed. Or ten spellings to practise. Or a maths sheet that neither of you fully understands the point of.

Most parents comply, because the school sent it, and schools presumably know what they’re doing. But here’s the thing: on the question of whether primary school homework improves learning, the research is remarkably consistent. And it doesn’t say what most schools would have you believe.

What the evidence actually says

John Hattie is a New Zealand education researcher who has spent decades conducting what is effectively the largest systematic review of educational interventions ever attempted — synthesising data from tens of thousands of studies covering millions of students. His Visible Learning project assigns each intervention an “effect size”: a measure of how much it moves the needle on student achievement. The average effect of any intervention is 0.40. Anything below that is producing less than typical school progress.

Homework at primary school level gets an effect size of 0.29. That is below average. It is, in statistical terms, barely distinguishable from not doing it.

At secondary school, the picture shifts. Hattie’s data gives secondary homework an effect size of 0.64 — genuinely meaningful, above the threshold where an intervention earns its place. The research distinguishes clearly between the two. What works for a fifteen-year-old reviewing material they’ve already understood does not translate to a seven-year-old completing tasks that were set without knowing whether they were ready for them.

Harris Cooper and thirty years of looking

If Hattie is the broadest voice on this, Harris Cooper at Duke University is the most specific. Cooper has been studying homework for over three decades and has conducted multiple meta-analyses on the subject. His conclusion, stated clearly in his 2006 synthesis of the research: for elementary school children, there is no consistent relationship between homework and academic achievement.

None. Not a weak one. Not a mixed one. No consistent relationship.

For middle school students, there is a positive but modest link. For high school students, the relationship is meaningful, though even there Cooper found diminishing returns beyond two hours per night. His research gave rise to what’s become known as the “10-minute rule” — endorsed by the National Education Association in the US and referenced by many UK practitioners — which suggests that at most, homework should amount to 10 minutes per year of school. Year 3 children: 30 minutes. Year 6: 60 minutes. Secondary students: no more than 90 minutes across all subjects. Most schools send considerably more than this.

Cooper’s explanation for why homework helps older students but not younger ones is intuitive once you hear it. Homework consolidates learning that has already happened. It doesn’t introduce new material effectively. For that consolidation to work, the student needs to have understood the material well enough to practise it independently, have the metacognitive skills to notice when they’re going wrong, and be able to sustain focused attention without a teacher in the room. Primary school children are still developing all three of those things. Sending them home with material to practise independently is, in many cases, asking them to do something they are not yet developmentally equipped to do well.

The Finland problem

If you needed international evidence to sit alongside the meta-analyses, PISA provides it. The Programme for International Student Assessment — the global benchmark of educational achievement across 79 countries — consistently shows that the countries which assign the most homework are not the highest performers. Finland, whose students routinely rank among the best in the world for reading and mathematics, assigns almost no homework at primary level. Its school day is shorter than the UK average. Its teachers have more planning time. The system is built around the premise that what happens in school should be sufficient, and that evenings belong to children and families.

This is not to say that Finland’s approach would translate directly to the UK. Education systems are embedded in culture, infrastructure and teacher training in complex ways. But it is a useful corrective to the assumption that more time on task automatically means better outcomes. It doesn’t. It means more time on task.

What homework does instead

The research on what homework actually produces — beyond the modest and inconsistent academic effects — is worth knowing about. Several studies have examined the non-academic outcomes, and they are not encouraging.

A 2015 Stanford University study of over 4,000 students found that excessive homework was a significant source of stress, with students reporting that homework was their primary stressor above examinations, grades, and social pressures. Stress at that level has documented effects on sleep, physical health, and motivation toward school. A separate body of research on primary-age children finds that homework assigned in excess of what the child can manage independently is associated with negative attitudes toward school — the opposite of what education is supposed to produce.

There is also the family dimension. Homework that children cannot do independently becomes homework that parents do. Research by Angeline Lillard and others on parental involvement in homework suggests that when parents provide direct help — telling a child the answer rather than guiding them to it — the academic benefit disappears entirely, and the interaction frequently generates conflict. The family evening, already compressed, becomes a battleground over something whose educational value was marginal to begin with.

So why does homework persist?

If the evidence is this consistent, why are primary schools still sending home spelling lists and reading logs and topic tasks every week? Several reasons, none of them about the research.

Firstly, parental expectation. Schools in many areas feel pressure from parents who believe homework signals academic rigour — that a school without regular homework is a school not taking learning seriously. This belief is not supported by the evidence, but it is genuinely held, and schools respond to it.

Secondly, habit and tradition. Homework has been part of the school structure for generations. Changing it requires a school to actively defend the decision, which takes effort and sometimes courage that individual headteachers may not feel they have.

Thirdly, the research hasn’t fully reached the classroom. Teacher training in the UK does not consistently include evidence-based education research as a core component. Many teachers assign homework because they were assigned it, because their school expects it, and because no one has presented them with a compelling reason not to.

There are schools that have made the change. Brighton and Hove primary schools began moving toward homework-free primary education based on the evidence over a decade ago. Several secondary schools have replaced traditional homework with structured reading programmes for younger year groups. The results, in each case, were positive or neutral — not the drop in standards that critics predicted.

What this means for your household

If your child’s school assigns homework, you’re unlikely to be able to opt out entirely. But you can be selective about how much family energy you spend on it, and you can prioritise the things that the evidence says actually matter.