Homework vs Movement: What Are We Actually Prioritising?

When a primary school child comes home at half past three, what they do with the next few hours matters. Most schools treat this as a settled question: homework is sent home because reinforcing classroom learning outside school hours is assumed to improve outcomes. Physical activity is good for health, but it is a family matter, essentially optional from the school's perspective. If there is a conflict between the two — between a child finishing a worksheet and a child playing outside — the worksheet wins, because it is the school's responsibility and the outdoor play is not.

This assumption is worth examining against the evidence. What does the research actually say about homework — particularly for primary-age children — and how does it compare with what we know about the effects of physical activity on the cognitive outcomes schools care most about? The answer is more lopsided than most educational culture currently acknowledges, and the implications are practical ones that schools can act on.

What the homework evidence actually says

Homework has a long and surprisingly contested evidence base. Meta-analytic reviews consistently find that the overall effect of homework on academic achievement is small — and that effect size is heavily moderated by age. For secondary school students with developed self-regulation and study skills, well-designed homework can meaningfully reinforce learning. For primary-age children, the picture is considerably less clear.

Cooper's influential meta-analysis of homework research — still one of the most comprehensive in the field — found correlations between homework and achievement that were near zero for primary-age students and small-to-moderate for secondary students, and noted that the quality and design of homework tasks matters enormously.¹ Poorly designed or excessive homework does not produce neutral outcomes. It increases cognitive fatigue, reduces intrinsic motivation, and can undermine the consolidation of learning it was intended to support. There is a real dose-response problem: more homework is not better homework, and at primary level, more homework may be actively counterproductive.

The equity dimension compounds this. Homework is not a level playing field. A child with a quiet space, reliable internet access, and an available adult to help has a fundamentally different homework experience from a child without those conditions. At primary level — where the academic effect is weakest — the main effect of homework may not be to reinforce learning but to reproduce and amplify existing advantage. Schools that set substantial primary homework are not raising standards equally. They are, in many cases, doing the opposite.

"At primary level, the main effect of substantial homework may not be to reinforce learning but to reproduce existing advantage. Schools that recognise this are not lowering standards. They are reading the evidence."

Annarie Boor

What the movement evidence says

The evidence on physical activity and academic outcomes is substantially stronger, more consistent across age groups, and more clearly mechanistic. A 2024 systematic review and meta-analysis in PLOS ONE examined school-based physical activity programmes across 42 studies involving over 30,000 children and adolescents, finding positive effects on academic achievement — particularly in mathematics and composite performance — across the age range.³ Classroom-based movement interventions consistently improve on-task behaviour and reduce off-task behaviour, with effects that appear immediately after movement and accumulate over sustained programmes.⁴

The mechanisms are well-established: movement increases cerebral blood flow, triggers neurotransmitter release, promotes BDNF production, and improves the function of the prefrontal cortex — the region responsible for the executive functions that underpin academic performance. These are not speculative pathways. They have been replicated across multiple research traditions, across age groups, and in SEN and mainstream populations alike. The cognitive case for physical activity is, at this point, as strong as the case for any single educational intervention.

For younger children — the primary-age cohort where the homework evidence is weakest — the movement evidence is particularly compelling. Executive function, attention, and emotional regulation are still actively developing in children aged four to eleven, and they develop through embodied, active experience. A child who runs, jumps, climbs, plays games with rules, engages in physical problem-solving, and sustains effortful physical activity is building the cognitive infrastructure that will determine their capacity for sustained academic engagement. A child sitting with a worksheet is practising a much narrower set of skills — and only to the extent that they can sustain the attention and motivation to engage with it at all.

The wellbeing and sleep evidence

There is a third dimension that is often treated as separate from the academic argument but is not. Sleep is among the most powerful determinants of learning, memory consolidation, emotional regulation, and cognitive performance. A 2025 study in BMC Psychology examining the relationships between sleep quality, physical exercise, academic stress, and wellbeing in students found that regular physical activity is positively associated with sleep quality and negatively associated with academic stress — and that both mediating factors independently predict academic outcomes.⁵

Heavy homework loads, particularly for younger children, are consistently associated with elevated stress, later bedtimes, and reduced sleep duration — especially in households where completing the homework requires significant adult support and produces family tension. The academic benefit of a completed worksheet needs to be weighed against the cost of the stress it created, the sleep it displaced, and the physical activity it crowded out. In many cases, that calculation does not favour the worksheet.

"The academic benefit of a completed worksheet needs to be weighed against the cost of the stress it created, the sleep it displaced, and the physical activity it crowded out."

Annarie Boor

What a better balance looks like

This is not an argument that homework has no value, or that schools should abandon it entirely. For secondary students, well-designed homework tasks — purposeful practice, applied problem-solving, genuine reading and research — produce real academic benefit. The issue is not homework in principle but the assumptions that currently govern how it is used and who it is prioritised for.

At primary level, the case for routinely prioritising physical activity over homework completion is strong. Schools that explicitly communicate this to families — that unstructured outdoor play, sport, and physical activity are not in competition with learning but are part of it — are not abandoning their responsibilities. They are acting on the evidence. The families who need to hear this most are often the ones who feel most anxious about their child falling behind, and who are most likely to be creating stress around homework completion at the cost of the outdoor play that would serve their child better.

The question schools should be asking

The starting question is not "how much homework should we set?" It is "what do we want children doing with their time outside school, and why?" If the answer — grounded in evidence rather than assumption — is that physical activity, unstructured play, adequate sleep, and genuine rest are the most developmentally valuable uses of an eight-year-old's out-of-school hours, then homework policy should reflect that conclusion rather than contradict it.

Schools that have made this shift report not a decline in outcomes but a shift in the conditions under which learning happens: children who arrive less fatigued, more regulated, more ready to engage. That is not a soft outcome. It is the precondition for every other outcome a school is trying to produce.

"Movement supports learning readiness and wellbeing. Well-designed homework reinforces specific learning. The evidence is reasonably clear about which of those effects is larger, more consistent, and more equitably distributed."

Annarie Boor