Homework: What the Research Actually Says and How to Design It Well

Homework is one of the most contentious topics in education, and one of the most misunderstood. Advocates cite academic benefits and habit development. Critics cite family stress, equity gaps, and limited learning outcomes. Both sides tend to cherry-pick the research.

Here's what the research actually says — and what it suggests about designing homework that's worth doing.

The Harris Cooper Meta-Analysis Findings

The most cited homework research is Harris Cooper's meta-analysis, which synthesized decades of studies. The findings are more nuanced than either side usually acknowledges:

For high school students, homework has a moderate positive effect on achievement — roughly equivalent to 35 additional minutes of classroom instruction per night.

For middle school students, the effect is smaller but still positive, roughly half the high school effect.

For elementary students, the research finds essentially no measurable effect on academic achievement. The case for elementary homework must rest on other grounds.

The 10-minute rule (10 minutes per grade level per night — 10 minutes in first grade, 100 minutes in tenth) comes from this research. Beyond these thresholds, additional homework time shows diminishing and sometimes negative returns.

What Actually Produces Learning in Homework

Not all homework is equal. The type of assignment matters more than the fact of its existence.

Practice of recently learned skills — applying something taught in the last few days — shows the strongest learning effects. Homework that precedes instruction (flipped learning models) shows weaker effects because students lack the conceptual foundation.

Reading is consistently the assignment with the best return at all grade levels. Independent reading for 20 minutes nightly produces vocabulary growth, fluency, and comprehension benefits that exceed most other homework formats.

Projects and multi-day work are valuable for developing sustained effort, but the benefits come from the cognitive engagement, not from being done at home specifically.

Busy work — completing long worksheets of procedures already mastered — produces frustration, compliance behavior, and no measurable learning. It's also the most common form of elementary homework.

The Equity Problem

A finding that research consistently shows but policymakers often ignore: homework effects are not evenly distributed.

Students with more resources — quiet study space, available adults to help, access to materials — consistently benefit more from homework than students without these conditions. Homework widens achievement gaps, not because low-resource students are less capable, but because the conditions for productive homework are unequal.

This doesn't mean homework is automatically bad. It means that homework design should account for the conditions students are working in, and policies that treat all students' home environments as equivalent are empirically wrong.

Designing Homework Worth Assigning

Assign with a specific purpose. What learning does this homework produce? If you can't answer that clearly, it's busy work.

Keep it completable in the recommended time. An assignment that takes 30 minutes for some students and 2 hours for others is inequitable by design. Preview homework before assigning it by attempting it yourself.

Make it low-stakes. Homework completion and accuracy should not carry significant grade weight. The purpose is practice, not assessment. When homework is high-stakes, it incentivizes dishonesty and parental over-involvement.

Require reading. Independent reading is the one homework practice that produces consistent, measurable benefits across grade levels and is accessible to students without significant resources.

Provide support for students who need it. If an assignment requires internet access, a printer, or adult help and some students won't have those, adapt before assigning.

What to Do Instead of Homework

For elementary teachers especially: the research does not support significant homework. The time might be better spent:

• In-class practice with feedback — the conditions for effective practice are better in school than at home
• Reading incentives — building reading volume through school-based reading programs
• Family engagement that builds connection rather than academic pressure — reading together, talking about school, building curiosity

For secondary teachers: homework works, but less is often more. Assigning an amount students can complete in 20 focused minutes produces better results than 60 minutes of compliance-driven completion.

Talking With Families About Homework

Families often expect homework as a signal that learning is happening. When teachers reduce homework, they sometimes get pushback.

Being clear and evidence-based with families helps: "Research shows that for 2nd graders, nightly independent reading has more impact on reading development than worksheets. Our homework this year is 20 minutes of reading. Here's how you can support it."

Families who understand the reasoning usually support it. The conversation is worth having.

LessonDraft can help you design assignments that produce genuine learning, including practice sequences and independent reading programs matched to your grade and subject.

Homework that students do without learning from it is not just ineffective — it damages relationships with learning and school. Every assignment should be worth the time it takes.