Research analysis: Getting the most out of homework

Written by: John Dabell | Published:
Image: Adobe Stock

What does the educational research say about the merits of homework and about how we should be using and setting homework in secondary education? John Dabell takes a look

If you have a spare couple of minutes and you are feeling mischievous then you can quite easily stir up a hornet’s nest by Tweeting about education’s sacred cow – homework. People have strong opinions on this and they do not hold back...

We have a difficult relationship with homework. Teachers, parents and students all have a view on its effectiveness and those views are often highly charged and pull in different directions (Hallam, 2006).

Some say it is a harmful practice that sabotages family life (and it does). Homework is a parental ball and chain that often leads to meltdowns, tears and slammed doors. It can create anxiety, limit learning, overburden and disengage overloaded pupils and can have a negative impact on wellbeing (Kralovec & Buell, 2000). Alfie Kohn (2006) in The Homework Myth thinks schools should set their default policy to “no homework”.

On the other side of the coin, those who back homework argue that it supports learning, practice and rehearsal, personal development, time management skills and preparation for later life.

But homework feeds on myths and things are not black and white. Much of what is said about homework is based on tradition rather than what we know about effective teaching and learning (Vatterott, 2008).

Decisions have to be made based on what the evidence is telling us and whether the claims for or against homework have a sound empirical basis.

What does the research tell us?

Homework has been extensively researched and studies from across the Western world tell us that it has no appreciable impact (low and moderate) on children’s learning and academic achievement and may lead to poorer outcomes overall. Well, in part.

If, however, we look at the Education Endowment Foundation’s (EEF) Teaching and Learning Toolkit evidence summary for secondary homework then we will find that homework is much more effective with older children. It states: “The evidence shows that the impact of homework, on average, is five months’ additional progress.

However, beneath this average there is a wide variation in potential impact, suggesting that how homework is set is likely to be very important. There is some evidence that homework is most effective when used as a short and focused intervention.”

However, the evidence is also clear that homework has zero effect on achievement for under-11s, as Professor John Hattie found in his Visible Learning meta-analysis (2009).

In 2014 in a BBC Radio 4 interview Prof Hattie said: “Homework in primary school has an effect of around zero. In high school it’s larger ... which is why we need to get it right, not why we need to get rid of it.

“Certainly I think we get over obsessed with homework. Five to 10 minutes has the same effect of one hour to two hours. The worst thing you can do with homework is give kids projects. The best thing you can do is to reinforce something you have already learnt.”

If we delve into Prof Hattie’s research a little further it shows that the effect size at primary age is 0.15 and for secondary students it is 0.64 (the average impact being 0.40). This shows that homework for secondary students has an “excellent” effect, if done well.

As always, what is measured has an impact on the scale of the effect and we are dealing with averages here – so some forms of homework are more likely to show an effect than others. And Professor Dylan Wiliam said at a ResearchEd event in 2014 that “most homework teachers set is crap” (YouTube, 2017).

We all know that homework can be token, poorly defined or even given as a punishment. Homework can be a public relations exercise to make a school look good and a crowd-pleaser to keep parents happy.

We also know that unless teachers ensure that the activities set are meaningful and relevant to current learning, they become largely redundant.

In other words, homework can be effective when it is the right type of homework and we should continue setting it (Kelleher, 2017). Indeed, Marzano and Pickering (2007) say that “teachers should not abandon homework, instead, they should improve its instructional quality”.

Homework does serve a purpose but it has to be purposeful. In Prof Hattie’s own words, when homework is not deliberate practice, it is pointless (The Conversation, 2016).

So what can we do?

MacBeath and Turner (1990) suggest a number of sensible and reasonable ideas:

  • Homework should be clearly related to on-going classroom work.
  • There should be a clear pattern to class work and homework.
  • Homework should be varied.
  • Homework should be manageable.
  • Homework should be challenging but not too difficult.
  • Homework should allow for individual initiative and creativity.
  • Homework should promote self-confidence and understanding.
  • There should be recognition or reward for work done.
  • There should be guidance and support.

Cathy Vatterott, aka the “Homework Lady”, has suggested that “there is a growing suspicion that something is wrong with homework”. In her book Rethinking Homework (2018), she argues that most teachers have never been properly trained in effective homework practices.

Vatterott has also identified five fundamental characteristics of good homework: purpose, efficiency, ownership, competence, and aesthetic appeal (Vatterott, 2010).

  • Purpose: All homework assignments are meaningful and students must also understand the purpose of the assignment and why it is important in the context of their academic experience.
  • Efficiency: Homework should not take a disproportionate amount of time and needs to involve some hard thinking.
  • Ownership: Students who feel connected to the content learn more and are more motivated. Providing students with choice in their assignments is one way to create ownership.
  • Competence: Students should feel competent in completing homework and so we need to abandon the one-size-fits-all model. Homework that students cannot do without help is not good homework.
  • Inspiring: A well-considered and clearly designed resource and task impacts positively upon student motivation.

So do we need to turn the age-old concept of homework on its head? Homework clearly needs greater attention and redesigning and that includes what Mark Creasy calls “unhomework” (2014), where children set their own learning and targets for homework and then it is self and/or peer assessed.

Or as Russel Tarr suggests, why not give students a choice, takeaway menu style? He says that “giving students the flexibility to choose the content and/or the outcome of their homework assignments increases engagement and promotes independent learning” (Tarr, 2015).

So assignments following the above suggestions and those found in the EEF’s evidence summary will go a long way to improving the image, intent, implementation and impact of secondary homework.

And finally...

Where effective schools do set homework, they guarantee that is it is in line with their global aims and vision for teaching, learning and assessment. In particular, both the level of challenge and the feedback are considered to ensure that homework promotes a greater love of school and interest in learning.

The Teaching Schools Council’s Effective primary teaching practice report (2016) outlined that schools employing homework successfully are clear about:

  • Its purpose: communicating with parents and sharing with them why their children do or do not have homework. The school makes sure that children clearly understand its purpose and no pupils lose out.
  • The impact on teacher workload: following up, but in a way that does not disproportionately add to teacher workload.
  • Limiting the time that children spend doing it: suggesting a cut-off point even if children have not completed everything. The US rule of thumb of “10 minutes per grade” is a sensible guide (this rule was suggested by researcher Harris Cooper – 10 to 20 minutes per night in the first grade, and an additional 10 minutes per grade level thereafter).
  • The level of challenge: making sure children can succeed without too many demands and without needing to ask their parents for lots of help.
  • The social context: ensuring that any homework set reflects different pupil experiences, background, and types of parental involvement.

Homework does not need to be abandoned but it does need far better management, especially in relation to how we communicate with parents. We need to get it right and start asking whether it is really making any difference.

  • John Dabell is a teacher, teacher trainer and writer. He has been teaching for 25 years and is the author of 10 books. He also trained as an Ofsted inspector. Visit and read his previous best practice articles for SecEd via

Further information & research

  • Homework: Its uses and abuses, Hallam, Institute of Education, UCL, 2006:
  • The End of Homework: How homework disrupts families, overburdens children, and limits learning, Kralovec & Buell, Beacon Press, 2000.
  • The Homework Myth: Why our kids get too much of a bad thing, Kohn, De Capo Books, 2006:
  • Homework Myths, Vatterott, The Homework Lady, February 2008:
  • Homework (Secondary) Evidence Summary, Teaching and Learning Toolkit, Education Endowment Foundation:
  • “Homework in primary school has an effect of zero”, John Hattie interview, BBC Radio 4, August 2014:
  • Visible Learning: A synthesis of over 800 meta-analyses on achievement, Professor John Hattie, 2009:
  • Why teaching will never be a research based profession, Dylan Wiliam presentation, ResearchEd, 2017:
  • How to shift a school towards better homework, Kelleher, The Learning Scientists, June 2017:
  • The case for and against homework, Marzano & Pickering, Educational Leadership, March 2007:
  • Speaking with: John Hattie on how to improve the quality of education in Australian schools, The Conversation, May 2016:
  • Learning out of school: Homework, policy and practice, Research study commissioned by the Scottish Education Department, MacBeath & Turner, 1990:
  • Rethinking Homework: Best practices that support diverse needs, Vatterott, ASCD, 2018.
  • Five hallmarks of good homework, Vatterott, Educational Leadership, September 2010:
  • Unhomework: How to get the most out of homework, without really setting it, Mark Creasy, Independent Thinking Press, 2014.
  • Why students should set and mark their own homework, Mark Creasy, Guardian, April 2014:
  • Takeaway Homework, Tarr’s Toolbox, Tarr, 2015:
  • Effective primary teaching practice, Teaching Schools Council, 2016:
  • The Institute for Effective Education’s Best Evidence in Brief has a selection of homework-related research which is worth a look:


Please view our Terms and Conditions before leaving a comment.

Change the CAPTCHA codeSpeak the CAPTCHA code
Sign up SecEd Bulletin