Homework for young children: Is it justified?

© 2019 Gwen Dewar, Ph.D., all rights reserved

Sepia image of young girl doing pencil and paper homework by ND Strupler flickr

Is homework for young children helpful? Some schools assign homework to kids as young as 5 or 6. But there isn't any compelling, scientific evidence in favor of the practice. Here's a critical look at the research -- and the difficulties that homework can pose for families.


Should teachers assign homework to preschoolers? Kindergartners? First graders? Second graders?

You might think the very question is outlandish. You might not have heard that homework for young children is a thing. But it is a thing.

In a survey of more than 2700 kindergarten teachers – all working in the United States circa 2010 – 40% said they believed that "homework should be given to kindergarten children almost every day" (Bassok et al 2016).

And the homework can be substantial. In a recent survey of nearly 1200 families living in Rhode Island, researchers asked parents how much homework their children were getting. The answers?

The average kindergartner was sitting through 25 minutes a night. First and second graders were averaging approximately 28 minutes per night.

The researchers in this study were dumbfounded. They had expected to hear that some 1st graders and 2nd graders were getting homework. But the amount took them by surprise.

And kindergartners? The researchers didn’t think kindergartners should be getting any homework at all.

In their paper, the researchers note that the average homework session – 25 minutes each day – "may be both taxing for the parents and overwhelming for the children." The researchers also warn that homework can displace important developmental experiences:

"[I]n a period of life when children are focused on early stages of socialization and finessing motor skills, we anticipate that an overload of homework will likely interfere with a kindergarten-aged child’s ability to play and participate in extra-curricular activities" (Pressman et al 2015).

The same concerns apply to older children as well. 

In an online essay for Edutopia2nd grade teacher Jacqueline Fiorentino notes that homework "causes a lot of stress and fighting in most families." It has the potential to turn young children against school. And kids "are are losing precious free time that could be used to engage in play and group activities like organized sports, music lessons, and clubs."

What happens when we give that free time back to children? Fiorentino eliminated mandatory homework, and her kids flourished. They actually started doing more – using that extra time "to explore subjects and topics of interest to them."

With free time, free choice, and encouragement, kids enthusiastically devoted themselves to voluntary projects, and shared the results at school.

Tell me about it

Fiorentino’s experience will probably resonate with many parents – parents who know there is more at stake than misspent time. Homework can also take a serious emotional toll on families.

Young children don’t complete their homework by themselves. They need help. Coaching. Intense supervision. Do parents have the time and skills to provide this? What if a child resists? What if the child is too tired, too restless, too distracted to concentrate?

A child’s abilities after school aren’t the same as a child’s abilities earlier in the day.

At school, kids do work in the context of a lesson plan. They have the immediate support of a teacher. Their minds are relatively fresh and ready to perform.

At home, kids have to begin their work outside the context of a lesson plan. The teacher who assigned the work isn’t there to interpret or answer questions. And the child has already spent 5-7 hours at school. It’s harder to focus.

So I’m not surprised to see Education Week reporting that some kindergartners are taking an hour to do their homework. Personally, I’ve seen first graders coping with even longer assignments. Homework can become an ordeal, creating conflict, bad feelings, and family stress (Pressman et al 2015).

How does this happen? Why are teachers assigning so much?

I’m guessing that they don’t realize how long their assignments will take. They may be misled by their experiences in the classroom, and assume that kids will complete homework assignments as quickly as they complete classroom work.

But whatever the cause, one thing is clear: Homework comes with a downside. And whenever a choice comes with costs, we need to ask ourselves if the benefits outweigh the costs.

What does the research tell us?

Surprisingly little.

In fact, if you dig into the research, it’s clear that studies don’t justify homework for young children. 

To see what I mean, consider first what a good study ought to look like. How can science help us determine the effects of homework on young children?

The best way is to conduct a controlled experiment.

What experiments need to show -- but don't


To design a rigorous experiment, we randomly assign kids to one of two treatments: homework and no homework. And we need to make sure that – except for homework – the groups are otherwise similar.

Are kids roughly the same age? Do they come from similar socioeconomic backgrounds? If some kids have attention problems or learning disabilities, are they equally represented in both groups?

Even randomized experiments can end up with inequality between groups -- especially if we are dealing with small sample sizes. So it's important to check for disparities. 

It's also important to make sure each group is starting the study with similar levels of background knowledge and academic preparation.

For example, if I want to test the effectiveness of homework for learning new words, I need to know if both groups begin with similar knowledge of those words. If I want to test the effectiveness of math homework, I need to know if both groups start with similar levels of math proficiency.

Thus, we should test kids before the study begins (to understand their baseline academic skills) and then re-test them at the end.

This will help us to determine if homework itself -- and not pre-existing differences between students -- is responsible for any group differences in outcome.

In addition, we must be mindful of the way groups are treated during the course of the experiment.

Are their classroom experiences similar? Is the quality of teaching the same? Are kids exposed to the same lesson plans? The same instructional techniques? 

If our groups experience different conditions in the classroom, this could influence the results. One group might outperform another because it had a more effective teacher. Or engaged in more effective classroom learning activities. 

So we should take care to control for these sources of variation too. Otherwise we'll be left scratching our heads. How much of the outcome was caused by classroom differences, and how much by homework?

Another crucial piece of the puzzle is the homework itself.


Homework can take a variety of forms, and it’s a sure bet not all forms of homework have the same effects.

Is it an open-ended essay, or a series of multiple choice reading comprehension questions? Is it a worksheet of arithmetic drills, or a single, thought-provoking story problem? Is it handwriting practice? A family game? An independent research project? Is it an assignment to create artwork? Prepare a speech? Engage with some interactive educational software?

An experiment might indicate that, say, after-school sessions with a specific mathematics learning app is helpful. But that wouldn't tell us that kids also benefit from writing after-school essays about whether cats are better than dogs.

It’s also obvious that the frequency, difficulty, and amount of homework could affect outcomes. 

So we need to keep track of what, exactly, homework consists of. We also need to know how often homework is assigned, and how long it takes kids to complete.

All of these steps should be par for the course. If we had a number of such well-designed studies, we might be able to make some inferences. But has anyone ever done it?

As far as I can tell, no such rigorous, experimental study has ever been published in a peer-reviewed journal. Not for elementary school students. And certainly not for the youngest children.

When Harris Cooper and his colleagues searched for experimental tests of homework, the researchers identified just four studies that targeted elementary school students (Cooper et al 2006).

All four studies were unpublished. Two were very small -- comparing outcomes in just two classrooms. And none of them met the criteria mentioned above.

For example, in a study of 2nd graders, the study author compared children enrolled in two different classrooms. One class received homework and the other did not (Finstad 1987).

At the end of the study, kids in the homework class performed better on a mathematics test about place value. But the study was small (39 students total), and kids weren't randomly assigned to their classes.

Moreover, the study author didn’t provide information about the homework itself. We don’t know how much work there was, how frequently it was assigned, how long it took children to complete (Cooper et al 2006).

As the study author herself noted, the results merited “further study on a larger scale” (Finstad 1987).

And we can say the same about the remaining three studies.

  • A small study of 3rd graders doesn't appear to have controlled for pre-existing differences between the two classrooms being tested (Townsend 1995).
  • Another study, of 5th graders, failed to control for the effects of different teachers (Foyle 1999).
  • The last study, of 3rd and 4th graders, failed to find any differences between classrooms assigned to either the homework or no-homework conditions. And when the researcher tried looking for evidence of improvement in individual children, the results were mixed: The homework had a negative effect on achievement on 3rd graders, but a positive effect on 4th graders (Meloy 1987).

So there is no ringing endorsement for homework here, and no information at all provided about kindergartners and 1st graders. At best, the evidence is ambiguous, weak, and contradictory.

What about other kinds of research? Non-experimental research?

There is another way to study homework. What if we simply look for correlations in the everyday world? See if there are any links between homework and academic outcomes? Do kids who happen to do homework perform better in school?

A number of studies have taken the correlational approach. But once again, the results leave us mostly in the dark.

In 2006, Harris Cooper and his colleagues performed a meta-analysis of existing correlational studies. They didn’t find any substantial links between homework and achievement among elementary school students (Cooper et al 2006).

More recently, researchers have re-examined the evidence – including a few, new unpublished studies – and found evidence that one particular measure of homework is correlated with school performance: Kids who routinely turn in their homework tend to perform better academically (Fan et al 2017).

A similar trend was reported by Spanish researchers in a peer-reviewed study of 9- to 13-year-olds (Valle et al 2006).

But these studies present serious interpretive problems. 

  • Only a few studies concern elementary school students (as opposed to secondary school students). I can find only one study (a study of 1st graders) that addresses very young school children.
  • The evidence is mixed, and some of the positive links come from studies with very small sample sizes.
  • Correlation doesn’t prove causation.

This last point is probably the most important.

Homework completion might be correlated with measures of superior academic achievement, but that doesn’t prove that homework completion makes kids more knowledgeable or skilled.

Children who complete their assigned homework don’t represent a random sample of the population. They probably have -- on average -- stronger academic skills. It’s easier for them to keep up.

In addition, they may tend to be more compliant in the classroom, or find the school work more engaging. They’re more likely to have academic support at home.

All of these factors could contribute to better school grades and higher test scores. So even if homework weren’t helpful, we’d expect to see homework completion linked with higher achievement.

Of all the correlational research I’ve seen, only one study -- dissertation research presented at an educational conference -- seems to take student characteristics into account. Tracking 143 students in the 3rd grade, researchers found that kids who completed math homework tended score higher on final mathematics tests, even after accounting for prior test scores, socioeconomic status, and personal attitudes about homework (Pelletier and Normore 2007).

This is promising. But I can't tell from the conference proceedings what kind(s) of homework were assigned, or how much time kids spent on homework. It's also unclear how applicable the results are to younger children, like kindergartners. We need more research to address these questions.

So -- overall -- the correlational evidence isn't very helpful. Especially when it comes to prescribing homework assignments to children in the earliest grades. 

What about the advantages of “spaced learning”? Haven’t researchers found that we learn better when our learning sessions are spaced apart in time? Doesn’t this suggest that homework is a good thing?

I’ve seen this argument on the internet. The reasoning runs like this:

  1. Experiments show that learning improves when learning sessions are spaced apart in time;
  2. homework allows us to schedule learning sessions in the afternoon or evening, therefore
  3. homework allows kids to enjoy the benefits of learning sessions that are spaced apart in time.

What’s wrong with this argument?

It’s true that experiments support the advantages of spaced learning.

But what is the optimal spacing between learning sessions for young children? We don’t yet know. 

As I will note in an upcoming post, some researchers are interested in this question. But they haven’t yet found answers that would support after-school learning sessions.

On the contrary, one study found that 1st and 2nd graders learned the most about a biology topic when they received no more than one brief lesson per day (Vlach and Sandhofer 2012).

That doesn’t prove that homework is detrimental. But it’s clear that we need more research before we can draw any conclusions.

Future studies may show that spaced learning is actually evidence against homework – not for it.

But hang on. This isn’t proof that homework can’t be helpful. Might future research report new discoveries? Reveal that certain types of homework really do contribute to higher achievement?

Maybe.

We know that young children benefit from play. They learn when they voluntarily and enthusiastically immerse themselves in their own, educational hobbies.

What if we design homework that feels like play? Assignments that are fun, brief,  age-appropriate, and attuned to the attention spans of children who've already spent the day at school?

Maybe it's possible to create homework that even kindergartners will enjoy and benefit from. Maybe, for example, we'll find out that most kids thrive when they spend a few minutes each day using certain well-designed, game-like, after-school educational apps. 

But no matter what we discover in the future, we will have to attend to the details.

If a study reports that 5th graders benefit from 20 minutes of math homework each night, we shouldn't assume that the same would be true for 1st graders. 

Nor should we assume that all forms of math homework would be equally effective, even among 5th graders. Research suggests that some forms of homework are better than others. Some have have been linked with superior test performance. Others have actually been linked with losses in test performance.

And no matter what we learn about achievement outcomes, we'll need to reckon the costs as well as the benefits. 

It won’t be enough to show that doing homework helps little kids score a few extra points on a test. We will also have to show that – on balance – the positives outweigh the negatives.

Meanwhile, I like the approach taken by Jacqueline Fiorentino. Give young children free time, and encourage them to follow their interests. Help parents support independent exploration at home. Don’t ask parents to play the role of homework cop. Don’t set families up for conflict and stress.

It seems to work in Finland, a country that produces some of the most high-achieving students in the world. Kids don’t begin elementary school until the age of 7, and they don’t normally receive homework until they are teenagers (Anderson 2011).


More reading

Are you looking for information about valuable developmental activities? See my Parenting Science articles about the benefits of exercise and outdoor play. In addition, check out my pages about fostering cognitive development in children.

For more information about developmental limitations on a young child's ability to concentrate, see my article about working memory in children.

And if you're interested in the debate about homework, check out the writings of Alfie Kohn. He is a former teacher and well-known critic of homework. He doesn’t just oppose homework for young children. He thinks schools should eliminate homework for older kids, too.

You can read about his ideas in his book, The Homework Myth. (Full disclosure: If you purchase the book using this link, I will receive small commission.) 


References: Is homework for young children justified?

Anderson J. 2011. From Finland, an intriguing school-reform model. New York Times. Published online December 11, 2011: http://www.nytimes.com/2011/12/13/education/from-finland-an-intriguing-school-reform-model.html?src=me&ref=general&_r=0

Barger MM, Kim EM, Kuncel NR, Pomerantz EM. 2019. The relation between parents' involvement in children's schooling and children's adjustment: A meta-analysis. Psychol Bull. 145(9):855-890.

Bassok D, Latham S, and Rorem A. 2016. Is kindergarten the new first grade? AERA Open January-March 2016, Vol. 1, No. 4, pp. 1–31

Cooper H, Robinson JC, and Patall EA. 2006. Does homework improve academic achievement? A synthesis of research, 1987–2003. Review of Educational Research 76(1): 1–62.

Cooper H, Lindsay JJ, and Nye B. 2000. Homework in the Home: How Student, Family, and Parenting-Style Differences Relate to the Homework Process. Contemp Educ Psychol. 25(4):464-487.

Fan H, Xu J, Cai Z, He J, and Fan X. 2017. Homework and students’ achievement in math and science: A 30-year meta-analysis, 1986–2015. Educational Research Review 20: 35-54.

Finstad E. 1987. Effects of mathematics homework on second grade achievement. Houston, TX: Sam Houston State University. (ERIC Document Reproduction Service No. ED291609)

Foyle H. 1990. Homework and cooperative learning: A classroom field experiment. Emporia, KS: Emporia State University, Faculty Research and Creativity Committee. (ERIC Document No. ED350285)

Fu AS and Markus HR. 2014. My Mother and Me: Why Tiger Mothers Motivate Asian Americans But Not European Americans. Personality and Social Psychology Bulletin 40(6): 739-749.

Hill NE and Tyson DF. 2009. Parental involvement in middle school: a meta-analytic assessment of the strategies that promote achievement. Dev Psychol. 45(3):740-63.

Hofmann W Schmeichel BJ and Baddeley AD 2012. Executive functions and self-regulation. Trends in Cognitive Sciences 3: 174-180.

Meloy LL. 1987. Effects of homework on language arts achievement in third and fourth grades (Doctoral dissertation, University of Iowa, 1987). Dissertation Abstracts International, 51, 725.

Omlin-Ruback H. 2009. A study of mathematics homework (Order No. 3384674). Available from ProQuest Dissertations & Theses Global. (304979029).

Pelletier R. 2005. The predictive power of homework assignments on student achievement in grade three (Order No. 3169466). Available from ProQuest Dissertations & Theses Global. (305350863).

Pelletier R and Normore AH. 2007. The predictive power of homework assignments on student achievement in mathematics. In S. M. Nielsen & M. S. Plakhotnik (Eds.), Proceedings of the Sixth Annual College of Education Research Conference: Urban and International Education Section (pp. 84-89). Miami: Florida International University. http://coeweb.fiu.edu/research_conference/

Pressman RM, Sugarman DB, Nemon ML, Desjarlais J, Owens JA, Schettini-Evans A. 2016. Homework and family stress: With consideration of parents' self confidence, educational level, and cultural background: The American Journal of Family Therapy. 43(4): 297-313,.

Rothbart MK and Rueda MR. 2005. The development of effortful control. In U. Mayr, E. Awh, & S. Keele (Eds.), Developing individuality in the human brain: A tribute to Michael I. Posner (167-188). Washington, DC: American Psychological Association.

Rueda MR, Posner MI, and Rothbart MK. 2004. Attentional control and self-regulation. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory, and applications (pp. 283-300). New York: Guilford Press.

Townsend S. 1995. The effects of vocabulary homework on third grade achievement. (Master’s thesis, Kean College of New Jersey, 1995). (ERIC No. ED379643).

Valle A, Regueiro B, Núñez JC, Rodríguez S, Piñeiro I, Rosário P. 2016. Academic Goals, Student Homework Engagement, and Academic Achievement in Elementary School. Front Psychol. 7:463.

Vlach HA, Sandhofer CM. 2012. Distributing learning over time: the spacing effect in children's acquisition and generalization of science concepts. Child Dev. 83(4):1137-44.

Title image of girl with pencil by ND Strupler / flickr

Image of brothers laughing by marco antonio torres / flickr

Black and white image of girl on the floor with homework sheet by Steve Depolo / flickr

Image of boys at computers Independence Learning Commons / flickr

Content last modified 11/19