What role should homework play in learning?

Issue 06 ・ March 10, 2016

Historical Background

In the U.S., debates about the value of homework have been steady over the past century. In the beginning half of the 1900’s, homework was not nearly as common. Many school districts banned homework at the elementary and middle school levels in the belief that it only facilitated rote learning. That changed in the 1950’s when the Soviet Union launched Sputnik, creating the perception that the U.S. needed to increase the amount of students’ homework to be competitive in the space race. Over the next 50 years until the present, the popular view of homework switched every 15 years or so between support and condemnation. Today research has enhanced our understanding, but the debate continues.

Homework: What’s the Point?

Numerous benefits are attributed to homework: better retention of factual knowledge, development of study habits, increased ability to manage time effectively, and heightened parental involvement. Evaluating exactly what of these benefits can be attributed to homework versus inherent differences in students or the quality of assignments presents obvious challenges. Further confounding research, homework often exacts costs in the form of increased stress and reduced sleep, which can negate potential benefits. Nonetheless, research to date provides guidance on homework’s role in achieving some of these goals.

Research Examples

How much homework should be assigned?

Although not unanimous, multiple studies on the relationship between homework time and academic achievement have shown that it is not a linear relationship: more homework does not equal more learning. One study of 7,451 13-year-old students administered a test of science and mathematics in addition to a survey of effort spent on homework, homework time, homework frequency, and the general circumstances around how homework was done. Results: student academic gains were associated with homework time up to 1 hour a day, after which increased homework time was associated with worse performance. This result varies with students’ ages.

How does the influence of homework change in different grades?

While empirical studies have not yet explicitly compared the role of homework in different ages, a review of the literature reveals clear trends. A meta-analysis of 4,400 studies up until 2003 found that homework time had different effects depending on the grade of students. Results: up to two hours of homework showed the most benefit for high school students, up to an hour of homework showed some benefit for middle school students, and almost no benefit was found among elementary school students.

What homework factors contribute to academic performance?

Studies that focus on the circumstances in which students complete homework have found that autonomy and effort are more important predictors of performance than homework time alone. One study of 483 eighth-graders from 20 classes analyzed students’ grades and standardized test scores in mathematics in relation to survey responses on homework effort. Students responded once in November and once in May of the academic year. Results: academic gains in grades and test scores were positively correlated to homework effort. However, other studies suggest that since effort and autonomy are highly correlated with students’ prior achievement, it’s difficult to say if this is a trait of the homework or of a certain kind of student.

How can homework best facilitate learning?

Practices from the “How can I improve long-term learning” edition of Gray Matters 01 should be implemented in homework design to facilitate learning. For instance, using homework to continue practice on material and skills learned in previous weeks uses distributed practice to improve recall. Homework that offers adequate practice of important skills also helps reinforce skills in a variety of contexts. A survey of school leaders around homework best practices found similar results: homework can facilitate engagement when it is an authentic, engaging extension of the class. Results: homework should be designed intentionally, with reference to most effective forms of practice.

Does the effect of homework change when instruction occurs at home and exercises are done in class?

In this case, the question is essentially if homework facilitates learning when it is done as a part class rather than at home. One model that represents this circumstance is the flipped classroom. Case studies done with ninth to twelfth-graders across a variety of subjects showed that achievement increased after implementing a model where students watched videos at home and practiced learning in class. In these schools the percentage of students passing standardized exams increased by up to 12%. Empirical studies comparing in-class work and at home work have repeated research design flaws, however, they generally indicate that in-class exercises have a positive effect for older students. Results: the research is inconclusive, but generally indicates that flipped classroom approaches have positive net benefits for high school students.

How frequently should homework be assigned?

The research on frequency is not conclusive, however most studies suggest that classes that receive homework frequently tend to do better on tests of achievement. One such study administered standardized mathematics examinations to 2,939 grade 7 and 8 students across 20 classes. In addition to the exam, students answered questions on frequency of homework and the average amount of time spent on homework each evening. Results: Classes that had more frequent homework assignments had overall higher averages on the achievement tests.

Does homework change students’ attitudes towards school?

Studies comparing the attitudes of students who received homework with those who did not receive homework generally found no significant results across a variety of grade levels. One study compared three different models for assigning arithmetic homework to 342 third-graders across twelve classrooms: teachers assigned no homework, assigned homework as usual, and were required to assign a constant amount of homework every night. Students answered questions measuring their attitudes towards school, teacher, arithmetic, spelling, reading, and homework. Results: no significant differences were found between the attitudes of the different groups.


The quality of homework and the students who complete it vary significantly, however the general research trends suggest that assigning homework does have benefits at the middle and upper school levels. Frequency of assignment and students’ autonomy in finishing homework were also correlated with student achievement. These results provide guidelines on the appropriate quantity of homework, but how best to incorporate homework content into an overall progression of students’ learning remains for teachers to judge.


“Adolescents’ Homework Performance in Mathematics and Science: Personal Factors and Teaching Practices” by Rubén Fernández-Alonso, Javier Suárez-Álvarez, and José Muñiz, Journal of Educational Psychology, 2015, 107(4), 1075.

“Ask the Cognitive Scientist Allocating Student Study Time” by Daniel T. Willingham, American Educator, 2002, 26(2), 37-39.

“Does Homework Improve Academic Achievement? A Synthesis of Research, 1987–2003” by Harris Cooper, Jorgianne Civey Robinson, and Erika A. Patall, Review of Educational Research, 2006, 76(1), 1-62.

“Does Homework Improve Learning?” by Alfie Kohn, AlfieKohn.org, 2006.

“Effects of Arithmetic Homework upon the Attitudes of Third Grade Pupils Toward Certain School-related Structure” by Norbert Maertens, School Science and Mathematics, 1968, 68(7), 657-662.

“Flipped Learning Model Dramatically Improves Course Pass Rate for At-Risk Students, Clintondale high School, MI: A Case Study” by Pearson Education, May, 2013.

“Flipped Learning Model Increases Student Engagement and Performance,
Byron High School, MN: A Case Study” by Pearson Education, June, 2013.

“Homework. Research on Teaching Monograph Series, Homework Versus In-class Supervised Study” by Harris Cooper, 1989, 77-89.

“If They’d Only Do Their Work!” by Linda Darling-Hammond and Olivia Ifill-Lynch, Educational Leadership, 2008, 63(5), 8-13.

“Research Trends: Why Homework Should Be Balanced” by Youki Terada, Edutopia, 2015, July 31.

“Synthesis of Research on Homework” by Harris Cooper, Educational leadership, 1989, 47(3), 85-91.

“The Homework–Achievement Relation Reconsidered: Differentiating Homework Time, Homework Frequency, and Homework Effort” by Ulrich Trautwein, Learning and Instruction, 2007, 17(3), 372-388.