Homework Research and Policy:

A Review of the Literature

by Harris Cooper, Department of Psychology, University of
Missouri-Columbia

Data show that homework accounts for about 20 percent of

the total time the typical American student spends on
academic tasks . . . considering this fact, it is surprising how little attention is paid to the topic of
homework in teacher education.

Homework plays a significant role in education in the United States. According to the National
Assessment of Educational Progress, two-thirds of 9-, 13-, and 16-year-olds reported doing homework,
and the percentage was increasing (Anderson et al., 1986). Thirteen-year-olds reported spending an
average of about one hour daily on homework (Walberg, 1991). These data show that homework
accounts for about 20 percent of the total time the typical American student spends on academic tasks.

Considering these facts, it is surprising how little attention is paid to the topic of homework in teacher
education. Most teachers in the United States report that in education courses they discussed
homework in relation to specific subjects, but received little training in how to devise good
assignments, how to decide how much homework to give, and how to involve parents.

Moreover, homework often causes a great deal of conflict among teachers, students, and parents.
Indeed, many doctors and family counselors indicate that problems with homework are a frequent
source of concern when children experience medical problems (Cooper, 1991).

In this article I describe the findings of a review of research on homework (Cooper, 1989). I examine
the efficacy of homework as an instructional method, develop a sequential model of the factors that
influence homework outcomes, and propose homework policy guidelines for teachers, schools, and
school districts (see also Cooper, in press). The review was supported by the National Science
Foundation and included nearly 120 empirical studies of homework's effects and the characteristics of
successful homework assignments.

The Role of Research in Improving Homework Practices

American researchers have studied homework for over 60 years. For example, Hagan
(1927) compared the effects of homework with the effects of in-school supervised study
on 11- and 12-year-olds' academic performance. The number and percentage of research
documents containing references to homework indicate that scholars are more interested
in homework now than ever before.

Researchers do not agree, however, on the advantages and disadvantages of homework

as an instructional tool. Although more than 12 reviews of the homework literature were
conducted between 1960 and 1987, reviewers' conclusions differed considerably. This
variance is due partly to a lack of overlap in the literature that reviews cover, to diverse
criteria for inclusion of studies, and to differing methods of synthesizing study results.
Defining Homework and Its Effects

I defined homework as "tasks assigned to students by school teachers that are meant to
be carried out during non-school hours" (Cooper, 1989, 7). Omitted in this definition
are

a. in-school guided study,

b. home study courses, and
c. extracurricular activities, such as sports or student newspapers.

The list of possible advantages and disadvantages of homework is long and often
surprising. These effects are given in Table 1. Among the suggested benefits of
homework, the most obvious is that it will increase students' retention and
understanding of the material it covers. Less directly, homework can improve students'
study skills and attitudes toward school and teach students that learning takes place
outside as well as inside of school.

Homework has numerous potential nonacademic payoffs as well; most of these involve
promoting student independence and responsibility. Finally, homework can involve
parents and the broader community in schooling, increasing their appreciation of
education and allowing them to reinforce students' achievement.

The possible negative effects of homework are perhaps more interesting. First, some
educators note that any activity can remain rewarding only for a limited time. It follows
that if students are required to spend too much time on academic material, they
eventually will become bored with it. Second, homework limits the time students can
spend on leisure-time and community activities that can impart important lessons, both
 academic and nonacademic. Third, parental involvement, however well-meant, often
becomes parental interference. Parents can confuse children if the teaching methods
they employ differ from those of teachers. Fourth, homework can lead to undesirable
behaviors such as cheating, either through copying of assignments or receiving
assistance with homework that involves more than tutoring. Finally, homework could
exacerbate existing social inequalities. Students from lower-socioeconomic homes are
likely to have more difficulty completing homework than their more well-to-do peers.
Poorer students are also more apt to work after school or may not have a quiet, well-lit
place to do assignments.

A Model of the Homework Process

Table 2 depicts how and in what sequence numerous factors affect the usefulness of
homework. The process begins with three factors-student characteristics, the subject
matter, and especially grade level-that determine the benefits of homework.

Characteristics of the assignment are important as well. Homework assignments can be

short or long, can have different purposes (such as the practice of old material,
introduction of new material, integration of skills, or extension of the curriculum), can
be designed for individual students or whole classes, and can be completed by
individuals or groups. The time and effort teachers take to develop an assignment also
varies.

When a student takes the assignment home, several factors will affect how it is
completed, including the student's other time commitments, the home environment, and
the involvement of others.

Finally, what the teacher does with assignments when students turn them in may affect
homework's utility. Some teachers may simply collect assignments, whereas others go
over them in class and provide written feedback, oral comments, or grades. Other
teachers may permit students to correct homework as part of the learning process and
even provide extra credit toward grades.

Is Homework Effective?
Three types of studies enable researchers to answer the question of whether homework
enhances students' achievement. The first type of study involves comparing the
achievement of students who receive homework with students given no homework or
any other treatment to compensate for their lack of home study. Of 20 studies
completed since 1962, 14 yielded results favoring homework, whereas six favored no
homework. Most interesting is the striking influence of grade level on homework's
effectiveness. According to these studies, the typical U.S. high school student, 14 to 16
years of age, in a class doing homework would outperform 69 percent of the students in
a no-homework class, as measured by standardized tests or grades. In junior high
school, students 11 to 13 years of age, the average homework effect was half this size.
In elementary school, homework had no effect on achievement.

In another group of studies, researchers compared homework with in-class supervised

study. The performance benefits of homework were generally about half what they were
when homework was compared with no treatment. Most significant in these studies was
the finding once again of a strong grade-level effect. For elementary students, in-class
study proved superior to homework. In junior high, homework was superior, and in high
school, homework's advantage was greatest.

Finally, in 50 studies researchers correlated the time students reported spending on

homework with their achievement. Many of these data were from state or national
surveys. A total of 43 correlations showed that students who did more homework had
better achievement, whereas only seven correlations indicated the opposite. Again, a
 strong grade-level effect was present. For elementary students, the
mean correlation between time spent on homework and achievement
was almost zero; for students in middle grades it was r = +.07, and
for high school students it was r = +.25.

Interpreting the Size of Homework's Influence

In the previous section I discussed the effect of homework across

differing grade levels, types of outcomes, and content areas. The
outcomes of homework can also be compared with those of other
teaching methods. This allows the consideration of homework in a
wider educational context, thus permitting a more accurate judgment
of its value.

Walberg (1986) described results of 11 reviews of research on

effects of instructional methods and teaching skills on student
achievement. The instructional methods included individualized,
special, and cooperative learning; ability grouping; direct and programmed instruction;
advance organizers; higher-level cognitive questioning; use of praise; use of pretests;
and television watching. For each strategy, Walberg calculated an effect similar to the
one I used previously (e.g., the average high school student doing homework
outperformed 69 percent of students not doing homework). In comparison with these
instructional strategies, homework's effect on achievement can be described most
accurately as above average. That is, homework's influence fell about in the middle of
the 11 strategies. If grade level is considered, the relation between homework and
achievement of elementary school students is comparatively small, but on high school
students homework's effect is large, relative to the effect of the other 11 instructional
techniques.

Another means of interpreting the size of an effect is to compare it with the cost of
implementing the treatment. Homework is definitely a low-cost treatment. The most
significant costs involved in assigning homework would be (a) a slight loss in
instructional class time because time must be set aside for dealing with homework, and
(b) additional outside-class preparation and management time for teachers.

Factors Affecting the Usefulness of Homework Assignments

In addition to studying homework's overall effectiveness, researchers also have

examined how variation in assignments might influence their efficacy. One obvious
question is whether homework is more effective for some subjects than for others.
Based on the three sets of evidence just described, no clear influence for subject matter
can be found. Still, homework probably works best when the material is not complex or
extremely novel.

Another important concern is the optimum amount of homework. In nine studies

researchers recorded academic performance as a function of increasing homework time.
Elementary students' performance did not improve when they spent more time on
homework. Junior high school students' achievement continued to improve with
increased homework until assignments took between one and two hours a night. More
homework than this was not accompanied by improved achievement. In contrast, high
school students' performance continued to increase through the highest point on the
measurement scales.
 Other findings were related to differences in homework assignments. For example,
there was considerable evidence that homework results in better achievement if material
is distributed across several assignments rather than concentrated only on material
covered in class that day. Homework that requires students to practice material already
taught, as well as assignments meant to prepare students for upcoming lessons, both
proved beneficial. The few (poorly designed) studies of parent involvement indicated
that giving parents a formal role in homework did not affect its utility. The same was
true for individualizing homework assignments, which had a minimal effect on
achievement but did significantly increase the time teachers spent on homework-related
activities. Finally, I located no study in which researchers compared a homework
feedback strategy (e.g., grading, instruction, or evaluative comments) with a no-
feedback approach. Several studies involved comparisons of differing feedback
strategies; no approach was clearly superior.

General Policy Guidelines

Based on results of research and over 100 other articles, I developed homework policies
for a representative (or average) American school district. These are listed in tables 3.1,
3.2, and 3.3. Although the policies would have to be adapted to any single district, they
provide a good basis on which to begin discussions.

My first recommendation is that coordinated policies should exist at the district, school,
and classroom levels. Some issues that must be addressed at one level are unique to that
level, whereas others overlap.

Districts should offer a clear and broad rationale for assigning homework, including
why it is sometimes mandatory, as well as general guidelines for the amount of how that
should be assigned. Schools need to provide more specific time requirements,
coordinate assignments between classes, and describe the role of teachers and
principals. Teachers should outline what they expect of students and why.

Although I will not discuss the policies in detail, a few underlying philosophical points
ought to be made explicit. First, elementary school students should be assigned
homework, though it should not be expected to improve their achievement. Rather,
homework should help young children develop good study habits, promote positive
attitudes toward school, and communicate to students that learning takes place outside
as well as inside school. Thus, assignments to elementary students should be brief,
should involve materials commonly found in the home, and should not be too
demanding.

The academic function of homework should emerge in junior high school. Its use as a
motivational tool should not be ignored, however. Thus, I advocate using both required
and voluntary assignments. The latter should involve tasks that are intrinsically
interesting to students of this age.

Teachers should never use homework as punishment. Using it in this way

communicates to students that schoolwork is boring and aversive.

High school teachers can view the home as an extension of the classroom. Homework
that involves practice and review of lessons previously taught and simple introductions
to material prior to its coverage in class is desirable. Assignments that require students
to integrate skills or differing parts of the curriculum should also be common.

Regardless of students' ages, the formal role of parents in homework should be minimal.
Parents vary in interest, knowledge, teaching skills, and time available. Clearly, parents
of young children should be more involved. In particular, they need opportunities to
express how much they value school achievement. Besides helping their children to
 practice of reading, spelling, and math skills, parents can express their interest by
having contracts with their children about study times, offering rewards for completed
assignments, or merely by signing homework before it is returned to school.

I also advise that teachers individualize few assignments within classes. Developing
individualized homework demands considerable teacher time and has few benefits.
Teachers who teach the same course to several classes that are progressing at different
rates might consider giving the same assignments to the top students in the lowest class
and the lowest-performing students in the highest class.

Finally, most homework assignments should not be graded. Teachers should not view
homework as an opportunity to test. Almost all students should complete assignments
successfully; thus, teachers should not differentiate much among performance levels.
Having students do homework out of fear of negative consequences turns a situation
ideal for building intrinsic motivation ("I must enjoy this; I'm doing it and the teacher
isn't standing over me") into one that implies that the teacher believes students need
rewards or punishment in order to complete assignments. Teachers should collect
homework, check it for completeness, and give intermittent instructional feedback. This
procedure shows that the teacher takes homework seriously and that it is purposeful.
The major purpose should be to identify individual students' learning problems.

Homework for Students with Learning Disabilities

Much attention has been given recently to the use of homework with learning-disabled
students. When I surveyed studies of homework conducted prior to 1986, I found few
that included students with learning disabilities. More research in this area exists today,
and a colleague and I recently reviewed this literature (Cooper & Nye, in press). We
 wanted to determine how homework practices and policies that produced positive
outcomes for students without disabilities might differ for students with disabilities.

First, we examined studies of the overall effectiveness of homework for students with
learning disabilities. This research indicated that the generally positive effects of
homework for students without disabilities also should appear for students with learning
disabilities.

Nevertheless, the characteristics of successful homework assignments for the two types
of students are different. For example, research has consistently shown that homework
assignments for students with learning disabilities should be brief, emphasizing
reinforcement of skills and class lessons rather than integration and extension of class
work. In addition, students who do not possess certain minimum skills in an area may
not benefit from homework at all.

It is vital that teachers monitor homework assignments for students with learning
disabilities. Monitoring might involve reviewing completed assignments promptly in
class, rewarding students for completeness and/or accuracy, and allowing students to
begin assignments in class so that teachers can make certain that students understand
assignments.

Parental involvement is essential for students with disabilities. These students tend to
have less developed self-management and study skills than their peers, and their ability
to study relies more on the provision of a proper environment, both physical and
emotional. Students with learning disabilities may need periodic rewards while they
work on assignments or immediately after completing assignments as well as more help
in finishing tasks. Research suggests that parents' involvement should be prolonged
rather than intermittent.

Summary

The relation between homework and student academic performance is influenced

heavily by grade level. The effects of homework on elementary students appear to be
small, almost trivial; expectations for homework's effects, especially short-term and in
the early grades, should be modest. Homework should be viewed as one of several
methods teachers can use to show children that learning takes place everywhere. For
high school students, however, homework can have significant effects on achievement.
Indeed, relative to other instructional techniques, and considering the minimal costs
involved in implementation, homework can yield a considerable increase in academic
performance at this level. Finally, homework can have significant benefits for students
with learning disabilities, but its positive outcome is contingent on (a) teacher
preparation and planning; (b) assignments appropriate to the skill, attention, and
motivation levels of students; and (c) appropriate involvement of parents.

Portions of this article were taken from "Synthesis of Research on Homework,"

Educational Leadership, October 1989. Used with permission from the Association for
Supervision and Curriculum Development. Thanks are extended to Gail Hinkel for
editorial comments.
References on Homework

Anderson, B., et al. 1986. Homework: What do National Assessment Results Tell Us?
Princeton, N.J.: Educational Testing Service.

Cooper, H. 1989. Homework. White Plains, N.Y.: Longman.

Cooper, H. 1991. Homework. Feelings and Their Medical Significance 33 (2), 7-10.

Cooper, H. In press. The Battle over Homework: An Administrator's Guide to Sound

and Effective Policies. Newbury Park, Calif.: Corwin Press.

Cooper, H., and B. Nye. In press. Homework research, policies and practices: A review
of the learning disabilities literature. Journal of Learning Disabilities.

Hagan, H. 1927. The value of homework as compared with supervised study. Second
Yearbook, Chicago Principal's Club, 147-49.

Walberg, H. J. 1991. Does homework help? School Community Journal 1 (1), 13-15.