Overall, homework does appear to result in higher levels of achievement for older students (at the secondary level).
For these students, more time spent on homework is associated with higher levels of achievement, although there is probably a level beyond which more is counterproductive (perhaps at three hours a day).
For students aged 11-13, homework appears to be of benefit, but not to the same degree as for older students.
For these students, spending more than an hour or two on homework does not result in greater benefit.
There is little evidence of benefit for students younger than 11, although it can be plausibly argued that small amounts of homework can have an indirect benefit for promoting good study habits and attitudes to learning.
The Suggested Benefits of Homework
The most obvious presumed benefit of homework is, of course, that it will improve students' understanding and retention of the material covered. However, partly because this (most measurable) benefit has not been consistently demonstrated, it has also been assumed that homework has less direct benefits:
- improving study skills, especially time management
- teaching students that learning can take place outside the classroom
- involving parents
- promoting responsibility and self-discipline
Probably the most obvious negative effect is the stress homework can produce in both student and parent. Homework can be a major battleground between parent and child, and in such cases, it's hard to argue that it's worth it. There are other potential problems with homework:
- homework demands can limit the time available to spend on other beneficial activities, such as sport and community involvement
- too much homework can lead to students losing interest in the subject, or even in learning
- parents can confuse students by using teaching methods different from those of their teachers
- homework can widen social inequalities
- homework may encourage cheating
Because homework has been a difficult variable to study directly, uncontaminated by other variables, research has produced mixed and inconclusive results. However, it does seem that the weight of the evidence is in favor of homework. According to Cooper's much-cited review of homework studies, there have been 20 studies since 1962 that compared the achievement of students who receive homework with students given no homework. Of these, 14 showed a benefit from doing homework, and six didn't.
The clearest point is the striking influence of age. There seems, from these studies, to be a clear and significant benefit to doing homework for high school students. Students 11 to 13 years of age also showed a clear benefit, but it was much smaller. Students below this age showed no benefit.
In 50 studies, time students reported spending on homework was correlated with their achievement. 43 of the 50 studies showed that students who did more homework achieved more; only 7 studies showed the opposite. The effect was greatest for the high school students and, again, didn't really exist for the elementary school students.For the students in the middle age range (11-13 years), more time spent on homework was associated with higher levels of achievement only up to one to two hours; more than this didn't lead to any more improvement.
TIMSS, however, found little correlation between amount of homework and levels of achievement in mathematics. While they did find that, on average, students who reported spending less than an hour a day on homework had lower average science achievement than classmates who reported more out-of-school study time, spending a lot of time studying was not necessarily associated with higher achievement. Students who reported spending between one and three hours a day on out-of-school study had average achievement that was as high as or higher than that of students who reported doing more than three hours a day.
Two British studies found that while homework in secondary schools produced better exam results, the influence was relatively small. Students who spent seven hours a week or more on a subject achieved about a third of an A level grade better than students of the same gender and ability who spent less than two hours a week.
A survey conducted by the United States Bureau of the Census (1984) found that public elementary school students reported spending an average of 4.9 hours and private school elementary students 5.5 hours a week on homework. Public high school students reported doing 6.5 hours and private school students 14.2 hours. Recent research studies by the Brown Center on Education Policy concluded that the majority of U.S. students (83% of nine-year-olds; 66% of thirteen-year-olds; 65% of seventeen-year-olds) spend less than an hour a day on homework, and this has held true for most of the past 50 years. In the last 20 years, homework has increased only in the lower grade levels, where it least matters (and indeed, may be counterproductive).
In America, NEA and the National PTA recommendations are in line with those suggested by Harris Cooper: 10 to 20 minutes per night in the first grade, and an additional 10 minutes per grade level thereafter (giving 2 hours for 12th grade).
In Britain, the Government has laid down guidelines, recommending that children as young as five should do up to an hour a week of homework on reading, spelling and numbers, rising to 1.5 hours per week for 8-9 year olds, and 30 minutes a day for 10-11 year olds. The primary motivation for the Government policy on this seems to be a hope that this will reduce the time children spend watching TV, and, presumably, instill good study habits.
TIMSS found that students on average across all the TIMSS 1999 countries spent one hour per day doing science homework, and 2.8 hours a day on all homework (the United States was below this level). On average across all countries, 36% of students reported spending one hour or more per day doing science homework.
There is some evidence that the relationship between time on homework and academic achievement may be curvilinear: pupils doing either very little or a great deal of homework tend to perform less well at school than those doing 'moderate' amounts. Presumably the association between lots of homework and poorer performance occurs because hard work is not the only factor to consider in performance -- ability and strategic skills count for a great deal, and it is likely that many very hard-working students work so long because they lack the skills to work more effectively.
By which I mean, what factors distinguish "good", i.e. useful, homework, from less productive (and even counterproductive) homework. This is the $64,000 question, and, unfortunately, research can tell us very little about it.
Cooper did conclude that there is 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.
There is no evidence that parental involvement helps, although it may well be that parental involvement can help, if done appropriately. Unfortunately, parental involvement can often be inappropriate.
A burning question for many parents!
A British study found that watching TV while doing homework was associated with poorer quality of work and more time spent. However, simply listening to the soundtrack did not affect the quality of the work or time spent. It's assumed that it's the constant task-switching caused by looking back and forth between the screen and the work that causes the negative effect. From this, it would also seem that listening to the radio should not be a problem. It's worth noting that we become less able to multi-task as we age, and that parents' objections to their children's study environment probably reflect their awareness that they themselves would find it difficult to concentrate in such circumstances.
You can read the TIMSS report at:
You can read an article on the motivational benefits of homework at:
And there are more articles about homework, with more details of Cooper's review at:
And a British review of homework research is available at:
April 2012: my update to this article.