Each day, seven million pupils sit down to write, log on and calculate, or curl up and read. Then they go home and do exactly the same thing. Or rather, they are supposed to do the same thing: some won't do it, or can't do it, or perhaps don't even have to do it.
"No one likes homework," says Susan Hallam of the University of London's Institute of Education, who has studied the subject, "but some pupils are more aware of the consequences of not doing it, not just in terms of punishments at school, but in terms of their attainment at GCSE."
Hallam's own analysis of studies going back 75 years, published in her 2004 book Homework: The Evidence, found that homework does boost achievement for secondary-aged pupils. The issue for secondary schools is the imbalance between pupils who do it and those who don't or can't. That is one reason why, for this age group, she thinks that non-compulsory homework clubs are a good idea.
Now the government has said that it is scrapping the guidelines on homework that have been in place since 1998 and leaving schools free to decide what to do, although it is not expecting pupils to now spend their evenings catapulting Angry Birds into pigs' houses, however much they plead that it really helps in understanding physics.
"Homework is part and parcel of a good education - along with high-quality teaching and strong discipline," the Department for Education statement read. "We trust headteachers to set the homework policy for their school. They know their pupils best and should be free to make these decisions without having to adhere to unnecessary bureaucratic guidance."
The decision was made after complaints from parents about homework in primary schools. At secondary level, the impact on pupils' grades has been backed up by research, but for primary-aged children there is no such conclusive evidence. The issue for primaries is how to set homework that strengthens the relationship between teachers, parents and pupils. There is also the question, what is homework? Does sending home reading books count? What about expecting parents to listen to children practising their times tables?
This year, the latest findings of one of the biggest long-term studies of education in England, the government-funded Effective Pre-school, Primary and Secondary Education (EPPSE) programme, confirmed a clear upward link between doing more homework and getting better grades.
The guidelines set by the Labour government 14 years ago suggested that when children start secondary school, they should be given 90 minutes' homework a night, rising to two hours in Year 9.
The EPPSE study found that the greatest impact of homework was on those 14-year-olds who appeared to be doing slightly more, between two and three hours a night. Pupils doing this amount were assessed at almost a whole national curriculum level higher in maths at the end of key stage 3 than similar pupils who didn't do any homework. The impact on science was slightly smaller: about three-quarters of a level, on average. On English, it was about 0.6 of a level.
But the researchers found that only around 5 per cent of pupils said they did between two and three hours. The most common amount, reported by 43 per cent, was between 30 minutes and one hour per night in Year 9. A further 22 per cent did less than 30 minutes and 5 per cent said they did no homework on a typical weekday evening.
Pam Sammons of the University of Oxford, one of the principal researchers, says: "The net impact of homework on attainment is pretty large. It is larger than the differences predicted by gender, free school meal status, family's social-economic status or mother's qualification level.
"If some of those who only spent 30 minutes to an hour increased their time on homework a bit, this could reap tangible benefits across the system."
But could the amount of homework children do just be an indicator of the quality of the pupils' school, rather than a cause of their results? The researchers were careful to see what the impact was once school quality was taken into account.
"Our data shows that the homework effect is not attributable to just attending a better school. While only a small number of pupils report spending very high amounts of time on homework - and some may overestimate - the results probably link with student effort and motivation to do well at school, parental expectations and expectations by teachers," Sammons says.
So it seems that there is a virtuous circle, meaning that children who are motivated and have parents and teachers with high expectations receive and do more homework, which leads to better results. But motivation and parental and teacher expectations are not enough - the homework needs to be done, and if a child who doesn't have the same parental and teacher expectations does the work, they will also gain the benefit.
Homework undoubtedly helps - and more homework helps even more.
In her book Battle Hymm of the Tiger Mother, Amy Chua recalls the day her eldest child, Sophia, then in fifth grade (Year 6), came second on a multiplication speed test.
"She lost to a Korean boy named Yoon-seok," writes Chua, a law professor at Yale University. "Over the next week, I made Sophia do 20 practice tests (of 100 problems each) every night, with me clocking her with a stopwatch. After that, she came in first every time. Poor Yoon-seok. He went back to Korea with his family, but probably not because of the speed test."
Pushing too far
So is it simply a question of putting in the hours ... and hours and hours? Actually, the EPPSE research found that studying for more than three hours a night had less impact than studying for two.
Hallam also found this effect. She likens it to learning to ride a bike. "If you imagine a child who is given a certain amount of homework to do to contribute to their understanding of a particular issue or gives them practice in problem-solving, then there comes a point when they understand it or have practised sufficiently. There is no point doing it after that," she says.
"It is like riding a bike. You practise until you can stay on; you don't need to keep practising to keep staying on it."
She points out that the problem with looking at time spent is that some children will have grasped the point within an hour and others will take three hours.
Brian Lightman, general secretary of the Association of School and College Leaders, says that this disparity was the reason he welcomed the scrapping of the homework guidance.
"Homework can be extremely effective in consolidating learning," he says. "But I don't think that you evaluate the quality of a school's homework policy by looking at how long children spend on homework. The issue is not quantity, it is quality."
But quantity can be a helpful rule of thumb, so the question is where to set it?
Harris Cooper, professor of social psychology at Duke University in Durham, North Carolina, began studying homework in 1986 and has carried out several syntheses of research.
He recommends that there is a "10-minute rule", with 10 minutes for each grade, or year group. With a later school starting age in the US, that would translate in England as 10 minutes per night for Year 2, 20 minutes for Year 3, growing to a maximum of two hours: almost identical to the now-scrapped guidance.
So while Mr Lightman welcomes the government's move on homework because it recognises that heads should make these decisions, in practice, he says, "I don't think it will make any difference at all."
What to set
Primary schools have been at the sharp end of the homework debate. In 2008, the ATL education union called for a ban on compulsory homework for primary-age children, saying that children should be able to explore, experiment and enjoy learning without feeling pressured.
"It is a complex, fraught issue. Out of all the areas of research over the years, it's the one thing I know raises hackles on every side, and everyone has a point of view," says Hallam.
"In some countries, primary homework has been banned. But at primary level there is no doubt that learning basic skills can be supported at home. But there is a lot of difference between parents reading with children or playing number games and a child coming home with an hour's worth of writing."
The research suggests that, as in the classroom, the type of task set for homework should depend on what the teacher wants the child to learn. However, researchers would recommend that homework is thought about seriously and the tasks varied, rather than simply asking children to finish off or practise something learned in the lesson.
"Teachers need to ask themselves, 'Is this going to make a real contribution to the learning of these children?'" Hallam says. "Why are they doing it? What is the purpose of it? If they cannot give a good reason why the children are doing it, then they probably shouldn't be doing it. If they've got a good reason - 'Ah, they need to improve their reading' - and think reading with parents will help, then by all means do it."
Setting work for pupils is one thing, but getting them to do it is another. One motivating factor appears to be choice, although this is clearly more appropriate for more flexible types of homework than, say, learning times tables.
The importance of choice
Erika Patall, assistant professor of educational psychology at the University of Texas in Austin, has studied the effects of choice on motivation to do homework. In a 2010 study co-authored with Harris Cooper and Susan Wynn of Duke University, she carried out a four-week experiment in which pupils were randomly assigned to a homework-choice group or a no-choice group.
Those in the homework choice group could pick one of two homework tasks. The teachers then matched the pupils in the no-choice group to pupils in the choice group, and gave the no-choice pupils the homework that their match had selected. The pupils were mixed up each time so that the no-choice pupil couldn't influence their homework through their partner, but this design meant that an equal number of pupils were given each homework choice.
The pupils in the choice group were likely to complete more of their homework and said they were more interested in and enjoyed their homework more. They also scored higher on the test at the end of this period compared with those who did not have a choice.
Choice also has a role to play in homework clubs at schools. Many secondary schools already run homework clubs, which effectively lengthen the school day for pupils who attend. But while the main reason to set up these clubs is to help pupils who may not have the facilities for study at home, important secondary benefits arise from the fact that this is study time that is not part of the usual school day.
In 2001, the results of a three-year longitudinal research project on study support led by John MacBeath, Tony Kirwan and Kate Myers of the University of Strathclyde were published by what was then the Department for Education and Skills. They had tracked 8,000 pupils in 52 schools and found that not only had pupils who attended homework clubs gained on average an extra GCSE (above grade C) than those who didn't, but it also had a favourable effect on attitudes to school.
But when asked why they went, the biggest factor mentioned by pupils was that they didn't have to. The researchers concluded: "Students like choosing to go. They value the relaxed, informal relations with staff, the opportunity to work with peers, more time and help to do work, access to learning resources and being treated as adults and being given responsibility for their own learning."
The push to extend the school day, now backed by all three main political parties, will continue to alter how pupils treat homework, whether or not they attend homework clubs.
Malsis School is an independent prep school in North Yorkshire which has a long-standing no-homework policy, but this does not mean that pupils are learning for fewer hours. In fact, the work is fitted into a school day that is far longer than is typical for a state school. Pupils are at school from 8.30am to 6.15pm on Monday to Friday, except on Wednesdays when there are no lessons in the afternoon, to allow time for sport. The school is also open on Saturdays.
But headteacher Marcus Peel has been careful not to simply bring the disadvantages of homework into school. "In other schools, the whole school may sit down from age eight or nine to do prep in silence. But over 20 years, I've watched pupils do that and they do 15 to 20 minutes of useful work and the rest of the time they are twiddling their pens or whatever," he says.
At Malsis, the study that is needed goes on in lesson times and, while older children (the school goes up to Year 8) have prep slots to prepare them for independent work, the younger children don't.
"You can push children along a bit, but the whole point is getting them to want to work and you can definitely overload youngsters. We want them to be excited about academic work, not put off it," Peel says.
That same thought will be shared by many schools now looking once more at how much homework they set.
The 1998 guidelines, now dropped:
Years 1 and 2: one hour per week
Years 3 and 4: 1.5 hours per week
Years 5 and 6: 30 minutes per day
Years 7 and 8: 45-90 minutes per day
Year 9: one to two hours per day
Years 10 and 11: 1.5-2.5 hours per day
Sammons, P. Sylva, K. Melhuish, E. Siraj-Blatchford, I. Taggart, B. Toth, K. Draghici, D. and Smees, R. Influences on Students' Attainment and progress in Key Stage 3: Academic Outcomes in English, Maths and Science in Year 9 (2011). Part of the Effective Pre-School, Primary and Secondary Education project (EPPSE 3-14). Institute of Education, University of LondonDepartment for Education.
Hallam, S. Homework: The Evidence (2004). Institute of Education, University of London
Cooper, H. The Battle over Homework, third edition (2006). Corwin, California.
Gill, B. and Schlossman, S. "A Nation at Rest: The American Way of Homework" (2003), Educational Evaluation and Policy Analysis. Vol. 25, number 3, pp319-337.
HOMEWORK WITH PURPOSE
Let parents know the purpose of the homework given. Do you expect them to help their child or not?
Vary the homework - sometimes drilling, sometimes discovering.
Never give homework as punishment. This implies that having to learn is a bad thing.
Usually, teachers should try to state:
the purpose of the assignment;
how the assignment is related to the topic under study;
how the assignment might best be carried out;
what the student needs to do to demonstrate that the assignment has been completed.
Online systems can help to inform parents of homework you have set, if their children have a habit of forgetting to do it or forgetting to tell their parents about it.