As the Government prepares the first national guidelines on how much homework pupils should do and pours money into homework clubs, academics from London University's Institute of Education have attacked the view that homework necessarily raises standards.
In a review of nearly half a century's research on homework, Dr Susan Hallam and Dr Richard Cowan also warn parents who help their children with homework that their efforts may be turn out to be counter-productive.
Homework tends to increase when schools are under pressure, says the paper to be presented at a British Psycho- logical Society conference on education. In 1883 after "payment by results" for teachers was introduced, time spent on homework rose sharply until parents and social reformers campaigned for a reduction.
In 1929, a leading education journal asked "Is homework necessary?" after a survey found that 11-year-olds were doing between one and 12 hours a week and those over 11 between 7.5 and 20 hours. (The Government proposes just over four hours for 11-year-olds and up to 12.5 for secondary pupils.)
By 1935, school inspectors were recommending a reduction for the under- 12s but in the Sixties a government report called for more after research revealed a divide between grammar school pupils, who did a lot, and secondary modern ones, who did very little.
The new paper argues that most research into the effect of homework on standards has been unable to separate it from the effect of other factors such as ability, the quality of teaching and teachers' expectations.
Much of the evidence in this country comes from inspectors and, most recently, the Office for Standards in Education (Ofsted). It is based on observations in schools and does not take into account individual achievements.
An Ofsted report three years ago commented cautiously of homework: "Where it is treated seriously by staff and pupils, it has the potential to raise standards." Dr Hallam's paper says that homework does appear to help pupils to pass secondary school exams. It continues: "Simply increasing pupil workloads will not of itself improve standards and in some cases where pupils become overloaded may have a negative impact on the performance of individual children."
What counts, say the authors, is quality not quantity, and there should be more investigation of what types of homework are most effective.
Many people see disadvantages in homework - it can increase negative attitudes to school and motivation, lead to cheating and copying and increase divisions between children from different backgrounds. But it also offers the opportunity to spend more time on a task, to memorise facts and to encourage independent study and time management.
Recent studies suggest that parents who spend hours helping their children with homework may be misguided. One found mothers' help had no effect on children's achievement. Another, from London University's Institute of Education, suggested that middle-class parents who help with homework may worsen family tension by comparing siblings.
Those parents infuriated by offspring who persist in writing essays in front of the television should keep cool. Background noise may be an aid to learning.