Why setting may be harmful
Grouping pupils by ability in schools has become more popular as the Government presses for higher standards. Both the Prime Minister and Education Secretary David Blunkett have called for more setting.
At present, most secondaries use sets for at least one or two subjects. A minority teach pupils in mixed-ability classes for all subjects.
Very few use streaming, where pupils are grouped into ability classes and have all their lessons with that class. But a study of more than 6,000 pupils in 45 secondaries has found setting does not necessarily enhance academic performance.
Dr Sue Hallam, from London University's Institute of Education, analysed the test results at 11 and 14 of pupils in mixed-ability groups and sets. She found progress in English and science was not related to setting.
And, in maths, pupils with lower attainment at the end of key stage 2 actually made more progress in schools with mixed-ability groups
Able pupils, however, flourished in schools which used setting.
The report, presented at last week's American Educational Research conference, also found that pupils' self-esteem was higher in schools where there was some setting.
Overall, 60 per cent of pupils said they preferred setting. Girls liked it more than boys and top-set pupils tended to prefer setting to those in lower groups.
Far more boys wanted promotion to the top group than girls. More girls than boys wanted to move down, particularly in maths where nearly a quarter of the girls wanted to drop to a lower set.
The study also looked at setting practices in primary schools. Teachers said that narrow ability ranges helped them to pitch their teaching at the right standard, but some were reluctant to teach low-ability groups because it was demoralising.
More than 60 per cent of the 765 primaries surveyed in this study had altered their system of grouping pupils since 1997.
"Ability Grouping in Schools: Practices and Consequences" by Dr Susan Hallam and Dr Judith Ireson.