Grouping pupils by ability in streams and sets has no significant effect on their performance, a major research study has concluded.
Even high-ability pupils do not always appear to benefit from streaming and setting. In fact it can harm them, by over-inflating their self-esteem,the report's authors say.
The findings will be challenged by many schools which have shifted away from mixed-ability teaching in recent years. They may also displease the Government, which has angered Labour party members by supporting ability-grouping.
Last year's White Paper, Excellence in Schools, claimed that setting had proved an effective way of teaching mathematics, science and languages. It stated: "Unless a school can demonstrate that it is getting better than expected results through a different approach, we do make the presumption that setting should be the norm in secondary schools."
Setting has also been endorsed personally by the Prime Minister, Tony Blair. "The modernisation of the comprehensive principle requires that all pupils are encouraged to progress as far and as fast as they are able," he said in a 1996 speech. "Grouping children by ability can be an important way of making that happen."
The new analysis of 20 research studies in the UK and the United States has, however, raised fresh doubts about the supposed benefits of ability-grouping.
It has also reaffirmed that the practice can have a negative effect on the attitudes, motivation and self-esteem of lower-ability pupils. Boys, pupils from working-class families, ethnic minorities and summer-born children are also more likely to be disadvantaged, the National Foundation for Educational Research says.
The NFER researchers did, however, find that within-class grouping in primary schools can improve pupils' attitudes, self-esteem and achievement,regardless of ability level.
The authors of the NFER report also point out that the way in which pupils are allocated to ability groups is often subjective and inconsistent. Furthermore, few schools enable pupils to move between streams, sets and groups.
Laura Sukhnandan, an NFER research officer, said: "The research in this area is complex. Different individuals react in different ways: some thrive within a competitive school environment, whereas others prefer mixed-ability classes.
"But some clear messages do emerge. Streaming and setting, compared with mixed-ability teaching, have had no overall impact on pupil achievement.
"They also reinforce social divisions. These are issues that need to be addressed."
The report concludes that schools, rather than grouping by ability, could try to focus on pupils' individual needs. "This could be achieved through greater modularisation of the curriculum, an increased emphasis on independent learning and improved library and information technology resources," they suggests.
Streaming, setting and grouping by ability, by Laura Sukhnandan with Barbara Lee, is available, priced Pounds 6, from the Communications Unit, NFER, The Mere, Upton Park, Slough, Berks SL1 2DQ. Tel. 01753 574123747281 Research Focus, 19
A HISTORY OF DIVISION
* Streaming - the practice of dividing children into ranked classes according to their general ability - was the standard form of pupil organisation in most primary and secondary schools in the 1940s and 1950s.
Its popularity nosedived in the 1960s - partly because there was little research evidence to justify its existence, but also because of growing concerns about equal opportunities.
After the 1967 Plowden report on primary education, streaming was increasingly replaced by mixed-ability teaching. But the latter approach also came under fire in the 1980s when it was blamed for pupil underachievement.
Since then the practice of setting pupils (grouping by ability for specific subjects) has enjoyed a resurgence.
In part this is a reaction to the tiered format of the national curriculum. But setting is also seen as a means of raising standards and has been popular with many teachers and middle-class parents.
Mixed-ability teaching is still commonplace in primary schools and the first two years of secondary. However, setting is increasingly used for maths and English lessons in primary school, and OFSTED inspectors' reports suggest that secondary schools have resorted to more and more setting over the past five years.