It is illegal in Sweden and leading education researchers vehemently reject it. So why do we set by ability in the UK? Every year, arguments break out in school departments over timetabling, with dejected teachers feeling the pressure of being given the lower set. Then, in September, millions of children start school disappointed at being classed as "low-ability".
Politicians across the political spectrum in the UK continue to back the long-debated setting process, in spite of the fact that those who have led years of research into the wider implications say setting does more harm than good.
Leading academics - including Jo Boaler, Marie Curie professor of maths education at Sussex University, and Professor Susan Hallam, dean of the faculty of policy and society at London University's Institute of Education - claim social pressure is holding our schools back as politicians bow to the snobbery of middle-class parents.
Academic research suggests that while mixed-ability teaching is undoubtedly harder for teachers, it boosts achievement for low-ability and middle-ability children. It also enables children to develop socially and avoids the psychological damage and feelings of inadequacy that setting can reinforce. Professor Boaler and Professor Hallam say setting compounds a historical cycle of poor performance among children from low socioeconomic backgrounds and reinforces the notion of boys as low-achievers.
While setting has been commonplace since the early 20th century, it fell out of favour in the 1960s before making a return in the Thatcher era. In 1993, all primary schools were encouraged to introduce setting by the Conservative government's Department for Education. In 1997, the new Labour government's Excellence in Schools white paper stated that "setting should be the norm in secondary schools", adding: "In some cases, it is worth considering in primary schools."
In 1998, the Ofsted chief inspector's annual report said the organisation of pupils into sets was increasing, especially in Years 5 and 6 for maths and English. Most primary schools used setting in those years only, with the proportion of pupils organised into sets for at least one subject falling steadily through the younger year groups. Ofsted now says it does not comment on decisions made by schools on setting by ability.
The social-class filter
In the UK today, an estimated 40 per cent of lessons in secondary schools are streamed. And in primary schools, it is one in seven lessons.
Once a pupil has been placed in a set, it is unlikely they will move to a different one: a 2002 report noted that 88 per cent of children organised into sets at the age of four remained in the same group until they left school.
Professor Boaler says research on ability grouping has persistently shown high correlations between social class and setting, with social class functioning as a subtle filter that results in the over-representation of working-class children in low-ability groups.
Meanwhile, Professor Hallam and her colleague Samantha Parsons have used information from the Millennium Cohort Study of 8,875 children born between 2000 and 2001 to explore setting (see box on page 7). Professor Hallam presented their most recent findings in September at the British Educational Research Association conference.
"Teachers and students hold a perception that ability setting is better for them, yet a high proportion of students say it's still too easy with setting," she says (to TES). "And politicians believe that the general public thinks setting is common sense and therefore promote it; if they advocated mixed-ability it would be a vote-loser at the moment and they dare not take the risk."
The pros of mixed-ability
In terms of the advantages of not setting, Professor Hallam says pupil feedback shows that they learn more from each other - and that parents who say "I don't want my child to be held back" are missing the point. "High-ability students can be greatly advantaged by mixed-ability teaching as this means that in group work they can take on roles such as leading the group, developing skills they might otherwise not get the chance to," she says.
And while mixed-ability classes do not necessarily raise academic achievement for all, Professor Hallam says schools should consider that setting can greatly affect pupils on a social and personal level.
"When you talk to headteachers, they say they set as it is best for their students and research tells them so," says Professor Peter Kutnick, chair-professor of psychology and education at Hong Kong University. "But if you push on this matter, very few headteachers can tell you which research they are referring to."
Professor Kutnick says his lifelong research has shown that if you compare mixed-ability classes with setted classes, pupils in mixed-ability groups make the most progress and gain more socially.
However, the research does not shy away from the fact that teachers across the UK feel more comfortable teaching by ability. A full timetable leaves little room for detailed differentiated planning, making mixed-ability teaching a far greater challenge. "Mixed-ability teaching is difficult and requires advanced knowledge and practice of pedagogy, but such grouping is more equitable and it must surely be worthy of a government's investment," says Professor Boaler.
For improvements to be made to mixed-ability teaching, and for it to be a more appealing option for teachers, Professor Boaler says investment must be made by the Department for Education. "At the moment, teachers don't have the materials they need and don't know what to do. They require weeks of pedagogical training in differentiation to help them cater for all students. A bank of materials needs to be developed."
And while Professor Boaler admits that teaching mixed-ability classes is harder, the child with lower predicted grades will be the winner. After all, why bother when you are told you are failing? "It's much harder for teachers to motivate students when they have the bottom set," she says.
Professor Hallam emphasises the importance of teachers feeling comfortable with mixed-ability groups. "If working with mixed-ability, teachers will find it easier to differentiate if they differentiate by the outcome: the resulting work of the lesson rather than the instruction.
"And if teachers are having to work with sets, departments should ensure that these sets are flexible and that students can be moved between sets assessment by assessment. But studies show that often class sizes effect that flexibility and class sets are not fluid."
According to Professor Kutnick's research, it is really important for teachers to think about their group work and the foundations for it. "While there's also a temptation to set by ability again in group work, students benefit from a collegiate personality in the classroom if the teacher can get everyone working together," he says.
He explains that a lot of security and early communication work should go into preparing for this, with the teacher reinforcing the equal importance of everybody in the classroom.
Self-belief improves performance
In September 2006, the Labour government's Department for Education and Skills published Grouping Pupils for Success, which provided guidance for schools on effective strategies for grouping and setting (http:bit.lyu41gbl).
But currently the DfE does not specifically direct schools on setting, saying they should make their own call. "We know that effective grouping of pupils can raise standards and better engage pupils in learning in some, though not all, subjects," a spokesman says. "Teachers are best placed to know the needs of their pupils, so quite rightly it is down to schools to decide how and when to group and set pupils by ability."
Much research shows that there is a big push from parents for setting because they do not want their children to be seen as low-achieving. Professor Boaler likens the mindset to "a class barrier". In research she carried out, one student compared his feeling of being setted to that of being in a "psychological prison".
"A lot of people think that the best way to teach kids is to narrow the range, rather than thinking about the psychological damage it is doing," Professor Boaler says.
She argues that pupils should be encouraged to change their thinking about themselves, so they realise their abilities are not fixed and can grow. "Research has shown that if children are encouraged to think outside of their own beliefs of themselves, and to believe that anything can happen, their performance improves," she says. Her research also shows that the later you set, the longer you give children to reach their potential.
Data from the Organisation for Economic Co-operation and Development indicates that countries which get strong attainment results in tests tend not to stream or to set until just before exams, when students are a lot older.
Professor Hallam says that, on average, when you compare attainment between high-ability pupils in sets and high-ability pupils in mixed groups there is no difference. But, worryingly, setting can limit attainment for low and middle-ability groups.
"The quality of teaching is crucial and attitudes of high aspiration of a teacher with all their students, whatever their ability, can help setting be successful," she says. "But too often teachers and students in the lower sets are demotivated, and this leads to a downward-spiralling lack of motivation - and often poor behaviour follows.
"Maths and modern foreign language teachers feel most strongly that setting is most necessary as these are subjects that build up step-by- step, but evidence has shown this perception to be inaccurate."
Many teachers and parents back setting as an opportunity to minimise the impact of bad behaviour on hard-working students. But Professor Hallam dismisses this and says organising pupils into sets can actually have a negative effect on behaviour. "The issue about behaviour has to be separated out as to whether you set by ability. Either way, schools have to deal with the behaviour, so it should be dealt with foremost in discussion with the student, outside of what ability a student is."
She says the pattern of placing badly behaved pupils in lower sets also adds to teachers' fear of taking on those low-ability sets and perpetuates an attitude of negativity and low aspiration in the classroom.
But Professor Hallam adds that schools could avoid the psychological impact of setting if they called sets by different names, rather than the obvious number or letter, and by ensuring that teachers have equally high expectations of every set. "They can also do this by embracing a school ethos where effort as well as achievement is rewarded equally," she says.
Cornelia Lucey is an English teacher at the Bushey Academy in Hertfordshire
Research by James Kulik and Chen-Lin Kulik of Michigan University found that gifted and talented pupils "achieved significantly more when grouped by ability compared with those who were not, but only when they were provided with programmes that were designed specifically to meet their needs".
The research also found that pupils who were grouped by ability for a specific subject had a better attitude towards that subject, but that grouping did not change attitudes about school in general.
Boaler, J. Experiencing School Mathematics: teaching styles, sex and setting (1997). Open University Press
Boaler, J. "Setting, Social Class and Survival of the Quickest" (1997). British Educational Research Journal, 23 (5), 575-95
Boaler, J. "When Even the Winners are Losers: evaluating the experiences of 'top set' students" (1997). Journal of Curriculum Studies, 29 (2), 165-82
Kutnick, P. Sebba, J. Blatchford, P. Galton, M. and Thorp, J. The Effects of Pupil Grouping: literature review (2005). The former Department for Education and Skills
Knowledge and skills for life: first results from Pisa 2000 (2001). Organisation for Economic Co-operation and Development
Hallam, S. Ability Grouping in Schools: a literature review (2002). Institute of Education, London University
Hallam, S. Ireson, J. and Davies, J. Effective Pupil Grouping in the Primary School: a practical guide (2002). David Fulton Publishers
Hallam, S. Ireson, J. and Davies, J. "Grouping Practices in the Primary School: what influences change?" (2004). British Educational Research Journal, 30 (1), 117-40
WORKING WITH SETS
What can teachers do to reduce the negative impact of organising pupils into sets by ability?
While there is very little research that shows setting is helpful for children, and none that suggests it improves performance, Professor Peter Kutnick of Hong Kong University says that if ability is part of a school's culture, teachers can support pupils by ensuring that group work is well-structured, differentiated and planned - and utilises the strengths of all pupils.
To make the best of ability teaching and group work, try these tips:
Do not let pupils sit with their friends, because children of similar ability levels will naturally converge.
Do not ask the teaching assistant to work with the lowest-ability child - they most need the support of the teacher.
Bring in early communication skills, planning lessons on topics such as "What did I do yesterday?", "How do I feel today?" and other spontaneous discussion points. These are easily accessible communication topics and encourage discussion between all pupils. Then teachers can think about building up to pair work before branching out to group work. This will develop more of a whole-class atmosphere.
When group work takes place, rules, groundwork, trust and sensitivity need to be constantly reinforced with pupils. According to Professor Kutnick, these skills need to be treated like a spiral. "After every half-term, students lose group-working skills and teachers need to reiterate the groundwork again. You should expect it to slacken over time like a spiral and to need to rebuild it," he says.
Around one in six Year 2 children are already in streamed classes, recent UK research has found. The report, based on 8,875 children born in 200001 surveyed as part of the Millennium Cohort Study, showed that 16.4 per cent of Year 2 pupils were in streamed classes.
Logistic regression analysis showed that the best predictors of a child being in the top set were whether they were born in the autumn or winter of 2000, their cognitive ability score and their parents owning their own home.
Predictors of a child being in the bottom set were being a boy, being born in the spring or summer of 2001, having a behaviour problem, being born into a lone-parent family and having a low cognitive ability score.
Hallam, S. and Parsons, S. Prevalence of Streaming in UK Primary Schools: evidence from the Millennium Cohort Study. A paper given at the 2011 British Educational Research Association conference.