I ASKED colleagues what they thought about setting. “That depends which set I’m getting,” said one. If offered the top set he would be in favour of setting; if offered the bottom set he wouldn’t.

Neatly, if jokingly, this describes the two-edged sword that setting and streaming represent. No one ever calls for the reintroduction of junior secondary (or in England, secondary modern) schools. It is always the “senior” version (grammar schools) that feature in headlines.

What is wrong with bottom sets is that they consist of individuals who have been told they are failures. Not just failures in their latest piece of work, but people not fit to get in the way of those who are going to succeed. “Life is like that,” you might say, “so they will have to get used to it.” School is not “life”, however. Pupils should be given the support and protection that are their due.

“Look at the effect on high achievers,” say supporters of setting. “They get on so much faster.” Well, eventually academic attainment will inevitably be very different for high and low achievers. What is at issue here is the stage at which such explicit differentiation becomes desirable.

It is the Scottish democratic tradition to leave categorisation as late as possible, and always to keep open the door of opportunity. We should build on this and be wary of labelling pupils too early. In many parts of Scotland there is already a kind of social apartheid, with working-class housing estates and middle-class suburbs.

Setting further increases this separation, to the detriment of both individuals and society. Its advocates often overlook the fact that learning is a social as well as an individual activity.

The often stated aim of the Government (indeed, of all politicians) is to raise standards. Let us consider an analogy here. If you are trying to raise your mark in a subject, you concentrate on your weakest not your strongest topic. So why, in seeking to raise standards, do advocates of setting or streaming almost always talk about the benefits for the strongest or highest achievers?

The answer, surely, is that targets and public perceptions of all sorts concentrate on figures such as the percentage of a cohort gaining three or five Highers, which leave beyond the pale the denizens of bottom sets. Targets should reflect what we want to see in our education system. Instead, a selection of targets is determining how the system is organised. This topsy-turvy logic is distorting the educational landscape.

The review of educational research carried out for the Scottish Office by the Scottish Council for Research in Education showed there was no clear evidence that setting produced educational advantages. In secondary schools, no effects of ability grouping on pupil achievement for any levels of ability were shown.

American research evidence was found unequivocally to refute the assertion that streaming increases pupil attainment in primary schools. A similar exercise by the National Foundation for Educational Research in England returned similar results. Research provides no support for separating pupils according to achievement as a way of improving the learning experiences of all pupils.

The fact is that those teachers who support setting in their subject do not do so because they have been convinced by the research evidence. They do so because they feel it benefits pupils and because of the job satisfaction of being able to teach their subject at a high level to those able to appreciate it.

With few exceptions, teachers and politicians alike do not discuss bottom sets when enumerating the advantages of the system. Setting “allows the best ones to work ahead”. Above all, perhaps, it is seen as “removing” troublesome pupils: those whose low levels of achievement fuel low self-esteem, disaffection and further lack of achievement.

Mixed-achievement teaching is demanding. Differentiation is built into the thinking and practice of classroom teachers. There are many ways of implementing it, and they are not all equally successful. But no teacher who took a set lesson without regard to the range of abilities of the pupils would feel they had done a good job. Management of a good lesson is a skilled task and it is sometimes impossible to satisfy every demand. Teachers want to do their best for pupils, and for this reason they see a reduction in the spread of attainment within classes as a way of being able to give a more tailored service.

A striking feature is the devastating effect on the self-esteem of pupils in bottom sets. Glib talk about low teacher expectations usually betrays a lack of experience. It is not the teachers’ expectations that are paramount, but the pupils’ often unshakeable belief that they are “thick” and have been rejected by the system. It is very hard to counter this self-evaluation, given that the school has explicitly placed them in the category of low achievers.

To pupils already locked into a vicious cycle of low self-esteem and low achievement are added those who have struggled and failed and those whose sometimes catastrophic home backgrounds have made success impossible. These last two often join the ranks of the disaffected, sometimes in a most aggressive way.

Achievement is often linked to social class. Research has shown that when setting is employed the link becomes much stronger. In determining exam success at age 16, social class becomes more important even than ability on entry to the secondary school. Premature setting emphasises and crystallises class diffusion.

Graham Dane is a member of the education committee of the Educational Institute of Scotland. He writes here in a personal capacity. References to quoted research are available from his e-mail address: dane@enterprise.net.