The latest figures on school exclusion make depressingly familiar reading (p4). There may - or may not - be reasons for rejoicing at the 11 per cent drop in the incidents of pupils given their marching orders. As with all statistics, it depends on how you read them. The total represents 57 incidents per thousand pupils, which means that schools are coping in the cases of 943 per 1,000. And while the headlines about increased exclusions for assaults using a weapon ignore the fact that this is a mere 0.9 per cent of known exclusions, that is no comfort if you are one of the 366 pupils or members of staff on the receiving end.

Perhaps there has been an overall improvement in exclusion rates because of the increasing use of alternative strategies and sanctions. Perhaps the additional resources and creative energies being invested in the problem are beginning to pay off. Perhaps teachers are being given more training. Perhaps changes in the curriculum are filtering through into behaviour. Certainly no area of school life has had more attention paid to it since the Pack committee on discipline reported in 1977.

But the really depressing feature of the figures is the stubborn correlation which persists between educational and social circumstances, reinforced by the familiar sight of Glasgow and East Renfrewshire at opposite ends of the spectrum. It remains the case that pupils are eight times more likely to be excluded if they are from deprived backgrounds compared to those from the least deprived areas, with a similar ratio if they are in care or not. Primary pupils with special needs are an astonishing nine times more likely to be excluded than those who do not require additional support.

If the 11 per cent fall in exclusions represents genuine progress, schools must be congratulated for their heroic efforts. But it is clear that substantive problems remain, and schools cannot deal with them alone.