Lunchtime classes, evening classes, even perhaps the best teachers, are now devoted to the D-grade student - that is to say, the student who, with suitable encouragement, stands a chance of gaining a C-grade and who is crucial to a school’s place in the academic rankings.

Meanwhile, runs the argument, both the strongest and the weakest are denied their rightful share of attention. Professor Carol Fitzgibbon, an expert on “value-added” analyses from Durham University, is one of many who believe the practice is deeply inequitable.

“This is a direct consequence of the way the league tables are set up, ” she said. “It’s not the fault of schools. These league tables do not weight each student equally.” She would prefer to see schools given an average grade score per pupil to make sure that all students’ results are thought to be significant.

Neglect of the weakest could prove more damaging if candidates who fail to do better than a G find themselves barred from the increasingly popular work-related GNVQ.

Not that there is much evidence to suggest that lower-ranking pupils have been neglected. On the contrary, says Professor Michael Barber, of London University’s Institute of Education. He points to a 1995 study of London comprehensive schools by Keele University which found that schools producing more A to C grades had also done better in producing D to F grade passes.

“If you were blatant , you might disadvantage some pupils,” he said. “But most schools aren’t blatant. For example, while a lot of schools target certain pupils, they say to the others, ‘If you want the same support, we’ll provide it even though we haven’t picked you out.’ That way there’s a sense of fairness. ”

Tricks such as offering extra coaching have attracted a lot of attention. But there is consensus among academics that such marginal devices are a distraction from the real business of improvement.

“These measures can certainly help if you have a few months to go before the exams; they are not to be sniffed at,” said Professor Barber. “But they don’t amount to a long-term improvement strategy. They won’t make up for weaknesses in the core job of the school.”

Professor David Reynolds, director of the International School Effectiveness Project at Newcastle University, takes up the point: “I’d want to draw a distinction between tactics and strategy. By tactics I mean things like playing around with entry patterns. These are short-term tactics that all schools appear to be using. But schools begin to sort themselves out only when they think beyond tactics to strategy.” And strategy, he says, is not necessarily a straightforward matter.

In the first place different schools probably require different sorts of help. “For example, if you’re not an effective school, it’s highly likely that the improvement has to be from the top down,” said Professor Reynolds. “But if you were to try that sort of method with highly successful schools, you wouldn’t last long. These schools probably need no more than clearer targets.”

Then there is the danger that teachers will take on more than they can handle: “Schools can attempt to do too much too quickly and exhaust themselves. That is fatal.”

A more sensible option is to generate momentum with uncomplicated initiatives offering an obvious impact. “A sensible school would go for something easily achievable, such as addressing the state of the buildings, or developing an attendance policy.” There is also a danger of re-inventing the wheel: it may be better to buy an off-the-shelf design for improvement such as “assertive discipline” or the American “Coalition of Essential Schools”.

If there is one obvious target for secondary schools with an eye on the league tables, Professor Barber believes it is the teaching at key stage 3. “A lot has been done on key stage 4, but the mine is largely used up. At key stage 3, however, there are an awful lot of pupils who aren’t being challenged. There needs to be a new focus on pedagogy at key stage 3, getting teaching and expectations sharper and more focused.”

For the longer term, there is general agreement that improvement will rely on schools’ ability to absorb the results of research and analysis. Professor Reynolds believes that, at present, such information is most notable by its absence.

“It seems to be that schools don’t get information about what the very best schools are doing directly circulated to them. Any other profession routinely identifies its leading edge. There’s no sense of that in British education. ”

There are some positive signs, however - not least from the Government’s agencies, if not the Government itself. The Teacher Training Agency is doing its best to direct research money on to the question of what works in the classroom.

Meanwhile, the School Curriculum and Assessment Authority has already announced plans to circulate massive amounts of information about the relative performance of schools - including information from OFSTED’s ever-expanding data base. Under SCAA’s regime of target-setting, each school will be able to set itself against nationally derived “templates” that take social and economic circumstances into account.

And this, says Professor Fitzgibbon, could prove sufficient. “It may be all that’s needed. If you give people evidence that others in a similar situation are doing better, they will find ways to improve.”