Both the Office for Standards in Education and the Qualifications and Curriculum Agency (formerly SCAA) are about to send schools information to help with this. OFSTED’s evaluation of how well a school is performing compared with other similar schools is known as a “Panda”, short for Performance and Assessment.

The QCA will be sending schools “benchmark” figures designed to help teachers and governors judge their pupils’ performance against other schools with similar intakes.

Using this information, schools will be required to draw up targets in the three core subjects - English, maths and science - aimed at improving results.

All this will be supervised by the standards and effectiveness unit at the Department for Education and Employment, whose guidance booklet Targets into Action last year set the scene. It contains suggestions on how targets could be set - as part of a five-stage cycle - and an explanation of benchmarking.

The QCA benchmarks are a set of tables which provide average attainment scores in national curriculum tests and GCSEs for different kinds of schools. Schools are categorised according to social factors affecting their pupils. Primary schools with more than 50 per cent of pupils for whom English is an additional language form one group of “similar” primary schools and there are five others determined by the percentage of children eligible for free school meals. For secondary schools there is a separate group for selective schools and six others based on the proportion eligible for free meals ranging from 5.5 per cent or less to 35 per cent or more.

For each group, a table shows the percentages of pupils who attained target levels in national curriculum tests or grades at GCSE in schools that are below average, average, about average and in the top 5 per cent (see box below). Schools are expected to use this information to set their own targets for improvement, though these will have to be approved by governors and the local authorities who have been set literacy targets of their own to achieve.

David Hawker, who is masterminding the benchmarking exercise at the QCA, believes heads and governors will welcome the information. But he admits there is a lot of explaining to do. Asked if they found the way the benchmarking tables were presented understandable, 54 per cent of heads in a survey said “yes”. But 30 per cent said “no”.

“The crucial factor is whether we and the DFEE and OFSTED can make sure we are giving consistent messages and that local education authorities can provide the right kind of support to schools to do their target setting successfully,” he says.

“It’s all very new, and inevitably some people are going to get it wrong to start with. We will have to work with them to make sure it is done successfully.”

The QCA, with OFSTED and the standards and effectiveness unit, is planning a series of conferences for local education authorities, starting next month, to explain benchmarking. Authorities will be expected to pass on this information to heads.

OFSTED believes the benchmarking is a breakthrough. “For the first time schools will have a management tool that draws on the wealth of inspection and other school data to help them in the process of performance review, setting targets and improving standards,” a spokesman said.

But in the depths of Whitehall, ministers must be asking themselves whether their own improvement targets are likely to be met. Kevin McAleese,head of Harrogate grammar school, for one, is sceptical. He points out that he has run his own benchmarking project for the past three years, involving more than 100 schools. Yet no one from the standards and effectiveness unit, OFSTED, or the QCA has asked him how it has worked so far. He does not believe anyone at the DFEE has much to offer schools.

“There’s no doubt that benchmarking and target setting will benefit schools,” he says. “But unless the school actually wants to take part, it’s in danger of becoming a bureaucratic exercise. Schools have to want to improve.”