By Gary Eason BBC News Online education editor

A large number of schools could be in for a shock - or a pleasant surprise - when a new sort of league table goes nationwide.

A pilot study of 495 of England's primary schools shows that many with apparently good test results are "coasting" - doing little if anything to raise their pupils' achievement.

And some with superficially poor results are shown to be working very hard to raise their pupils' achievement between the tests they take when they are seven and those taken at 11.

An official evaluation of the pilot study is beginning straight away, with a view to nationwide publication of "value added" scores in next autumn's league tables.

The new measure uses a similar technique to that first tried in secondary schools last year.

The issue

The 495 schools chosen to take part are, the Department for Education says, "broadly representative of schools across the country".

Ten of them were special schools - for children with more severe learning difficulties of one sort or another.

Teachers have long complained that league tables using "raw" test results are unfair in that they take no account of how well pupils are already doing when they enter a school.

The progress schools help their pupils to make is usually referred to as "value added".

How it works

The department's statistical experts worked out a value added score for each pupil by comparing their Key Stage 2 performance with the middle performance of other pupils with the same or similar prior attainment at Key Stage 1.

Then they averaged these individual scores to give a score for the whole school which is a number based around 100 - for example, 98.5 or 103.9.

The resulting measures were divided into five bands: A, B, C, D and E.

The experts constructed these deliberately so that, nationally, band A relates to the top 5% of schools, band B the next 20%, band C the middle 50%, band D the next 20% of schools and band E, the 5% of schools with the lowest value added measures.

Results

Because all the schools took part voluntarily, it would be unfair to rank them in any way.

But analysing the results throws up some interesting findings, especially when their "value added" grade is compared with their "aggregate" score out of 300, the basis of the usual league tables.

The A band schools range from 300 down to 258, B from 300 to 186, C from 300 to 173, D from 274 to 138 and E from 261 to 67.

So there is a tapering-off of performance - the ones showing the worst value added also produced the worst "raw" scores.

But even a school getting just 186 on its aggregate score makes it into band B on a value added basis.

And some with very respectable test results - an aggregate of 261 is not bad - nevertheless fall into band E on a value added basis.

'Secret' data

These findings might not surprise the schools' head teachers, who get a package of data each autumn which shows them how well they are performing compared with other similar schools.

But this is rarely shared with parents and often not even with school governors, who might therefore read next year's national "value added" tables with considerable interest.

Springfield primary in West Berkshire is a "typical" primary school.

It had 45 pupils taking the tests this May - the national average - with 22.2% of those having special needs, just slightly above the national average.

In the English, maths and science tests, 67%, 76% and 87% of them achieved at least the expected level. The national averages this year were 75%, 73% and 86%.

It had an aggregate score of 229 - not precisely the sum of the three percentages, because of rounding - ranking it 35th out of 47 schools locally in the conventional league tables.

But with a value added measure of 101.1 it is in band B - showing that it is above average in bringing on the children from their earlier test results.

So on that new measure the school is doing well by its pupils.

Special efforts

Special schools, on the conventional measure of their test results, make a poor showing, not surprisingly - if they have any results at all.

The "best" in the pilot, Orrets Meadow School in Moreton, Wirral, had 34%, 24% and 59% of its pupils achieving the expected level in their English, maths and science.

Yet with a value added measure of 104.4 it is shown to be an A-grade school.

And none of the special schools for which results are published rates less than value added band C.