Value-added league tables for primary schools are not going to be possible for a long time, if at all, according to a report commissioned by the School Curriculum and Assessment Authority.
The interim report, written by the Curriculum Evaluation and Management Centre at the University of Newcastle upon Tyne, says early evidence shows comparisons of pupils' key stage results with GCSE results offer a suitable basis for value-added measures, but small classes and transient school populations make it more difficult in primary schools.
Peter Tymms, co-author of the primary school document, said the four-year gap between key stages 1 and 2 created difficulties. He said the research showed that looking at averages and scores for whole schools would not work because of the variation caused by having different school populations at each stage of the analysis. There is also concern about the validity of current KS2 results.
There may be more of a case for secondary schools where the numbers involved make statistical analysis more meaningful. Mr Tymms said: "There is great dissatisfaction at the league tables for secondary schools, which have been around for a few years now. Schools want there to be a fairer system and to be able to put the record straight."
He does believe that both primary and secondary schools, using information based on pupil performance, will be able to judge their progress. While there may be a large difference between the levels reached by different schools, the progress made from pupil entry to exit may not be so different.
Michael Barber, dean at London's Institute of Education and one of the SCAA value-added working party, agrees that value added has to be based on data about individual pupils.
But, according to the report, because almost half of children in the sample changed schools before KS2, then a cohort of 60 would be needed to produce satisfactory KS2 value-added indicators. Only 15 per cent of schools have cohorts of 60.
If value-added results are to be published, the issue of which children to include becomes contentious. If a child leaves halfway between KS1 and KS2, can the two schools each claim half the credit?
"In theory value-added is worth doing for management information purposes, " said Professor Barber. "But I suspect that if we are ever going to be in the position to have national value-added league tables it will be a long time hence."
While the SCAA report says value added is more feasible for secondary schools, it was still difficult within a two-year gap to match results of pupils from KS3 to GCSE. Professor Carol Fitz-Gibbon, one of the report's authors, believes that secondary schools will be able to do a meaningful analysis of their own results at this stage. "Any maths department will be able to process the information on to a spreadsheet," she said. However, the jump to national comparisons is more complex and further work needs to be done.
The problem the Newcastle team had at this stage was the paucity of data; only one local authority, Avon, had kept the records of individual pupils' KS1 scores. The data used was that of the Performance Indicators in Primary Schools, a project supported by the National Association of Head Teachers, in Bradford and Solihull.
In his foreword to the report Sir Ron Dearing, chairman of SCAA, said: "At the outset we must recognise that there are many ways in which schools 'add value' to their students - personally, socially, culturally and spiritually - which this report does not tackle. We are dealing with just one aspect, the attainments of pupils in the school as measured by the end-of-key-stage assessments. The others are no less essential and must feature as part of the context of the whole school within which any value added may be interpreted. "
The next stage of the project will involve using a sample of schools to try out the different systems. Work is also being undertaken to develop a school improvement index so that schools will be able to monitor their own progress year on year.