We have attempted to do this by looking at the 1998 and 1999 GCSE data for 3,151 English secondaries and the number of pupils who had statements for SEN and were at stages 1-3 of the SEN code of practice.
From the 1998 data, we used the GCSE average point score for each school, because this is based on grades A*-G, and calculated the percentage of pupils at each school with statements.
We divided the schools into 10 groups according to GCSE scores, enabling us to examine the percentage of pupils with statements in the highest-attaining group of schools, the second highest group and so on.
We found the average proportion of pupils with statements in these 3,151 schools was 2.7 per cent. By contrast, there was only 0.9 per cent in the highest-attaining group of schools and 1.8 per cent in the second group.
The average percentage of pupils with statements increased as the attainment level of the group of schools decreased. In the lowest-attaining 10 per cent of schools, 3.9 per cent of pupils had statements. This is more than four times the percentage for the most academically-successful schools. This inverse correlation between GCSE attainment and percentage of pupils with SEN was even more pronounced when the number at stages 1-3 of the code of practice was considered.
To identify the few schools bucking the trend, we also divided the schools into 10 groups according to the percentage ofpupils with statements. None of the schools in the top 10 per cent for attainment was in the group with the highest percentage of pupils with statements (top 10 per cent). Only 19 schools were in the top 20 per cent for both attainment and statements, while 94 schools were in the top 30 per cent of both categories.
The 94 included an above-average proportion of co-educational schools but there was a standard mix of local education authority-maintained, grant-maintained, and voluntary-aided schools. We found almost 80 per cent of these schools were in county authorities. Shropshire, Cornwall, Devon, Lancashire, Suffolk and Norfolk each had five such schools. Not one was in inner London. Nor did we find any schools in LEAs well-known for inclusion policies.
A similar picture emerged when we used 1999 data. But this time the lowest-attaining group of schools had more than five times the percentage of pupils with statements than the highest-attaining group.
Our study therefore challenges the assumption that having more pupils with significant special needs is compatible with excellent GCSE results.
But there are exceptions. We are contacting the 43 schools which were in the top 30 per cent for both attainment and SEN in 1998 and 1999. We will ask about their policies and practices but we will also be trying to find out why so few of them are in densely-populated areas.
Brahm Norwich is a professor at the School of Education, University of Exeter. Dr Ingrid Lunt is a reader at the Institute of Education, London University. Their book, "Can effective schools be inclusive schools?", is available, price pound;6.95 plus pound;1pamp;p, from the Institute bookshop, 20 Bedford Way, London WC1H OAL