Burgess S, McConnell B, Propper, C and Wilson, D, (2004)

'Sorting and Choice in English Secondary Schools'

CMPO Working Paper 04/111, CMPO, University of Bristol

• Examines the extent of pupil sorting by ability, ethnicity and income/disadvantage in the English secondary school system.
• Uses a comprehensive dataset containing test scores, postcodes, ethnicity and free-school-meal eligibility for every pupil in England.
• Most Local Education Authorities (LEAs) in England operate a ‘neighbourhood schooling’ system in which pupils’ default school is their nearest one, and distance from the school is an important admission criterion. A few LEAs, however, have retained a ‘selective’ system in which children are explicitly allocated to schools on the basis of ability.
• The authors investigate the degree to which sorting is affected by the degree of choice available in an area, and the assignment rule (by distance or by ability) used.
  

Key results:

• On average, secondary schools in England have more than 6 schools within ten minutes drive of themselves. In London the mean is 17 while in rural areas it is just over one.
• 45% of children go to the school nearest to their home, suggesting that just over half are “exercising choice” in the sense of not going to their nearest school.
• The levels of ability and poverty segregation are generally not high. Three quarters of LEAs have ability and poverty segregation measures of 0.32 or less (with 0 indicating no segregation and 1 indicating ‘total’ segregation).
• For ethnic segregation the picture is different, with high average values and very high values in some LEAs.
• While ‘selective’ admission is correlated with sorting at school level by income and ethnicity, it is negatively correlated with neighbourhood segregation by ability. This may be because residency in such areas is divorced from school choice.
• In areas where there are a bigger number of schools to choose between, school segregation is high relative to neighbourhood segregation.