Ethnic segregation in schools and neighbourhoods in England has neither a negative nor a positive effect on how pupils from different ethnic groups perform at school, according to new research from the University's Centre for Market and Public Organisation.

Using data on 1.6 million pupils, drawn from the Pupil Level Annual School Census (PLASC), a census of all children in state schools in England, the researchers looked at the difference in GCSE results between White pupils and pupils from various minority ethnic groups (Black Caribbean, Black African, Black Other, Indian, Pakistani, Bangladeshi and Chinese). It also looked at the performance of students in each of these minority ethnic groups across cities with varying levels of segregation.

On average, White pupils score 41 points at GCSE. This is higher than some groups (33 for Black Caribbean, 38 for Black African, 38 for Pakistani and 40 for Bangladeshi pupils) but lower than others (48 for Indian and 55 for Chinese pupils). This mixed pattern shows that of the two most numerous minority groups, one scores higher than Whites (Indian) and one less (Pakistani).

There are many potential explanations, some unrelated to school factors, such as poverty, social class and family background, and others due to school factors, such as the quality of teaching and the school itself.

The differences in GCSE results between White and minority ethnic students is largely unaffected by segregation and, controlling for differences in prosperity and geography, the research shows that segregation has no overall impact on differences in test scores. This is in stark contrast to findings in the US, where the equivalent study shows that segregation does impact on test results.

The PLASC data were gathered from GCSE students in 2002, 2003 and 2004 but it should be noted that, although the data are on 1.6 million pupils, 88 per cent of pupils are White, so the overall percentage of minority ethnic children surveyed is relatively small. Only the following ethnic groups have more than one per cent of pupils: Black Caribbean (1.4 per cent), Black African (1.3 per cent), Indian (2.6 per cent) and Pakistani (2.5 per cent).

It should also be noted that minority ethnic communities are concentrated in a relatively small number of cities, for example London, Birmingham and Manchester (Black Caribbean) and Blackburn, Bradford, Leicester, London, Manchester and Oldham (Indian, Pakistani and Bangladeshi).

Data on segregation in neighbourhoods were gathered using the full postcode of each pupil’s home address and the definition of ‘neighbourhood’ was reached by using Middle-Layer Super Output Areas (MSOA). There are just under 7,000 MSOAs in England, with an average population of 7,200 and a minimum of 5,000. To characterise where people live, the researchers used the Indices of Deprivation 2004 (ID 2004) produced by the Office of the Deputy Prime Minister (ODPM). The Indices cover income, employment, health and disability, education and crime.

The results show that White, Indian and Chinese pupils live, on average, in the least deprived areas and Bangladeshi and Pakistani pupils in the most deprived.

According to Professor Simon Burgess, the lead author:

‘We can speculate on the reasons for the difference between the results in England and in the US. One possible reason is the centralised education funding system in the UK, which actively attempts to equalise educational spending per head. Another is that the nature of the academic performance of the relevant minority groups is very different. Thirdly, levels of school and neighbourhood segregation are lower in England than in the US.

‘The fact that our results suggest that levels of segregation do not impact – either positively or negatively – on test score outcomes adds weight to the call for increasing integration of different ethnic groups at school in order to increase the potential for improved social cohesion.’