| Summary: | <p>As England becomes increasingly ethnically diverse there have been concerns over perceived ethnic divisions within society (Cantle, 2013; Casey, 2016; May, 2016). Schools have been a focal point for these discussions, and the government is now committed to ensuring state schools become places where students create “relationships with others from different ethnic... groups” (HM Government, 2018a, p. 26). </p>
<p>Unfortunately, academic literature on ethnic school segregation suffers from three issues. First, it is fairly small and often dated. Only a limited number of studies have measured ethnic school segregation across England, and many rely on data from the turn of the century. Second, no national study has measured segregation within local educational markets across England and many frame ethnic identity as “White” or “Non-White” which does not reflect the diversity of the school population. Finally, there is limited evidence on the predictors of ethnic school segregation, which hinders our ability to understand why different patterns exist. </p>
<p>Using school census data, this study provides the most in-depth, up-to date analysis of ethnic school segregation in England. It maps patterns in detail, investigates how segregation varies between areas with different characteristics and explores whether certain types of state schools contribute more to school segregation than others. Additionally, using a bespoke extract from the National Pupil Database on the application choices of year 6 students, the study charts how segregation shifted during the 2016/17 school admissions cycle and explores how segregation may have differed, had students been allocated to state schools using alternative approaches. The thesis also helps pioneers three methodological techniques. It is the first national study to use a multi-group measure of ethnic school segregation (H Index) to calculate the level of segregation between five major ethnic groups. Additionally, by grouping sets of roughly six schools into “pseudo-neighbourhoods” using a clustering algorithm the study provides a new measure of ethnic school segregation calculated within local educational markets. Finally, it used simulation analysis to explore the links between school admissions and ethnic school segregation. </p>
<p>The study found that between 2006 and 2019 the median level of ethnic school segregation for pseudo-neighbourhoods in England fell by 25%, from an H value of 6.8 to 5.1, which suggests that ethnic groups became more evenly spread across local state schools during this period. After controlling for the number of schools within each pseudo-neighbourhood, segregation was found to be higher, on average, in pseudo-neighbourhoods with lower average house prices and higher rates of entitlement to Free School Meals. Ethnic school segregation also tended to be higher in primary school pseudo-neighbourhoods with a greater proportion of religious schools and secondary school pseudo-neighbourhoods with a grammar school. Subsequent models found that faith schools tended to contribute more to ethnic school segregation, on average, than non-religious schools in the same pseudo-neighbourhood, with schools linked to minority religions contributing particularly disproportionate amounts. Finally, school segregation was found to rise between each stage of the admissions process and all seven simulations exhibited lower median H values than observed in practice. Simulations that allocated students to schools using academic banding within the current school choice system, or that used a strict home-school distance rule without any parental choice, produced some of the lowest median H values. The implications of these results for policy making, public debates and school practices are discussed.</p>
|
|---|