Chapter 3 illustrated the clear patterns in educational participation and outcomes linked to the characteristics and backgrounds of learners. The indicators of possible educational disadvantage may vary over time and place, but these kinds of patterns are clear in all such studies worldwide. How do they arise and what can be done about them? This chapter starts looking at the possible reasons and solutions by considering who goes to school with whom.

My concern with school intakes started in the 1990s when I looked at issues of school choice (Gorard, 1997a, 1998a, 1999), including the choice to use fee-paying schools (Gorard, 1996, 1997b), and the operation of a ‘market’ or quasi-market in education (Gorard, 1998b). This chapter is chiefly about a theme that developed from this – the extent to which children of similar social and economic backgrounds are then clustered within the same schools, the reasons this happens, the impact on attainment and the wider damage this causes. This kind of clustering is referred to here as ‘segregation’ between schools. I have looked at levels of segregation, the causes and results of segregation and its links to attainment (Gorard and Taylor, 2002b; Gorard, 2006c).

Although all state-funded schools in England are ‘choice’ schools in the sense that any family is entitled to express a preference to attend any of them, this does not mean that all preferences are met. Schools have a planned fixed number of places. A small number of these schools are selective, taking only those children scoring above a certain level on an entrance test or in terms of aptitude. Many more schools continue to have a faith basis, and can restrict the number of children accepted who come from families apparently without that faith. Even more commonly, popular schools or their admissions authorities use over-subscription criteria such as the proximity of home to school to decide who gets contested places. Because of the segregated nature of housing in parts of England, due to differences in cost and availability, children can then turn out to be clustered by their family backgrounds into particular schools. This is measurable in terms of a range of characteristics including low attainment, poverty, ethnic origin, immigrant status, disability or learning difficulties.