Disability, special educational needs, class, capitals, and segregation in schools: A population geography perspective

This paper investigates the spatially variable schooling of young people with special educational needs and disability (SEND) and interconnections with class and capitals, using analysis of the School Census and interviews with 64 educational professionals and parents in three areas in Southeast England. Three key original findings emerge. First, high proportions of young people with SEND come from poor backgrounds; how-ever, most young people with SEND labels are not poor. Second, social class, capitals, and SEND intersect in ways that relatively advantage young people from more affluent and educated families, who gain access to specific labels and what is locally considered the “ best ” education. Third, we conceptualise school spaces as differently “ bounded ” or “ connected, ” providing different opportunities to develop meaningful relationships and qualifications, or social and cultural capital, rather than focus on the type of school ( “ special, ” separate schools for students with SEND; or “ mainstream ” local schools). What are locally considered to be “ the best ” school spaces are connected and porous, providing opportunities to develop social and cultural capital. Other school spaces are containers of both SEND and poverty, with limited opportunities to acquire social and cultural capitals. Overall, we suggest that the intersecting experience of SEND, class, and capitals can (re)produce socio ‐ economic inequalities through school spaces.

but also, as this paper demonstrates, by the ways in which bodies (minds and emotional states) are socio-spatially positioned, as well as corporeally lived, as "different" with material affects in specific spaces. These processes are socio-spatially constituted and intersect with other "axes of power," including class and capitals, in school spaces.
As a key institution of social reproduction, population patterns and processes of schools have resonance for population geographers; nonetheless, increasing critical interest in geographies of and in schools (Collins & Coleman, 2008;Holloway, Hubbard, Jöns, & Pimlott-Wilson, 2010;Nguyen, Cohen, & Huff, 2017) has not fully filtered into population geographies, although studies of mobility and migration for education have (Prazeres, 2013;Smith, Rérat, & Sage, 2014;Waters, 2017).
Drawing upon a descriptive analysis of Schools Census Data from the National Pupil Database and 64 qualitative interviews with professionals and parents, we explore patterns and processes underpinning segregations of young people in schools in three different local administrative areas, local authorities (LAs) in England. We examine different school spaces that young people with SEND are educated within, which are characterised by specific connections and configurations of "special" and "general" education institutions-segregated special (separate schools for young people with SEND), mainstream (where students with and without SEND are educated alongside each other), and special units or facilities within mainstream schools-and different opportunities for acquiring cultural capital afforded to young people in these school spaces. We examine how young people attending these spaces are differentially positioned according to "poverty," class, and capitals (Bourdieu, 1986). We argue that certain school spaces act as spatial containers of SEND and socioeconomic disadvantage with limited opportunities for acquiring cultural capital. Other schools are networked, connected porous spaces, which provide opportunities for acquiring cultural capital, with fewer children from poor backgrounds. Rather than a dualism between mainstream or special schools, these differences reflect local particularities of "powers and resources" (Philo & Parr, 2000) of special and mainstream education institutions, which are situated within specific spatial contexts.
The paper proceeds through four further key sections. Next, we contextualise our discussion in emerging literature on diagnoses of SEND and educational inequalities. After outlining the methods and presenting background information, we move on to explore intersections between class, capitals, poverty, and SEND using descriptive analysis of the National Pupil Database and interviews with professionals and parents/carers. We emphasise complex interconnections between class, poverty, capitals, and SEND. These sections emphasise both connections between poverty, which has geographical and intergenerational aspects and how families with higher levels of social, cultural, and economic capitals deploy these to relatively advantage their children. The final section presents a discussion and conclusion.

| EDUCATION, INEQUALITIES, AND (RE) PRODUCING PRIVILEGE: THE ROLE OF SEND
Critical population geographers have an enduring interest in social (im) mobility (Dorling, 2015). Many political arguments focus on the potential of schools to enhance social mobility (e.g. "Schools that work for everyone," Department for Education, DFE, 2016). Rather than facilitating social mobility, education is often a key mechanism for the "ongoing creation" (Youdell, 2010: 14) of class-based, racial/ethnic inequalities and privileges. Although much policy focuses upon improving home environments of young people from disadvantaged backgrounds, scholars, often drawing upon Bourdieu (e.g., Bourdieu & Passeron, 1990) have highlighted that norms and expectations that permeate school spaces are implicitly classed, advantaging middle-class and certain ethnic groups (Ball, 2017;Reay, Crozier, & James, 2011). Scholars argue that insidious institutional frameworks (e.g., curricula) and everyday practices reflect middle-class ways of knowing and conduct "symbolic violence," denigrating knowledges of other groups (Hollingworth, 2017).
These processes are expressed spatially. Quantitative studies illuminate that state-school education is segregated by intersecting axes of class (Burgess & Briggs, 2010), race/ethnicity (Khattab, 2009), and "ability" (Burgess, Crawford, & Macmillan, 2017). Coldron, Cripps, and Shipton (2010) claim that mixed student intake enhances school effectiveness. Further, Coldron et al. go on to emphasise that segregated/polarised schools reduce interactions between "children from different social backgrounds," leading to "the injustice of malrecognition and denigration" (p. 20). There is an implicit suggestion that "encounters" between different social groups can reduce social divisions (Valentine & Waite, 2012); although as we have emphasised, this is not automatic and will depend upon the contexts of encounters and children's agencies (Holt, Bowlby, & Lea, 2017). Complex interconnections exist between educational and residential segregation (Burgess & Briggs, 2010), and middle-class parents deploy their capitals to ensure their children access "the best" educational spaces (Butler & Hamnett, 2011). Nonetheless, school segregation is often more entrenched than residential patterns (Johnston, Wilson, & Burgess, 2004).
Interest has recently turned towards diagnoses of SEND and segregation. Gorard (2016) found SEND-based segregation decreased between 1989 and 2014, reflecting policies during this period, which advocated "inclusion." In practice, this was characterised by a partial shift in the location of students with labels of SEND from special schools into mainstream schools. Much debate surrounds the appropriate school spaces for young people with SEND. Inclusion policies have been challenged in England and Wales (DFE, 2014) and elsewhere globally (Armstrong, Armstrong, & Barton, 2016), with a move towards educating higher proportions of students with SEND in special schools. In the United Kingdom, currently, there is a mixed economy of education. Many young people attend (and even at the height of inclusion attended) special schools.
Other young people with SEND are educated within mainstream school spaces, which often fall short of "inclusive ideals" (Azorín & Ainscow, 2018). In the context of neoliberal state education in the United Kingdom, significant numbers of young people are being "excluded" in hidden ways from/in school spaces (Education Select Committee, 2018;Titheradge, 2018). Parents of young people with SEND agonise over the "best" place for their child (Runswick-Cole, 2008). Scholars have noted that young people with SEND benefit from educational and leisure interactions with other young people with and without SEND labels and an educational offer of meaningful qualifications and subjects (Shah, 2013).
There is a quantitative link between SEND, socio-economic disadvantage, and the reproduction of educational inequalities (Parsons & Platt, 2017). Keslair and McNally (2009) use the 2006 Schools Annual Census to emphasise that relatively high proportions of young people in all SEND categories in England are entitled to free school meals (FSM), an accepted, though imperfect, measure of poverty (Ilie, Sutherland, & Vignoles, 2017). This relationship varies according to SEND label; young people on the "autistic spectrum" (AS) fall only slightly above national average FMS eligibility, approximately 15%, whereas the percentage of young people labelled with what would now be social, emotional, and mental health difficulties (SEMHD), moderate learning difficulties (MLD), and severe learning difficulties who are eligible for FSM stand at 32.3%, 31.13%, and 29.71%, respectively (Keslair & McNally, 2009).
Further, critical disability scholars have highlighted that disabled people are more likely to experience poverty, marginalisation, and socioeconomic exclusion (Soldatic & Pini, 2009), and poverty may stem from demands of caring for a child with SEND in a context of inadequate and decreasing benefits and services (Bradshaw & Main, 2016).
Despite these headline figures, the majority of young people with SEND do not face poverty and hardship. Therefore, intersections between class, capitals, socio-economic background, and SEND are complex and warrant further exploration. Spatial impacts of different school settings in specific administrative areas have been underexplored and require further investigation, because poverty, capitals, and SEND intersect in specific ways in particular spaces and opportunities to develop social and cultural capital are spatially differentiated.
Exclusion is not just experienced economically but also in relation to embodied capital (see Holt, 2008). Bodily and mental states of young people with mind-body-emotional differences or SEND can be experienced as difficult or troubling; however, diagnoses are also about "ableism"-These young people fall outside socially situated norms of bodily, emotional, mental, or learning expectations of development (Hodge & Runswick Cole, 2013;McLaughlin & Coleman-Fountain, 2014). Importantly, norms in comparison with which young people are (dis)abled are not neutral. They reflect performances of those above the "precariat," particularly those in established and technical middle class (Savage et al., 2015). They are tied to broader operations of school-level education, which are key mechanisms of social reproduction of capitalist and increasingly neoliberal societies (Ball, 2017). In this context, it is arguably unsurprising that young people from poor backgrounds, certain racial and ethnic groups, and boys are more likely to be diagnosed as having SEN(D), because they fall outside these norms of learning and competence (Youdell, 2010). We have emphasised that, rather than a single homogeneous ableism (Campbell, 2009), there are intersecting and socio-spatially shifting ableisms, with potentials for difference to be interpreted in other, more enabling ways (Hall & Wilton, 2017;Holt, Lea, & Bowlby, 2012). Therefore, the experiences of young people with SEND, access to capitals, and the subjectivities they embody can vary in time and space.
We examine interconnections of SEND diagnoses and capitals in specific local authority (LA) and school spaces, drawing upon Bourdieu's (1984Bourdieu's ( , 1986 concepts of social, economic, and cultural capital, habitus, and fields (see also Hollingworth, 2017). Bourdieu's theories of capitals explain how social and cultural, along with economic, aspects of life have "value," which are interconnected to power and workings of capitalist economies. Bourdieu was preoccupied by how wealthy classes reproduced their advantage intergenerationally through both direct (e.g., handing down of wealth) and indirect means, such as cultural capital (embodied, objectified, or institutionalised), and through social capital-the value of social networks and relationships, which are viewed as a mechanism for continuation of advantage (Bourdieu, 1986). Whilst pertaining to inequalities tied to distributions of economic, cultural, and social resources, "capital" helps to explore how economic, cultural, and social domains operate in intersecting yet distinct logics. Bourdieu's original conceptualisations focus particularly on class-based differences and have been critiqued for retaining a historical-material focus on the economy and over-emphasising social reproduction rather than transformation and social mobility, which he himself embodied. Nonetheless, we have suggested elsewhere how using Bourdieu as a starting point can be used to help understand a diversity of experiences of inequality and privilege (Holt, Bowlby, & Lea, 2013), in line with sociologies of education (Ball, 2017).
Here, we explore how capitals intersect with SEND in experiences of schooling of young people.

| Methods and methodologies
Data presented in this paper come from an ESRC-funded project and focus upon young people defined as having SEND in five schools: three special schools and two mainstream high schools (with students aged 11-16) taken from research in three different LAs in Southeast England, given pseudonyms "Coastal," "Rural," and "Urban" LA. These LAs and schools were selected to express a range of spatial settings in relation to proportions of young people with SEND in mainstream as opposed to segregated special schools, affluence, and ethnic diversity.
We carried out descriptive secondary quantitative analysis of the controlled-access National Pupil Database, which draws upon the School Census data, completed by most schools termly (Dent, 2016) to identify case-study schools and examine characteristics of the schools and LAs. We also conducted descriptive analysis of Office for National Statistics data about ethnicity and indices of multiple deprivation, 2 and qualitative secondary analysis of Office for Standards of Education (Ofsteds) school inspection reports, school and LA policy documents and websites. We cross-tabulated SEND data against FSM data to examine how our case-study schools and LAs were situated within broader patterns of association between these two variables discussed above.
Research in the schools was qualitative and included participant observation and research with adults and young people. Findings discussed here emerge from semi-structured interviews with 64 adults, including key educational personnel, such as heads or deputy heads of SEN provision, teachers, teaching assistants, managers of charities and NGOS, educational psychologists, senior teachers, head teachers, special educational needs co-ordinators, and parents/carers. We interviewed a total of 40 "professionals" and 24 parents. Interviews were recorded and transcribed in full and analysed via a thematic approach. An abductive approach was taken with a combination of "a priori" themes, driven from previous literature and our own theories, and "in vivo" themes emerging directly from participants (Mason, 2018).
The key themes examined in this paper are were selected because they were discussed with a high level of frequency in adult interviews and spoke to, and sometimes challenged, emerging theories of connections between SEND, class, and capitals. The themes were intersecting exclusions-poverty and SEND, which has geographical and intergenerational aspects; problematising the link between poverty and SEND; how cultural and economic capital mediates experiences of SEND; schools as spatial containers of SEND and poverty; schools as networked hubs or spatial containers with different potentials for acquiring cultural capital.

| Background to the case studies 3
The two high schools had "special units," which some young people with SEND attended for some or all of the time. The Coastal High School had specific class for young people who needed more support who had levels of literacy and numeracy in line with primary (elementary) school expectations. The Rural High School had an "Inclusion Unit" for young people with "SEMHD" and a unit of young people on the AS. 4 The two high schools had "good" Ofsted reports, although the AS provision in the rural school was commended as "outstanding." The special school in the Coastal LA catered for young people with "complex needs" (see Table 3). The school was rated "outstanding" by Ofsted. The special school in the Rural LA was a school for young people with "complex learning difficulties" (see Table 4). The school was rated "good" by Ofsted. The student population of the Urban Special School was aged 7-11, and all the young people in this school had SEMHD, often with additional diagnoses.
The Coastal LA had a mixed socio-economic profile, with slightly above national mean indices of deprivation, below national mean ethnic diversity, and spatially concentrated pockets of high levels of relative socio-economic deprivation. High proportions of young people with SEND diagnoses attended mainstream schools compared with national levels, and it was in the top quartile for the proportion of young people with SEND diagnoses in mainstream schools. In interviews with parents and professionals, the Coastal LA was identified as well organised and supportive of young people with SEND. Schools were connected and shared resources and skills, and there were a variety of well-organised and resourced cross-LA facilities provided by NGOs in collaboration with the LA, including after-school and leisure facilities, a counselling service, and a parents' "voice" organisation, who ran the statutory Parent Partnership Service.
The Rural LA was relatively affluent, with isolated pockets of high levels of deprivation and low ethnic diversity by national and regional standards. The LA had just above the national median proportion of students in mainstream schools. Interview data suggests that some children with SEND were educated in special schools in other LAs; therefore, more young people attended special schools than would show in LA figures. There were after-school clubs and leisure facilities provided by NGOs, along with alternative curricula, with evidence of resource pressures. There was less evidence of connections between schools in relation to sharing of knowledge and resources. The Rural LA was spatially more extensive than the Coastal LA, and schools were more dispersed. In this LA, there was more discussion about parents having to "fight" for "appropriate" support than the other LAs (see Section 4.2.3).
The Urban LA was broadly reflective of national averages in terms of socio-economic and ethnic composition and therefore had slightly elevated levels of both in the Southeast England context. The high school system was ability selective, with grammar schools acting as spatial containers of relative academic competence and affluence. It was in the third quartile for the proportion of students with SEND diagnoses in mainstream schools, with a relatively high proportion of students in segregated schools. In an endeavour to provide joined-up working between health, social care, and education sectors in line with government policy, the city had been divided into geographical demarked "children's action teams." Professionals felt that the reorganisation was ill considered and boundaries arbitrary, with an unintended consequence being that professionals in same field were unable to collaborate.        These discourses reflect the statistical over-representation of young people from socio-economically excluded families among young people diagnosed with MLD and SEMHD discussed above and the amorphous and challenging diagnoses of MLD (Norwich, Ylonen, & Gwernan-Jones, 2014).

| SEND and free school meals: Patterns for the LAs and schools
There was a geographical element, with the poverty and multiple deprivation that is tied to SEND being closely associated with particu- Similar comments about intergenerationality and association with particular spaces that had close-knit and often related communities were made in the other LAs. Certain mind-body-emotional differences and/or labels of particular kinds of SEND might be an important mediator in social exclusion and poverty. Understanding a link between diagnoses of SEND, class, and poverty locates the "problem" and "cause" of SEND, within families. Professionals emphasised that SEND can be tied to poverty, and a gamut of family issues, ranging from intra-family conflict, drug use, and family breakdown to social services intervention, reflective of "troubled families" discourses (Crossley & Lambert, 2017). These are part of broader neoliberal tendency to blame poor people and specifically poor parenting for their problems (Jensen & Tyler, 2012) rather than exploring structural underpinnings.
The following quote is reflective: This quote, which was one of many along a similar theme, emphasise that parenting and family problems are viewed as a key cause of SEMHD. Nonetheless, some professionals were critical of the tendency to "blame" SEMHD on poor parenting.

| Problematizing connections between poverty and SEND
Some professionals were critical of a tendency to blame parents and parenting for certain diagnoses of SEND and highlighted the variable workings of the SEND system: Some professional discussions were akin to Bourdieu-inspired (e.g., Bourdieu & Passeron, 1990) interpretations of schools (and social services) as (re)producing bourgeois norms and doing "symbolic violence" to those who fall outside of these norms. Indeed, interviewees suggested that young people from families regarded as "troubled" (or trouble) are more likely to be excluded from school than young people from families who can behave in ways expected by professionals.

| Class, capitals, and securing (specific) diagnoses and "the best" place for the child
A complex relationship exists between membership of socio-economic groups, SEND label, and being given a statement of SEND, 5 which provides legal protection and resources (Keslair & McNally, 2009).
Although those from lower socio-economic populations are more likely to be labelled as experiencing SEND, those identified as having a SEND from higher socio-economic groups are more likely to gain a statement of SEND. This means that children from lower socioeconomic populations are less likely to gain resources and powers of a statement (Riddell & Weedon, 2016). One reason for this is the shifting "norms" in different schools, wherein some schools (such as the primary school discussed in Section 4.2.1) have a "norm" of a high level of learning and emotional differences. This is problematic, as a key way that children without impairments become identified as having SEN is falling below norms of learning or behaviour compared with peers in class (Department for Education, 2015). This emphasises that spatially shifting operations of SEND diagnoses can compound existing disadvantages (Galloway, Armstrong, & Tomlinson, 2013 In some contexts, families discussed how they had to fight for the provision their child required in an adversarial SEND system (Lewis et al., 2010 In addition to deploying their resources to gain ["the correct"] diagnoses or statement, families with higher capitals used these to gain "the best" provision for their child. Journeys of children into the current school varied and had often involved exclusion from mainstream schools. Many parents of children in special schools and special units discussed how their child had been marginalised and excluded in mainstream settings, and this is an important backdrop of the "choice" of parents to send their child to a segregated space (Runswick-Cole, 2008) in a separate school or a unit in a mainstream school. Some parents (usually mothers) gave up paid employment to deploy their cultural capital in order to gain, what they perceived to be, the best provision for their child. This mother fought for provision out-ofcounty in a residential school for her daughter: Gaining a statement (now EHCP) is critical in securing preferred school options-some facilities will only accept children with statements (EHCPs), and mainstream schools have a requirement to prioritise children whose statement (EHCP) names the school. There were significant differences between schools providing specialist provision in terms of the level of cultural capital available to young people.

| "Networked hubs" or "spatial containers"-Different contexts and opportunities for acquiring cultural capital
The units in the mainstream schools were mixed in relation to level of poverty as measured by FSM. The units, particularly the AS unit in the rural mainstream school, provided open and connected spaces with young people spending some time in mainstream spaces, rather than acting as "separate worlds" (Webster & Blatchford, 2015). This facilitated young people in undertaking national-level qualifications, whilst also operationalising "normalising" power, with therapies and interventions to change the young people to facilitate inclusion into mainstream spaces, rather than changing the mainstream spaces (Holt et al., 2012).
The special schools, particularly in the Rural and Urban LAs, were spatial containers with high concentrations both of relatively poverty and SEND (see Section 4.1). The two special high schools were different in the opportunities for cultural capital acquisition and level of connectedness to the broader educational institution in the LA. In the Rural Special School, young people were offered mostly entry level qualifications and life skills, radically different to national level GCSE and A Level qualifications. Senior staff and Ofsted argued that the curriculum reflected limited student potentials; nonetheless, this also limits opportunities for cultural capital (Shah, 2013 (Bourdieu, 1986 By contrast, the Rural Special School acted as a supportive and nurturing spatial "container" for young people, with most academic and leisure opportunities situated in the school or in specialist spaces. Young people did not travel independently to other schools for curricula activities.
Links were made with mainstream schools, although these were largely for social reasons and in order to educate wider society about disability.
The senior teacher (Rural Special School) emphasised: We also have done projects with some of our youngsters from here going into primary schools to do a specific project … so that the primary pupils actually understand that our children are not that much different from them … I think we need to do a lot of work on making people out in the community aware of disability.
A key difference in porosity of the schools was tied to young people's independent transport; in the Rural Special School, there was limited opportunity for young people to travel independently to other schools, given both limited local public transport links and students' perceived vulnerability. As a parent volunteer for a local charity in the Rural LA (white, female) emphasised: "This LA covers a vast rural area with poor transport links." By contrast, many young people in the Coastal Special School travelled independently by bus to go to other schools or colleges.

| DISCUSSION AND CONCLUSION
In this paper, we have explored complex intersections of capitals and SEND. The paper complicates dominant emerging academic accounts that emphasise the interconnections between poverty, hardship, and SEND. We found that young people in our case-study areas with SEND diagnoses are more likely to come from "poor" backgrounds, being eligible for FSM, in line with broader arguments in social science (Parsons & Platt, 2017). Conversely, our findings show a complicated intersection between economic and cultural capital, SEND diagnoses and education, in which the spatial contexts of specific LAs and schools play an important part.
Professional and parent interviews highlighted some of the mechanisms that can help to explain why those from higher socio-economic groups who are diagnosed with SEND are more likely to gain a statement of SEND (EHCP), providing legal protection and often resources (Galloway et al., 2013). The role of cultural and economic capital was highlighted, and problems faced by parents without knowledges or dispositions to effectively negotiate the SEND system was emphasised. The reasons behind the link between particular diagnoses of SEND and poverty were also interrogated and, in many cases, challenged. Key professionals emphasised that poverty and problematic families could create specific difficulties, particularly SEMH and MLD. This is often spatial, tied to specific areas, and intergenerational; children with certain SEND diagnoses often have parents with SEND.
On the other hand, professionals were critical, emphasising how parents with higher levels of capitals worked to gain, not just any diagnosis, but a specific diagnosis of being on the AS or specific learning differences, rather than SEMHD or MLD. Families with higher levels of cultural and economic capital deployed these resources in gaining what was perceived to be the best school placement for their child, which was often facilitated by acquiring a statement (now EHCP).
Critically, the cultural capital to which young people have access, in terms of both formal qualifications and embodied capital, is influenced by specific socio-spatial contexts of schools. Parents with higher levels of capitals strategised to gain access to what they considered "the best" provision for their child. The rural and urban special school was a segregated space of relative socio-economic disadvantage, along with being a space of SEND segregation. Special schools have been criticised for providing limited access to the cultural capital of recognised qualifications (Shah, 2013 The paper contributes to geographies of education, children's geographies, and population geographies, along with broader social and geographical literatures by highlighting how important schoollevel education is to social reproduction of enduring inequalities tied to both socio-economic and educational differences and how they intersect with mind-body-emotional differences. The discussion reminds scholars that geographies of education are highly differentiated and that children and young people's geographies of education are "structured" by broader social and economic processes and by adult everyday practices. Broader socio-economic inequalities are (re) produced through everyday social practices, and the enduring patterns of inequality that can be observed through quantitative analysis are continually being recreated and generated anew through everyday practices in specific spaces. Ultimately, this continued (re)creation also provides opportunities to challenge and change these enduring inequalities.
The paper emphasises bodily (mental and emotional) differences in population and educational inequalities, which have tended to be underexplored. It is crucial that these inequalities are addressed, because young people with labels of SEND who come from poorer backgrounds, from families with lower capitals, are frequently not reaching their potential in education. These inequalities are a problem at the individual level in relation to everyday experiences of education, inclusion, segregation, and marginalisation in the present and for future socio-economic trajectories of these young people. These inequalities also are a problem for society, because these young people are not being given the skills and resources to participate fully in society and the economy.

ACKNOWLEDGEMENTS
, and we use this for clarity, although in principle, we prefer the term "differences" for young people who are labelled as experiencing difficulties in school spaces due to falling outside of (below) age-related normative expectations of learning, emotional development, behaviour, or because they have an impairment. Butler and Parr (2005) coined the term mindbody differences to emphasise the interconnections and mutual coconstruction between social and spatial experiences of exclusion and the corporeality of the experience of difference. We added "emotions" to include the experience of young people with emotional differences.
2 The official measure of deprivation of small areas in the United Kingdom. 3 The data from this section are from ONS indices of Deprivation (2007); School and LA Ofsted Reports (2007Reports ( , 2008, and the National Pupil Database (2008).