‘Faith’ schools are deeply embedded within mainstream state education in England, with 28% of primary school children (aged 4-11) attending them: 18% in Church of England (CoE) schools, and 9% in Catholic schools. To an extent, these ‘faith’ schools are a historical artefact and remnant of the formation of the English education system. However, recent governments have also actively championed and promoted their continuation and expansion, including under agendas of ‘choice,’ ‘competition’ and the ‘quasi-market’ (DfES, 2001; DfE, 2018).

It is well established that, as a result of this ‘market’ and negotiation of ‘choice,’ ‘faith’ schools have tended disproportionately to educate pupils from more ‘advantaged,’ higher-income families (proxied by eligibility for free school meals [FSM]) (Long & Danechi, 2019; West & Curry, 2008). As West and Curry (2008) note, ‘faith’ ‘[s]chools are funded by the state for the benefit of the whole community’ and should be held to account accordingly, and on ‘social justice grounds.’ If they disproportionately serve children who are more ‘advantaged,’ what part do they really play in English state education?

This research expands empirical evidence on the children served by ‘faith’ primary schools by exploring whether they also under-serve children who may be conceived of as ‘disadvantaged’ according to another key dimension: being recorded with ‘special educational needs’ or disabilities (SEND).

Previous analyses of data for older children have suggested that ‘faith’ schools may contain proportionally fewer children with SEND (Andrews & Johnes, 2016). However, issues around inconsistent denotation of children’s needs within a system of perverse (dis)-incentives and unevenly distributed funding (Campbell, 2021; Hutchinson, 2021) mean that it is not clear whether this unambiguously reflects a skewing of ‘faith’ school attendees towards children without ‘special educational needs’ and / or disabilities, or, alternatively, reflects differential tendencies in formal attribution and reporting. The methodology employed here directly addresses these issues and expands investigations, to present a clearer sense of the under-concentration of children with SEND in ‘faith’ primary schools in England.

De-identified individual-level administrative National Pupil Database (NPD) census records are used for this research. Information is collected for all children in state-funded education in England, from the beginning of pre-school through to the late teen years. It is available to researchers through a secure setting after approval and vetting. Information for each pupil can be linked longitudinally, and includes variables indicating eligibility for FSM (which very roughly proxies family income); whether/when they are denoted to have SEND; other demographic characteristics; the governance of the school they attend, and whether it is a ‘faith’ school; and school and local area information and identifiers. Data for selected cohorts of four/five-year-olds entering the first – Reception – year of primary school across the decade 2010-2020 are currently being analysed. Initially, data for those entering in the 2017-18 academic year has been scrutinised, and key preliminary results are summarised below. The sample here is thus all children in the Reception year-group in mainstream state primary schools in England in the 2017-18 academic year (N=640,066). In order to attempt to address difficulties with accurate documentation of children with SEND within the school system and the NPD data, the focus is on two groups. Firstly, children who were recorded with SEND in the pre-school years, and secondly, children who already have a higher-level, statutory Education, Health and Care Plan (EHCP) – funded by the Local Authority and conferring recognition of and provision for their needs – in their Reception year. The key assumption is that conferral of these children’s SEND would have been established outside of the distortions of measurement within the education system, prior to school entry: and that their needs would be manifest and salient at and prior to the point of application for admission to primary school. Descriptive and regression analyses at the levels of the pupil, the school, and the Local Authority (LA; the area within which admissions take place in England) are used, to build a comprehensive sense of patterns of state school provision, and of the under-admission of children with SEND to ‘faith’ schools in England.

All LAs in England have a mixed quasi-market of state primary school provision in 'faith' and non-'faith' schools. Patterns vary by area: in the LA with the most 'faith' schools, in 2018, two thirds of Reception places were in these schools; in the LA with least, fewer than 10% of places. In the 2017-18 sample, nationally, ‘faith’ schools have lower proportions of children with both pre-school SEND and higher-level (EHCP) SEND. On average, children with SEND are less likely to attend their first year of primary school in ‘faith’ schools. School-level regressions, where respective outcomes are proportion Reception children with pre-school SEND / proportion Reception children with an EHCP, suggest that it is not covarying area-level factors that explain these patterns. In fact, controlling for these factors, the disproportionate under-concentration of children with SEND in Catholic schools becomes more pronounced. Child-level regressions (logistic and linear) explore the odds / probability of an individual child attending a ‘faith’ school. Controls include FSM, birth month, gender, ethnicity, home language, area deprivation, urban/rural indicators, LA. There is some attenuation by controls, but each measure of SEND continues to be strongly associated with lessened probability of attending a ‘faith’ school. SEND is also interacted with FSM. Both recorded eligibility for FSM and recorded SEND are associated with reduced odds of attending a ‘faith’ school, independently of one another, with additive effects. Analyses so far indicate that, nationally, ‘faith’ schools of all denominations seem disproportionately to under-serve children with SEND, as well as those from poorer families. Have they, through the ‘quasi-market,’ become hubs of advantage? What ‘choices’ and ‘competitions’ are really taking place? Does the existence of these schools ‘benefit the whole community’? And what place should they hold in state education in 21st Century England?

Andrews, J. & Johnes, R. (2016). Faith Schools, pupil performance, and social selection. https://epi.org.uk/publications-and-research/faith-schools-pupil-performance-social-selection/ Campbell, T. (2021). Special Educational Needs and Disabilities within the English primary school system. https://sticerd.lse.ac.uk/CASE/_NEW/PUBLICATIONS/abstract/?index=8089 DfES (2001). Schools achieving success. https://www.gov.uk/government/publications/schools-achieving-success DfE (2018). Schools that work for everyone. https://www.gov.uk/government/consultations/schools-that-work-for-everyone Hutchinson, J. (2021). Identifying pupils with special educational needs and disabilities. https://epi.org.uk/publications-and-research/identifying-send/ Long, R. & Danechi, S. (2019). Faith Schools in England: FAQs. https://commonslibrary.parliament.uk/research-briefings/sn06972/ West, A. & Curry, P. (2008). School diversity and social justice: policy and politics. Educational Studies, 34(3). https://doi.org/10.1080/03055690701811362