Joint memorandum submitted by Professor Anne West and Rebecca Allen


This memorandum provides an overview of several pieces of work carried out by Anne West and colleagues at the London School of Economics and Political Science and Rebecca Allen at the Institute of Education, University of London. The common themes across the papers are:

· secondary school admissions policies and processes;

· the social and religious composition of schools;

· social and ability segregation across schools.

It is important to note that the research used data obtained prior to the 2006 Education and Inspections Act and the 2007 School Admissions Code. Research relating to secondary school admissions criteria and practices for admission in September 2008 is currently being carried out at the LSE for RISE (the Research and Information on State Education Trust) with funding from the Esmée Fairbairn Foundation.

Religious composition and admissions processes of secondary schools in London (Pennell et al., 2007)

In 2006, the LSE was commissioned by Comprehensive Future, with funding from the Joseph Rowntree Reform Trust Limited, to carry out a small-scale research project to examine the religious composition and admission processes of publicly-funded secondary schools with a religious character in London. No information was available about the composition of these schools in terms of the religion of the pupils enrolled. London was chosen as the location for this research given its religiously diverse population and the high proportion of publicly-funded religious schools in the capital.

A short questionnaire was sent to all voluntary-aided secondary schools and academies in London that were classified by the Department for Education and Skills as having a religious character (N=106). This asked for the numbers of pupils on roll at the school that were of different religions or no religion. Fifty schools/academies (47%) provided useable information. A sample of supplementary forms used by schools with a religious character were analysed. The admissions processes used by schools that appeared to be inclusive of other faiths (or no faith) were also examined. Key findings were:

· In Church of England schools, around seven out of ten pupils were reported to be Christian; just under one in ten were reported to be Muslim and a similar proportion to be of no faith (for the remainder, no information was available).

· In Roman Catholic schools, over nine out of ten pupils were reported to be Christian. Very small percentages were of other religions or no faith.

· In the three Jewish schools, all pupils were reported to be Jewish. In the two schools of other Christian denominations, around eight out of ten pupils were reported to be Christian.

Supplementary forms were available for 24 of the 44 Roman Catholic and Church of England schools. All required a reference from a priest/minister/religious leader to confirm that the information provided by parents on their religious background and practice was accurate. Eight out of ten forms sought information on church attendance; half on involvement in the church and a third asked for proof that a child's religious milestones, such as baptism or first holy communion had taken place.

Comparison of the supplementary forms with the local authority forms found that the former, in general, were more complex than the latter: they were longer and more space was provided for parents to give reasons why they wanted their child to attend the school. In some cases personal information was sought such as parents' occupation, details of the schools attended by all other children in the family or other schools they were applying to. However, some forms were brief and simple to complete: they asked for basic details about the child, their parents/carers and a church contact so that a religious reference could be sought.

Further analyses were undertaken of the admissions processes used by schools that appeared to be inclusive of other faiths. It was found that they tended to set aside a proportion of places for those of other faiths/no faith, by dividing the available places between 'foundation' and 'open places.' However, it was noted that the school with the highest proportion of non-Christian faiths was a Church of England school that did not set aside places in this way.

Schools that were inclusive of other religions were not necessarily inclusive in other respects. In particular, an analysis of admissions criteria and supplementary information forms used suggested that, in some cases, they offered schools opportunities for social selection.

Changes in admissions criteria and practices between 2001 and 2005 (West et al., 2008)

This research focused on admissions criteria and practices between 2001 and 2005 and examined how they had changed in London secondary schools following the introduction of the 2003 Code of Practice on School Admissions.

In community and voluntary-controlled schools, where admissions are controlled by the local authority, the proportion of schools giving priority to children in care increased from 4% to 95%. And the percentage of schools giving priority to pupils whose parent was an employee of the school decreased from 13% to 5%. For voluntary-aided schools there were a number of differences between the two years. There was an increase in the percentage of schools giving priority to children in care (from zero in 2001 to 74% in 2005), to medical/social needs (42 to 54%) and to special educational needs (18 to 26%), along with a decrease in the proportion of schools giving priority to children of former pupils (14 to 4%). However, compassionate factors increased from 8 to 12%. There was a very marked reduction in the use of interviews, which dropped from 52% in 2001 to 6% in 2005 (including pre-admission meetings in 2005); this is significant as the 2003 Code of Practice, to which admission authorities had to 'have regard', stated that religious schools should not carry out interviews, unlike the 1999 Code which had allowed religious schools to carry out interviews to assess religious commitment. The percentage of schools giving priority to pupils with a religion other than that of the school concerned increased from 32 to 48%.

Composition of faith secondary schools (West and Hind, 2007; Allen, 2008)

Allen (2008) provides a summary of the socio-demographic characteristics of pupils educated in voluntary-aided (VA) religious schools compared to foundation and community 'comprehensive' (i.e. non-grammar) secondary schools in England. The key observations are that:

· VA religious schools had 13% free school meals (FSM) eligibility, which is slightly lower than community comprehensives (15%) but higher than foundation comprehensive schools (10%).

· VA religious schools had a higher proportion of pupils scoring in the top quarter nationally in KS2 tests at the end of primary (28%) than community (21%) or foundation (24%) schools.

· VA religious schools had a lower proportion of White British pupils (78%), mostly due the increased numbers of Black African and Caribbean ethnicity pupils.

West and Hind (2007) carried out an examination of the composition of London state-funded 'comprehensive' secondary schools that were voluntary-aided (mostly religious) and compared them with foundation schools (which like voluntary-aided schools are responsible for their own admissions, but rarely have a religious character) and community/voluntary-controlled schools. They also found that schools controlling their own admissions admitted pupils with higher levels of prior attainment, on average, than those that did not control their own admissions and had, on average, lower proportions of children from low income families and with special educational needs.

The paper also looked at the relationship between admissions criteria that select based on religion or religious commitment and ethnic composition of the school. Because religious affiliation is likely to vary according to ethnicity, certain pupils are more likely than others to be admitted in the event of the school being oversubscribed as the religious criteria will be used to prioritise who should be admitted - in particular, Black pupils in England are more likely to be Christian than are pupils of Pakistani or Bangladeshi origin, who are more likely to be Muslim. This London analysis confirms that there were indeed, on average, more Black pupils in voluntary-aided 'comprehensive' schools than in community/voluntary-controlled schools. There were more Bangladeshi/Pakistani pupils in community/voluntary-controlled schools than in other types of school.

The school versus neighbourhood compositions of faith schools (Allen and West, 2007; Allen, 2008)

These papers compare a school's composition with the characteristics of pupils living in the immediate neighbourhood. Allen and West (2007) use a very narrow definition of neighbourhood (e.g. the 150 pupils who live closest to the school if the school cohort size is 150); Allen (2008) combines this with the definition of neighbourhood that covers a wider geographical area. Allen and West (2007) analyse religious comprehensive schools in London and find:

· Almost all religious schools in London have a FSM proportion that is lower than the FSM proportion in their immediate neighbourhood.

· Religious schools have only about 85% of the FSM pupils they would have if they admitted the pupils who lived closest to their schools.

· By contrast, non-religious comprehensive schools in London have only 75% of the pupils who scored in the top quartile nationally in KS2 test that they would have if they admitted the pupils who lived closest to their schools.

The differences between religious and non-religious schools presented in Allen (2008) for England as a whole are less pronounced:

· Religious comprehensives have about 10% fewer FSM pupils than they would if they admitted the pupils from their immediate neighbourhood (and community comprehensives have about 30% more).

· Religious comprehensives have about 25% more pupils scoring in the top quartile nationally in KS2 tests than if they admitted the pupils from their immediate neighbourhood.

Since neither of these papers had access to information on who applied to religious schools in the first place, it is not possible to use this information to ascertain whether or not the admissions policies/procedures of religious schools favour higher ability or more affluent children (whether overtly or inadvertently in the processes of selection by religious adherence). Allen and West (2007) try to overcome this problem by looking at sorting within the group of pupils who attend religious schools in London, and who are therefore likely to have sufficiently proved their religious adherence (the analysis is carried out separately for RC and CofE schools). The paper shows that there exist 'élite' RC and CofE schools which appear to systematically favour more able or affluent religious pupils over FSM-eligible or lower ability pupils (of the correct denomination) who live close to the school.

Contribution of faith schools to socially segregated schooling (Allen, 2007; Allen and Vignoles, 2007)

Many academic research teams have used administrative data (the Annual Schools Census from 1989 onwards and the National Pupil Database from 2001 onwards) to look at the association between the number of faith secondary schools in an area and the level of social segregation between schools. Social segregation is usually defined as the extent to which pupils who are eligible for free-school meals (FSM) are unevenly distributed across schools in an area.

All papers in this field, regardless of the year of data used, find an association between the proportion of pupils in voluntary-aided (VA) schools and the level of FSM segregation in a local authority (Allen, 2007; Allen and Vignoles, 2007; Goldstein and Noden, 2003; Gorard et al., 2002; 2003). However, it should be emphasised that this is an empirical observation with no suggestion of causality. Goldstein and Noden (2003) go further and demonstrate that between the years 1994 and 1999 there was a greater increase in segregation in local authorities where a larger proportion of schools controlled their own-admissions. Allen and Vignoles (2007) report similar findings for the years 1999 through 2004. They show that areas with a higher proportion of pupils at VA schools in 1999 have seen greater growth in segregation. Where these VA schools have grown in size, increasing their share of pupils in the local authority, this is again associated with increasing segregation. Once again, it would be unwise to attribute causation of this phenomenon to the behaviour of VA schools.

Allen (2007) investigates the characteristics of areas where the level of school segregation is significantly greater than the underlying level of residential segregation (calculated by re-allocating pupils to their nearest school). The paper finds that having a high proportion of pupils in the local authority educated in VA schools is associated with a larger gap between school and residential segregation. Again, it is not possible to show why this should be the case using these administrative data. The paper also shows that, although VA schools do appear to be associated with raised levels of school segregation, school segregation caused by the housing market remains the most important contributor to stratified schooling in most areas.

Overall conclusions

There are clear differences in terms of the composition of secondary schools of different types. Schools with a religious denomination can be shown to have a more able and affluent intake than community comprehensives, especially once the characteristics of the local neighbourhood are taken into account. This means that areas with many religious schools have higher levels of school segregation. However, we do not know the extent to which this results from patterns of applications made by parents or offers made by schools.

There is a complex interaction between admissions criteria and practices, preferences made by parents and offers made by schools. Taking the potential sources of stratification separately, if parents from particular social backgrounds are not applying to particular types of schools is this because they do not want their child to go to a particular school (maybe because the school is not perceived to be for 'people like us') or are they for some reason discouraged from applying? There are many ways in which parents may be discouraged. For example, do they believe that their child will stand little chance of being admitted, because they do not fulfil the criteria? Alternatively, have they found the admissions process too complicated? Are they concerned about travel costs to particular schools?

If, on the other hand, parents from lower social backgrounds applied to a school but were not offered a place, is this because they were not religious or were they unable to achieve a high enough score on the school's measure of 'religious adherence'? For example, were they aware of the appropriate feeder primary school or local church to attend, and did they attend the correct number of Sunday services over the length of time specified by the school?

The current measurement of religious adherence on a 'continuum' can be seen to justify the collection of additional information from parents, giving religious schools the means to socially select pupils, should they wish to do so. However, there is no proof that such selection is actually taking place in schools - the apparent social selection may be an entirely inadvertent side-effect of selecting by religious adherence.

One way to simplify the admissions process for these families would be to establish a nationally agreed binary criterion of 'religious adherence' that families are deemed to have either met, or not met. Once this is established, religious schools could then rely solely on the presence of a signature on a form from a religious leader to decide who has priority in the admissions process. This would avoid the need for the schools themselves to collect additional background information. Thus, a policy such as this could simultaneously remove the means by which covert cream-skimming is possible, while simplifying the admissions process for parents.

Monitoring is needed to determine which pupils apply to which schools and which are admitted. State-maintained schools in London are publicly-funded, yet access to a significant number of schools is restricted for various reasons: on account of selection by religion and selection via other admissions criteria or practices, all of which privilege some pupils over others.

More generally, if community cohesion is to be fostered, schools with a religious character should be inclusive of all religions (or no faith). At present this is not the case. Major tensions arise in balancing policies that aim to increase the number of faith schools and promote religious inclusion. These are not easily resolved in a pluralist society, but given that public money is used to fund schools with a religious character there is a strong case to be made for such schools to be open to the wider community in the interests of enhancing social cohesion.


Allen, R. (2008) 'Do own-admission schools cream-skim? School intake versus neighbourhood characteristics'. Working paper presented at Sheffield Hallam University on 30th January.

Allen, R. (2007) 'Allocating pupils to their nearest secondary school: the consequences for social and ability stratification'. Urban Studies 44 (4) 751-770.

Allen, R. and Vignoles, A. (2007) 'What should an index of school segregation measure?' Oxford Review of Education 33 (5) 643-668.

Allen, R. and West, A. (2007) 'Religious schools in London: school admissions, religious composition and selectivity?' Paper presented at the British Educational Research Association Conference, London. September.

Goldstein, H. & Noden, P. (2003) 'Modelling Social Segregation'. Oxford Review of Education, 29, 225-237.

Gorard, S., Taylor, C. & Fitz, J. (2003) Schools, Markets and Choice Policies London: RoutledgeFalmer.

Gorard, S., Taylor, C. and Fitz, J. (2002) Markets in public policy: the case of the United Kingdom Education Reform Act 1988. International Studies in Sociology of Education, 12, 23-41.

Pennell, H., West, A. and Hind, A. (2007) Religious Composition and Admission Processes of Faith Secondary Schools in London. London: Comprehensive Future.

West, A. (2008) 'Quasi-regulation and principal-agent relationships: Secondary school admissions in London, England'. Submitted to Educational Management Administration and Leadership.

West, A. and Hind, A. (2007) School Choice in England: Characteristics of Students in Different Types of Schools. Peabody Journal of Education, 82 (2-3), 498-529.

March 2008