EPI research: faith schools, pupil performance, and social selection by Jon Andrews and Rebecca Johnes
The Government’s green paper, Schools that work for everyone, proposes removing the restriction on faith schools which currently limits faith based admissions to 50 per cent. This proposal is based on the premise that the majority of faith schools are high-performing, have good Ofsted ratings, and support increased social mobility. We test this premise by analysing the:
- overall attainment and progress made by pupils, including disadvantaged pupils, in faith schools;
- characteristics of pupils in faith schools, including levels of deprivation and special educational needs.
This enables us to make an impartial assessment of the impact of faith schools, and therefore judge whether higher performance is due to faith schools being more effective, or whether it is the result of admitting more affluent, high-attaining pupils than non-faith schools.
At Key Stage 2, we find that:
- 83 per cent of pupils in Church of England schools, and 85 per cent of pupils in Roman Catholic schools achieved level 4 in reading writing and mathematics, compared to 81 per cent in non-faith schools.
At Key Stage 4, we find that:
- 60.6 per cent of pupils in Church of England schools, and 63.2 per cent of pupils in Roman Catholic schools achieved five good GCSEs, including English and mathematics, compared to 57.4 per cent of pupils in non-faith secondary schools.
However, we also find faith schools have proportionately fewer pupils with challenging needs, compared to non-faith schools. We find that faith schools:
- educate a lower proportion of disadvantaged children (12.1 per cent at KS2 versus 18.0 per cent; 12.6 per cent at KS4 versus 14.1 per cent);
- educate a lower proportion of pupils with special educational needs (SEN) (16.8 per cent at KS2 versus 19.7 per cent; 14.4 per cent at KS4 versus 16.6 per cent); and
- enrol a larger proportion of high attaining pupils (28.4 per cent at KS2 versus 23.7 per cent; 27.4 per cent at KS4 versus 24.5 per cent).
In addition to analysing how representative faith schools are compared to non-faith schools, we also look at how representative they are of their local communities through a ‘social selection’ index. A score of 1.0 means that a school draws in from its catchment exactly the same proportion of poor students who are represented in the catchment area. Using this, we find:
Grammar schools are the most socially selective schools, with an average score of 0.2.
- This means that on average the odds of a pupil in a grammar school being eligible for free school meals are one fifth of those for all children in their local area.
- Of the 100 most socially selective schools in England, 65 of these are grammar schools (including some ‘faith grammars’).
Secondary faith schools have, on average, a socially-selective score of 0.7.
- This means that the odds of a pupil in a secondary faith school being eligible for free school meals are around two thirds of those for all children in their local area.
- In the top 100 socially selective secondary schools, 30 of these are faith schools and 17 of these are non-academically selective faith schools – raising concerns about their admission arrangements.
- Primary faith schools have similarly socially selective intakes.
Because the intake of pupils in faith schools are not, on average, representative of their local areas or of the national picture, to impartially assess the impact of faith schools, the report analysed the performance data of faith schools controlling for deprivation, prior attainment and ethnicity. We find that:
- The difference in attainment between faith and non-faith schools at Key Stage 2 is largely eliminated after controlling for prior attainment and pupil characteristics – and is so small as to be educationally insignificant.
- At Key Stage 4, also adjusting for pupil characteristics, pupils in faith schools achieved the equivalent of around one-seventh of a grade higher in each of 8 GCSE subjects. This is a relatively small attainment gain.
Using this analysis, the report concludes that given that the average faith school admits fewer pupils from poor backgrounds than the average non faith school, there is a risk that increasing the numbers of faith schools would come at the price of increased social segregation, with a risk of lower social mobility.
The full report can be read here.