The proportion of children receiving free school meals across the whole school population is 17 per cent, falling to 13.5 per cent at key stage 4; the figure for Catholic schools is 15 per cent, and for Church of England schools 11 per cent, falling to 10 per cent at key stage 4. Faith schools, then, may be different.

Indeed, and as the report notes, the handful of Muslim schools go against the general grain by admitting far more pupils who receive free school meals than either Church of England schools or schools as a whole. This does not mean that Muslim schools have more generous admission policies. It reflects the fact that the small number of Muslim schools in Britain are located in economically disadvantaged inner cities. By contrast, Church of England and Catholic schools are usually found in smaller towns and suburban locations. So they serve different constituencies.

The fact that 17 per cent of schools are faith-based gives some credibility to the claim that they are middle-class institutions. However, the crude measure of free school meals is of limited value in determining whether Church of England schools have a predominantly, and disproportionately, middle-class intake.

To do that, we would need information that is hard to find. We would need to know what proportion of Catholic, Jewish or Church of England parents are middle class, and what proportion of their children are entitled to free school meals. We would also need to establish whether there are different attitudes to claiming free meals in faith schools. Only then would we be able to confirm whether these schools are disproportionately serving the middle class.

It is interesting that Church of England schools are more middle class, in terms of the figures presented, than Catholic schools. Not for nothing is the Church of England known as the Conservative party at prayer.

The concern that church schools are catering for a narrow, privileged clientele is certainly contestable. After all, the difference between the percentages claiming free school meals is quite small.

A high performing secondary school (probably a former grammar) is more likely to take pupils from the immediate location. Thus intake is primarily decided by the ability to buy houses in the area; and since houses in such areas attract a significant premium, the overwhelming proportion of pupils will be middle class.

However, religious belief is not constrained by class. And admissions policies of faith schools are likely to favour pupils and parents who share a religious belief over geographic proximity. Consequently, they will almost certainly draw from a wider area.

So, faith schools allow a reasonable proportion of places for less affluent pupils, whether or not they are entitled to free school meals. What we find, therefore, is that faith schools actually admit notably more working-class pupils than similar, non-faith, comprehensive schools. An analysis by the Sutton Trust of the 200 state schools that performed best in terms of student achievement (and some of these will of course be faith schools) found an average free school meal entitlement of 3 per cent.

So, rather than having a socially restricted entry, faith schools are one of the best avenues of social mobility around, at least since the demise of the grammar school.

Graham Fowler

Graham Fowler is a researcher, writer and consultant