Entry to grammar schools in England for disadvantaged children

Published on 8 November 2013

New work by IFS researchers, funded by the Sutton Trust, suggests that grammar schools are disproportionately unlikely to admit students who are eligible for free school meals, even when conditioning on their academic performance in primary school. They are by contrast disproportionately likely to admit children who have attended private schools before age 11.

There are 164 grammar schools in England, which select their pupils on the basis of performance in entry tests in Year 6. These schools educate 4% of the Year 7 pupils in England. They are concentrated in selective local authorities such as Kent, Buckinghamshire, Slough and Trafford, although a third of grammar schools are in non-selective local authorities (including London). Despite their small number, grammar schools attract a significant amount of attention. New research published today by IFS researchers, which was funded by the Sutton Trust, investigates the extent to which children from disadvantaged backgrounds are disproportionately unlikely to attend grammar schools.

Using administrative data on all children in state schools in England, the research investigates whether pupils who are eligible for Free School Meals (FSM) are particularly unlikely to enter a grammar school. FSM eligibility is a good measure of coming from a poor household, as it indicates that a parent is in receipt of a means tested out of work benefit.

Comparing the proportion of children eligible for FSM in grammar schools to the proportion eligible for FSM in the surrounding areas shows that relatively few disadvantaged children go to a grammar school. In selective local authorities, 3% of grammar school entrants were eligible for FSM, compared with 17.5% at other state schools.

Given the selective nature of grammar schools, if pupils eligible for FSM have poor academic performance, they are unlikely to pass a selective entrance test. Unsurprisingly our research found that higher academic performance in Key Stage 2 tests in Year 6 dramatically increases the probability of getting a place at a grammar school. But, as has been shown before, FSM eligible children are far less likely to have high academic performance. According to the latest statistics, about 14% of disadvantaged pupils (mostly those who are or who have ever been eligible for FSM) achieved Level 5 in both English and maths at age 11, compared with around a third of other pupils. As well as having lower attainment at age 11, pupils from poorer families are also more likely to experience other potential disadvantages, such as having English as an additional language, or going to a more challenging primary school.

In areas operating the grammar system, we find that two thirds of children who achieve level 5 in both English and Maths at Key Stage 2 and who are not eligible for free school meals go to a grammar school. This compares with only 40% of similarly high achieving children who are eligible for free school meals. Furthermore, even after allowing for a wider range of factors that may depress pupils’ academic achievement, as well as pupils’ level of prior achievement in Year 6, this difference remains at just over 12 percentage points . Therefore it is not purely low attainment that prevents FSM children from attending grammar school. We found similar patterns in London local authorities and in grammar schools located in local authorities that do not operate a full grammar system.

Eligibility for free school meals only allows us to look at the poorest group of families. To get a handle on whether these socio-economic differences extend further up the income distribution, we compare pupils’ chances of getting into a grammar school according to the level of deprivation in their local area. This shows that 4% of pupils in grammar schools live in the poorest fifth of neighbourhoods, around 21% come from the middle quintile and 34% live in the richest fifth of neighbourhoods. Children living in the richest areas are disproportionately likely to attend a grammar school compared with those living in areas close to the national average, who are in turn more likely to attend grammar schools compared with those living in the poorest areas. The differences do thus seem to extend further up the income distribution.

While pupils from more disadvantaged backgrounds are less likely to attend grammar schools, a further striking result from the research is that large numbers of pupils who enter grammar schools come from primary schools outside the English state education system. Most of these students coming in from outside the state system are likely to have attended private schools. Children attending private schools, who tend to come from more affluent families, make up 13% of Year 7 entrants to grammar schools. In comparison, on average 6% of 10 year olds are enrolled in a private school nationally. Unfortunately we do not have any measures of prior achievement for pupils in private schools, so it is impossible to compare the likelihood of enrolling in a grammar school of state and private school pupils with similar levels of prior achievement.

This evidence indicates that some parents may use private primary schooling to try to increase the probability of their child attending a selective state secondary school. Of course others may purchase individual tutoring for their children to help them prepare for the entrance exams. Surveys of individual grammar schools have suggested that a high proportion of the students in those particular schools had been tutored for the entrance exam, though the extent of tutoring nationally for this purpose is not known. Clearly these are advantages which children from poorer backgrounds are unlikely to benefit from.

What drives the differential in attendance at a grammar school between pupils from different socioeconomic backgrounds? Unfortunately, this research is unable to pin down the exact mechanism for the under-representation of poorer pupils in grammar schools. It may be that children from poorer backgrounds are less likely to apply to a grammar school in the first place, or that those who do sit the exams are less likely to pass the entrance exam, perhaps due to lack of preparation for that kind of test, or a combination of both.

Future research is needed to determine exactly what is driving this grammar school attendance gap between richer and poorer pupils. This is an important research question as if it is the result of low applications then the policy response should focus on encouraging greater numbers of applications. If applications are similar, then the focus should be on the nature of the test and levels of preparation.


This paper, Entry into Grammar Schools in England is authored by Jonathan Cribb (IFS), Luke Sibieta (IFS), and Anna Vignoles (Cambridge and IFS).

It was published as part of a larger report, Poor Grammar: Entry into Grammar Schools for disadvantaged pupils in England by Jonathan Cribb, Luke Sibieta, Anna Vignoles, Amy Skipp, Fay Sadro and David Jesson. The report has been funded by the Sutton Trust.

IFS receives significant funding from the ESRC through the ESRC Centre for the Microeconomic Analysis of Public Policy at the Institute for Fiscal Studies.