Evidence check: Grammar schools Contents

1Introduction

Background

1.Selective schools admit some or all of their pupils based upon certain selection criteria. Grammar schools are an example of selective secondary education in England, where entry is dependent upon measured academic attainment in examinations at age 11 years (the 11-plus). In contrast, non-selective schools accept pupils regardless of their attainment (although they may use admissions criteria when oversubscribed). Non-selective schools may stream, track, or band pupils by ability or aptitude, but this generally takes place within a school rather than across different schools.

2.English grammar schools have existed in some form since before the 17th Century,1 but the modern grammar school concept dates back to the Education Act 1944.2 This provided for free secondary education after the age of 14, and split schools into three principal types: grammar schools, which focused on academic studies; secondary modern schools, which were intended to prepare children for trades; and secondary technical schools. Very few technical schools were established, undermining the structure and leaving—in effect—a two-tier system.

3.The number of state grammar schools peaked at almost 1,300 in the mid-1960s when around one-quarter of all pupils in state secondary schools attended grammar schools. While in Opposition, the Labour Party rejected the principle of academic selection, instead adopting a policy of ‘comprehensivisation’.3 The then Education Secretary Anthony Crosland requested plans from local authorities to convert all secondary schools into comprehensives,4 leading to a decline in the number of grammar schools, which continued through subsequent governments, with the fastest period of decline in the 1970s. The proportion of pupils in grammars fell to below 20% early that decade, below 10% by 1976, and has been 5% or less from the late 1970s onwards.

4.Grammar schools select almost all of their pupils based on examination of their academic ability, usually at age 11.5 Section 104 of the School Standards and Framework Act 1998 provided for the designation of maintained schools as grammar schools where the Secretary of State was satisfied that a school had selective admission arrangements at the beginning of the 1997–98 school year, the upshot of which being a ban on new grammars. There are currently 163 grammar schools in England with a total of 167,000 pupils.6

The Government’s proposals for selective education

5.On 12 September 2016 the Government launched a consultation entitled Schools that Work for Everyone, which sought “views on proposals to create more good school places”.7 The Green Paper comprises four sections, including proposals relating to allowing selective schools to expand, or new ones to open, while making sure they support non-selective schools. This report focuses on those proposals, which attracted a great deal of public interest, and were a significant departure from the education policy of the past two decades.

Box 1: The Green Paper

The Government intends to apply conditions on new and expanded selective schools, and has proposed the following menu of options:

  • “Take a proportion of pupils from lower income households;
  • Establish a new non-selective secondary school;
  • Establish a primary feeder in an area with higher density of lower income households to widen access;
  • Partner with an existing non-selective school within a multi-academy trust or sponsor a currently underperforming and non-selective academy; and
  • Ensure that there are opportunities to join the selective school at different ages.”

Source: Department for Education, Schools that work for everyone (12 September 2016)

Our evidence check

6.On 8 November 2016 we held an oral evidence session in order to examine the evidence surrounding academically selective schools, or grammar schools. Evidence checks are a form of select committee inquiry previously used by the House of Commons Science and Technology Committee and the Education Committee, and have been used to take a closer look at the evidence base for a high-profile policy. The November session gave us an opportunity to hear cases for and against grammar schools, and their effects on academic attainment, social mobility, and the education system as a whole.

7.We heard from Dr Rebecca Allen, Reader in Education, University College London and Director of Education Datalab; Professor David Jesson, Professor of Economics, University of York; Mr Luke Sibieta, Programme Director, Institute for Fiscal Studies; Professor Anna Vignoles, Professor of Education, University of Cambridge; Nick Gibb MP, Minister for School Standards, Department for Education; Dr Tim Leunig, Chief Scientific Adviser and Chief Analyst, Department for Education; and Donna Ward, Chief Analyst, Department for Education. This report summarises the evidence we heard in one oral evidence session in relation to the Government’s proposals. It does not offer a comprehensive analysis of all the issues, which are well documented elsewhere.

8.We are very grateful to all of our witnesses; our standing specialist adviser on education, Professor Jo-Anne Baird, Director of the Department for Education at the University of Oxford; and the Parliamentary Office of Science and Technology, which was a great aid to our work.8

Aims of the policy

9.The aims of the policy set out by the Government fell broadly into three categories: to increase parental choice;9 to create more good schools;10 and to decrease the attainment gap between children from high and low socio-economic groups.11 This is expected to be achieved by repealing Section 104 of the School Standards and Framework Act 1998.

Parental choice

10.Parental choice was a recurring theme throughout the evidence check. We asked the Government how many new grammar schools might emerge from this policy by the anticipated end of the present Parliament in 2020. The Minister told us that growth was not pre-determined and “will be driven by demand from local communities”12 and that new school creation “will ultimately be driven by local demand and what parents want.”13 Donna Ward told us that the Government would consult with local communities before the expansion of existing grammar schools or the creation of new grammar schools was permitted.14

11.We asked the Minister what would happen if a local community did not make such a request.15 There is still a degree of ambiguity around the response to such a potential situation—the Government did not explain whether an external figure (or body) would decide on behalf of communities if the expansion was deemed to be in their best interest. The Minister did tell us that the aim was to create schools that have:

the ability to provide a very rigorous academic education for high-ability pupils [and that] they can then provide the support and knowhow to the other schools in the area about how to raise academic standards for their most able pupils.16

Creation of more good schools

12.The second reason given by the Minister for the introduction of the policy was that “grammar schools are good schools”.17 He noted that 99% of them are rated at least “good” by Ofsted and 82% are “outstanding”, while only 20% of non-selective schools are rated “outstanding”18 under the same measure. The Minister told us that:

one of the conditions for [grammar school] expansion [ … ] would be that you have to work with local neighbouring schools, either to establish an honest, non-selective school or to work with an underperforming selective school.19

He told us that the Government intends to:

bring the DNA of a grammar school, how they educate the most able pupils, to take that expertise about the curriculum, and about teaching methods, to those other schools to ensure that they are delivering similarly high-quality education.20

13.Professor Vignoles told us that she was struck in the green paper by the statistic that 99% of grammar schools were rated “outstanding” or “good”: “That is a very high proportion. We think that that is down to the students who attend those schools or the teachers who teach in them or a combination of both.”21 She noted that it was “debatable” that “grammar schools know better how to educate [ … ] pupils, and therefore, could they be the ones to provide the assistance to the non-selective schools, as it appears to be proposed”.22

14.Professor Vignoles also emphasised that “there is an assumption in the Green Paper [that] the grammar school is a superior quality school and that they are doing things that provide those children with a better education.”23 We understand that—indeed, in many cases—current grammar schools do provide an excellent quality of education for the children who attend them. That said there is a “big assumption”,24 in Professor Vignoles’s words, that the same model of schooling in these good and outstanding grammar schools would be as effective when applied to a school with a more varied intake. This is an area for which the existing evidence is inconclusive. Luke Sibieta told us that “grammar schools do not necessarily need to be part of the equation.  It is offering high-quality academic education and high expectations [that matters].”25

Narrowing the attainment gap

15.The pupil premium is additional funding made available by the Government for publicly-funded schools in England, with a view to raising the “attainment of disadvantaged pupils of all abilities and to close the gaps between them and their peers”:26 closing the attainment gap. In oral evidence, the Minister cited a recent report into schools in Knowsley—an area that has historically low levels of educational achievement—which concluded that grammar schools “had the potential to transform education in the area.”27 In 2016–2017 financial year, schools receive £1,320 per registered pupil in reception to year 6, and £935 per pupil in year 7 to year 11, where those pupils were eligible for free school meals at any point in the previous six years.28

16.We asked the witnesses whether, assuming the same prior attainment, pupils from lower socio-economic backgrounds achieve better in grammar schools than they would do in non-selective schools. Professor Jesson explained that—from the current evidence base—“we find very little difference between the performance of those children with that measure.”29 Dr Allen highlighted the uncertainty that exists in this area, noting that the data is imperfect:

We are grouping this huge group of children who are eligible for the pupil premium together [ … ] Yet, when we look at the data, we can see that those who are eligible for the pupil premium in grammar schools are less disadvantaged. They have spent less time on free school meals.30

17.Professor Vignoles concluded that she did not think one could “sensibly argue that grammar schools are a force for social mobility.”31 Professor Jesson went on to say that:

the increase in difference in outcomes between the advantaged or the ordinary child, and the children who are disadvantaged—with free school meals—has been fairly stable over the last few years [and] I believe there is a disadvantage about the selective system, notably because of its very poor record at recruiting youngsters from disadvantaged backgrounds.32

This is a point that the Government has indicated it wishes to tackle in the development of this policy.33

18.Members of the first panel were in broad agreement that the evidence that pupils from lower socio-economic backgrounds did better in grammar school was weak. They told us that historical and current evidence did not indicate that grammar schools had narrowed the attainment gap.

19.The policy aims set out by the Government—especially with respect to closing the attainment gap—differ significantly from the characteristics of grammar schools of the past and present.

Historical and international context

20.The Minister emphasised the importance of the Government’s proposed reforms not being a return to the 1950s and 1960s ‘binary model’, “where the alternative to a grammar school was a very weak secondary modern”.34 Instead, the Government proposes that the alternative to going to a new grammar school would be for a child to attend “a very good comprehensive school”,35 which would be effective at preparing them for modern life. The Minister reassured us that “we are taking seriously the improvement in skills so that the individual can fulfil their potential, but also for the needs of our economy.”36 He noted that our economy is short of engineers, mathematicians, and physicists, and that this was part of the drive for the Government’s proposed education reforms.

21.The first panel of witnesses discussed the current labour market,37 noting its demands for different skills than those that were required in the past. The witnesses told us that—when children are tracked by ability—the quality of the provision for children who do not pass the 11-plus can ameliorate the disadvantage for this group. Dr Allen drew attention to examples from abroad, noting that the economic gains of academic subject-specialisation can have significant effects on a country’s output.38 We were told that, internationally, the move towards delayed tracking of students—delaying academic selection—is consistent with the changes that have taken place in the UK economy.39 However, in some countries with high-performing education systems, tracking has proven beneficial. Dr Allen told us that in Germany, Hungary, and Austria, for example, tracking is used extensively. Luke Sibieta noted that the benefits of selection are most noticeable after 18, “because we think there are large gains from specialisation at [that] age”.40

22.However, wide-scale selection in the 1950s and 1960s was predicated largely on the expected benefits of early specialisation. Since then, the UK economy has modernised and having a substantial supply of individuals able to fill unskilled jobs no longer matches economic demand. As Mr Sibieta noted, “we are moving to a society where we have a need for a more general level of skills, which suggests that selection at age 11 is less productive.”41 Professor Vignoles agreed: “we have moved to a labour market that requires a broader range of general skills [which] pushes you towards a more general education for longer”.42

23.OECD analysis reviewed for us by the Parliamentary Office of Science and Technology found that, in highly differentiated systems, the impact of a student’s socio-economic status on their educational outcome was stronger than in less differentiated systems. This is perhaps due to disadvantaged students tending to be grouped into less academically-focused tracks. The Minister agreed that the UK economy requires a higher standard of education across the board than in the past, but made clear the Government’s determination that the current proposals would not replicate the then weak alternative provision of the past. Tim Leunig, the Department’s Chief Scientific Adviser, offered an example from abroad which suggested that early academic selection and overall educational performance were not necessarily mutually exclusive:

I think the most interesting country is the Netherlands. It is a much more similar country to ours [ … ] They select at the age of 12 [ … ] It is a tripartite of academic, semi-academic and technical. They are then stratified within each stream and in total they have nine different strata, so it is a very selective system. They do better than us on PISA, so clearly selection is compatible with doing well.43

24.The Government’s proposals must take account of the needs of the economy for a broadly skilled workforce, recognising that generally technical specialisation occurs later in a student’s education, and take into account the UK’s competitiveness in a globalised economy. This will involve having regard to international trends and the performance of other countries’ education systems, which do not always point towards earlier specialisation within school systems, and attention should be paid to the Dutch model, which is overall a successful system and one that includes selection. If England is to take this course, it would be important for the Government to demonstrate clearly how this policy will meet the requirements of the Industrial Strategy.


1Private Education from the Sixteenth Century: Developments from the 16th to the early 19th Century’, in JS Cockburn, HPF King and KGT McDonnell, eds, A History of the County of Middlesex (London, 1969), pp 241–255

3 Grammar Schools and Selection, Lords Library Note LLN 2016/0049, House of Lords Library, October 2016

4 Comprehensive schools were implemented more widely in Scotland than in England, Wales or Northern Ireland.

5 Grammar School Statistics, Standard Note SN01398, House of Commons Library, November 2016

6 Ibid

7 Department for Education, Schools that Work for Everyone, September 2016

8 Academic Evidence on Selective Secondary Education, POSTbrief 22, Parliamentary Office of Science and Technology, December 2016

9 Qq 57, 59, 72, 75

10 Qq 11, 20, 44, 46

11 Qq 2–3, 15–16, 37

12 Q57

13 Q59

14 Q75

15 Q72

16 Ibid

17 Q44

18 Ibid

19 Q46

20 Ibid

21 Q11

22 Ibid

23 Q20

24 Ibid

25 Q16

26 Department for Education, ‘Pupil premium: funding and accountability for schools’, accessed 9 February 2017

27 Q72

28 Department for Education, ‘Pupil premium: funding and accountability for schools’, accessed 9 February 2017

29 Q37

30 Ibid

31 Q2

32 Q3

33 Department for Education, Schools that Work for Everyone, September 2016

34 Q65

35 Ibid

36 Q76

37 Qq 5, 10

38 Q5

39 Ibid

40 Q33

41 Ibid

42 Q10

43 Q84. The Programme for International Student Assessment (“PISA”) is a triennial international survey designed to evaluate education systems worldwide. In 2015, half a million pupils from 72 countries took the internationally agreed two-hour PISA test, results and rankings from which were published in December 2016. See OECD, ‘What is PISA?’, accessed 9 February 2017




10 February 2017