Earlier this week David Willetts sought your questions on grammar schools. We said at the time that ConservativeHome would pick ten of those questions and David Willetts supplied his answers to those questions yesterday afternoon. Our apologies to those questioners who we didn't choose. David's answers are published below...
"First of all, many thanks for the Tim Montgomerie for this opportunity to engage with these issues. It is a bit frustrating that these questions are all about grammar schools when this speech ranged very widely, but they do help explain what my speech was all about."
N: "You base your argument that grammar schools do not promote social mobility on statistics relating to the number of pupils entitled to free school meals. Do other indices of poverty corroborate this finding? 30% of the pupils at my former school are from the most deprived quintile. 50% of pupils are ethnic minorities, many of whom speak English as a foreign language. This might not be typical, I concede, but it is one example of a grammar school that does not entrench privilege."
"There has been some misunderstanding of the evidence on free school meal eligibility (FSM) and grammar schools. I cited evidence from the Sutton Trust (The Sutton Trust, 'Rates of Eligibility for Free School Meals at the Top State Schools', October 2005) that across the nation as a whole 14% of pupils are eligible for free school meals, whereas in areas where the grammar schools survive, the figure is 12%. However, only 2% of the grammar school children are eligible for free school meals. (Incidentally, these are figures for free meal eligibility, rather than claims).
Some of the children who are missing out in this way are of high-ability (measured in terms of their Key Stage 2 score). It is not merely that poor children are underrepresented at grammar schools, high ability children from poor families are underrepresented at grammar schools. A high ability FSM child has a 32% of passing the 11+ whereas a more affluent high ability child has a 60% chance. (A. Atkinson, P. Gregg, B. McConnell, 'The result of 11 plus selection; An investigation into opportunities and outcomes for pupils in selective LEAs', April 2006, Working Paper No. 06/150, CMPO).
I know that this is uncomfortable evidence, but we need to confront it. Some people have written to me to explain that they were from modest backgrounds and that they owe their achievements in life to grammar schools. I have no doubt that grammar schools truly helped those people. Grammar schools remain excellent schools for those people who get into them; this is why we are committed to keeping them. However, as society has changed so it has affected the diversity of experiences of childhood and made social selection more of a problem. Many of those people from poor backgrounds who got into grammar schools in previous decades would not do so now.
Going back to your question, you are correct both that the FSM eligibility is not a perfect indicator of poverty, and that there are many people who are not FSM-eligible who are - by any reasonable standards - poor. However, it is a useful short-hand proxy and the short answer to your question: "does other evidence corroborate the FSM findings" is "yes". The following quotation by one of our leading researchers in this field - Dr. Anna Vignoles from the Institute of Education - is a good summary of the evidence:
"There is a substantial body of evidence that suggest that children from higher socio-economic backgrounds are more likely to access grammar schools, for a given level of ability. This results in children from wealthier backgrounds being disproportionately represented in grammar schools. Although most research on current data relies on using the somewhat limited Free School Meal measure as an indicator of the socio-economic background of the child, earlier studies that used rich information on parental income and/or social class, also find that children from more advantaged backgrounds are more likely to be enrolled in grammar school than similarly able but less well off children."
(Cited with permission of the author).
Robert T: "Could Mr. Willetts please explain why it is acceptable to separate children by ability within schools but not to separate children into different schools on the same grounds?"
"In addition to the social selection issue outlined above, it is far easier for children to move between sets than it is to move between schools. Any child who for whatever reason is behind at 11 or 13 but develops later is unlikely to get into a grammar school. In streamed schools there is always the opportunity to move up to the grammar stream. In such a system it is less likely that children with potential are left behind."
1AM: "Mr. Willetts, as we all agree Grammar schools help the children who attend them, do you believe:
a) That children in an area with grammar schools, who themselves don't get in to the grammar school, are disadvantaged?
b) That children in an area with grammar schools, who themselves don't get in to the grammar school, are NOT disadvantaged?
If a) then surely we should abolish all grammar schools - as we are harming the education of the vast majority to benefit a small minority. We should also seriously look at abolishing private schools - which also have too much of a middle class bias and elitist standards. I wonder how many children on free school meals are at Eton?
If b) then we should surely be arguing for the expansion of grammar schools as they benefit the 25% who attend without any loss to those who do not attend.
Your defence of the status quo is illogical and confused."
"I am not defending the status quo.
There is a modest disadvantage to children in selective areas who do not get into grammar schools. (A. Atkinson, P. Gregg, B. McConnell, 'The result of 11 plus selection; An investigation into opportunities and outcomes for pupils in selective LEAs', April 2006, Working Paper No. 06/150, CMPO)
However, you have presented a false dilemma, choosing between comprehensives and grammar schools.
As it happens, the academically successful comprehensive schools are also socially selective. (The Sutton Trust, The Social Composition of Top Comprehensive Schools - Rates of Eligibility for Free School Meals at the 200 Highest Performing Comprehensives , January 2006, S. Burgess, A. Briggs, School Assignment, School Choice and Social Mobility, November 2006, Working Paper No. 0.16/157, CMPO)
In my speech, I have tried to move the debate on to how we spread the opportunity and advantages which have historically existed inside those top schools to those children unlucky enough not to get into them. In short, how we create more good school places.
We are committed to keeping the grammar schools where they exist, but reform in the future will come from unleashing the power of pupils choosing schools rather than schools choosing pupils."
Piers Thompson: "In Northern Ireland where almost half of children are educated at Grammar Schools, 42% of children from poorer backgrounds go to university compared with 28% in the rest of the UK. If the Conservatives are serious about improving social mobility shouldn't we learn lessons from Northern Ireland and aim to have more Grammar Schools in the rest of the UK?"
"Northern Ireland has a significantly different economic structure to the rest of the UK. The Troubles have in many ways preserved a type of society which has disappeared in England. The religious composition is different, household and family structure is different, the average age is lower, marriages are more likely to last and the labour market structure of the province is very different to the rest of the country. Even comparing social mobility in Northern Ireland to the rest of the UK, given the sudden improvement in the political fortunes of the Catholic population over the past 30 years, is impossible. If the school system in Northern Ireland is genuinely more successful, it may be because (as I argued in my speech) the social structure there is like the structure which sustained grammar schools in England in previous decades.
These thorny issues are why we need serious pupil-level research - such as that produced by the University of Bristol - which can help us. They have analysed various selective and non-selective LEAs in England; an approach which is much better since it avoids a lot of the problems of using Northern Irish data. Whilst they have found that grammar schools are excellent vehicles for the children who get into them, they are also socially selective. This is not an isolated result, nor is it a result confined to the UK’s experiences with selective education. (A. Atkinson, P. Gregg, B. McConnell, 'The result of 11 plus selection; An investigation into opportunities and outcomes for pupils in selective LEAs', April 2006, Working Paper No. 06/150 CMPO; A. Manning, J. Pischke, Comprehensive versus Selective Schooling in England & Wales: What do we know? June 2006, Centre for the Economics of Education and (for example) PISA-Konsortium Deutschland, PISA 2003, Ergebnisse des zweiten internationalen Vergleichs Zusammenfassung, 2003, OECD; C. Meghir, M. Palme Educational Reform, ability and family background, IFS Working Papers, no.W04/10 (2004) The Institute for Fiscal Studies).
On the specific statistic you have quoted, the CPS's claim is actually that 42% of university entrants in Northern Ireland are from "less privileged backgrounds". That is quite different to 42% of less privileged Northern Irish children going to university. (CPS Briefing Note – Grammar Schools – 19th May 2007)
I cannot find the original source for this statistic, but if an area has a higher proportion of people from low income households than the average, all else being equal, one would expect a higher proportion than average of university applications from that area to come from low income households. Until I know more about how they are defining poverty, I would be loathe to analyse that statistic. However it is certainly possible that, under the definition of poverty being used in this statistic, differences in the socio-economic structures could account for all of the difference. For example, there are one and a half times as many out-of-work benefit claimants in Northern Ireland per capita than in the rest of the UK, and using some definitions of poverty, Northern Ireland is very significantly poorer."
Jonathan Powell: "I saw you on the Andrew Marr program on Sunday, and you said the problem with grammar schools was that by the time poor children were 11 they had fallen too far behind, so it tended to be the middle class children that got into the grammar schools and thus you claimed they did nothing for social mobility. Surely, if this is true, the answer is not to give up on grammar schools but to improve primary education, say by having a voucher system, or perhaps having selection at a younger age?"
"Thank you for engaging with the central argument that grammar schools have not changed, but English society has. I agree, part of the answer is to improve what has happened before the age of 11. However, I have to deal with the world as it is, not the world as I would like it to be. Improving primary education and tackling root disadvantage are very important. However, we cannot write off those generations of school children who will come through in the meantime (nor should we not pretend our secondary schools are doing as well as they might)."
DavidTBreaker: "Why do your children not attend local state schools? I've no problem with private schools, I support them fully, but if you think grammars harm social mobility aren't private schools even more damaging? Why did you even mention grammar schools? Surely you knew the uproar it would create? Why didn't you discuss private schools which far outnumber grammars?"
"I fully support parents doing everything they can for their children. Aspiration is the engine of a strong and healthy society. However, the question is how we deal with the consequences of it; namely social selection. At the moment, parents are all trying to get their children into private schools, grammar schools and the academically successful comprehensives and so it is a zero sum game. Instead of simply arguing how we parcel out a fixed number of good places, we are thinking about how we improve all the other schools. Far from getting rid of them, the question is how we spread the advantage.
Our proposal for improving our educational system is to increase the power of parents over local schools; a pretty strong endorsement of aspiration!"
Ted Broadbent: "When one of the emerging themes of Conservative policy is localism, how does the pledge to remove the local referendums on whether to keep existing grammar schools fit in with this theme? And if the party was really keen on localism shouldn't Conservative policy be to expand the referendum system to allow parents to vote to set up grammar schools if that is the will of the local community?"
Andrew Lilico: "In 2001 we had a free schools policy, and my understanding from Nadine Dorries is that the education subgroup of Cameron's Public Policy Commission may be inclined to recommend that at least some schools are free to determine their own admissions criteria. Following your speech, would it be feasible for you as Education Secretary to permit schools, choosing for themselves, to adopt academic selection as their basis of admission?"
“The official position of the Policy Group appeared in a letter to the Daily Telegraph on 24th May from the head of the education group, Baroness Perry.
The 1979 Conservative Government introduced legislation which would allow local authorities to reintroduce grammar schools which remained in force until 1998. Only one Conservative council tried to use it (Solihull in 1984) and they abandoned it in the face of a parental revolt. We did have the opportunity to create more grammar schools and it did not happen. In fact, the number of grammar schools fell by 96. The CPS's own polling explains why: setting and streaming in mixed ability schools are more popular than grammar schools. (CPS Briefing Note – Grammar Schools – 19th May 2007).
Setting the right balance between local discretion and a national framework is very difficult. However, across the world, state of the art school reform operates with a fair funding formula and no selection by academic ability. Sweden (which had a history of selective education) banned selection by ability. In the successful US charter schools, they have not allowed selection. (for a review of these experiments, see C. Hoxby, School Choice and School Productivity, Working Paper, February 2001, Harvard) It was certainly no part of Milton Friedman's vision for vouchers than the producers should be in a position to select the consumers.
The think-tank 'Reform', whilst in broad agreement with our policy, has cited the example of the Dutch school system which does allow selection in a choice model. However, designing the rules of the market is very different if you are starting from a situation where there is a deficit of good school places (as is the case in the UK and as was the case in 1990s Sweden or the USA) as against if you are in a situation where there is a surplus (as is the case in the Netherlands). It really functions as a sorting rather than as a rationing system (which is how it would function here)."
Simon N: "How was this policy reached? And why was it announced this way? The party hasn't so far felt the need to reveal policies on anythign much, and there was no clamour for a policy on schools last week. Yet reports suggest this major policy was rammed through a sceptical shadow cabinet without the leader present, and with little discussion, just weeks before the relevant policy commission was due to report. No official explanation has been given, but someone told the Daily Mail the aim was to embarass Gordon Brown by highlighting his differences with Blair."
"This is something which David Cameron referred to during the leadership campaign on 10th November 2005. He wrote "if all we have to say on education is 'Bring back 20 grammar schools what do we have to say to the parents of children in the 26,000 schools where standards are falling and Labour are failing?" (The Guardian). He announced a policy of not extending selection on 9th January 2006. On 10th January 2006, it was attacked by the Telegraph as “repudiating the most successful educational model in our history"; a full 14 months ago.
The new elements in the speech were on opening up the supply side of education, using academies to drive up standards across the country, to support federations of small schools and driving diversity of provision."
dogides: "Assume that I live in an area where there are two schools, one known to be relatively bad and one known to be relatively good. I realise that we would like all schools to be equally good but I hope you appreciate that this will never be so. Under the system that you propose, is there any way at all that by working hard and getting good grades that I can increase the chance of getting into the good school, or will it be determined purely by chance? I am only partially interested in setting. I realise that by working hard I will get into a better set. But I know that the top set in one school simply isn't as good as the top set at the other school. I know that as a top set student in the poorer school I will get bullied for being a swot. Because if I can't earn some level of control by working hard then how can I realise my aspirations? How can I be what I want to be?"
"In your scenario, you have assumed away a few issues. Under current rules, there would be no places going spare at the other school and - if the other school is a grammar school - you have assumed away the problem of the entrance procedure. You also assume that your current 'bad' school uses setting or streaming. It is also a rather strange assumption that it makes more sense to move you and the other swots to another school rather than take on the bullies!
However, your question does drive to the heart of this issue.
First, there is no way fairly to apportion a fixed number of good school places. Ideally, choice schemes require that good schools be allowed to expand and have sufficient capacity so that parents can choose the school which suits them. If the gulf between your current school and the neighbour was significant, the neighbour could expand and you could join it. At the same time, your moving would issue a challenge to that bad school that it should raise its standards. In a choice scheme, the quality of all schools rises - and those which are the worst rise fastest and furthest. (for an overview, see C. Hoxby, School Choice and School Productivity, Working Paper, February 2001, Harvard)
Second, on life inside that second school, standards, setting and streaming are crucial.
Third, we need a focus on discipline. There is good evidence that through tried-and-tested techniques such as setting and streaming, children are better behaved, enjoy better outcomes and even have higher self-esteem. However, we need to allow heads to deal as they wish with their misbehaving students. This requires more work to be done on the Pupil Referral Units and on provision for expelled pupils."
***
“I suspect those answers are rather longer than you expected! Many thanks again to Tim.”
