Education Committee
Oral evidence: Selective education, HC 780
Tuesday 8 November 2016
Ordered by the House of Commons to be published on 8 November 2016.
Members present: Neil Carmichael (Chair); Lucy Allan; Suella Fernandes; Lucy Frazer; Lilian Greenwood; Catherine McKinnell; Ian Mearns; William Wragg.
Questions 1 – 106
Witnesses
Dr Rebecca Allen, Director, Education Datalab, Professor David Jesson, Professor of Economics and Education, University of York, Luke Sibieta, Programme Director, Institute for Fiscal Studies, and Professor Anna Vignoles, Professor of Education, University of Cambridge.
Mr Nick Gibb MP, Minister for School Standards, Dr Tim Leunig, Chief Scientific Adviser and Chief Analyst and Donna Ward, Chief Analyst, Department for Education.
Examination of Witnesses
Dr Rebecca Allen, Professor David Jesson, Luke Sibieta and Professor Anna Vignoles gave evidence.
Q1 Chair: Good morning and welcome to our session today. It is about selective schools. The purpose of our session is to review the evidence relating to the Government’s proposals for selective schools, as outlined in the Green Paper entitled “Schools that work for everyone”. We have a set of questions that we want to ask you, our panellists. Before we begin, can you please introduce yourselves and say briefly who you represent, starting with Luke?
Luke Sibieta: Good morning. My name is Luke Sibieta. I am Programme Director of the Education and Skills Sector at the Institute for Fiscal Studies.
Professor Vignoles: I am Anna Vignoles from the Faculty of Education at the University of Cambridge. I am Director of Research there.
Dr Allen: I am Rebecca Allen. I am Director of Education Datalab. We are a research centre that helps support schools and policymakers in making use of data. I am also on leave from the UCL Institute of Education.
Professor Jesson: I am David Jesson from the University of York, recently retired from the Department of Economics where I was Director of the Centre for Performance Evaluation and Resource Management. I recently joined the Department of Education at the same university, as a retired member.
Q2 Chair: Fabulous. That is excellent. My first question is, is the evidence base on academic selection a sound basis for the selective schools policy proposed by the Green Paper?
Professor Vignoles: There are three separate issues that we need to think about here. First, do pupils do better in grammar schools? Secondly, are grammar systems better than non-selective systems? Thirdly, is a selective system or grammar system good for social mobility?
Do pupils do better in grammar schools? Generally, high-attaining students do do better in grammar schools, but the effect is modest and some estimates are zero.
Is a grammar system better than a non-selective system? There does not seem to be much evidence to support that as a strong finding.
Finally, are grammar schools a force for social mobility? The answer on that is definitively no, both if you look historically, in terms of the likelihood of students accessing grammar schools in the past, if they came from disadvantaged backgrounds, and then, looking at the current system, it is also true that students from poor backgrounds are relatively less likely to access grammar schools. We can then perhaps talk about what a future system might look like and whether you could change that relationship, but looking at the historic and the current data, I don’t think you could sensibly argue that grammar schools are a force for social mobility.
Q3 Chair: Rebecca, if we have a grammar school system and children go to grammar school and, as Anna suggests, ones who are academically confident will do okay, what about the ones who are not and don’t go to the grammar school and end up in a different type of school? What happens to them?
Dr Allen: My judgment of the literature is that these students lose out compared to going through a comprehensive system. Admittedly, the size of the damage to their GCSE outcomes is small. We might talk about something like a 10th of a grade per subject, but there are a great many students in this group. There are three times as many on average, typically, in secondary moderns than there are in grammar schools. Not only that, but some studies find that this impact is what is known as statistically insignificant. We cannot be confident that it is different from zero.
This happens when you have a small negative impact and it is imprecisely estimated, but given that not a single study claims that these students benefit from a secondary modern and many or most claim that they are worse off in a secondary modern, I think we have to say that the balance of evidence is that they are worse off in a secondary modern.
The question for us is: if the data shows that, is there a plausible reason as to why they are worse off? The research that we carried out provides at least one explanation as to why there are losers in this kind of system. The current selective systems in England—by that I am talking about areas such as Kent or Lincolnshire or Buckinghamshire—have far more pronounced inequalities and access to suitably qualified teachers than do comprehensive or non-selective systems. What do I mean by that? Grammar schools are more likely to have fewer unqualified teachers, far more experienced teachers than in secondary moderns, more teachers with an academic degree in the subject that they are teaching, and less churn of teachers. This provides one piece of evidence as to why secondary moderns find it so difficult to function within a selective system.
Chair: David, what are your thoughts?
Professor Jesson: Interestingly enough, the term “secondary modern” is being used quite a lot these days. It does go back to a previous Minister of State for Education in the 1980s, Rhodes Boyson, who made the point—it may have been uncomfortable at the time—that any school that was in the neighbourhood of a selective grammar school was nothing but a misnamed secondary modern school. I take it from his own words that that was a word of abuse and, to an extent, I do think that the words “secondary modern” and “second rate” very often tend to go in people’s eyes as being the same thing and that is most unfortunate.
Subsequently, beyond that, I do think there is a strong issue 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. Advantaged local areas, which have comprehensive schools, have a gap of about 25% in the performance of the youngsters who are—I would say straightforwardly—without free school meals and those who are on free school meals. I am sorry to say that, within selective areas, recent evidence is that it is well over 30%. In that sense, I believe there is a disadvantage about the selective system, notably because of its very poor record at recruiting youngsters from disadvantaged backgrounds.
Q4 Chair: You are an economist, so you will be interested in the impact that those sorts of figures have on our productivity as an economy, as well as the social mobility issues, so what do you think they are?
Professor Jesson: Sorry, would you repeat the question?
Chair: What do you think those figures mean for economic productivity if they are accurate?
Professor Jesson: Very interestingly, the likelihood of achieving a good job is enhanced by good examination results. There is little doubt about that, and if it is that disadvantaged children are at a further disadvantage by being educated in selective systems I am pretty sure that that has an influence on their subsequent earnings.
Q5 Chair: Rebecca, do you want a quick comment before I ask Luke something?
Dr Allen: The international literature written by economists on what they call tracking and we call academic selection, explains this quite well. They talk about the gains of specialisation for children for the economy against the costs of tracking and the costs of mistracking as well. What they say is, “This is a really important decision.” They talk about potentially having big effects on country output; by that we mean 2%, which is a really big figure.
What they also say is that, as demand in the country shifts towards more general skills, there are lower benefits of specialisation. That is why it is natural that across the western economies, across the developed world, we have seen a move towards delayed tracking of students, delaying academic selection. That is consistent with the changes that take place in our economy. They say that, overall across the world, by their estimates, it is undoubtedly true that some tracking is beneficial. In this country we accept that we track children at 16. We are okay with that. Lots of countries do it at 15.
What they also say is that there are many countries who appear to be tracking for longer than is optimal—and the obvious examples of that are Germany, Hungary and Austria—and that there is this clear trade-off that, across the international literature, wherever you see tracking and, in particular, early tracking, you see that the impact of family background on educational attainment and on labour market outcomes strengthens as a result.
Q6 Chair: Luke, your organisation is already in the news today with its forecast about loss of revenue from tax and you have come up with quite a significant figure, which I suppose concerns us all. Do you think that the direction of travel, as paved by the Green Paper, is going to help us narrow that gap or will it widen it?
Luke Sibieta: There are two important points to make from that. Building on what everyone else has said, the effect of grammar schools, in terms of skill base and individuals’ earnings and their ability to pay taxes, depends crucially on what the offer is to those who are not able to attend grammar schools.
If we look at a couple of examples, there is a paper looking at grammar schools in Aberdeen in the 1950s and 1960s. There were big losses for women who were not able to attend grammar schools. There was not a fairly decent offer of education for them. For men who were not able to attend grammar schools they did not experience a big loss, because there was a high-quality level of vocational training and fairly clear occupational routes for those men.
If you then look at children in Yorkshire in the 1960s and 1970s, children who were not able to attend grammar schools experienced more of a negative effect, in terms of not being able to go, because there was no route for them to go to university. Almost none of the children who were not able to go to grammar schools in Yorkshire in the 1960s and 1970s went to university because they did not have access to O levels and could not go on to university. A similar thing happened in Northern Ireland.[1]
What I am trying to say is that if you want a skill base that is focused on the number of élites, and highly skilled individuals, the selective education system worked very well in the 1950s and 1960s. As Rebecca has just said, we are probably not in that era of technology any more. We are in an era where we have more massification of skills and decision making across firms and across the economy, and if you want an economy that has a mass level of skills you need an education system that delivers that as well.
Q7 Chair: Anna, could you comment on the effects—I think it is an important question—that academic selection has within the school system?
Professor Vignoles: This is where we need to be careful about using evidence from the past to inform a policy that might look quite different from the grammar systems of old. In the old system, you had a very structured system where slightly more than one-fifth of students ended up in grammar schools. That was organised by local authorities. As I understand the proposals in the Green Paper, if you were to allow any school to admit by ability that would end up in quite a different situation. One issue would be that every school has an incentive potentially to try to admit students by ability, in order to improve their intake and compete with their neighbouring schools. One can imagine that if you allowed schools to admit by ability across the piece—and when I call it “ability”, I am talking about a measure of achievement at the age of 11—you could end up with a perfectly sorted education system where children are ranked into different kinds of schools.
The consequences for the schools at the bottom of that system, I think we would all agree, would be dire. Not least, teachers, presumably, would be less willing to work in such schools. Further, there is a hidden issue perhaps that very low-achieving students generally require smaller class sizes. If you had schools where the only students enrolled were of generally lower achievement, you would have this issue that you would have either larger class sizes for them than they would receive under a comprehensive system; thereby the quality of their offer would be lower.
Inside that, there is this conundrum that we face whenever we talk about selective schooling, and that is that what might benefit one student comes at a potential cost to another student. That becomes somewhat of a normative judgment as to where that line is.
Q8 Lucy Frazer: Anna, may I pick up on something you said: have you done a similar study in relation to independent schools? You described the effect whereby you get better teachers in a grammar school, so is there a similar effect in independent schools or have you not done that study?
Professor Vignoles: We haven’t, simply because the data isn’t there. We cannot collect data on teachers who are in independent schools.
Q9 Lucy Frazer: What I am hearing from all of you is that people in the grammar schools do better but everyone else around the grammar school does worse. Is that not an argument to improve the education everywhere else, rather than take away the good education from those in grammar schools? Is it possible to do that?
Professor Vignoles: You have to ask: why are those schools doing worse? If it is something to do with either resource sorting—and by that I mean we have a fixed stock of teachers, some of whom are going to the grammar schools and, by virtue of being at the grammar schools, we cannot put them in the secondary moderns—then, no, we cannot create this system that can somehow make secondary moderns better-functioning schools while we have grammar schools.
If it is something different, something about the policies that have been put in place that mean that we can do material things to improve all schools, then, sure, we could potentially have a grammar school sitting alongside well-functioning secondary moderns. But I would argue it is difficult to run a school where, not only do we have problems of teacher sorting, but we have problems of pupil sorting too. The impact of that is not well understood. Children who arrive in secondary moderns have potentially lowered expectations, damaged self-esteem. They come from lower social class backgrounds. We have the problems of lowered parental social capital enveloping that school and supporting that school as a result. We may have concentrated problems of behaviour and disadvantage at those schools. All of those things are things that manifest themselves because they are grammar schools, not because these schools happen to be sitting alongside grammar schools.
Q10 William Wragg: I want to follow up on something Professor Vignoles mentioned about the “old system”. The old system was a failure, in the sense that the intention of the 1944 Education Act was to have a tripartite system. The technical colleges, which were meant to be that third aspect, only ever made up roughly 2% of the school-going population. Do you think that references to the old system are valid considering today’s much more varied education landscape?
Professor Vignoles: That is a very good point. The tripartite system never really existed. It was always essentially grammars and secondary moderns, with the notable exceptions of some very good technical schools. We might want to ask why that is. It comes back to Rebecca’s point that as we have moved to a labour market that requires a broader range of general skills, both parents and employers want a higher level of general skills, and that pushes you towards a more general education for longer. Again, this comes back to Luke’s point. What would be an alternative offer if you did revert to a selective education system? What is it that we would be trying to provide for those in secondary moderns?
If I may add one point, in response to your query about the independent sector, work that Professor Jesson, Luke and I have done indicates that a substantial fraction of those in grammar schools are currently coming from the independent sector—around double the rate that one would expect given the proportion in the independent sector. In some areas it is as high as one-third of the intake of individual schools or local authority grammar schools. That also may explain some of the performance, because your peer group is quite different. It also begs the question, what would happen when you roll a system out far more widely than it currently is? That is an issue that we need to think about.
Q11 Lucy Frazer: Is that an advantage or a disadvantage that you are getting?
Professor Vignoles: I was struck in the Green Paper by the fact that 99% of grammar schools are 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. The question, I think, in the Green Paper is: is it that grammar schools know better how to educate all pupils, and therefore, could they be the ones to provide the assistance to the non-selective schools, as it appears to be proposed? That is a debatable point.
Dr Allen: On that point of private schools, it is important to make the case that introducing grammar schools does not reduce the size of the private sector. So a county like Kent has a secondary private school sector that is as large as we would expect, given the demographics, in any other sector and comprehensive schools. It is just a different mix of students. It is students who failed the 11-plus but come from relatively well-off backgrounds. You have a different type of private sector. It is just as large. In fact, I would argue that the primary sector in Kent is larger than we would expect. That is consistent with the idea that to go to a private primary school is a good way to prepare for the 11-plus.
Q12 Ian Mearns: We have already heard this morning that, for the youngsters at the higher end of the academic range, grammar schools are doing marginally better, but even within a grammar school that is selective there is a range of ability. How are grammar schools doing for the youngsters who are not at the very high end of the academic range but still pass the 11-plus?
Dr Allen: I would argue that any claims that they are more or less effective for children of different abilities within grammar schools are not robust. There are good reasons why. We do not have access to 11-plus test data. That means we are mis-measuring ability. There are studies that suggest that those at the bottom of the ability distribution who enter grammar schools do better than others. I think that that claim is not correct. Equally, there are other studies that argue the opposite and that, for those who just pass the 11-plus, there are relatively few gains to getting into a grammar school. That is what the literature states.
Professor Vignoles: The only thing I would add to that is that when we looked historically—and I have said that historic data might not be very relevant but one aspect of it might be—we did find what looks to be quite a lot of misclassification. In other words, children who are in our dataset have high levels of achievement that you would expect to find in grammar schools, who were not in grammar schools, which talks to—and perhaps Professor Jesson and Luke want to come in—there is a social selection mechanism for who ends up in grammar schools and it is not entirely determined by your achievement at age 11.
Q13 Catherine McKinnell: That leads in quite well to the questions I was hoping to put to you. You have already touched on it to some extent, but the research does seem to indicate that social segregation between schools is much higher in areas where there are grammar schools. It would be helpful if you could explain for the Committee why that is the case, if that is something you agree with.
Luke Sibieta: I will start on that. The evidence is clear that, in terms of the social mix and socio-economic mix, only 3% of children who attend grammar schools are eligible for free school meals, which compares with about 17% of children in the area as a whole. We do not use the indicator of free school meals because we think that is the sole measure of poverty. It is the best that is available.
There are two main reasons why less well-off children are not able to attend grammar schools. The first is because they demonstrate lower levels of academic achievement by age 11. There are big socio-economic gaps in achievement and attainment, in terms of ability, even by age 3 and age 5. That widens throughout childhood. It is not surprising that children from less well-off families are less able to go to grammar schools.
Even if you look among children who have the highest levels of attainment at age 11—they get level 5, the top level, in both English and maths—about two-thirds of children who are not eligible for free school meals will go on to a grammar school and 40% of those who are eligible for free school meals will. Conditional on having high levels of ability, you are less likely to if you are eligible for free school meals. That could be because they are less well prepared for the test or perhaps they are less likely to apply because they don’t think they are going to get in.
Professor Jesson: I was going to touch on the point, which the Chairman raised, about the issue of youngsters in the middle range of achievement within grammar schools that overlap with those levels of attainment in the other range of schools. I have recently done some work on this on the last full data that we have, on 2015, and it does seem very clear—as Anna and Rebecca have pointed out—that there is a degree of overlap between those that does not show any significant advantage for youngsters who attend grammar schools, as opposed to those who are in the other great majority of schools in the country.
May I add a little point here? There are 151 local authorities in the country and 115 of those do not have any grammar schools; 12 of them, out of that 151, have 103 of the 163 grammar schools. Those are what we would call highly selective authorities. In those contexts, that is where the issue about differentiation between social backgrounds is profound. Something like 13% of the pupils who are in these highly selective grammar schools come from independent schools. Just under 3% come with any measure of disadvantage, the proxy that we use being eligibility for free school meals. My point here is that it looks as though there is an increasing element of differentiation, in terms of the social mix, the higher up the selective level you go. To that extent, the idea of reintroducing selective systems seems to me to be quite obtuse—perverse.
What I would say is that we ought to look a little more carefully at those relatively few authorities—24—who have partial selection. Within those there are some very interesting areas, such as Lancashire, Gloucestershire—your own constituency, I believe—and Warwickshire. There are a number of grammar schools there of ancient traditions and the funny thing is that, among those, the proportion of youngsters that go into the grammar schools with any level of disadvantage is under 2%, whereas those who come from independent schools is well over 14%. To my way of thinking, the problem is that that is exacerbated when you have a set of fairly isolated grammar schools within an already existent comprehensive system. I might have more to say about that in a moment but that is a very important point about social mix.
Q14 Catherine McKinnell: The evidence base seems to suggest that it does decrease social mobility. Therefore, the assumption I can make is that either the Government do not see that as a priority in coming forward with this proposal, or they see some way of mitigating that as a factor. Is that something that you have looked at? Can you see some way of mitigating that consequence of the expansion of grammars, somehow increasing social mobility, going forward?
Professor Jesson: The interesting thing here is that if you look at the primary schools—either independent or state primary schools—that are recruiting students for the grammar schools, it is quite remarkable that, year on year on year, the same schools provide the majority of pupils for the grammar schools.
There is one very interesting example that is going on in Birmingham at the present time with the King Edward VI school, which has so enhanced its rapport and its engagement with its local schools that something like 40 out of the 120 pupils that go to this independent school are from state primary schools and 10% of those are pupils who are on free school meals. My concern is here that, if we are not careful, we will replace selection at the age of 11, without reform of the way that schools recruit their pupils, with recruitment at the age of five or, at the latest, seven.
Q15 Catherine McKinnell: Do you believe that expanding the current policy in terms of grammar schools will potentially have that effect?
Professor Jesson: I think so, without a clear commitment to requiring new or existent grammar schools to look very carefully at how it is that they support the schools in their community, for youngsters in schools which may have no track record of providing youngsters for grammar schools. This is such an important issue. It is touched on in the Green Paper and I really rate that suggestion. I believe it is an important one that we should take very seriously in any moves to increase the potential number of grammar schools within this country.
Luke Sibieta: I want to make the point that the evidence on grammar schools does have lessons for ways to improve social mobility, but it is not necessarily about whether you expand or decrease the number of grammar schools. If you compare what was offered at grammar schools, particularly in the past, and what is offered at secondary moderns, it is either the offer of a high-quality, high-expectations academic education or the offer of CSEs or perhaps a few routes to an academic education. Unsurprisingly, those who were offered the academic education did far better as a result.
If you look, say, at Northern Ireland, 80% of children who went to grammar school got an A-level or more and 60% of those who did not go to a grammar school got an A-level. It is a completely different system of education. What the evidence on grammar schools says is that offering an academic education and high expectations to more children is a good thing for those children. If we look in other areas of England, particularly in London, London has done exactly that. It has offered high-quality academic education, high expectations, and children have benefited as a result.
Q16 Catherine McKinnell: That is without the grammar school system?
Luke Sibieta: Grammar schools do not necessarily need to be part of the equation. It is offering high-quality academic education and high expectations.
Q17 William Wragg: To talk about the free school meal entitlement percentage figures, because they are bandied around quite often, I wonder, is not the danger of overreliance on this figure that we are focusing on very limited areas of the country? Correct me if I am wrong, as I am sure you will. Are those areas where we have grammar schools generally in areas of relative affluence and, therefore, the percentage of entitlement to free school meals is generally lower than the country as a whole? Is that correct?
Dr Allen: I have the figures here. We currently have 2.5% eligible for free school meals in grammar schools. In the areas in which they are located, free school meal entitlement is 9%. Currently in the country as a whole, it is 13%. The free school meal children are significantly underrepresented.
Q18 William Wragg: But it is still lower than the national average in those areas. I am not disputing what you are saying.
Dr Allen: That is right. The figure of 9% is lower than 13%. You are right, yes.
William Wragg: Correct, yes. It is indeed.
Professor Vignoles: Can I add one point? We have talked about the fact that in grammar schools in some areas they are taking in large numbers of students from independent schools. Obviously, the other point to think about is that, when we did our research, grammar schools were also taking in students from outside their local area disproportionately. When you add those two groups together, many grammar schools cannot be described as catering for their local school market and so those numbers become quite problematic as to what you might expect.
Q19 William Wragg: A further point on social mobility, perhaps touching on, Professor Jesson, what you were saying. If there were a requirement for a fixed level of pupils who are entitled to free school meals who must attend a particular grammar school, what effect on social mobility do you think that would have?
Dr Allen: I can talk about the experience of Birmingham, because the King Edward VI schools are an example of a situation where they have managed to significantly increase the number of pupil premium children in their schools. They set the target that, at their schools, 20% of students in most of their schools and, in one, 25% would be pupil premium children, and they succeeded in reaching those goals within a year or two of setting them.
There was a small problem. Because King Edward VI does not run all the grammar schools in Birmingham, one consequence was that there was quite a large displacement of pupil premium children from some of the other grammar schools, and one in particular, into the King Edward VI grammar school. When we think about the impact on the number of pupil premium children who are managing to go to a Birmingham grammar school, the positive impact was not quite as big.
We also need to put it into context. Let me say a couple of things about Birmingham. The first is that 20% sounds fantastic for the number of pupil premium children, but this is a city where 50% of students are eligible for the pupil premium. In that context, it does not look quite so great. Moreover, I would argue that there are demographic reasons why there are relatively high numbers of children eligible for free school meals who are capable of passing the 11-plus in Birmingham. These are largely related to the ethnic composition of Birmingham.
If we look at the Birmingham experience and say, “We could do that across the country. We could have dedicated places for free school meal children,” some areas would find them very difficult to fill. By that I mean they would have to dip so far down the ability distribution that they would be admitting pupil premium children who did not look much like the other children in the grammar school as a consequence. They would not be high-achieving at the age of 11. We might be all right about that. We might think it does not look like a selective school any more. We might also worry about the consequences for those free school meal children, who would ultimately be ability-tracked into a lower stream comprised entirely of pupil premium children within a grammar school. It is a difficult proposition in areas where there are few high-attaining pupil premium children.
Q20 Lucy Frazer: Is there any answer? Let’s assume that there would be a grammar school system but that some of the measures in the Green Paper came into force. Let’s assume it is not a zero-sum game. What is the best measure? We have talked about mandating a certain percentage of pupil premium, or is it a grammar school joining a multi-academy trust and sharing teaching and excellence? Is it something else? What is the one measure that you would identify that would improve education for everybody in an area?
Professor Jesson: The important thing here is a point I made earlier about the contact about a given grammar school, because we are talking about the re-emergence of grammar schools, with the community that surrounds it. The significant thing about what is happening in Birmingham is that, although in one respect there may have been a re-sorting of those pupils, the fact is that, where schools engage with their communities, they bring to light the possibility of a grammar school education for children, for parents and even, I may say, for primary school teachers, for whom it was never a conceivable option. I believe that is one of the great successes of the King Edward VI school, which is an independent school but which I believe is in the process of beginning to form a multi-academy trust with a number of other schools within Birmingham.
To my way of thinking, the improvement here is to ensure that the widest possible range of young people are given the opportunity to understand what the benefits are of a grammar school, and I believe there are certainly some, and secondly that it could be right for them. I believe that is a very important issue that we should continue to press, and I am pleased—as I have said already—that the Green Paper does make that point. I would like it to be made more firmly and to be implementable if this does come about.
Chair: We will put that to the Minister when he is next before us, which is in about an hour’s time.
Professor Vignoles: I will come back on one point there. I would not disagree with that but there is an assumption in the Green Paper—and indeed in previous policies by previous Governments—essentially that the grammar school is a superior quality school and that they are doing things that provide those children with a better education; likewise with the independent sector, that it is a superior quality education. Indeed, I am sure in many cases it is. However, across the piece, it is a big assumption that, were you to take those very same teachers and the very same methodology and apply it to a much more varied intake school, you would necessarily produce gains. It is rather dangerous to assume that, not least of which it might also upset teachers in the other types of schools. That is something that we need to think about before we assume that simply requiring grammars, independent schools or indeed universities to assist schools in this way would necessarily bring about an improvement gain.
The other thing I would say is that it is not just about social mobility. Rebecca alluded to issues around ethnicity. Not only are poor children underrepresented in grammar schools but white British children are underrepresented in grammar schools. The nature of the grammar excludes. We need to be very clear that that is a worry.
Q21 Lucy Frazer: Luke and Rebecca, I would like a positive one thing.
Luke Sibieta: I am very happy to give a positive thing. You talked about what the one thing is we need to do given the potential disadvantage mentioned in the Green Paper, particularly around admissions. We should not be taking as given the idea that free school meal children are always going to perform less well at age 11 than children from more advantaged families. Essentially, the focus should be on ensuring that the attainment gap between children from less well-off and more well-off families is closed earlier. The focus should be on improving primary school quality to make sure that does not happen by age 11. If we look at areas of the country that have achieved that, they are areas like inner London to some extent, Birmingham and Manchester. The Government’s White Paper or Green Paper earlier this year—I cannot remember which—talked about the patchy geographical variation in quality across the country. We should not lose sight of that.
Q22 Lucy Frazer: What about having a selective system much earlier? We are talking about 11, 14 or 16, and you are talking about narrowing the gap earlier. Is it possible to narrow the gap for those disadvantaged children for whom the gap gets so big they never make it at 11, possibly in a comprehensive system or—I am throwing out to you—in a selective system?
Professor Vignoles: You have a very large socio-economic gap in measured achievement. That is probably a better description than ability necessarily. You have that large gap by age three and age five. Anything that embeds those gaps early and cements them in different schools is going to create an issue with—let’s call it—educational mobility. That does not apply necessarily, for example, to streaming within schools. There is something very specific about placing children in separate schools from a very early age because the movement across those two schooling systems is inevitably quite limited.
Q23 Lucy Frazer: Did you have one thing you want to say, Dr Allen?
Dr Allen: One positive? Because I am so concerned about resource sorting and how difficult it is for secondary moderns to operate, I would mandate that every grammar school in the country is forced to take into a multi-academy trust three or so secondary moderns, the correct number to be representative of the population. The advantage of that is they will have to bear the consequences of any of the negative impacts of resource sorting around finance and around teachers. Then we set up an accountability system that punishes them for standards in those secondary moderns that they are running.
Q24 Lucy Frazer: Do you think people who have special educational needs and people who have additional language needs would lose out from a selective system?
Professor Vignoles: Very broadly, one thing we know about special educational needs and additional needs is that that changes over the child’s schooling career. Again, you have issues around later development for some children. Even if you did nothing, one of the downsides of having the decision point quite so early is that you would identify children who look like they are lower-achieving but then who later have high achievement.
Dr Allen: In the case of children who have English as a second language, they are well represented in grammar schools. There are no issues there. For special needs children, there are around 4% in grammar schools compared to 13% nationally. Is that a problem? They are well-represented given their measured attainment at the age of 11. This reveals the fundamental problem with the idea of academic selection. Why on earth would we allow lower qualifying scores for children who have some sort of special educational needs to give them priority access to a grammar school, when the whole idea of the grammar school is that we only let people who measurably have been an academic success into the school? To me it does not make sense. If we do it for them, why not do it for anybody else who so far in their career has had limiting factors that have held them back in what they have been able to achieve so far?
Q25 Lilian Greenwood: It is probably too big a question but if I could have your, kind of, one thing. The panel as a whole seems to be saying that grammar schools are not the best way to improve social mobility for young people from low-income backgrounds. What would you recommend that the Government should be doing instead if they do want to tackle social mobility?
Professor Vignoles: In terms of things that we know are effective, we do need to improve—continue to improve, I should say, because there has been improvement—primary school education for the most disadvantaged students. We know that teachers and teaching quality is the key input there. Therefore, we need to continue to be focused on ensuring a strong supply of good-quality teachers. There are a lot of policy developments that you could do around that.
Then there is increasing evidence that, even when you have students in secondary schools from disadvantaged backgrounds that are lower achieving, at that point there are interventions that might not be quite as focused on academic achievement but, none the less, could help them with other aspects of their non-cognitive skill, if you want to call it that. In other words, that might help them in the labour market even if they do not necessarily encourage them to get higher levels of academic qualification.
Luke Sibieta: To add to that, it is important to look before school as well. If you look at, say, Norway in the 1970s, they introduced high-quality universal preschool education. If you track those children into the labour market, their employment and earnings, it was children far down the income distribution that received the very biggest benefits as a result of universal high-quality preschool education.
Q26 Lucy Allan: On the 11-plus as a measure of assessment, we all would agree that it inherently favours children who have been coached or have had early advantage and have reached a certain standard in English and maths. That is what is being tested. It clearly does not measure future potential. Is there a way of devising a test that will measure future potential irrespective of where a child has reached at that age? David, would you like to kick off with that one?
Professor Jesson: This is one of the key reasons why the grammar school movement, following the 1944 Education Act, finally collapsed. The evidence that was supposed to lead to high-quality education for some and less good quality for others was shown to be completely meaningless—first of all, if I may say, in the County of Leicestershire, which was never a bastion of socialist rhetoric. What actually happened was parents simply rebelled against the idea that you should select children at the age of 11 in order to go to grammar schools, when it was clearly using instruments that were flawed. It is not one instrument. I know that the National Foundation for Educational Research produces some kind of test, but there are a whole range of tests that can be used. I think that each of them suffer the same disadvantage, that they do not do what you are asking them to do, and that is give good, predictive accounts of what it is that a given child at a given age should be able to do.
What they did in Leicestershire was to ensure that every school became open to everybody. They transferred into a high school at the age of nine. They stayed there until 14. If they wanted to go to a grammar school, they transferred to the grammar school at 14. It was a completely open parental, children and teacher choice, and the significant part of that was the child’s own choice. I think that the evidence in the background about effective 11-plus tests is pretty poor.
Dr Allen: In thinking about whether we can design smarter tests, we have to think about what it is we are purporting to measure. You talk about this desire to measure future likely academic achievement. That very idea presupposes that ability is something that is fixed. I think that everything we are starting to learn about the brain and about cognitive neuroscience suggests that is not right. I think only this week a really lovely piece of research came out from the University College London Institute of Cognitive Neuroscience. They said that older teenagers and young adults were actually fantastically good at being trained to improve their fundamental maths skills and their reasoning abilities. They were better at improving than younger children, which provides evidence that the brain is malleable; it is even malleable for us as adults.
Because of that, the notion that we can measure likely future academic achievement is not one that is, in my opinion, really meaningful. I think all we can do is measure current academic achievement. I could go on and talk about some of the differences between the two different 11-pluses that are operating and—
Q27 Lucy Allan: I wonder if I could come back on what you said, which was very interesting. There are some top-achieving schools, both state and independent, that use measures other than 11-plus to select for future potential. Are you saying that those schools are probably not delivering on what they say they are?
Dr Allen: It is undoubtedly true that past academic achievement is a good predictor of future academic achievement.
Lucy Allan: I am talking about things like computer tests and looking at their school reports, doing non-verbal reasoning—anything other than maths and English 11-plus.
Dr Allen: Sure, but even so-called IQ tests are still measuring a set of competencies and skills that a child may or may not have, and it is that that is the reason why we can coach for these tests. We can intervene to do something to make people better at these tests, and that is why the vast tutoring industry around the 11-plus exists. It is not that parents are being irrational and that there is nothing we can do to prepare; it is that there is a lot we can do to prepare for the 11-plus and it is completely rational that you invest a year or more of money on weekly tutoring sessions. That is the correct thing to do.
There is contrast within the system. There is the historic—what you talk about—non-verbal/verbal numerical reasoning tests that you may be familiar with. But 40% of the 11-plus is now taken with a different test, a test known as the CEM tests. These tests were devised because there was concern that the non-verbal reasoning tests were very poorly aligned with what pupils are taught in primary schools and therefore it was necessarily true that you needed special preparation for the 11-plus. They say, “Let’s design a set of questions that are better aligned to the key stage 2 curricula. Moreover, let’s make the question types and the structure of the test quite unpredictable, to make it really, really hard to tutor.”
Now, they are striving to make the selection process fair, but there is no evidence at all that students are not being advantaged by tutoring in these tests. In fact, in 2014 the huge area of Buckinghamshire switched to using these tests. When they switched, an area that already had very, very low free school meal proportions in grammar schools, the proportion of free school meal children fell. It fell in 2014 and 2015. There is a logical reason why. If you have a very narrowly defined predictable test, the amount of money you need to invest in private tutoring is relatively limited. If somebody says to you, “You are going to be tested on something, something in the primary curriculum, but we are not going to tell you why or how,” as a parent my response would be to put my child through tutoring for years to ensure that all bases are covered. This is the difficulty with the idea of designing a tutor-proof test.
Q28 Ian Mearns: Buckinghamshire had a massive increase in the number of private coaching institutions opening up as well.
Dr Allen: Yes, some allege that it has.
Q29 Lucy Allan: Luke, would you say it is not possible to devise any measure of future ability in making these decisions?
Luke Sibieta: I would agree with everything that has been said, and I add to that that I think it depends on what you are using the test for. If a school is using a test to inform their judgment with regard to a pupil—what stream to put them in, what their likely level of achievement is going to be, how to target resources at particular pupils—then that is fine because if you realise you made mistakes you can correct them. You can see how a pupil is doing. You can say, “Actually, they should probably be in the top set for maths. They are doing really, really well.” If you use it to put them into different schools, it inevitably means that it is that much more costly to correct mistakes. If you are using it and you can then adapt as you go, then it is fine. If you are using it for a one-shot decision, I find hard the idea that you would get a tutor-proof test.
Q30 Lucy Allan: Anna, do you think if teachers were being asked for their input from the primary sector, so you can look at a class and say which children in your class are likely to be the best performers going forward, is that not a potential measure?
Professor Vignoles: Two points. I am not a psychometrician, so I don’t know if it is possible to invent something that might be tutor-proof, but all the evidence suggests that it would be extremely difficult and, further, that there is a huge incentive for parents to find ways of preparing their children for it.
Could you ask teachers? In fact, if we go back far enough, significant numbers of grammar schools used to use parent and teacher suggestions as to which children go to a grammar. Anything that involves teacher choice, own choice or parental choice is likely to exacerbate, not improve, the social mix. Own choice and parental choice is perhaps obvious because the more parents are involved in it, the more likely it is that their social background kicks in. Teacher choice is slightly subtler. The pressure brought to bear on teachers and head teachers from different types of parents would be quite striking, I think, so I would suggest that is not going to be a likely solution.
Luke’s basic point is correct. If you are using tests on a regular basis to inform about the progress of a child, that is absolutely fine. A one-off decision of where to place them, in terms of their schooling, is a lot more problematic, particularly when there is a lot of incentive for some children to be better prepared.
Chair: I am going to have to pick up the pace a bit because we have more questions, but Catherine has another.
Q31 Catherine McKinnell: One supplementary, particularly on what you said, Rebecca, which is quite alarming: if we do see an expansion of grammars, which seems to be the Government’s policy, and we do see attempts to make the 11-plus test tutor-proof, in fact the result could potentially be an even greater selection in terms of social mobility and ability to pay for that preparation than we currently have under the current system. It could make the education system even more selective, not just on an expanded basis but on an individual school-by-school basis as well.
Dr Allen: Ironically, it could improve attainment at the age of 11. We certainly see that, in areas that have grammar schools today, primary schools score very well on the SATs test compared to the place we expect them to be, given their demographics. We put that down to the private tutoring that goes on in those areas, so we see rising standards in primary schools but far sharper inequalities in achievement.
Chair: I think David has a quick contribution—a small one, though.
Professor Jesson: Just a point: one thing about the Green Paper, it suggests that there is nothing set in stone about selection at 11. I think this is a crucially important element. New systems that may arise do need to take very carefully into account what it is that other members of the team have said today about the difficulty of ensuring that prediction, on the basis of narrow tests, is the basis of recruiting children into grammar schools. The whole idea of expanding the age at which children may select or be selected or be encouraged to ask for accommodation in a selective school is important.
Chair: That is basically Lilian’s question, so let’s see how Lilian develops it.
Q32 Lilian Greenwood: One of the longstanding concerns about the old selective system was that, essentially, children’s whole futures were decided on the basis of how they performed on one particular day, and that that set their future. There is being mooted the idea that perhaps selection could happen at a later stage or that there would be opportunities to take some sort of test at other points in their school career, so not just at 11 but perhaps at 14 and 16. If children had the opportunity to move at a later date, having been allocated to schools at 11, so perhaps at 14 or 16, what does the research say the impact would be on them, of moving children between schools after they have settled into secondary? Is there any evidence or any research that tells us how that would operate?
Professor Jesson: There is historical evidence, and that was the beginning of the whole enterprise of moving away from selection, as I mentioned, in what was called the Leicestershire plan way back in the 1960s and 1970s. One of the interesting things that has already been taken into account by a couple of Governments recently has been the setting up of the university technical colleges, which offer an opportunity for people not to have necessarily a grammar school academic education.
I would suggest that one of the ways that we move ahead is to broaden the context into which young people know that they are moving. If we have a single academic line, which is the top gun, then I am afraid that the whole process will not be advantaged by sticking at the age of 11. There are examples of post-11 selection, or even encouragement and selection, and I would suggest that, again, the Green Paper makes a very good point about ensuring that it is not all back to the age of 11.
Dr Allen: To talk generally about the impact of delaying until the age of 14, we delay the damage caused by selection and in that sense it is a good idea. There will be less misallocation of students. Those who go to secondary moderns eventually at the age of 14 will spend less time in them. This is good. We will not see more free school meal children passing a 14-plus than an 11-plus. We will probably see fewer because, in general, free school meal children fall further behind the longer they are in school, so it will not fix that.
You talked about two different options that are radically different. One is having a grammar school that starts taking pupils in at the age of something like 13 or 14 and the other is that we have secondary intakes. To talk about the idea of selection at the age of 14, the evidence from things like the Kent grammar school called Cranbrook, which was selecting in and starting at 13 and is now changing to 11 because it did not work, and the evidence from UTCs that have struggled to recruit—it has been bad for the comprehensives around them who have lost students and it has had a financial impact—is do not introduce any sort of school that has a later date of selection into a system that is largely an 11 to 16 or 11 to 18 system.
In terms of the second intake argument, we already do it in part. Age 16, 20% of those in the grammar school leave; 25% of those in a grammar sixth form are new. So, age 16 selection to grammar schools is already there; it is already happening. 2% a year in the grammar school sector as a whole join in year 8 or in year 9. It is much, much higher in many grammar schools. Some of the local authorities organise co-ordinated 12-plus or 13-plus entrance exams. I suppose my concern about these entrance exams is that, essentially, they are ad hoc entry points. What we know is that whenever we have ad hoc entry points, some families are more motivated than others to enter at that time.
Chair: We have an entry point, which is to get the next panel going, so let’s have swift straight-to-the-point answers. Lilian, do you have anything else to ask?
Q33 Lilian Greenwood: Picking up on what you were saying in terms of the ad hoc entry points, do they improve the social mobility or do they make it even worse because then people are coaching to have that second chance?
Dr Allen: We deal with some misallocation: that some people who should have passed the 11-plus—but didn’t—manage to pass the 12-plus or 13-plus. In all likelihood, they are not going to be students from disadvantaged families.
Luke Sibieta: To add to Rebecca’s point, it is important to remember that we do select by ability and aptitude on a very large scale in the UK. It is at age 18 when we go to university. We do that because we think there are large gains from specialisation at age 18. When we had selection by ability across the UK in the 1950s and 1960s, we did it for that reason because there were gains from specialisation and from, say, having an élite with a high level of skills and a large number of individuals with low levels of skills who then moved into unskilled jobs. Those unskilled jobs do not exist any more. 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.
Q34 Lucy Allan: Is the change at the 16 entry a benefit or a disadvantage?
Chair: “Yes” or “No” will do.
Lucy Allan: You say 20% and 25%. Is that good or bad?
Dr Allen: To my mind it is good.
Professor Vignoles: It is no different from the movement that you get at 16 between schools, FE colleges and sixth-form colleges. It is just a point at which, as Luke says, people start to specialise and choose as young individuals.
Q35 William Wragg: This is a question about the achievement of those going to grammar schools, which I suppose is the whole point of our discussion this morning. It is a straightforward question: do pupils tend to achieve better results at grammar schools than they would have achieved at non-selective schools? Would each of you briefly answer that question?
Professor Jesson: I made this point a little earlier about those youngsters who overlap in terms of their attainment between the majority of those at comprehensive schools and selective schools. There is no real evidence of an enhancement in the outcomes of the youngsters who are in grammar schools, when we take account of the entry attainment with which they enter the school. To that extent, I think the myth that because a grammar school gets 100% of its children getting five A*-C and the secondary school down the road gets 70%, that does not tell you that the grammar school is better. It simply tells you that most of the youngsters that go to the grammar school are at the higher end of the attainment profile and those in the other school are not. When you compare like with like—which is a critical factor in looking at how schools work with their own youngsters—that is when you are quite clear that the advantage that is claimed for grammar schools does not seem to exist.
Chair: Does anyone have anything to add to that?
Luke Sibieta: Yes. I would say the evidence is clear that children on the cusp of just either side of getting into a grammar school, if they just get in they are much more likely to get academic qualifications than if they did not get in. That is because those who did not get in were never going to be offered them in the first place.
Q36 William Wragg: If I throw into the mix Progress 8—I know some of the panel have mixed views on Progress 8—does the Progress 8 data not demonstrate the contrary, perhaps, of what was just said then? I think Rebecca would be keen to answer this question.
Dr Allen: I should preface it by saying I am a huge fan of Progress 8. I think it is a great accountability measure, but when we look at the Progress 8 figure for grammar schools and it says that they have a Progress 8 figure of about a third of a grade, there is one particular problem with that. There are many children who are in grammar schools who have quite modest key stage 2 scores but they clearly passed the 11-plus, so we suspect they are much more able than their key stage 2 scores suggest.
Then we observe these children with low key stage 2 scores in grammar schools making enormous progress in the grammar schools. It is most likely the case that the grammar schools are being wrongly rewarded for the progress that they made before they reached the age of 11, because we are understating their attainment. The reverse is true for secondary schools. I would say we are overstating how good grammar schools are by saying the Progress 8 figure is a third of a grade.
Q37 William Wragg: My colleague to my left must forgive me because my supplementary question then touched on the next substantial question, which will deal with Progress 8 entirely. If I could close by asking: assuming the same prior attainment, do pupils from lower socio-economic backgrounds achieve better in grammar schools than they would do in non-selective schools? Admittedly, we have heard earlier it is a low percentage, but do they achieve better than they would do in non-selective schools?
Professor Jesson: When we look at the question of Progress 8, there are obviously newnesses about that, which I think are still being understood. The fact about the old measure, which no doubt is flawed—the percentage of youngsters who achieve this, that or the other—that still has a purchase with many parents, with the public, and it is very important that, when we look at what it is that schools are achieving and what their individual children are achieving, we come back and recognise that a measure that is important to the population and to parents is the one that is going to be most recognisable. It is in that context that, when we look at the same background of youngsters in grammar schools and in non-grammar schools, we find very little difference between the performance of those children with that measure.
William Wragg: Thank you. Briefly from Rebecca; I think she wants to answer on that point.
Chair: A small answer because we have run out of time.
Dr Allen: I would say the answer is: we are not really sure, and one reason is we are always using administrative data. We are grouping this huge group of children who are eligible for the pupil premium together. We are saying, “You are all alike.” 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.
Chair: Thank you. Suella, you have a quick question.
Q38 Suella Fernandes: It is to pick up on a point that Dr Allen made, that many of you have made, that there is damage caused by selection. I want to challenge that, because there is a lot of selection that is currently part of our system, whether it is through postcode, whether it is through particular aptitudes in music and technology, and the deeper point that “selection” and “competition” are not dirty words. They are healthy incentives for better performance in the academic context and failing is not to be stigmatised. Children who do fail these exams or don’t pass them are not going to be on a scrapheap. They are going to be having the option of attending good, non-selective schools. I want to see what the panel’s thoughts are on changing this cultural attitude to selection and competition and, actually, part of the need for improving the way we will improve standards is to bring back and re-inject a bit more competition and testing.
Professor Vignoles: I think you are right on many levels. First of all, there is a lot of selection in the system anyway, by postcode and by parental income in terms of the houses that you buy near to good schools. There is also nothing wrong with, and there is a lot that could be said is very good about, high academic achievement, and there is certainly nothing wrong with competition in that respect. I don’t think the issue is: should we be striving for better standards, higher levels of achievement? Higher levels of achievement for all children, absolutely. The question is whether or not having a particular point at age 11, which has consequences for those who do not succeed at that particular point, is the best way to bring that about.
You could indeed do what you are suggesting, which is: have a cultural shift that emphasises high achievement in every school. The Government has done a very good job in terms of introducing Progress 8 because, essentially, that is what it is supposed to do. It is a recognition that in the school system we need to encourage high achievement of all students, including those at the very top. That seems a very sensible thing to do to me.
Q39 Ian Mearns: Rebecca, earlier you seemed very enthusiastic about Progress 8. I am wondering whether the rest of the panel concurs with Rebecca’s enthusiasm.
Chair: “Yes” or “No” will do. We are getting really short on time here.
Professor Vignoles: Very quickly, in a year’s time you need to look at the consequences of Progress 8. There are some unfortunate consequences, including subject choice particularly around science. As with any measure that you are using for accountability, you need to keep on top of the way that schools respond to it.
Chair: That is the longest “Yes” or “No” I have heard.
Professor Jesson: In principle, yes.
Chair: Right, good. Luke?
Luke Sibieta: Yes.
Chair: Spot on; you get a gold star for that. Last question, Ian.
Q40 Ian Mearns: Do you think that the expansion of selective schooling could have any significant effects on neighbouring schools? We have talked about selective schools but it seems to me, in an area where you have selective schools, the number of youngsters who are selected to go to that school—that is, a selective school—by its very nature the other schools around are also selective, because they include the children who have not been selected for the first one. Are there any potential negative impacts on the schools in an area where we are going to expand the selective school system?
Professor Vignoles: The two big ones are peer effects, because you are taking out students and parents from the neighbouring schools that have high achievement. Then there are teacher selection effects because you may be drawing teachers out of those schools into the grammars. Those would be the two main ways in which you could have a negative consequence.
Professor Jesson: There is one other effect, which I think is important. That is, that the gap between pupils in selective areas, between those who are not disadvantaged and those who are, is greater in areas of selective schools than it is where there are no selective schools. To that extent, the answer to your question is, yes, there is an effect and it is palpable.
Chair: Another quick answer from Luke; and this is the last answer we are getting.
Luke Sibieta: There is a lot of unpredictability, given the policy context at the moment. We used to have a system where whole local authorities would have selection by ability. We are not going to have that system if it happens. It will be individual academies deciding whether to select by ability or not. I don’t think we know what would happen in such a system. Indeed, the individual selective schools across England are probably the most socially selective of them all.
Chair: I want to thank you all very much. You have certainly given some interesting answers. We have taken it all in and we will be turning some of those answers that you gave us into questions for the next panel. Thank you all very much indeed for what has been a very enlightening hour and a quarter.
Examination of Witnesses
Mr Nick Gibb MP, Dr Tim Leunig and Donna Ward gave evidence.
Q41 Chair: Welcome to our inquiry into the evidence behind selective education. It is quite clear what we are getting at here: is the idea that we have more grammar schools going to help or hinder our progress, particularly in terms of social mobility? Also this question, which I threw in earlier, about economic productivity, which is clearly relevant as well. Donna, would you like to say who you are and what you represent? Nick and then Tim can follow.
Donna Ward: Hi, I am Donna Ward, Chief Analyst in the Department for Education. I do a job share/split with Tim Leunig.
Chair: Great. Nick, I think we know who you are but just for the viewers.
Mr Gibb: For the record, I am the Minister of State for School Standards in the Department for Education.
Dr Leunig: I am Donna’s job share partner as Chief Analyst. However, this area falls under Donna and I am here with my other hat as Chief Scientific Adviser.
Q42 Chair: That is absolutely excellent. We have a lot of multi-hatted people from the Department and you are another on that list, so thank you very much for coming along today. Tim, Donna, if a fairy godmother turned up and said, “I have an opportunity here to change the education system and you are either going to be able to change structure or teaching in the classroom,” which would it be, Tim?
Dr Leunig: Everybody wants to change teaching in the classroom as their first thing. The aim of changing structures is to enhance the quality of teaching in the classroom. Nobody thinks that structures per se do anything other than facilitate good teachers teaching brilliantly in schools.
Q43 Chair: Donna, are you in agreement with that?
Donna Ward: Yes. It is about teaching, but I think if you mean by structures as well offering an academically stretching education, then that would also work.
Q44 Chair: Now, Nick, you have said, “The Department has reviewed a wide range of external research and evidence”. The question is: which specific research forms your main base of evidence?
Mr Gibb: There is a lot. The purpose of this Green Paper is to increase the number of good schools in our system. Over the last six years, we have 1.4 million more children in “good” and “outstanding” schools than in 2010, but there are still 1.25 million children in schools that are not “good” or “outstanding”. This is part of the plan to get more good schools throughout the country, and grammar schools are good schools; 99% of them are rated “good” or “outstanding”; 82% are “outstanding” while 20% of non-selective schools are rated “outstanding”. There is your first piece of evidence. They are very popular with parents; they are 50% more popular in terms of first preference choices than other schools. Children travel three times as far to get to a grammar school than they do to any other school.
The Progress 8 measure in grammar schools is 0.33 compared to the national average of zero and for non-selective schools -0.03. They are closing the attainment gap between those from less privileged backgrounds compared with their wealthier peers. Our concern is that there are 65 local authority districts—one in five—where less than half of secondary school children have access to a “good” or “outstanding” school within 5 kilometres, so this is what we are trying to do.
If you take, for example, the Sutton Trust report of 2008—we had a lot of discussion in the earlier session about the effect of grammar schools on the other schools in those areas—the Sutton Trust report says that there is no adverse effect on the non-grammar schools in those areas, although I admit that the evidence is mixed.
The other interesting evidence is that level 5 pupils, those pupils who leave primary school, those high-ability children with a level 5, 78% of those children will go on to achieve the EBacc—the core academic GCSEs—when they are 16, compared to 55% of those same, similar high-ability children leaving primary school if they go to a non-selective school, so there is a starter for 10 for some of the evidence we have been looking at.
Of course we are consulting. That is the purpose of this consultation process—to gather more evidence over the three months of the consultation process.
Q45 Chair: Is it your contention, Nick, that having more grammar schools is really about increasing the capacity of good schools?
Mr Gibb: That is one of the contentions, yes. We need more good school places in our system and grammar schools are good schools. Therefore, we want them to expand. We want them to set up new grammar schools in grammar school areas, and in other areas that are non-selective at the moment, also, where there is a wish in the local community from parents to have a selective option, we want them to have that option. We want to remove the ban on creating more good schools, more good grammar schools.
Q46 Chair: We have heard a bit of evidence about the impact that grammar schools have on neighbouring schools. Have you had any thought on that matter?
Mr Gibb: Yes, we are interested in that evidence. As I said, the evidence is mixed, but the purpose of the consultation document is to address that problem. Those people that are shouting loudest about these problems are doing the least to address it. We want to address that issue by saying, “If you want to expand a grammar school or you want to create a new grammar school, then one of the conditions for that expansion or approval of a new grammar school 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.” We want 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.
It worries me, Chairman, that, if you look at the list of local authorities around the country—and not looking at any selective areas—taking the non-selective areas, Blackpool, the EBacc figure in Blackpool, 9.2% of pupils achieving the EBacc combination of GCSEs compared to the national average of 24%. Knowsley 10.4%, Middlesbrough 10.4%, Isle of Wight 13.3%, Oldham 15.7%. We need those figures to be more comparable to Southwark, another non-selective area, 35.6%, York 35%, a selective area, Sutton, 45.8%. These are the kinds of figures that we need.
I know there are all kinds of reasons why some of those areas will have higher EBaccs than other areas—children crossing borders and so on—but it is unacceptable that if you live in Middlesbrough, only 10.4% of the pupils in that borough will achieve the core academic combination that makes up the EBacc.
Q47 Chair: Have you been thinking about the performance of primary schools in those areas that you have just been listing as not doing too well in secondary schools?
Mr Gibb: All the reforms over the last six years are about raising academic standards in the primary sector and in the secondary sector. We have introduced the phonic check at the age of six. We now have 147,000 more six-year-olds reading more effectively this year than they would have done had we not introduced phonics and had a real focus on phonics. We have reformed and rewritten the primary curriculum. If you look at the Ofsted annual report for last year, Sir Michael Wilshaw is very praising about the effect that the phonics policy has had on improving reading in our primary schools. There is more to do, absolutely. We have, of course, reformed the curriculum in secondary schools. We have improved behaviour in secondary schools. The policy set out in this Green Paper is one part of our reform programme. It is building on that reform programme over the last six years to increase diversity, to increase choice, and to increase the number of good school places throughout the country.
Q48 Chair: Tim, Donna, does the Department intend to undertake further research into the potential effects of widespread academic selection? That is something that we touched upon in the last panel and it is quite an interesting issue, is it not?
Donna Ward: Yes. As the Minister said, we have looked at the existing evidence that is out there to get as far as we have with the Green Paper. Of course, we will be doing evaluations of any policies that come forward, specifically looking at how successful new or expanding selective schools are in reaching a greater diversity of pupils, what happens to the composition of attainment and the effects on schools in the area. I think all of our data has been used by the academics that we heard from earlier. We have also done some of our own analysis, which has been referred to by the Minister, to get us this far and then, of course, we will be doing evaluation of the policy as we go on.
Q49 Chair: Tim, you are a scientific adviser and you are used to controlled situations and being able to extrapolate data from that, but this is not quite so controlled, is it? And it is going to be very hard to take into account other factors, such as the areas that Nick was talking about: is it actually the primary school problem or is it something else? How are you going to try and isolate the impact of selection on school systems, given that we have a range of different systems to look at?
Dr Leunig: Useful additional information that we could collect in future, which we do not collect now is: who takes the 11-plus and what score do they get? It is easy for me as an education analyst to say, “We would love this data and therefore the Government should mandate it.” I am not saying that, because that is of course an additional burden on schools, so there is both a benefit of collecting it and a cost to the system for doing so. In terms of working out whether grammar schools are effective, what you really want to know is not the key stage 2 results but who is the child who just passes the 11-plus, who is the child who just fails the 11-plus and what are their trajectories? In terms of building a stronger evidence base, that is the information that would make this closer to an answerable question with a degree of certainty.
Q50 Chair: That is only about the period around the test. It is not about the run-up to the test, and that is very distinct, as I think Lucy was referring to earlier. There are issues at the beginning of school time and certainly at the early stages of primary school. My question is: how are you going to assess those impacts against the ones you have just talked about?
Dr Leunig: There are some things that I think are next to impossible for us to assess. It is really unlikely that we are ever going to get data on how many private lessons a child has had, if you think about the data collection, making it a legal obligation for parents to declare that.
Q51 Chair: You are still talking about the test, though. You are not talking about the impact of primary school performance as a basis for whatever school a child will end up in.
Dr Leunig: That one is relatively easy because we know quite a lot about primary schools already. We know that in places where children from poor backgrounds achieve well at primary school they are more likely to pass the 11-plus. It is not a slam dunk that they will, but they are more likely to. I refer you to Luke’s answer earlier. What is the best thing we could do to improve the system? It is to improve the quality of primary schools, particularly for poor children. We have a range of policies to address that, some of which, such as phonics, Nick has already referred to. Those sorts of things we already know about primary schools, so of all the things I would be worried about, that would be low on my list.
Q52 Ian Mearns: The 1.4 million more pupils in “good” and “outstanding” schools—the Secretary of State used that statistic a number of weeks ago when she was in front of us. Do you have a breakdown, Minister: of that 1.4 million more pupils who are in “good” or “outstanding” schools, what proportion of them are in academies, what proportion are in free schools, what proportion are in maintained schools, voluntary aided schools, voluntary controlled schools, or university technology colleges? What are the proportions? Is this all down to academies or is something broader happening here?
Mr Gibb: We can send you that breakdown, but what I am saying is that the reforms over the last six years in primary schools, to the curriculum, the improvements in behaviour that we have instituted over the last six years, and the academies programme and the free school programme, have all contributed to the fact that there are 1.4 million more pupils in “good” and “outstanding” schools today than there were in 2010. Now we are not happy with that. We are not resting on that laurel, because we want to move on to deal with the 1.25 million pupils who are not in those good schools.
The flexibility of allowing new selective schools to be established is part and parcel of that broader programme: to get more good schools in more parts of the country that currently do not have them.
Q53 Ian Mearns: Can you tell me, how do Knowsley, Blackpool, Oldham and Middlesbrough compare in terms of the number of youngsters in receipt of free school meals as opposed to the national average?
Mr Gibb: Again, I can give you those figures but, intuitively, you would assume that they were higher in those particular areas.
Ian Mearns: I am pretty certain they are higher.
Mr Gibb: But that should not correlate. What we are trying to end is this correlation—this link—between disadvantaged backgrounds and academic performance. We have got to break that link, and none of us should be happy in excusing poor EBacc performance in local authority areas with high levels of deprivation. We want to break that link. That is what is driving our reforms; and you, Mr Mearns, I would hope would be behind us in that objective.
Q54 Ian Mearns: Absolutely, Minister, but there are some authorities—because there are good authorities and bad authorities in terms of the overall make-up of the schools—with equally high numbers of pupils on free school meals as the ones that I mentioned, that are outperforming the national average.
Mr Gibb: Yes. Take Newham for example. Newham, which is an area of very high deprivation, their EBacc figure is 30.4% on this schedule. If you take Newham’s phonics results, I think they are one of the highest-performing local authorities in terms of their phonics, 87% of pupils in that borough achieving the expected standard in the phonics check compared to 81% nationally.
Q55 Ian Mearns: Is the comparison between the four authorities that you have mentioned with Southwark, for instance, fair, given the resources of the London Challenge and so on?
Mr Gibb: What I am demonstrating is that we have to address these areas that are not achieving a high enough academic standard. That is the purpose of listing those results. What I am expecting will happen is that to establish a selective school, if that is what the community wants in those areas, will provide an example of what can be achieved for high academically able children, which is to shake up the system in those areas. I hope the result will be that we will get higher EBacc figures in those authorities as a consequence of a grammar school being established in that area.
I want to give you another figure. If you take the Oxbridge entries between 2012 and 2014—take Oxford first of all—one in five of the state school students that went to Oxford between those years came from a grammar school. If you look at Cambridge in 2015, of the state school pupils who went to Cambridge in 2015, 682 came from comprehensives and 589 came from grammar schools. So the 163 grammar schools are sending almost as many pupils to Cambridge as all the other comprehensives put together.
Q56 Lucy Frazer: Minister, you said “if the community wants it”, and I would like to ask you about that. Take for example the area that I represent just north of Cambridge, so in the south of my constituency, we have a lot of village colleges, which are community-led and play a big part in the community. They go up to 16 and then we have some excellent sixth-form provision. Those schools are “good” and “outstanding”, so what if the community wants it? What does that mean? Who is the “community” and how does that grammar school then come into being?
Mr Gibb: These are the issues that we are consulting on in the Green Paper. What this means, in terms of the free school application process, is that every application for a new grammar school will have to be approved by the Secretary of State. The issues she will be looking at will be: what parental demand can you demonstrate as part of your proposal? You will also have to demonstrate the impact that your school will have on other schools in the locality. These are the sorts of issues that we are consulting on in the Green Paper to apply to the new grammar school process.
Q57 Chair: What sort of figure do you have in mind for the number of new grammar schools that might emerge from this policy, let’s say, by 2020?
Mr Gibb: We don’t have any figures in mind. This is about removing the ban on creating more selective schools and academies. That is what it is about, and it will be driven by demand from local communities.
Q58 Chair: If the local communities don’t go down the route and there are only perhaps one or two extra grammar schools, will you consider that a success or a failure?
Mr Gibb: I want to see more choice for parents and more diversity in the system. That is the drive behind this policy. We are not going to set targets for what is deemed to be good and deemed to be bad.
Q59 Chair: I am not asking for a target. I am asking for an assessment of the number of schools you think might be new grammar schools by 2020.
Mr Gibb: We do not have such an assessment. This is about allowing flexibility in the system. We are consulting on that. It will ultimately be driven by local demand and what parents want.
Q60 William Wragg: Minister, my suggested question links into educational excellence everywhere, but I don’t think that is the current catchphrase, so I shall rephrase the question to say: how will increasing academic selection lead to schools that work for everyone?
Mr Gibb: This consultation document, which is its proper technical term, is about creating more good school places. That is allowing more grammar schools to be established, because we know that 99% of those schools are “good” or “outstanding”. It is also about encouraging universities to help us establish more good schools. It is also about the independent sector using their expertise to help establish more good schools as well. That is what drives this consultation document.
Q61 William Wragg: On the particular aspect of academic selection, though, how will that improve education in this country?
Mr Gibb: It will improve it by giving greater diversity and choice. It is about having a grammar school in an area that has historically underperformed, for example; providing an example of what a good academic education can look like, and also using the skills of that grammar school to help improve other schools in the area. That will be one of the conditions that will apply to establish a new grammar school.
Q62 Ian Mearns: Could we not do that in Kent now—get the existing grammar schools in Kent now to improve the standards in the schools around them?
Mr Gibb: I agree with that. Your Government presided over the situation in Kent for 13 years and did not do anything to address the problems that were highlighted by the panel. This Government are addressing those problems, for new grammar schools, for expanding grammar schools and, indeed, in terms of the social mix for existing grammar schools.
Q63 Ian Mearns: Is it not a fantastic opportunity, though, for an educational pilot? You have an existing grammar school system in some places, Buckinghamshire, Kent, Tyneside. Could we not get some grammar schools in those places and get them to do that sort of work to see if it works?
Mr Gibb: Yes, I agree totally with you. It is already happening in parts of the country where grammar schools are working with local comprehensive schools together to raise standards, and we want to see more of that.
Q64 Chair: William’s question: I think we need more detail in the answer, because it is all very well saying it is going to increase competition and choice. The question is: will it improve outcomes, not just for those who get to a grammar school but for everybody else? The Green Paper—
Mr Gibb: The consultation document.
Chair: It is everyone, so are you convinced, and what evidence do you have to back up any statement you might make about outcomes being improved for children who are not necessarily going to grammar schools but are going to be affected by new grammar schools being created?
Mr Gibb: First of all, the Education Policy Institute recently reported that those children going to a grammar school achieve one-third of a grade in each subject higher than similar pupils in non-selective schools. I have already cited the level 5 children in terms of their EBacc performance. That is what happens to children who go to those schools. In terms of what happens to the other schools in grammar school areas, the evidence is mixed, as we have heard. The Sutton Trust report in 2008 says that there is no negative effect on other schools, but that is the past. That is the current situation. Our consultation document is talking about new grammar schools and expanding grammar schools having to work—as a condition of being allowed to expand or be established—with other schools in the locality in terms of raising the academic standards in those schools, working with them to help them, to support them. But also, in terms of improving the social mix of the intake of those new or expanded grammar schools, there will be a requirement that they have to demonstrate how they are going to improve the social mix, for example along the lines that we have seen in the Birmingham evidence.
Q65 Chair: That is the point we were going to get to, as I promised. Can I ask one more question in connection with this? What is your answer to the point made by Luke from the Institute of Fiscal Studies, where he noted that we have a different economy where skills are of a different type and that everybody does need to have a higher standard of education than they would have had, let’s say, in a grammar school/secondary modern system?
Mr Gibb: I totally agree with that. That is what has been driving all our reforms in primary education and the curriculum in secondary education. We are not going back to the binary system of the 1950s and 1960s, where the alternative to a grammar school was a very weak secondary modern. We now have a system where 80% to 85% of our state schools are either “good” or “outstanding”. We want to continue that drive, so that all schools, whether grammar or non-selective, are “good” or “outstanding”. The alternative to going to a grammar school would be going to a very good comprehensive school that would cater for all abilities.
Q66 William Wragg: To come back to my own question, Minister, and play the devil’s advocate and invite you perhaps to deal with this loaded question: does this policy benefit the few who are inherently more able rather than nurturing those who need more support to fulfil their potential?
Mr Gibb: It is about making sure that able children from disadvantaged backgrounds have the same opportunities as able children from more privileged backgrounds, whether it is being able to buy a house in a catchment area of a good school, or whether it is being able to pay for independent education. We want to end this gap between those children from poorer and wealthier backgrounds.
Don’t forget, the grammar school policy is just one element of our overall reform programme. It is even only one element of this consultation document, where we want more good schools and we want to harness the expertise of the university sector, the independent school sector, the faith school sector, to create more good school places. On top of that, there are all our other reforms in terms of the academies programme, the free schools programme, which is increasing diversity, the new curriculum that we have introduced in secondary schools, the new GCSEs, the better behaviour standards, more power for teachers to maintain good behaviour in our schools and, of course, all the reforms of primary education. It is just one element of that overall reform.
Q67 William Wragg: Very briefly, what would you say if perhaps we were to abandon the language of those who are more able, those who need more nurturing, segregation, and talked about aptitude of pupils? What would you say to that?
Mr Gibb: Yes. We have selection by aptitude now. Any comprehensive school is permitted to select up to 10% of its pupils according to the aptitudes set out in the legislation, so yes, I do not disagree with your use of language.
Q68 Lilian Greenwood: I want to follow up specifically this issue about whether selective education widens or narrows educational inequalities. The IFS said that it widens educational equalities, and this morning in evidence Professor David Jesson said there was a 25% gap between free school meals children and others where there are comprehensives, and that is over 30% in selective areas. How do you expect it to tackle that issue around social mobility? What evidence are you basing that assertion upon, when the evidence we heard this morning suggests just the opposite?
Mr Gibb: This is evidence based on the current system. What we are saying in this consultation document is that any grammar school that wants to expand, or any proposer who wants to establish a new grammar school in an area, have to support other schools in those areas to improve their academic standards. That is the difference between what we are proposing in this consultation document and the existing system. This existing system has been presided over by Labour Governments, by previous Conservative Governments. It is this Government who are seeking to tackle that issue.
There is some dispute about the effect in a grammar school area on the other schools, and it ranges from the Sutton Trust report saying this has no negative effect on those other schools to Dr Allen saying there is a small negative effect of one-10th of a grade per subject. I accept that the evidence is mixed, but the drive behind this consultation document is that we want to address even that small negative effect. We want those grammar schools to be working with those local schools to raise academic standards for all the pupils of all abilities in those schools.
Q69 Lilian Greenwood: What is the evidence that exists that your plan will work? What are you drawing on to assume that that is going to deliver what you say?
Mr Gibb: I am drawing on the fact that those pupils that go to grammar schools will do better than they would have done had they not gone to a selective school: one-third of a grade per subject higher according to the EPI report; the level 5 pupils getting better EBacc results; the Oxbridge entry results. The fact that you are twice as likely, if you are from a disadvantaged background, to go to a Russell Group university than a child who goes to a non-selective school, and you are three times more likely, as a disadvantaged pupil at a grammar school, to go to a Russell Group university than a disadvantaged pupil at a non-selective school.
Q70 Lucy Allan: Minister, if this proposal goes ahead as planned, will I see in my constituency, which is an area of low socio-economic background on the whole, more academic selective schools? That is what I would like to hear and I would like some commitment towards that.
Mr Gibb: That will depend on whether proposals come forward from proposers of new grammar schools in that area. If there is no community demand for a grammar school, even if a group of people got together and proposed one, if they were not able to demonstrate to the Secretary of State that there was parental demand for that school, on the arrangements that we are consulting on at the moment that school would not be approved.
Q71 Lucy Allan: Currently, we have all good and outstanding primary schools and there is nowhere for the most able children from those schools to go to a secondary school that is “good” or “outstanding”. I think that is inherently wrong, because it will create disadvantage in what is already a disadvantaged area. I would like some sort of confirmation, perhaps from the whole panel, that that type of inequality will be tackled by these proposals.
Mr Gibb: That is the drive behind not just the consultation document but everything we are trying to achieve in our education reforms, but I will hand over to Donna.
Donna Ward: I would expect in areas where there is no selection already and there is disadvantage, that is the sort of area the consultation document was pointing out as being ripe for extra good secondary schools that might want selection.
I return to one of the previous questions—when do selective schools become detrimental to the rest of the system? That tends only to be where the system is really polarised and there is an awful lot of selection in the system, so we would expect in areas where there are no “good” or “outstanding” schools, and no selection currently, that would be exactly the kind of area that these proposals are aiming at.
Q72 Lucy Allan: What if the community does not come forward and say, “Give me a selective school”? That may happen. They may just accept what is currently on offer.
Mr Gibb: If I may, I will cite the ResPublica report that came out last week into schools in Knowsley. Phillip Blond said they had the potential to transform education in that area. That is what I would expect would happen if there was a proposal from the community to establish a grammar school in an area, like Knowsley or indeed Telford—that it would have the potential to transform education in that area. For the first time you are getting a school, which has the ability to provide a very rigorous academic education for high-ability pupils, providing an example for other schools in the area to provide a similar curriculum. It is generally the curriculum that is the issue, as we have seen in the EBacc figures, and 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.
Q73 Chair: Minister, if pupils are travelling three or four times as far to get to a grammar school as another school in their area, does that not suggest that, if you had more grammar schools, the grammar schools benefiting from those long-distance learners would have a diminished group of people to choose their cohort from?
Mr Gibb: It may mean that people do not have to travel as far to a grammar school if there were more grammar schools in the area. One of the interesting figures is that children on free school meals do not live any further away from a grammar school than children not on free school meals.
Q74 Chair: I was really asking about the impact of other grammar schools on existing grammar schools.
Mr Gibb: They would provide further competition for those grammar schools, which I think is a good thing. The question was asked, do we need to raise education attainments in this country because we are now in a modern economy with different demand for different skills? Yes, we do. We need to raise academic standards, I believe, right across the ability range, if we are to prepare young people for life in modern Britain. We need to increase those standards for low-ability children and for high-ability children and we need to push even the most successful schools further up that achievement ladder.
Donna Ward: The degree of selection that is already in an area or close to an area, the composition of schools already there and the likely impact on the remaining schools, will all be part of the considerations in deciding whether new grammars can open.
Q75 Chair: I thought it was the community that was going to be deciding whether or not they wanted grammar schools.
Donna Ward: They propose it but the Secretary of State will eventually decide.
Mr Gibb: Part of the evidence that they will have to bring forward will be the impact it has on other local schools, which is similar to the free school system we have at the moment.
Q76 Ian Mearns: I can understand the ambition but you have talked about academic skills and academic attainment, and yet we have been conducting a different inquiry looking at skills in the broader sense because of skill shortages across the piece. Inherently within this there seems to be an ambition that our university population should continue to grow, so what would your target be, in that case, in terms of where the university population should be growing to? If your experiment, when it is a national experiment, was to become very successful and many more youngsters, particularly those from poor backgrounds, would have much better levels of academic achievement, what would the university population grow to, in your estimation?
Mr Gibb: What we want is for every child to go through our education system and fulfil their potential. That means at the top end of the ability range, which we are discussing today, we want those children, from whatever background they come from, to be able to achieve to the best of their ability. We also want children, whatever their aptitude or their interest, to achieve to the best of their ability as well, and that is why we are investing so much in the skills agenda. We want 3 million apprenticeship starts in this Parliament and we have improved the quality of those apprenticeships. We are taking seriously the improvement in skills so that the individual can fulfil their potential, but also for the needs of our economy. Our economy demands all these things. We are short of engineers, mathematicians and physicists and we are short of all the skills that drive what is a very modern and demanding economy.
Q77 Ian Mearns: But even youngsters who are possibly destined to go down the apprenticeship route need to be properly prepared to go on to that when the appropriate time comes. I am not convinced yet. What are the other parts of the equation from the proposals here, in terms of what would be done to the system to make sure that the youngsters who are going to go down the vocational route are properly prepared to go into an apprenticeship when the appropriate time comes?
Mr Gibb: That is what we want the schools to achieve by the time those youngsters leave school. Whether it is leaving school at 16 or 18, we want to make sure they have the literacy skills, the mathematics skills, the knowledge of the science that they need to have to go on to a good apprenticeship. That is what is driving our school reform. That is why we have reformed the GCSEs, and for any child who does not get a good pass in English and maths at GCSE, when they start their technical education at the age of 16, there is a requirement that they will continue to study English and maths so they can achieve a GCSE in those subjects.
Q78 Catherine McKinnell: Minister, you say that the evidence is mixed but the evidence that has been presented to the Committee seems to overwhelmingly suggest that selection does not have an effect on overall attainment. It does for those who get to the grammar school, but for those who don’t and for the schools that are left surrounding it, it has a detrimental impact. What are you going to do to ensure that that is not the case?
Mr Gibb: Again, it is mixed and it does depend, as Donna Ward said, on the proportion of selection in the area. The EPI report goes into great detail of the different levels of selection and the Sutton Trust report in 2008 says that there is no negative effect on those other schools. Having said all that, that is about the system as it currently exists. The proposals in the consultation document require any new grammar school to be working with local schools to improve the quality of academic education in those other surrounding schools. That is part of the requirements.
Q79 Catherine McKinnell: How will that be ensured?
Mr Gibb: If you read the consultation documents, there are lists of different proposals of the sanctions that would apply if that was not delivered. It could ultimately lead to the school being no longer able to select by ability, or result in funding being limited so they can’t recruit more pupils going into the future. There are all kinds of sanctions that we are consulting on in the document.
Q80 Catherine McKinnell: One thing there is no mention of is special educational needs. The Government have an ambition to create educational opportunities for all children, so where do special educational needs children fit into this?
Mr Gibb: The more good grammar school places we have the more opportunities there will be for those children with special needs, whether physical or in their attitude, to get into a grammar school. We want to create more good school places right across the system. That will benefit children with special educational needs as much as it will benefit all children.
Q81 Catherine McKinnell: Does the Minister accept that perhaps the Government should ensure that that is the case, rather than simply hoping it forms part of the consultation?
Mr Gibb: We are not just hoping. We have taken a range of measures in the last six years to improve the quality of outcomes for children with special educational needs. The introduction of the education, health and care plans, for example, is just one element of the reforms that we have introduced over the last six years and continue to introduce to improve the educational opportunities for children with special needs.
Q82 Chair: Do you see new grammar schools going into MATs or creating MATs, in order to deliver what you think is an important part of their responsibility, which is to improve education beyond their own school?
Mr Gibb: Yes, I do very much. We want grammar schools to be part of a multi-academy trust. That is the direction of travel for schools going into the future. It means you can have more collaboration. You can, for example, share teachers of shortage subjects, of subjects that have fewer pupils taking them, between a grammar school and a comprehensive.
Q83 Chair: You are talking about sanctions for not helping other schools, but you have not really mentioned the structures to enable grammar schools to definitely help other schools. One would have thought becoming a member of a MAT or forming a MAT would be the way forward. Should that be something that is prescriptive or just an option?
Mr Gibb: These are the kinds of ideas that we are listening to when people respond to the consultation document. What we are saying is that it absolutely will be a condition of approval that they have to demonstrate how they are going to help other schools in the area, and also work with feeder primary schools or establish feeder primary schools, establish a feeder primary school in an area of deprivation to ensure that all the children in that area have proper access to that grammar school, which of course does not happen in many areas at the moment. We want to address that shortcoming in the current system.
Q84 Suella Fernandes: To follow on from the point Catherine has made about evidence, there is considerable evidence in the consultation document but I also want you, Minister, and the other panellists to talk a little bit about the international evidence. The Financial Times recently reported about Singapore and Hong Kong who topped the Programme for International Student Assessment tables and noted that there are more than 80% of their children in academically selective schools. Is there anything that you would like to say about international comparisons and evidence in that context?
Dr Leunig: I will speak on that one. I think the most interesting country is the Netherlands. It is a much more similar country to ours than, say, Singapore. They select at the age of 12 through something that would look recognisably like selection here. 20% of people go into the most academic stream and those people in that school are then explicitly destined for conventional research universities. The next 20% of people go into the next group of schools. I am not going to give you their names; I can’t pronounce them. We can write to you afterwards with their names. They are essentially destined for what we would have once called polytechnics, applied vocational universities. The remaining 60% of children are in a less academic stream that is explicitly technical. It is not the sort of Rab Butler tripartite. 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. Their system is a little more fluid than ours. They judge slightly later than us. The older a child, the more reliable selection is in general. They also take into account parental and school views, which have both strengths and weaknesses. It can be helpful if a child has performed unexpectedly poorly but—as Anna said earlier—that sort of thing tends to favour the middle class, and it does in the Netherlands. You can essentially choose your school at age 12 but at the end of your first year in secondary school, the school has the right to say, “This school is not the one for you,” and essentially send you to another one. So it is a bit more fluid than our system has been.
If the Committee wanted to study selection somewhere else, that would be the place that I would look, if I were you, at both the strengths and weaknesses of what is, overall, a successful system and one that has selection. I am not claiming selection has no downsides there but it is one that you could usefully look at.
Q85 Suella Fernandes: Thank you. That is very helpful. Moving on to the proposal itself, there are various conditions attached to opening a new grammar, and there is also considerable reference to involvement of the independent sector and universities and other state schools. Could you talk globally about the conditions attached to this proposal, and also the involvement of the other parts of the school sector?
Mr Gibb: As I said earlier, we want to tap into the expertise of the university sector and to encourage them to establish a new school or to work with an underperforming school, and the same with the independent sector. They gain considerable benefits from charitable status and we would like them—if they are big enough and their resources are sufficient—to establish a new state-funded school and use that expertise to establish that school. If they are much smaller, then we would like them to be involved in some way in contributing to state education in our system.
Then there is the issue of faith schools, which are very popular. There is still a 50% cap on a newly established academy that can select by faith if it is over-subscribed. That is inhibiting the establishment of a number of new faith schools, particularly Catholic faith schools. Given that we want to create more good school places in our system, it seems absurd to have a regulation that, in effect, prevents the Catholic Education Service from setting up new schools.
Donna Ward: I think the common theme running throughout the consultation document is that it is not a zero-sum game, which I think the academics did make it seem like in the hour before. If you can create the right incentives to open “good” and “outstanding” school places, and harness expertise that is already out there, then the total amount of expertise and leadership skills in the system can increase.
Q86 Lilian Greenwood: I want to come to how children get to go to grammar schools or not go to grammar schools. The concern has been expressed that, when it comes to the 11-plus, it simply reinforces disadvantage, because children from less well-off backgrounds maybe do not even take the test but then tend to perform less well. The Prime Minister has spoken about smarter tests that assess true potential. What are you suggesting that a selective test should look like?
Mr Gibb: You can look to examples, such as the King Edward VI Foundation in Birmingham that has different criteria for children from more disadvantaged backgrounds. That has been successful in increasing the proportion of children from those backgrounds attending those schools. Also, some of the reasons why children from disadvantaged backgrounds do not apply are because, for example, the primary school does not encourage them to apply or the parents may feel that this is not a school for them. They may think it would be socially difficult for those children. We need to ensure that the new grammar schools or the grammar schools that are permitted to expand are working with the primary schools, engaging in outreach programmes to dispel these myths and encouraging the parents to apply for these places.
Q87 Lilian Greenwood: When the Prime Minister talks about a smarter test, what is your expectation about what those tests look like?
Mr Gibb: That is what we are consulting on. We listen very carefully to academics, such as those we had in the previous panel, about how you could tweak the system or amend the system of selection to try to remove the inherent advantages that come with having sufficient money to be able to pay for private tutors. But that is not the only inhibition on children from disadvantaged backgrounds going to grammar school. One of the key inhibitions is that the primary school they attend does not encourage applications, or they are not used to sending children to a grammar school and the parents have an instinctive fear, if you like, of what a grammar school would be like for their children, and that is something we want to overcome.
Q88 Lilian Greenwood: But isn’t it the case that what happens in primary school is absolutely key, not to which school children go on to but their overall attainment? Should that not be the focus, rather than teaching to pass a test?
Mr Gibb: I totally agree. A major part of my work over the last six years has been about improving reading ability of children in primary schools, and improving the mathematics and the arithmetic of children in primary schools. We have engaged in the phonics programme. We have engaged in the Shanghai maths programme. We want to get more textbooks back into primary schools, and we have rewritten the curriculum. It is a much more demanding curriculum than it was, and that has been in place since 2014. We have introduced a quite tough, quite demanding national curriculum for grammar, punctuation and spelling. All these things will help children as we see standards rise in primary schools.
Q89 Lilian Greenwood: Finally, can we come back to this issue around a test, because no matter how good primary schools are, if you are going to have a test—do you contend that it is possible to produce a tutor-proof test?
Mr Gibb: That will be the Holy Grail. Grammar schools are trying to do that. There are examples all over the country of grammar schools that have—
Ian Mearns: People have spent a long time looking for the Holy Grail in this panel.
Q90 Chair: My Holy Grail would be excellent education for everybody.
Mr Gibb: Yes, and I am with you on that, Neil. Absolutely.
Q91 Lilian Greenwood: Rather, what evidence is there that you can produce a test that tests potential as opposed to testing prior achievement?
Mr Gibb: That is the objective—what we would like to see delivered. That is something we are consulting on. We want to receive evidence from specialists in this area. But ultimately, as I said, it is not the only factor that is preventing children from disadvantaged backgrounds going to grammar schools.
Dr Leunig: I will add something, which is that the King Edward experiment is interesting here because, essentially, what they have is one pass mark for children not on free school meals, and a slightly—but not hugely—lower pass mark for children on free school meals. If we judge that tuition and private tutoring add six marks to a child’s score, it would be perfectly possible to add six marks to the score of any free-school-meal child on the grounds that it is unlikely, although not perfectly, that they will not have had that sort of tuition.
It is possible to level the playing field. We do it already with age. The 11-plus has a different pass mark for people according to which month you were born in, because children born in August do less well than children born in September. It would not be quite as easy in the case of poverty, because a poor family, if they are really determined, can still find that money. But you can go a long way to answering that question, even without a perfect, tutor-proof test, which, as you imply, at least as yet does not exist. Although we hope through the consultation we will get a better handle on it.
Q92 Suella Fernandes: On this point about working with the primary schools, the single biggest predictor of how a child will do at 16 is the education of his or her family. It is really important, according to some studies. I am seeing Dr Leunig shaking his head. If we were trying to identify children at an early age in primary school, what do you think about the proposal made by some organisations that there should be mentoring by grammar schools of children in primary school who are earmarked for the 11-plus exam, to assist preparation?
Dr Leunig: That is a classic example of a test of a hypothesis. We should not rule it out nationally, because then we will never know whether it is effective. We should run a good, randomised, control trial with a control group. The EF is very good at that. They celebrate their fifth anniversary today. That is exactly the sort of thing they can test and give us robust evidence whether it works.
Chair: There speaks a Chief Scientific Adviser.
Dr Leunig: What other answer would you expect of me?
Q93 Suella Fernandes: Lastly—I think this is mainly for the Minister—what potential do you think there is for grammar schools to become a beacon of social mobility, whereby they enable children from more disadvantaged backgrounds to flourish in a context that fosters high-quality learning and enables them to escape poverty through education?
Mr Gibb: I do think they have that potential. As I said, the main driver is to get more good school places into our system. When you look at some of the evidence of children that do go to grammar schools from poorer backgrounds, they do well. In fact, they do better in grammar schools than the children at those grammar schools who are not from a disadvantaged background. That is where you get these very high Progress 8 figures. That is why you get the gap between children from disadvantaged backgrounds and their wealthier peers closing in grammar schools. I am particularly struck, though, by the Oxford and Cambridge figures. We need to get more young people to Russell Group universities and to Oxford and Cambridge from disadvantaged backgrounds. I think they will help in delivering that as well.
Q94 Catherine McKinnell: I have three issues that I would be grateful if the panel could address. One is: why does this have to be about segregation? I want my children to do well at school, but I don’t particularly want them to be plucked from a system and put elsewhere. Why can that not be achieved in a more comprehensive school? Why do they have to be separated into separate schools? Secondly, has the evidence looked at the issue of malleability and cognitive function and the fact that, at 11, you still have a huge amount of malleability in terms of your cognitive function? You may well not reach your peak until 16; you may reach it younger than that. Every child is different and, therefore, to have a set time for an academic assessment and to make such a permanent change to their life is quite a significant change for them. The third I cannot remember, but if you answer the first two I will think of the third.
Mr Gibb: First of all, it is not about segregation. It is not going back to the binary system of the 1950s and 1960s. We live in a completely different school system now. We do not have the secondary moderns that provide a very poor-quality education that we had in the 1950s and 1960s, where almost no pupils in some of those schools were taking any qualifications. We now have a system where 85% of comprehensive schools are “good” or “outstanding”. The consequence of not going into one of these selective schools is you will go to a very good comprehensive. We want to finish that 85% and get it up to 100%. That is part of what is driving these policies.
Q95 Catherine McKinnell: Minister, do you not realise that by creating—
Ian Mearns: Minister, for a school to be comprehensive it has to have a comprehensive intake, including all ability ranges.
Catherine McKinnell: —grammar selection and segregation you are, by virtue of that, creating a secondary modern, in effect.
Mr Gibb: Yes, but the secondary modern schools themselves were just not good schools, and that is not the position now. Even in areas with selection, there are very many “good” and “outstanding” schools. The consequence of not going to those schools in an area with selection is not as it was in the 1950s and 1960s. These are very good schools. We have an ambition for more pupils—
Q96 Catherine McKinnell: Why do you need grammars, then? Why do you need them?
Mr Gibb: Hang on; I want to come to your second point about the grammar school ethos. We want there to be a very high quality of academic education in all our schools, whether they are selective or non-selective. That is what is driving our English Baccalaureate policy. We want more young people to be taking that combination of GCSEs. Those subjects that are A-level are regarded as facilitating subjects, as defined by the Russell Group of universities. As I cited at the beginning of this session, there are far too many local authority areas where schools are simply not entering enough pupils for the EBacc combination, and not enough children are achieving that combination. I believe that by having more grammar schools in those kinds of areas, in particular, it will help to drive up, to focus, to change minds, and to change attitudes towards a core academic curriculum in those other schools, not just in the grammar school.
Secondly, you raised the issue of 11. It may be too young for some children, which is why the consultation document talks about there being flexibility in the age of entry into a selective school. It talks about 14 and then 16. Of course, 16 is permitted now. We do need to make sure that any proposal that comes forward does have facilities within it to allow intakes at different ages.
Q97 Catherine McKinnell: The final, third point was the evidence from Dr Allen, which was that the introduction of an attempted, tutor-proof test may exacerbate the problem of those with financial resources to tutor, and may well widen the gap between—when you say the six marks—those who can afford tutoring and those who cannot. It may widen that gap because more intensity will be put into the tutoring in order to beat the tutor-proof test.
Dr Leunig: They introduced it in Buckinghamshire for exactly the right reasons. I don’t want to cast any aspersion on their reasons for doing it or on the people who designed the test but, as Becky said earlier, it didn’t work. But designing the best test is an empirical matter. The only thing you can do is to try it and see whether it works better or worse than the predecessor. If it works better you build on it. If it works less well, you withdraw it and go back to what you had before. That is true with so many things in life. That is the scientific—
Chair: Okay, we have to leave it there. Lucy wants the last question.
Q98 Lucy Frazer: I was very interested in your international comparison. I want to hear a little bit more about it, so if you could just tell us a little bit more about it that would be very helpful.
Dr Leunig: The Dutch system has been established for a long time. They have a national test. It is called the CITO. Not every child takes it, but the overwhelming majority do, and it is a discussion between the child, the school and the parents whether they take it. The result then comes out, and that informs the parents’ and the teachers’ and the child’s decision as to where they go next.
Q99 Lucy Frazer: That is at 12?
Dr Leunig: That is at 12. It is one year later than us. Then they go into what they call a bridge year, which has a common curriculum across all types of school. In fact, they do that for the first two years. At the end of the first year some children—not a large number, but some children—move, sometimes because they realise this is not the school for them, sometimes because the school says, “I am sorry, Sunshine, this really isn’t working, and we are afraid, we are sorry, you are out of here.”
Q100 Lucy Frazer: It is highly academic, less academic and technical/vocational?
Dr Leunig: Yes. And within the five strands of the mainstream education, the bottom one is called a practical school, and they do a much more limited academic curriculum, and it is very job-focused.
Q101 Lucy Frazer: Is there a stigma attached to which school you go to?
Dr Leunig: Yes. That would be my reading of the evidence. I think there always is.
Q102 Lucy Frazer: What is the level of wellbeing and happiness of children in Dutch schools?
Dr Leunig: I don’t know the answer to that one. We could look in to see if there is good evidence on that. The Netherlands as a whole is a fairly happy society, so I would be optimistic, but I am well beyond my knowledge base here.
Q103 Lucy Frazer: Is the teaching different—the stigma attaching to the teaching in the particular schools?
Dr Leunig: The teaching is different, in that the most academic—
Q104 Lucy Frazer: Is the quality of the teaching different?
Dr Leunig: It depends what you mean by “quality” here. To compare the quality of teaching of ancient Greek with the quality of teaching a practical subject is something that Ofsted may want to tell you is possible, but I am not going to say that that comparison would be sound.
Q105 Chair: Today’s search for evidence to support this policy must come to an end, but before it does, can I ask you, Tim, in your capacity as Chief Scientific Adviser—with that particular hat on—do you think there is sufficient evidence to rate this policy, and how would you judge it? I am going to use the Ofsted system. Do you think it is outstanding, good, requires improvement, or inadequate?
Dr Leunig: Clearly, you are asking me a question that no politician should put to a civil servant. So let me give you a diplomatic answer—
Chair: I am putting it to the Chief Scientific Adviser.
Dr Leunig: I know you are. I am going to give you the answer that was given to me by an excellent head teacher in an outstanding school, who said, “All schools require improvement, and the moment we think we don’t require improvement we are doomed.” This policy, like all policy, requires improvement. That is why there is a consultation document. That is why we will look at the evidence and, hopefully, we will deliver that improvement that we need.
Q106 Chair: It is the evidence we are looking at, and that is what we are trying to find, and we will continue our search another day.
Dr Leunig: That is an excellent thing for you to do.
Chair: Thank you very much for your contributions. You too, Nick, and you too, Donna. It has been a great two-and-a-quarter hours, and I think we have had what has been very revealing in the answers, and sometimes the non-answers. Thank you.