Do the numbers add up for Croydon grammar schools?

By - Thursday 30th March, 2017

A modest increase in grammar school places would be good for Croydon

Photo public domain.

Money set aside in this week’s budget is another step towards more grammar school places for Croydon, as strongly supported by Croydon South’s MP Chris Philp and Steve O’Connell, our local GLA member. This won’t happen any time soon – but the sooner the better for those of us who support a modest increase in the number of grammar school places.

No policy is without its pros and cons, which one might hope would be discussed rationally. Yet grammar schooling is an issue on which this has proved nigh on impossible due to some opponents being viscerally opposed to selection of any kind (except perhaps selection by geography). It is a point of view, but one that they avoid defending by claiming a monopoly of supporting evidence. They are wrong. I shall endeavour to explain.

The question is whether there is a level of grammar school provision (compared to non-selective schools) that benefits our community. But how to measure that benefit?

Performance of schools is difficult to measure

Measuring the performance of individuals is relatively simple. Success in examinations (GCSEs) does so nicely. The performance of schools is more difficult. A school might produce better results because it had better material to work with (more able children). To measure the performance of different types of schools, we must somehow correct for these factors.

This isn’t simple, and is subject to error, but the reports that I shall quote do just that. Let’s assume that they have done a decent job.

We are not there yet, though. Assuming a positive benefit for grammar school pupils, we must consider the effect on the others. That effect, positive or negative, must be weighed against the positive effect on the grammar school pupils. But let’s say that there was a level of grammar school provision at which these children did better and the others did no worse – is that ok? Visceral opponents would, if pushed, say no.

Sacrifice excellence in favour of uniformity

Their view is that one group doing better, no matter that others do no worse, is inherently unfair. They are prepared to sacrifice excellence in favour of uniformity. In a sporting analogy, their view is that it is better to have a team of runners who are more equal in performance than to have some Olympic medallists and national champions. That is the heart of the argument, but one that they are keen to avoid.

They much prefer a bit of stereotyping. Stereotyping is normally an anathema to the visceral opponents, but in this case they feel able to indulge themselves because it is to stereotype ‘pushy’ (read ambitious) parents, probably living in ‘leafy’ (read Tory) suburbs. Whether there is any truth in it is another matter.

Their evidence is that children on free school meals (an easy and reliable parameter to collect) are less likely to get into grammar schools. This is true, although those that do get on extremely well, better even than their better-off peers.

Grammar school children do better because they are at grammar school

But does such a sweet spot exist? Yes it does.

The authoritative report commissioned by the Sutton Trust in 2008 found that better performance of grammar school children “does not seem to be fully explained by their higher ability or their tendency to live in more socially advantaged areas”; translation: grammar school children do better because they are at grammar school.

They also “failed to find any evidence of collateral harm to any other schools, arising from the existence of grammar schools”. Translation: grammar schools do not damage nearby comprehensives.

Scope for expansion

More recently (2016) The Education Policy Institute (EPI) found the ‘tipping point’ at which a negative effect emerged for those who did not attend grammar schools. This occurred when places were available for more than 70 per cent of high-attaining (top quartile) pupils. According to the House of Commons library, fewer than five percent of children currently go to grammar schools. That leaves a lot of scope for expansion.

Some opponents in their heart of hearts already understand the weakness of their argument. I have seen the introduction of additional factors; for example, to exclude areas where there is not currently evidence of clear public demand. The argument is that this is a criterion put forward by the government. Take out the word currently and I agree. But is it surprising that there is currently less support, given the many years when such an option was legally forbidden? Times change.

We can do still better. The pupil premium, introduced in 2011, targets disadvantaged pupils. Its effects have not yet filtered through into the data. It is also the government’s intention to require selective schools to play a greater role in raising standards at other schools.

Plans in place to close the gap

So higher performance for some, no loss for others, and the plans in place to do better still in closing the gap for poorer children. What’s not to like? If you are a visceral opponent of selection, everything.

I recognise that there are decent, well-meaning people who oppose grammar schools. That’s fine; let’s have the argument. Just let’s not pretend that the data is on their side.

Robert Ward

Robert Ward

Engineer and project manager, started work on the railway but most of career in oil exploration and production. For the last fifteen years specialised in helping businesses improve their performance. Conservative Party candidate to represent Selsdon and Addington Village on Croydon Council. He tweets as @moguloilman.

More Posts

  • Tig

    I don’t know many anti-selection people who argue that all pupils are the same and should have the same lessons. Most argue that a short test at ten and a crude % divide is a very poor deal for those who don’t get selected.

    You quote one report that says grammars have no impact on other local schools. There are many more that show subject choice changes, the schools find it harder to recruit teachers, and that a lack of ‘stretch’ through top sets produces worse results. Not to mention the fact the schools around grammars have a much higher proportion of time consuming pupils with special educational needs and poor english. Grammar schools are very ‘bad neighbours’ in their community of schools. They are in no way inclusive.

    I live in Kent and have a ‘high school’ up the road. It offers no top sets and hardly any A levels. I know many children who hope to go to university would do a whole lot better in a good comprehensive with sets. Instead we have a crazy system where most parents I know pay £1,000 per child for 11+ tutoring. Not because the grammars are so good, but simply to avoid the high schools.

    There is no parity of esteem among local schools. Where there are high achiever schools there are low achiever schools. More than 40% of children go on to university and a selective system like Kents, taking 28% of children for grammars, bears no relation to modern needs for many more children to have an academic focus.

    I believe that the report you highlighted showed that 22% of pupils are incorrectly judged by the 11-plus, based on eventual GCSE results. It feels terribly flawed to be expanding a form of education that defines our school places so inaccurately. If we use a test that judges potential so poorly then it is creating unfair admissions to our schools. And this is before we even consider that tutoring makes grammars more about parental keeness that true ability.

    We have tests in school already, schools don’t need an 11-plus. A clear plan for high achievers, monitored by Ofsted, would do a whole lot more good than patchy new grammar schools. Good comprehensives work better than a grammar/secondary modern divide. As Kent’s results prove.

    Perhaps if you had a bright child, given a choice of Requires Improvement secondary moderns due to a test score, you might feel as I do about expanding selective education. Many of my degree educated friends are in the same boat. There are many children of broadly similar ability divided by a crude old-fashoned test. If the test is not up to the job (and it isn’t) then grammar schools should stay in the past.

    • Robert Ward

      Thank you. I will endeavour to respond to some of the many points you raise:

      I am not proposing a return to the universal eleven plus as it is currently practiced in Kent. Please read the article.

      I am not familiar with the “many” reports you claim that show an adverse impact. The Sutton Trust report is the only rigorous report I have been able to find, but I am willing to look at others.

      No examination is perfect. Demanding a perfect examination is just a surrogate for a visceral opposition to selection. What is important is that a child who falls just below whatever line you choose to draw does not suffer significant adverse impact. The ‘tutored to death’ line is another such surrogate.

      Evidence is that having English as a second language is not a significant impediment, indeed some immigrant groups do systematically better than the average. Funding for children with special and other needs follows the pupil so less of such children means less funding.

      Having looked at the data on Kent and Medway one thing that stood out for me is that Kent has relatively low funding compared to some nearby, very accessible London Boroughs. Much of the Inner London improved performance that is used in comparisons ignores this fact. If I were you I would lobby my MP on the education bill currently going through Parliament. The current proposals do not sufficiently address this issue in my opinion.

      • Tig

        Isn’t your support for grammar schools based on your visceral opinion that they get better results? The proof for that idea is very hard to come by.

        Grammar schools contain children from wealthy families (14% from the independent sector) they contain children proven to have parental support (they entered an academic test & most practised for it) and they all come from families that value education. The schools contain few dyslexic children, few children with SEN and few disadvantaged children.

        It seems impossible to judge grammar school’s success as being some factor to do with a ‘seperate building’ rather than it being about the type of pupils they assemble. These same pupils with great support, ambition and wealth, are likely to achieve exactly the same results in any good mixed ability school.

        I would be very interested to hear ‘why’ you think grammar schools get better results?

        Is it because they have better teachers?
        That leaves other local schools with less good teachers. See:

        Is it because the children do homework and have great parental support?
        They get that without a need for a seperate building.

        Is it something to do with children who enjoy learning?
        Could we have a test for a pupil’s interest in school instead of an IQ test? This is also problematic if it places a burden on other schools with less of these children.

        Is it lessons pitched at a higher level?
        This is achieved by setting in most schools, and it is also a problem if one school takes all the high achievers. This stops other local schools having enough pupils to create top sets.

        You state you don’t want a system like Kent… this is telling. How many new grammar schools does it take in your area to create a Kent balance? Is it really worth the risk?

        You seem to admit it causes problems, but don’t see that one grammar school in a community creates three nearby secondary moderns. You simply can not put all the high achievers in one school and expect the other schools to go on just the same.

        You seem willing to risk the secondary modern problem. To me it seems like a risk that is not worth taking. If I were you I would question your local schools about their results if you are worried about them. Although having taken a quick look some seem to be getting better progress for high achievers than my local grammars!

        Do you really want two tier education? Once there is a high status school there are low status schools. Local parents will spend £900 to £1,200 per child in tutor fees.

        The report you linked to in your article concluded that pupils in grammars get the same results in good comprehensives.

        I have yet to hear a teaching professional make a case for academic selection. This is mostly because of the impact on other schools.

        So I still fail to see why you support a plan that you admit carries risk of damaging local schools. You seem prepared to trust one report. Well come to my Kent town and I will show you how the schools change.

        • Robert Ward

          I suggest you look at the report that I quoted. Find it here:

          This is the only report that is in my view rigorous and objective. Even less objective reports conclude that there is a positive grammar school effect, indeed I am not aware of one that doesn’t. Again happy to read anything you may have.

          I have no visceral objections here. I am relying on the evidence. For that reason I have deliberately not speculated on why grammar schools get better results. Again please read my article in which I point out that the models attempt to correct for the differences in the intake. They may not be perfect but where the results are statistically significant that is the best evidence we have.

          • Tig

            I have looked at the report a number of times. It is produced by CEM who are the largest provider of 11-plus tests! :) So much for your ‘objective’ point.

            It is a shame that you don’t explain the WHY of grammar school success. It seems very clear that the factors that make grammar successful correlate with factors that cause problems in secondary modern schools. Why would you accept one and not the other?

            The reason this is a problem is because grammar schools educate mostly the advantaged, while secondary modern schools mostly educate the disadvantaged. So most people who care about social inequality think selective systems are a bad idea.

            If you want up to date properly objective data on the secondary modern effect you can find it here in figures 7 and 8:


            It is also worth noting the subject choice changes in schools with a lower ability intake. I see that in Kent with no triple science being offered to my daughter. It is lucky she didn’t want to be a doctor.

            I respect that your argument is ‘grammar schools get better results’ but then it all gets hazy because not only do we have to try to adjust for pupil ability, we really need to to adjust for parental background. But no one has you have to admit that no report has looked at parental motivation and family background as a factor. Yet we know that many pupils who get in are tutored. These are parents who would probably pay if their child was getting a C at grammar. Whose to say that this isn’t some part of the ‘grammar effect.’

            Nor did the report you quote look at this, it just said, “There may be important but unmeasured differences between grammar and non-grammar school pupils and this
            somewhat undermines our confidence in these estimates of a ‘grammar school effect’.

            Ah, so they’re not even sure.. and this is the best we have!?

            So we have schools that ‘might’ be better, that cost parents tutor cash, that stop other schools offering top sets and three sciences, and that suck up the best teachers. Oh, and they’re almost entirely full of rich kids.


  • Patrick Blewer

    Hi Robert. As a 3rd Generation ex grammar school boy and first to go to university I have somewhat mixed feelings about the concept. I disliked much of my experience of it, despite accepting that I might not have got to where I am today without attending a well known grammar school on the borders of Sutton and Croydon.

    I’m not against selection per se, nor am I against excellence. It’s pretty how the adult corporate world works, but with more politics and inter personal relationships having a greater effect.

    However I’d argue that we’re looking at the wrong end of education if we really want to enhance the education of the entire populace. The investment needs to made earlier into pre school and infants education.

    I found the attached studies around this issue – that being pro or anti grammar isn’t necessarily the point.

    I’d argue that the current system isn’t about the purity of competition because a significant proportion of the 10 year olds taking the exams are fighting with one hand tied behind their back, with either the competition coming from schools where 11+ support is part of the curriculum, or external tutoring which is way beyond. Improve early education, and this problem might be at least partially removed.

    So in other words, I’d argue that if we are going to have selection (and it’s not necessarily coincidental I recently moved back to Sutton) infants and junior school need investment so that selection starts on the most even playing field possible, so as to avoid the accusation that grammars are basically middle class havens for those too mean or ideologically minded to shell out for independent education.



    • Robert Ward

      Thanks Paddy. There is much there that I agree with.

      The low numbers of certain groups such as free school meal children entering grammars, often used as a criticism of grammars, is in my view a measure of the failure of schooling (and parenting) before that age. Educational disadvantage is already there before children enter school. I heard a head teacher of a school that took in children who still wore nappies and who could only construct two word sentences. Education closes some of the gap, but not all.

      For precisely that reason much recent educational funding has been targeted at early years and pre-school, with the pupil premium targeting the more disadvantaged. Dealing further with that issue does get us into difficult arguments about the boundary between parental responsibility and the responsibility of the state. That argument gets ducked by reaching for the stereotypes (pushy parents, tutoring, etc).

      What grammars can do to help is to partner with other schools, open in more disadvantaged areas and seek to attract pupils from more disadvantaged backgrounds. All of these are I understand part of the plan.

  • Gary Coy

    An interesting discussion thread and I can see both sides of the argument. I am Founder and Chair of the Rowdown Inspire to Aspire Foundation, a small charity supporting gifted and talented children, including the academically able, at Rowdown Primary School (where I was Chair of Governors) in New Addington where their parents are not financially in a position to do so. As part of this, we have run a tuition programme for 4 or 5 pupils in Years 3, 4 and 5 (and into the first term of Year 6) in order to help these already bright children have a chance at the 11+ entrance exams for the best grammar and independent schools in the area. In the two Year 6 groups (9 pupils) that we have tutored so far (the first for only 1 year), 8 have reach a secondary school through the selective system including one to each of Old Palace, Trinity, Wallington County Grammar School, Whitgift and Wilson’s. Those that have a place at fee-paying schools have received a combination of scholarships and bursaries to make these schools affordable to their parents. For instance, the girl that has just won her place at Old Palace has done so with a 25% scholarship plus a bursary which means her parents are left with £200 pa to pay.

    I went to a comprehensive school at a time when Surrey had grammar schools and made it through that system to attend a top-10 UK university going on to have a 20+ year career in the City. I believe that the system can work occasionally. I would rather not have to exploit the two – or probably three – tier secondary education system in Croydon (and nationwide). But in New Addington, not only are most families economically disadvantaged, but their children are educationally disadvantaged: there is no Ofsted-rated Outstanding comprehensive school where New Addington is in the catchment area. Therefore children living in New Addington cannot get access to an outstanding secondary education. Schools may be improving in the area but, in the meantime, there is a reality that children there are not reaching their full potential academically. This is where the Rowdown Foundation seeks to level the playing field with those more affluent parents that can get their children to grammar schools by coaching at prep schools or private tutors. The success levels we have seen so far prove (albeit with a small sample size) that there is latent academic talent in New Addington that is prevented from reaching its true potential by a broken state education system, a system that will take decades to fix. Why should these children be held back from a life-enhancing opportunity offered by these schools because of a broken ideology that every child must have the same opportunity? Should the few suffer for little or no proven benefit for the majority? It is interesting that many of the parents involved in the tutoring scheme have said that they would not have looked outside the local state school offering had the Foundation not been tutoring their child. So it is also a lack of inspiration and aspiration inherent in areas like New Addington that we have to work on. From my meetings with many of the local grammar and independent schools, they all really want to encourage social mobility, probably themselves encouraged to do so by the recent Government consultation “Schools that work for everyone” that suggested (between the lines) if they don’t do it willingly, they may be forced to.

    The good news – in my view at least – is that success breeds success. The Foundation has established good relationships with grammar and independent schools in order to facilitate social mobility for children in New Addington. We have just received an offer from an independent boys school in Croydon to set up a Saturday morning extension learning programme for up to twenty Year 5 (into Year 6) boys and girls from New Addington free of charge to parents, the primary schools and the Foundation. We are also discussing with another school the prospect of a similar scheme for Years 4-5. And a Sutton-based grammar school has also offered extra tuition in the run up to the summer holidays ahead of the Sutton schools SET in September. Suddenly, rather than 4 children each year achieving access to a quality secondary education that might otherwise have been denied, we could have 20 and it is New Addington wide. Three other primary schools have now signed up to the scheme. These lessons will not be specifically designed to pass the 11+ exams – I am sure they will still help the children in this but there will be no compulsion on them to sit these exams – but will stretch these children in an enjoyable fashion to make sure they remain engaged in learning for the future, something that many local schools fail to do with low or negative Progress 8 scores at KS4 (cf Wallington CGS +0.66, Wallington Girls +0.70, Meridian High -0.08, Quest 0.18, Charles Darwin 0.14: these three state schools are where the vast majority of Rowdown pupils go on to). This is progress remember, not attainment. It is a measure (albeit some might argue a little flawed) of measuring fulfilment of potential.

    The proposal for a grammar school in Croydon is divisive. It would need to aid social mobility without damaging the education of those of less academic ability remaining at the comprehensive schools. It is a fine line but one that can, I believe, work under the right controls and management. However, an alternative is to ensure primary school children from a less affluent background (and not just those that qualify as “disadvantaged” through eligibility for the Pupil Premium Grant) have a chance of competing with their richer neighbours for places at the existing grammar and independent schools: and it can be done cheaply and efficiently by engaging primary schools, charities and the very schools that they are aspiring to attend.

    Sorry for the long comment!!!