Who and what are the Education Policy Institute? What do they say about Grammar Schools?
An influential document
One of the most cited documents in the current grammar school debate is a report by the ‘Education Policy Institute’ http://epi.org.uk/report/grammar-schools-social-mobility/
Full report here http://epi.org.uk/wp-content/uploads/2016/09/Grammar-schools-and-social-mobility_.pdf
A little later in the post I will examine parts of the report (not all - one particular section of it requires consulting hard-to-find books and will take longer, but I hope to get round to it).
Independent? Impartial? Evidence-based?
The report itself asserts in its text that the EPI ‘is an independent, impartial and evidence-based research institute that aims to promote high quality education outcomes, regardless of social background.’
Yet even the slightest investigation shows that it started life as a Liberal Democrat think tank, ‘CentreForum’ (itself the inheritor of an earlier Liberal Party think tank) and that many of its leading figures are Liberal Democrats such as David Laws, EPI’s Executive Chairman, and Cameroon Tories – such as Lord (David) Willetts, a member of EPI’s advisory board - plus one full-power Blairite, Baroness (Sally) Morgan of Huyton, a Trustee of EPI. I see no sign of any guiding figure outside the Blairite-Cameroon consensus on education, which now of course focuses on ‘academisation’ and is dismissive of academic selection.
Its luminaries also include persons closely linked to ‘Academy’ schools and trusts notably Sir Paul Marshall, Chairman of ARK Schools and also Chairman of EPI, and Sir Theodore Agnew, a Trustee of EPI and also chairman and sponsor of a multi-academy trust based in Norfolk. Academies have long seemed to me to be the most dogged opponents of a revival of academic selection, alongside the similar but much rarer ‘Free Schools’ pioneered by Toby Young.
(All these details of affiliations are in fact clearly displayed on the EPI website and can be found there without much effort, which makes the claim of independence even odder.)
I really do not see, especially when grammar schools are concerned, how such a body can describe itself as either independent or impartial. The Lib Dems, plainly the political origin of the EPI, have for years been against academic selection, as have Blairites such as Sally Morgan and Blairite Tories such as David Willetts and Michael Gove ( as discussed in detail here http://hitchensblog.mailonsunday.co.uk/2014/05/more-details-of-michael-goves-attitude-towards-grammar-schools.html0
Academies may be the main foes of Grammar Schools now
The ‘Academies’ which in my view badly need a thorough examination by a truly impartial research body, generally assert that they do not select (always questionable in fact) and generally do not select academically. Their political progenitors and supporters, especially Lord Adonis and Michael Gove, do not approve of academic selection. I am aware of the many anomalous existing grammar schools which have converted to ‘Academy’ status, presumably because this provided them with access to more generous funding, and more control over their own governance, than they had under local authorities. But they are anomalous. New ‘Academies’ do not select academically, and LEA comprehensives which convert to ‘Academy’ status do not select academically either.
My most virulent opponent at the Kent grammar schools debate (about which I wrote here recently) was the head of a local ‘Academy’ school, itself part of one of the many Academy Trusts which now perform the same role in our education system as train operating companies do in the privatised , yet state-subsidised railway system. The taxpayer pays the piper but does not call the tune. She gave the impression that she regarded academic selection with a particular dislike and said she believed that every school could be good, with every appearance of sincerity. When debating the academic selection issue I often have my attention drawn to the ARK school in North London, as a supposed example of success without selection. I have indeed paid it some attention, and will at some point write about it, but this is not the place for that.
Perhaps not as damaging to the grammar school cause as some think
Now to the report itself
It seems to me that it has turned up several pieces of evidence which favour the opening of new grammars, but in all cases it discounts this evidence. Is it right to do so? Is it is critical of evidence which points in the anti-grammar direction? Let us see.
But first some points that I think need to be made, absolutely, before we go any further into this. I strongly believe, and think the evidence shows, that grammar schools lifted an unknown but significant number of bright boys and girls from poor homes out of the narrow worlds in which they might otherwise have stagnated, and into a wider universe where their talents could flourish. Few became famous, but some did - Joan Bakewell (‘lower middle class’ by her own description) , Alan Bennett (son of a butcher), Alan Rickman (son of a painter and decorator) and Raymond Briggs (son of a milkman) spring to mind.
Thousands more (this was before the weird age when millions went to alleged ‘universities’ often in truth nothing of the kind) went on to happy and satisfying careers in the police, the professions and education who otherwise, I suspect, would not have done. The key thing is that these people contributed, to the full, the abilities which they had been given. The whole nation and in my view the world benefited from the resulting talent, competence, responsibility, knowledge and reasoning which they learned in schools. So they were much more use to others, and more fully able to be themselves, in those scores of areas where a truly free society’s fate is decided.
As they go into retirement, and are replaced by the new generation, I personally feel the loss all the time. I simply don’t believe, for instance, that a grammar-educated elite could have messed up our power generation as badly as it has now been messed up.
Thousands of anecdotes, but no actual research, support this impression. The research shows only a steadily growing number of children from grammars and their close allies, the Direct Grant schools (which could not exist without academic selection and died soon after it was abolished) , getting into Oxford, outclassing the ‘Public’ boarding schools as year succeeded year. This stopped around 1970, and the process could only be restarted through what amount to quotas.
This argument about opportunity is important as a counter to the absurd claim that comprehensives have in some way benefited the poor.
Grammar Schools were never supposed to be Fairy Godmothers
But it is not in fact the main or central argument for the restoration of grammar schools. The part of the population which is always bound to benefit most from academically selective schools is the lower middle class and the skilled working class. Why shouldn’t they? It is this class which holds the country together and repeatedly saves the sum of things when the upper crust has run away or failed or shirked its responsibilities. Everyone in the armed services knows this very well.
This group (to which I do not belong but which certainly describes my father and grandfather) cannot afford private schooling now, never will and never could. The current sharp rise in the cost of private schooling now means that much of the middle and upper middle classes are also unable to afford school fees for their own children, even though their (nominally less affluent) parents could afford to pay fees for *them*, a fact which will prove politically important in the years to come.
Yet, partly because of heredity and partly because of nurture, it is bound to contain a significant number of boys and girls who will benefit from an academic education, and won’t get one without academic selection. The seriously and very poor classes will also contain a number ( I would guess a smaller proportion, but here we trespass on areas of nature versus nurture about which there is much dispute) of such children, whose hopes may be dashed very early on by poor home circumstances or inadequate primary schools. An academically selective system, *may* help them. It will certainly help them more than a comprehensive system. But it will often not do so alone, or without other interventions – individual teachers or other non-family benefactors spotting a talent and nurturing it, special help in primary school, even being entered for exams at all, which is currently only universal in Buckinghamshire.
The point I am making here is that the real purpose of academic selection is not social engineering, or the alleviation of class distinction, good as these things are in this case. It is the discovery and nurture of talent for the good of the nation and of the world (and I would add, for the glory of God, whose grief over the waste of talent is so well expressed in Christ’s Parable of the Talents).
The argument about the undoubted benefits to many bright children from poor homes is mainly important as a counter to the lazy and unproven claim that comprehensives somehow helped the poor to get better schooling. I have yet to find research which can show, one way or the other, how much or how little academic selection, between 1944 and 1965, helped children from families who could never afford private schools to rise above their backgrounds.
If I did find such research, I would not solely be concerned with the children of the very poor. What about the children of what might be described as the ‘NCO Classes’ – the children of police officers, non-commissioned officers in the services, state-school teachers, laboratory assistants, junior clerical workers in businesses and town halls, local newspaper journalists, owners of small businesses, managers of small shops, draughtsmen, engineers – people who could never conceivably afford a private education but whose children might well possess talents valuable to the nation which would be undiscovered, neglected or wasted in bog-standard comprehensive schools? Actually, in modern Britain, good private education is rapidly becoming unaffordable to all but the very rich indeed, one of the reasons why this issue is not going away.
Do people who attack Secondary Moderns realise, by the way, that many ‘comprehensive’ schools in poor areas do not have sixth forms or the capacity to teach at ‘A’ level? Knowsley, on Merseyside, has no state school sixth forms.
Disproving a claim that was never made
So let us not try to prove a claim that supporters of grammar schools never made, that they were fairy godmothers capable of transforming the lives of all poor children. They were not. They could not have been. No school system could be.
Even so I think there is good evidence that they were a lot better at finding and encouraging talent than the great bulk of the comprehensive system, which has since 1965 ceded most of the best jobs and careers to privately-educated pupils.
Why GCSEs are a bad measure of a Good School - telescoped results.
Then there is the question of ‘What is a good school’. I have discussed here before the strange collapse of the old GCE ‘O’ level, an exam which lay like a long shadow over my own early teens. I had spent several years at a private prep school preparing for the ‘Common Entrance’ in those days (the early and middle 1960s) a very severe examination taken at 13, requiring the ability to translate from English in Latin and French with huge accuracy and grammatical correctness, large vocabularies in both these languages, a considerable knowledge of arithmetic, geometry and algebra, a detailed knowledge of large parts of English history and of global geography, and an ability to express oneself coherently in writing at length.
Some personal memories of exams as they were before 1967
Your marks in this exam determined what sort of Public (i.e. private) secondary school you would attend, and our teachers left us in no doubt that there was a great and strict hierarchy, in which some schools were far better than others. I was under great pressure to do well, and ended up taking and passing a scholarship exam, so saving my parents a good deal of money on the fees of the rather odd public school where I unexpectedly ended up.
This school had some fine qualities, and others less admirable, mainly stemming from the fact that it was a boarding school - for which I was (especially at that age) temperamentally unfitted. I simply was not prepared to endure so much interference in my private life. So many things which ought to have been voluntary were made compulsory. There was so little privacy. You might have thought we were at war, or being punished for some terrible crime.
Yet I especially recall the very determined and persuasive mathematics teacher who got me through elementary Maths ‘O’ level , a real struggle of sweat and headache and revision; and also the dry, witty man, physically resembling George Orwell, who taught us English literature and Latin. But goodness it was hard work, even more rigorous than the Common Entrance had been. You had to know things and remember them, and write literately and at length about them. And I ended up, despite my scholarship, with (I think) three grade ones (then the top) a grade two, a grade three and a shameful grade six (a bare pass) in my best and favourite subject, History. The last, which would have been a grade one had I been allowed to take it early, as I wanted to, was a symptom of my (by then) rapidly souring relations with the school. But the point is that nobody expected me to get a full set of grade ones, even though I was a scholarship holder in the top stream. A scattering of grades was perfectly normal, and my miserable history score would not have attracted much attention if I hadn’t won my scholarship by a good performance in that subject.
The exam, its shape and general form, would have been quite recognisable to a schoolboy of the Edwardian era or in the 1930s, though I suspect it was less severe than it would have been in either of those eras.
But it would baffle and overpower most modern schoolchildren of that age, and probably leave many of them with Post Traumatic Stress Disorder. No doubt modern exams require a lot of *work*. I would be driven crazy by a GCSE course (Yes, I have seen the papers and coursework) because of this mismatch between work done and knowledge gained. But they are wholly different things.
Why they had to get rid of O levels
And within a few short years of the destruction of grammar schools, almost all state secondary schools began to struggle badly with the ‘O’ level. At first it was watered down and regraded, to make it easier both to pass and to score well. But this was not enough, and in 1988 the O level was finally put out of its misery (by Sir Keith Joseph, of all people) and the GCSE introduced. The EPI report notes on 12 that ‘The introduction of GCSE qualifications with the 1988 Education Reform Act also had significant implications for the selective system. The previous dichotomy in curricula between grammar schools (in which students tended to take O-level exams) and secondary moderns (whose pupils generally took CSEs) was substantially weakened by the replacement of the two qualifications by GCSEs.’
Another way of saying this is to state that all schools now had much lower targets to meet, in which grammar schools were no longer able to show how much better they were than non-grammars ( and also how much better they were than many independent schools, who, being selective, can waltz through these exams by discipline and slog as they could not with the old O levels).
So the current measure of the academic goodness of a school, its performance in obtaining good grades in certain GCSEs, is telescoped. We simply cannot tell from it how much better grammar schools are than the socially-selective comps which appear to rival them in performance. But if we really did restore the old pre-1965 'O' levels, as we should, we should pretty quickly find out. I think the difference would be huge.
It is also a target of no real national importance. It makes no difference to the country how many *schools* meet this arbitrary and not very-indicative score. What matters is how many boys and girls are educated to the full limits of their ability, which is simply not happening now. This is why I am so unimpressed with complaints that grammar schools ‘cream’ the best pupils away from other schools.
What's so bad about 'Creaming'?
Of course they do. That is the point of them.
Bright children are not a resource for boosting a school’s OFSTED rating. They are a national resource, and they are also valuable individuals whose talents should be nurtured to the full for their own benefit. If they do better in academically selective schools than in non-academically selective schools, that is where they should be, and to blazes with the league tables.
So, with that in mind, to the report itself (see above for link)
Beginning on page 6, after a foreword by the impartial and independent Liberal Democrat opponent of academic selection, David Laws, its executive summary notes that pupils travel much further to attend grammar schools than they do to attend comprehensives.
‘Pupils travel, on average, twice as far to attend a selective school as a non-selective
school and a quarter of pupils in grammar schools cross local authority boundaries to attend
(compared to 9 per cent in non-selective schools.) Whilst grammar schools are only found in
36 of 152 local authorities, over 40 per cent of pupils are within a reasonable travel distance
of at least one grammar school.’
Well now, haven’t I always said that the few remaining grammars were subject to a much-increased pressure on places, which is why arguments that they have too few poor pupils tell us nothing about an evenly-spread national system, such as I advocate? One to me, I think.
It's not just grammar schools that select, dear
But, apparently not noticing this or not caring to reflect in its meaning the EPI immediately goes on to labour this point: ‘Pupils who are eligible for free school meals are notably under-represented in grammar schools, with only 2.5 per cent of grammar school pupils entitled to these free meals, compared with 13.2 per cent in all state funded secondary schools, and 8.9 per cent in the areas that they are situated in.’
If you close grammars in poor areas, they won't have many poor pupils. Obvs.
Interesting here to observe that the areas in which grammars are situated have a *much* lower rate of FSM take-up than the country as a whole. What do we learn from this? That spiteful Labour councils in the 1960s and 1970s destroyed almost all the grammars in areas where poor families are concentrated. You cannot blame the grammars for that. Indeed we know from here http://www.suttontrust.com/newsarchive/non-grammars-socially-selective-state-schools/
that elite non-grammars are more socially selective than grammars, and from here http://www.suttontrust.com/wp-content/uploads/2010/03/Worlds_apart.pdf
that the Free School Meals problem is just as great, if not more so, in socially selective comps as it is in the remaining grammars.
Now, there’s another interesting matter here that will re-emerge later. There is, as Sir Michael Wilshaw pointed out in his January 2016 speech much cited here
https://www.gov.uk/government/speeches/ambitions-for-education-sir-michael-wilshaw....a special problem with the children of the ‘White working class’ (his phrase). They are the ones whose performance falls off most sharply between 11 and 15, victims of what you might called ‘subtracted value’ in non-selective secondary schools, where poor discipline and mixed ability teaching cause them to fall behind.
The report notes without comment that ‘Some ethnic groups such as Indian and Chinese pupils are over represented in grammar schools.’
Indeed. And the issue of such communities placing a higher value on education than our own indigenous working class culture, and so helping schools achieve better results, is one of the great unexplored issues of our education system. The secondary-school performance of poor children is *hugely* better in London, increasingly a city of recent migrants, than in the rest of the country. There’s a similar phenomenon, not quite so great but noticeable, in Slough, also an area very heavily populated by young families of recent migrants.
The Great Mystery of the London Effect
This supposed ‘London effect’ is often attributed to various bureaucratic initiatives of the education establishment, such as academisation. But it appears to have predated them. I think we need research, but there is an increasingly good case for suggesting that this is actually caused by the much higher value given to education in these communities, than in traditional British poor areas, where the culture is, to put it mildly, not especially keen on education. Whenever any education statistics are given which involve London, or indeed national averages in which London is included, I think it wise to at least suspect that this influence may be operating.
I discovered this after Alan Milburn, the ‘Equality Tsar’, compared achievement rates among poor children in academically selective Kent with the national average and with comprehensive London. https://www.theguardian.com/society/2016/sep/07/grammar-schools-expansion-disaster-social-mobility-tsar-alan-milburn
In fact Kent’s figures were not significantly worse than those of any other comparable county, and were almost identical to those in comprehensive Wiltshire and Portsmouth. Had London been excluded from the national average, the figures he cited would not have been anything like so striking. Much of Kent, despite its ‘Garden of England’ image , is quite poor and deprived and working class - abandoned coalmines, shutdown naval dockyards and declining Channel ports and seaside resorts - and it suffers from that class’s decline as much as anywhere in the country.
Of course we can agree on this
The EPI report then (rightly) in my view, attributes much of the performance of Grammar Schools to their selection, in the first place, of those best able to take advantage of a good education.
It wrongly (in my view) then spends a lot of time saying that good comps do just as well . as I argue at length above, this is only because of the severely telescoped possibilities of our exam system.
A clue to what might happen if a less-telescoped standard were applied was given in the fascinating revelations, ten years ago, about which state schools were supplying the great majority of Oxbridge state-school entrants. The report from the Sunday Times is behind a pay wall but is referred to in detail here : http://hitchensblog.mailonsunday.co.uk/2015/07/a-new-discussion-about-school-selection-.html
Grammars surpassed even the very best ‘comprehensives’. Oxbridge colleges are still able to select much more rigorously than any other institutions of higher education, thanks to their use of interviews, essays and other means of probing beyond the exam results. Despite being no more socially selective than the top comps, the grammars still do better than them at getting their pupils into Oxbridge. And this is an entirely level contest, as both sorts of school count as state schools for quota purposes.
The EPI seem unaware of, or uninterested in, these facts. They have instead (p.9) invented a different category of so called ‘high quality non-selective schools.’ These are said to be ‘be in the top 25 per cent based on value-added progress measures, and represent good quality schools operating at large scale.’
That London Effect again?
I don’t know how many of this group are also highly socially selective. But I did ask the EPI how many of them are in London. They replied : ‘around 198 schools (in London) out of a total of 653 "high quality non selective schools" in England.’ That means almost a third of these are possibly undergoing the so-called ‘London effect’ which I discussed above and which may have nothing to do with the quality of the schools and much more to do with the culture of the pupils. Given that London is so untypical, and so over-represented in these numbers (London’s population is roughly 8.7 million, that of England , 53 million, much fewer than one third). I think we need to be careful with making too much of these figures. In saying this, I am only adopting the approach of the EPI whenever anything comes up which suggests that grammars may actually be a good thing.
Take this interesting passage on page 7 :
‘At a national level, and adjusting for pupil characteristics, there appears to be no overall impact of selective schooling, either positive or negative. Taking selective areas as a whole, this conclusion applies on average both to children from low and high income backgrounds. This conclusion is likely to disappoint both the advocates of more grammar schools and the critics of selective schools expansion (for example, the present Leader of the Labour Party has claimed that “Grammar schools depress overall educational achievement”1 ). This result suggests that additional grammar schools are not a good intervention for raising average standards across a schools system.
‘However, we do find positive attainment effects for pupils attending grammar schools (adjusting for pupil characteristics). Pupils attending grammar schools achieve, on average, an estimated one third of a GCSE grade higher in each of eight GCSE subjects, compared with similar pupils in non-selective schools in comprehensive areas.’ (My emphasis)
And then, on the same page ‘ This positive attainment effect varies by socio-economic background. For children entitled to free school meals and attending grammar schools the estimated effect is larger than for non-FSM children – at around half a grade higher in each of eight GCSEs. (My emphasis).’
But this of course has immediately to be explained away thus:
‘However, it is important to note that this is based on just 500 grammar school pupils out of almost 90,000 FSM pupils in any one year group. And the characteristics of typical FSM pupils who gain admittance to grammar schools are extremely different from FSM children who do not gain admittance to grammar schools. It is therefore probable that this positive effect is an overestimate and that the real effect of grammar schools on FSM pupils is smaller.’
Perhaps, but this looks a bit like a fact (perhaps unwelcome) being explained away by a guess
Then we get : ‘The gap between all children on free school meals (attaining five A*-C GCSEs, including English and Maths) and all other children is wider in wholly-selective areas than in nonselective areas - at around 34.1 per cent compared with 27.8 per cent., but coupled with the absolutely correct qualification that ‘ This is not surprising because grammar schools attract a larger number of high-attaining, non-FSM pupils from other areas and so, in selective areas, we have a disproportionately large number of high-attaining, non-disadvantaged children, who we then compare to disadvantaged children from across the attainment distribution.’
On page 8 we get : ‘It appears to be more difficult for poor children to access grammar schools, even when prior attainment is taken into account. Pupils eligible for free school meals make up 6.9% of those with high prior attainment near selective schools, but only 2.4% actually attend selective schools.’
Compare this with the Sutton Trust’s 2010 measure (on a different measure of deprivation) that ‘the country’s top 164 comprehensive schools took only 9.2% of children from income deprived homes although they drew pupils from areas where about 20% were income deprived. The 164 remaining grammar schools, also drawing their pupils from areas where 20% were income deprived, were found to be more inclusive, admitting 13.5% of children from poor homes.”
I could go on, and shall return to this report when I have researched its interesting claim that ‘In the late 1950s and early 1960s, new research which highlighted the barriers faced by working class children in accessing grammar schools combined with scepticism regarding the reliability of selection tests meant that the selective school system started to be side-lined to make room for a new, comprehensive system.’
I am seeking this ‘new research which highlighted the barriers faced by working class children in accessing grammar schools’.