Poorer pupils less likely to get into grammars than richer classmates

Roger Titcombe's picture

This Guardian article reports research arguing for the existence of a class-based conspiracy to prevent the children of poor families getting into grammar schools. There are many very good reasons for opposing the re-introduction and expansion of grammar schools, but this is not one of them.

“High-achieving children from disadvantaged backgrounds who perform well at primary school have less chance of getting into a grammar school than their more affluent classmates who perform less well, according to new research. The study, by a team from Bristol University, Warwick University and University College London, says access to grammar schools is “highly skewed” by a child’s socio-economic background.”

The ‘Socio-Economic Status’ (SES) analysis is very thorough, but the conclusions of the study are flawed because of the assumption that KS2 SATs data provide a valid measurement of academic ability.

“Let’s look at two children – one from the poorest SES quintile and one from the least deprived SES quintile – both performing at the 80th percentile of the Key Stage 2 distribution. Despite the same level of academic attainment, our analysis shows that the most deprived pupil has only a 25% chance of attending a grammar compared to a 70% chance for the least deprived pupil.”

But children are not admitted to grammar schools on the basis of ‘academic attainment’ measured by KS2 SATs. The 11+ is an IQ test. It is designed to be a snapshot of general intelligence, which is a far better predictor of ultimate academic attainment than curriculum-specific attainment tests, of which SATs are an example.  Grammar schools have always done this because within the educationally flawed ‘fixed intelligence at birth’ paradigm, it works for them.

The 11 plus is not the only IQ test widely used in the English education system. There are also Cognitive Ability Tests (CATs). These are currently provided by GL Assessment, a commercial company that doesn’t like the CAT being described as an intelligence test. CATs are used on a large scale by many Academies and Multi-Academy Trusts as the basis of ‘Fair Banding’ admissions systems, which successfully ensure that comprehensive schools really are comprehensive in that they contain a balanced mix of pupil abilities. CATs test data produce the same social pattern as the 11 plus results, in that the lower bands are dominated by pupils from poor SES families inviting the same false conclusion that the schools that use the tests are somehow discriminating against children from ‘disadvantaged’ backgrounds.

The fact is that poor SES postcodes are characterised by a significant excess of children with lower average IQ test scores. The 11 plus and Y6 CATs are simply the evidence that this is the case.

Despite the objective truth of this relationship and its profound implications for the national education system, rarely is it admitted, let alone taken into account, even in research carried out by prestigious university institutions like Bristol, Warwick and UCL.

It is why ‘competition’ for pupils in our marketised education system has morphed under commercial institutional pressures to become competition for the most intelligent pupils, in which schools located in posh areas with higher concentrations of more intelligent pupils have an inherent league table advantage over schools in the middle of poor SES council estates, even though the quality of teaching and learning in the ‘posher’ schools may well be poorer.

These facts and their implications are discussed in Part 4 of my book, ‘Learning Matters’, which is about the success of Mossbourne Academy and the Hackney Borough secondary education system and in this article.

There are three types of relationships that secondary schools have with IQ-based admission tests.

Grammar schools – Where the tests are used to ensure that the pupils they admit are intelligent enough to succeed despite often mediocre standards of teaching and learning. There is an important secondary advantage of ‘keeping the riff-raff out’ and so re-assuring snobby parents that their children will not suffer from too close contact with lower class children. This is the real reason for expensive uniforms and rituals that have little beneficial effect on learning.

Fair-banded comprehensive schools – Where the tests are used for the opposite and educationally desirable purpose of achieving mixed ability, diverse social class/ethnicity intakes.

Neighbourhood comprehensive and secondary modern schools – Where there are no admission tests, resulting in the well established pattern that the schools in ‘posh’ areas become successfully established at the top of local school league tables and the schools in poorer areas become labelled as ‘failing’ and drift towards the bottom, in constant fear of OfSTED and academisation. If they have already been academised and fail to ‘improve’, as is likely if they fail to address the low mean cognitive ability of their pupils, they face being ‘transferred’ to whichever MAT can be persuaded and/or bribed by the DfE to take them on. Curiously in our marketised education system ideologically designed to support parental choice of schools, parents have no say in this process whatever.

It is the middle group; comprehensive schools with all ability, mixed social class/ethnicity intakes, that are best placed to successfully confront the flawed ‘fixed intelligence at birth’ paradigm and so transform the life chances of all of their pupils including those from poor communities with low intake IQ scores. This is something that the former Chief Inspector, Sir Michael Wilshaw, well understood even if he was wrong about so much else, including his obsessions with academisation, school uniforms and obsessive discipline.

Grammar schools recruit pupils of IQ 115+ (more than 1 SD above the mean) so there is little need for them to further develop the intelligence of their pupils in order to retain their league table superiority and social class differentiation compared to neighbouring non-selective schools. While some of the best grammar schools may be challenging their already able pupils to further develop their general cognitive levels, the popularity of selection with the parents of 11 plus successes provides no incentive to change traditions of complacently poor practice in many of these schools.

Comprehensive schools/secondary moderns in poor areas are always fighting against the market pressure of league tables and the ever present threat from OfSTED regimes driven by ‘floor targets’ and dodgy performance measures.

What is the educational solution for schools located within low SES communities?

Urban schools can move into the second category by adopting the Hackney solution of CATs driven banded admissions systems. This does not prevent the operation of the market competition, but it does place limits on the numbers of pupils that can be admitted to each ability band. For example Mossbourne Academy has four admission bands based the quartile-defining CATs scores: Band A 110+, Band B, 100 – 109, Band C 90 – 99 and Band D 89 and below. At Mossbourne there are 50 places in each band in each new Y7 intake.

The poor SES Mossbourne local community results in a large excess of applicants for Band D, while Band A can be ‘topped up’ with pupils from more affluent areas further from the school or in neighbouring Boroughs. The unsuccessful Band D applicants will find places in nearby schools operating the same LA managed system. In the absence of banded admissions Mossbourne would fill up with low SES Band D pupils that live closest to the school, with no spaces left for the more able pupils that live further away. This is what sunk Mossbourne’s predecessor, Hackney Downs School.

Under the Hackney banded admissions system all of the schools are protected from becoming low ability sink schools. Within the Borough the more popular schools can still attract a higher mean IQ admission cohort and so, other things being equal, secure ‘better’ GCSE results, but not by much, which is probably the best that that can be achieved in a marketised system.

The result has been the huge improvement in outcomes for all pupils in the Hackney secondary system. It is only possible because the Hackney Academies are far sighted enough to recognise the advantages to all of the schools of ceding their Academy-based admissions powers to the Local Authority. This is regrettably, so far a unique development.

More rural areas have schools with large natural catchments that are inherently reasonably mixed in terms of the ability/social class profile. Banded admissions would bring few advantages to such schools with the major disadvantage of unsuccessful applicants being faced with having to attend schools many miles from their homes.

Inner urban comprehensives still have to work hard to combat cultural challenges to the establishment of a mutually supportive and vibrant learning community of co-operatively assertive pupils. This article explains how this was achieved in my inner-urban headship school from 1989 until my retirement in 2003.

What is the explanation for the Bristol/Warwick/UCL study conflating KS2 SATs driven ‘academic attainment’ with IQ/cognitive ability?

Poor SES communities characterised by low average IQ children are served by primary schools that are permanently threatened with closure (and the sacking of the head) for failing to meet floor targets. Such schools can fall victim to educational snake oil salespersons that promote behaviourist, knowledge-based, abusive discipline approaches, that the Conservative Party, the DfE and Donald Trump all love.

These methods do not develop cognitive ability or much else besides Gradgrindian conformity and obedience. They may actually make children dimmer and therefore less likely to do well on IQ tests used for secondary school admissions including grammar schools. Their methods work with SATs but not CATs or other IQ tests. Hence the illusion of ‘high academic attainment’ that does not translate into passing the 11 plus. Worse still, there are now secondary Academy Chains and individual schools that use these behaviourist methods to cram pupils for GCSE, with the same claims for success. See this article and my comments to it.

In our primary and secondary schools the DfE seems determined to make the quality of teaching and learning worse and the experience of our children and their teachers even more dismal.

In summary, grammar schools do not discriminate against children from poor backgrounds. They just insist on an IQ test threshold for admissions. It has long been established that the children of well-educated parents do better at school. The CATs data show that, whether by nature, nurture or both, such parents tend to produce higher IQ children. The higher the level of parental education, the less likely are the families to live in areas characterised by poor SES. They can afford to move ‘up the housing ladder’, so they do. Therefore high IQ children will be scarcer on the ground in low SES areas resulting in the pattern observed in the research.

It was not the case in the 1960s where I grew up.

It is worth pointing out that this pattern was hugely intensified by Margaret Thatcher’s disastrous policy of selling council houses at a huge discount to tenants. Fifty years later a high proportion of these well built and roomy homes with front and rear gardens are now owned by private landlords resulting in a historically low level of home ownership. Increasingly poor and aging children are still living with their parents. We children of the 60s were launched into a high degree of independence in our late teens/early 20s to enjoy swinging Britain, free university education supported by maintenance grants (not loans) and full employment in proper pensionable jobs.

There is therefore no class-based discrimination in admissions to grammar schools, just ongoing market based corruption and degradation of the primary school curriculum combined with costly, ideologically driven fragmentation of the secondary school system that damages the opportunities of all pupils to attend inspirational schools  that prioritise the development of individual cognitive and other abilities through co-operative endeavour.

Academisation, Free Schools and now this proposed reintroduction of grammar school selection to those parts of England that are well rid of it, will just waste £millions when per-pupil funding of established and successful comprehensive schools is being drastically cut for the first time in many decades.

I set out a step by step plan to really raise both educational attainment and the mean IQ of our school leavers here. The latter is the essential prerequisite for the former and is essential if our democracy is to reject the wave of economic and political populism that is threatening all of our precious public services.

I welcome comments to all my articles including from those that disagree with me.

Share on Twitter Share on Facebook

Be notified by email of each new post.


agov's picture
Tue, 14/03/2017 - 10:44

"I welcome comments to all my articles including from those that disagree with me."

Hmm. Up to a point perhaps.

Not sure you're being entirely fair to this study. I couldn't find any allegation of conspiracy in the study or the article.
Also, it wasn't a study into grammar schools but into partially selective schools. The report states they looked at "secondary schools which admit children to Year 7 at age 11 using some form of ability or aptitude assessment" but not into fully selective schools albeit that it sometimes refers to grammar schools.
Although there may be a degree of confusion within the report it is not as unaware of limitations and distinctions between SATs and other selection methods as you suggest. They are reasonably clear that they don't really know what schools mean by selecting on 'academic ability' nor how it is assessed. They say that as "selection test data is not available, we are unable to tell which pupils in the cohorts studied entered via a selective route" and that "Government should consider requiring all selective and partially selective schools and local authorities to make 11+ test results available for research purposes".
None of which necessarily invalidates the factual claims they actually make.

Roger Titcombe's picture
Tue, 14/03/2017 - 11:52

Hello again agov. This is what the article states.

"Our analysis shows clearly that it is mostly the very affluent that make it into grammar schools – and that there is a dramatic difference in access to selective schools depending on the pupil’s background."

The second part of the statement is seriously misleading. It implies that pupils applying for access to grammar schools are disadvantaged by their home background. They are not. They are disadvantaged by their IQ as measured by the 11+ test.

It is easy to distinguish between the two factors.

1.  Would a child from a poor SES background with a high 11+ test score get in? - Yes

2.  Would a child from an affluent background, but with a low 11+ test score get in? - No

Therefore access to grammar schools is not on the basis of Socio-Economic Status as most readers of the Guardian article would assume to be the outcome of the research.

Please correct me if I am wrong but no-one except me appears to be pointing out the error in the Guardian interpretation - hence my article.


agov's picture
Wed, 15/03/2017 - 09:21

I would agree that there is some confusion in how they express themselves and in that they frequently refer to grammar schools having explicitly said the report is not about fully selective, i.e. grammar, schools. Nevertheless I take your quote ("Our analysis shows...") to be entirely consistent with your argument that intake tends to be from the more affluent as they are more likely to have higher IQs. In fact they seem very confused about what their report is actually demonstrating but they do not claim there is any conspiracy against the poor albeit the report may well prompt such a supposition from Guardian readers.

Roger Titcombe's picture
Wed, 15/03/2017 - 11:11

The more important issue raised by me, but ignored by the Guardian and the mainstream 'liberal left' as well as the authoritarian right, is the conflation between high Y6 attainment (as judged by SATs and OfSTED) and intelligence/cognitive ability. As you know I argue that schools at all Key Stages should be prioritising the development of the latter. I am encouraged by the emergence of the 'Growth Mindset' movement that rightly rejects the notion of fixed intelligence and attainment potential, but depressed by its corruption by the 'Gradgrind' Academy/Free School/Hirsch tendency into a justification for creating a school culture of 'anybody can achieve anything if they work hard enough' supported by abusive, authoritarian disciplinary regimes.

I discuss in this article


Neil Postlethwaite's picture
Sat, 10/06/2017 - 14:48

Some further background info I came across when listening to a podcast when walking the dog.

Although far from firm evidence, there is some compelling/growing research in the area and if nothign else it makes some good reading and food for thought on why poor kids are less likely to success coming back to it all being about their parents again, and not really the money.


Add new comment

Already a member? Click here to log in before you comment. Or register with us.