North-south schools divide not supported by evidence

Roger Titcombe's picture

This is the title of a Guardian article which appeared on 11 September 2018. The following (in italics) are excerpts.

Major study of 1.8 million pupils also challenges ministers’ claims children do better in academies and grammars.

Claims that schools in the north of England are worse than those in the south are based on myth and bad data, according to a large-scale research project that calls into question the education policies of successive governments.

The study also challenges the idea that selective grammar schools or academies are more likely to improve pupil progress overall than community comprehensives, tracing the progress of 1.8 million pupils, their social, family and economic backgrounds and the type of schools they attended.

The analysis of three annual cohorts of 600,000 pupils each was carried out by Prof Stephen Gorard, director of the Durham University evidence centre for education, who says he found no evidence that schools in the north or north-east are differentially effective or ineffective with equivalent pupil intakes.

This is important support for my arguments that the ‘attainment gap’ is an illusion and that northern schools are being unfairly criticised for low educational standards compared to schools in the south of England. The ‘north/south attainment gap’ claimed to exist by The Sutton Trust, The Social Mobility Foundation, the DfE, the National Schools Commissioner and virtually the entire English educational establishment is a fallacy. The actions taken by the government to ‘close the gap’ through market pressure on allegedly under-performing Northern schools from league tables and OfSTED  is counter productive and having the opposite effect to that which is intended. Professor Gorard’s conclusion is that the exam results of northern schools compared to those in the south accurately reflect the intake characteristics of those schools. He writes as follows.

“What my new analysis suggests is that schools in some areas are not doing a worse job, they simply do not have an equivalent mix of children,” he says. “The most important factor that determines school test and exam results is not the quality of teaching or leadership but who they teach, the proportion of pupils who are disadvantaged through poverty, family circumstances or special educational needs and most crucially the length of time they have been disadvantaged.”

Although Professor Gorard is only half right, this half which is correct is of enormous importance. For decades, both Labour and Conservative governments have dismissed such arguments as ‘making excuses for failure’ and that what is needed is for schools to appoint strong ‘Executive Principals’ to independent, market-oriented Academy and Free Schools, freed from the namby-pamby interference of Local Authority Education Departments and their ‘inspector advisers’ who have wasted their careers studying how children learn at university and applying it in their often considerable teaching experience. Professor Gorard has found that, contrary to decades of pro-Academy propaganda in which OfSTED has been deeply complicit (see Section 3.2 of my book, ‘Learning Matters’), these schools have performed no better than more poorly funded and resourced LA schools, and in some cases, spectacularly worse. The following is where Professor Gorard has, in my view, been misled.

“The most important factor that determines school test and exam results is not the quality of teaching or leadership but who they teach, the proportion of pupils who are disadvantaged through poverty, family circumstances or special educational needs and most crucially the length of time they have been disadvantaged.”

 It seems to be a classic example of, ‘putting the cart before the horse’. Schools that get poorer aggregated GCSE results do indeed take in a higher proportion of disadvantaged pupils, but in terms of school attainment, they are disadvantaged by lower cognitive abilities. It is not that poverty reduces cognitive ability, but that lower parental cognitive ability results in poverty, and damaged family circumstances and special educational needs and lengthens the period of time for which cognitive development is likely to be stunted.

 I grew up on a huge 1950s South Birmingham Council Estate covering several square miles. Nobody told me I was a social housing child and there was no suggestion that living in a council flat disadvantaged me at school.

 I can see why any researcher would be led into this error. Whereas Socio-Economic Status (SES) data is everywhere, cognitive ability data is hard to find: hence there are any number of studies based on SES, but very few based on cognitive ability. But, the data are out there. The largest database is currently owned by the private company, GL Assessment,  which markets the Cognitive Ability Tests (CATs). This company also acquired decades of data built up by NfER-Nelson, which previously marketed the CATs. GL Assessment is not bound by the Freedom of Information Act, and keeps its data secret. However, a valuable and highly illuminating nugget was posted on-line in 2010.

P10 of this report features valuable data giving mean cognitive ability scores linked to SES, ethnicity and special needs data.

During my Cumbria headship from 1989 until my retirement in 2003, I was  part of an LEA working group that administered the Y7 CATs testing of all Cumbria pupils. The LEA produced this chart that plotted the GCSE performance of the school against the mean intake CATs score. (I first published this chart elsewhere, changing the name of the county and my headship school.) The key features of the chart are the high correlation between intake CATs scores and GCSE performance and the fact that from local knowledge it was clear that the position of the school on the CATs score axis reflected the relative affluence of the school catchment.

In the last six months John Mountford and I have been researching the link between Free School Meals eligibility and SATs and CATs scores. CATs data have been hard to come by in the north of England as there are comparatively few schools that take them. However John was able to find a lot of data from schools along the generally prosperous M4 corridor. The relative affluence of school catchment areas makes little difference to the general pattern, which is that SATs scores are inflated compared to CATs and especially so in schools that have high proportions of FSM pupils. It was also almost always the case that the CATs scores compared to SATs, for FSM pupils were lower (and often much lower) than for non-FSM pupils. This is discussed in my article refuting the claims made by Dr Rebecca Montacute of the Sutton Trust. Some of the SATs/CATs data, that we obtained from schools using Freedom of information, are set out in this article.

Unsurprisingly, describing a cognitive ability deficit as a ‘disadvantage’ is discomforting, but cognitive ability can be raised, which is the hugely optimistic consequence of the recognition of the plasticity of general intelligence.

Professor Gorard concludes as follows.

“The data shows that despite all these whizzo ideas for subsets of schools, underneath it all we have reasonably good teachers and reasonably good schools across the country. Once we know the effect of long-term disadvantage on attainment, then we can direct money to help those children get to what should be a universal entitlement, for example, based on the length of time they have been eligible for free school meals.

“We have to determine through evidence what works best and provide it for everyone.”

Certainly we need to direct resources to combat disadvantage. But this means changing the priority of teaching and learning in all schools and for all pupils, from hitting exam results targets for league table purposes or to avoid falling below high stakes ‘floor targets’, to enhancing cognitive ability, targeting especially those pupils most in need of general cognitive gains.

This article describing a breakthrough in Finland shows what could be achieved.

Returning to the Cumbria LEA policy of Y7 CATs testing for all pupils, this was introduced in 1990 as part of the Local Management of Schools (LMS) formula. This contained a provision for ‘Non-statutory Special Needs’. All LEAs then provided significant extra funding to support pupils with learning difficulties short of those requiring a formal Statement. Unlike other LEAs this element of the Cumbria formula was driven not by FSM eligibility, but by CATs scores. There was a sliding scale of ‘pupil premium’ for all pupils with CATs scores of less than one Standard Deviation below the mean of 100 (85). A high proportion of these pupils were eligible for FSM and had other SES issues, but it was the low CATs scores which the school could directly address that resulted in their SEN, not their SES issues that the school had little influence over.

CATs scores can also draw attention to ‘specific learning difficulties’ like dyslexia. Pupils where there were large differences in scores on the three CAT sub-tests were always further assessed by our SENCO, supported by excellent LEA SEN advisers. This often led to Statements for specific learning difficulties. The use of FSM as a proxy for low CATs scores has driven the ‘attainment gap’ fallacy that Professor Gorard has only partially debunked.

Teaching approaches for the acquisition of deep understanding and cognitive development are well established in the UK. However, they are not promoted by the DfE, nor are they adopted by most Academy and Free Schools. Part 5 of my book, discusses many of them. I also recommend the ‘Lets Think‘ and ‘Slow Education‘ websites.

Share on Twitter Share on Facebook

Be notified by email of each new post.


John Mountford's picture
Tue, 18/09/2018 - 10:29

I wrote to Professor Gorard recently. I indicated my general agreement with the main findings of his research. However, there was an important issue I felt was overlooked in the final analysis of the data.

The issue to which I refer concerns the description of what constitutes an 'equivalent mix of children' when comparing schools. Whilst I accept the description of what disadvantage means in an educational context, I believe that a more complete description must include reference to cognitive ability.

I accept the very act of referring to an intelligence quotient is at best questionable for many. For some it is a 'no-go area'. But, like every other human trait and capability, intelligence has a genetic component. Even the most ardent skeptics can agree to that extent. My feeling is that the notion that something as complex as intelligence can be reduced to a set of digits may seem far from helpful. But, by giving it a value it enables those of us who are open-minded enough to discuss it, the opportunity to use IQ as a convenient short-hand in doing so.

The fact is, whether you agree with the practice of segregation by ability or not, generations of children have been treated thus on the outcome of IQ testing. I failed my 11+ and moved on to a technical grammar school in the Rhondda Valley so have personal experience of how the process is fundamentally flawed. Used in the way it has been has caused problems for generations of people like me who took longer to blossom and reach their true potential. My greatest objection to the practice, however, is the belief that IQ testing is a dependable tool for predicating future outcomes in isolation from other factors. I maintain the same level of skepticism still but cannot dismiss the idea that some measurement of cognitive capacity can and does help produce a more complete picture of educational deprivation.

It is my belief that, if combined with Professor Gorard's latest findings, evidence from the small scale research project you and I have been engaged in recently, Roger, presents a more complete picture of deprivation. It offers a new insight to evaluate the alleged North/South divide and offers an opportunity to transform the national education agenda. To date we have failed to attract the attention of other researchers with a view to expand the research. Having got close to getting backing from a number of academics to take our research forward, interest and support have been withdrawn. It seems most of academia is still too uncomfortable with looking too closely at whether and how aspects of cognitive capacity are influenced by our genes. I hope readers will note that I have used the word influenced rather than determined because that is where the growing body of evidence is taking the debate if we allow it to take place.

Roger Titcombe's picture
Tue, 18/09/2018 - 11:00

John, you are right, but nothing changes in practical school-based issues like those tackled by Professor Gorard whether the inherited contribution of cognitive ability/IQ is 0.25, 0.50 or 0.75 and all but the purest Marxists accept that the truth almost certainly lies in this range.

As for the penicious effects of the 11+, you are so obviously right and we must all know lots of of people who failed it who have gone on to worthwhile professions and/or done amazing things.

As you say, it is important to point up the advantages of CATs testing. Unlike the 11+ there is absolutely nothing judgemental or life restricting about these tests as they are neither qualifications nor high stakes hurdles of any kind. On the contrary they are powerfully diagnostic of indivual learning needs, while also providing mechanisms for fair and balanced admissions and as the basis for statistically sound approaches to comparing educational effectiveness.

That is not say that raising the general cognitive ability of society is not important, and never more so than at the present. See

Add new comment

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