Some Thoughts On ‘Some thoughts on education and political priorities’ by Dominic Cummings

This is likely to become a landmark document in literature on education. It is heady stuff.

I have only skimmed the surface of a paper that is festooned with footnotes, charts and hyperlinks. The links themselves, which I intend following up when time allows, will lead into further heady avenues.

Cummings cannot be said to wear his learning or his research lightly. He has a message to get across but he is out to impress along the way. Quite a number of the charts go uncommented and, to be honest, for the purposes of this quick review, I didn’t pay much attention to them. Mea culpa.

Let’s not beat about the bush, this paper is both challenging and a challenge.

It is a challenge principally to politicians and to educators.

Cummings has a pretty poor view of the current state of education.

“The education of the majority even in rich countries is between awful and mediocre.”

What he means by ‘mediocre’ will become apparent as we look at some key quotes from the paper.

Cummings is particularly scathing about the low level of understanding of mathematical concepts essential to modern debate – exponential functions, normal distributions, the bell curve, conditional probability – which he reckons less than 10% leave school having received formal training in.

This is something he returns to again and again.

The ideas presented in his paper are aimed, he says, mainly at 15-25 yr olds. How many 15-25 yr olds are likely to take notice is a moot point. Cummings says it’s the kind of paper he wishes he had stumbled on when he was 15. Which tells us a lot about the kind of 15-year-old he was, and the kind of 15-25 yr old he envisages as his audience.

His vision of an ‘Odyssean’ education comprises seven broad fields, the first of which is Maths & Complexity and (herein lies the challenge) Cummings wants Britain to become ‘the leading country for education and science’.

He thinks 15-25 yr olds today have no real choice. They either continue their education in “dysfunctional” institutions run by “middle-aged mediocrities” or become despised and unemployed.

He quotes E. O. Wilson explaining the importance and effectiveness of teaching from the perspective of Big Questions.

He acknowledges his lack of qualification to write such a paper: “This paper is very crude and written by someone with no relevant expertise in any subject except politics – it is a bad version of something I wish I had been given aged 15.”

He writes at length about the successful ways in which mathematical models have helped us predict (for example in weather forecasting) the behaviour of complex systems.

“This combination of a) precise real world data, b) mathematical modelling software that captures dynamic processes, and c) the potential for simulations of large numbers of virtual samples that vary according to real world data, has the potential to revolutionise our understanding, prediction, and control of complex systems and improve how we cope with failure.”

“However, there is also a danger in the use of statistical models based on ‘big data’ analysis – ‘overfitting’ models and wrongly inferring a ‘signal’ from what is actually ‘noise’. We usually a) have a noisy data set and b) an inadequate theoretical understanding of the system, so we do not know how accurately the data represents some underlying structure (if there is such a structure).”

This he believes is particularly the case when it comes to Education.

He extols the advances in “biomedicine” and believes GM crops could help feed the world.

The aspect of Cummings’ paper which has been most leapt on by the press and most misreported is what he has to say about genetics. A matter which will not have surprised him:

“Those working at the cutting edge in genetics are understandably reluctant to involve themselves in contentious debates and partly because, for understandable reasons given our history, people are wary of discussing the importance of genes in explaining human behaviour.”

He makes much of

“‘Missing heritability’… Twin and adoption studies have showed that many fundamental human qualities are at least 50% heritable (i.e at least half the variance between individuals is caused by genes): e.g. ‘g’ or ‘general cognitive ability’ (~60-80% heritable for adults) and some personality characteristics. After the success of the Human Genome Project over a decade ago, there was much hope of finding ‘the gene for X’. However, while some rare conditions and disorders (including learning disorders) are caused by a mutation in a single gene, most common traits and disorders (including learning disorders) are caused by many genes with small effects, making them hard to find.88 This became known as ‘the missing heritability problem’: we know that genes are responsible but we cannot find many of the specific genes. The combination of whole-genome sequencing and ‘Genome-Wide Association Studies’ (GWAS) is likely to allow us to make progress in finding this ‘missing heritability’.

When he reaches the section on Education he repeats his emphasis on the importance of Mathematics:

“Unavoidably, the level of one’s mathematical understanding imposes limits on the depth to which one can explore many subjects. For example, it is impossible to follow academic debates about IQ unless one knows roughly what ‘normal distribution’ and ‘standard deviation’ mean, and many political decisions, concerning issues such as risk, cannot be wisely taken without at least knowing of the existence of mathematical tools such as conditional probability.”

One smiles to think of the frustrations he must have experienced when debating and discussing such matters with political and civil servant colleagues who did not have the ‘required’ level of understanding.

“There is widespread dishonesty about standards in English schools, low aspiration even for the brightest children,118 and a common view that only a small fraction of the population, a subset of the most able, should be given a reasonably advanced mathematical and scientific education, while many other able pupils leave school with little more than basic numeracy and some scattered, soon-forgotten facts.”

After spending more than 36 years in education I have to agree with Cummings’ comments on the lack of objective scientific research.

“There is hostility to treating education as a field for objective scientific research to identify what different methods and resources might achieve for different sorts of pupils. The quality of much education research is poor. Randomised control trials (RCTs) are rarely used to evaluate programmes costing huge amounts of money”

He is clear where blame lies.

“Westminster and Whitehall have distorted incentives to learn and improve, have simultaneously taken control of curricula and exams and undermined the credibility of both, and have then blamed universities for the failures of state schools and put enormous pressure on Universities and academics not to speak publicly about problems with exams, which has made rational discussion of exams impossible.”

Here he is again bemoaning the level of mathematical understanding. We can choose to argue with Cummings about how important this knowledge is, but he’s probably correct in saying that only 20% of students currently leave school in command of these concepts.

“Because of how courses have been devised, ~4/5 pupils leave England’s schools without basic knowledge of subjects like logarithms and exponential functions which are fundamental to many theoretical and practical problems (such as compound interest and interpreting a simple chart on a log scale), and unaware of the maths and physics of Newton (basic calculus and mechanics).

Supporters of Sure Start and similar programmes will not like his assertion that billions of pounds have been wasted (and the “no real gains” he talks about are only one kind of gain), but there can be no doubt that there has been a woeful lack of control trials to validate government spending and that many interventions successful in one context have been enthusiastically copied and have had funds lavished upon them with very little generalised benefit.

“Overall, there is great political pressure to spend money on such things as Sure Start but little scientific testing, refinement, and changing budgets to reinforce demonstrated success, therefore billions have been spent with no real gains. The billions now spent in Britain should be tied to ‘randomised control trials’ and there needs to be more experimentation with how to expand experiments that seem to work (lots of small things work because of specific individuals and therefore do not work when expanded).

I agree with the following.

“Some argue that ‘rote learning is a waste of time /damaging.’ … lots of things require practice: who thinks that teaching the commutative law of addition would produce successful children without practice? Moving things from the short-term memory (which is extremely limited even for the cleverest) to the long-term memory needs building proteins in the brain and this needs practice (except in extremely unusual people who instantly remember things permanently). [My emphasis]

There has been a marked reduction in repetition and practice in primary education. The way the curriculum has been constructed (particularly in the teaching of Maths) has meant that lessons across a period of time move on at a pace that does not allow sufficient consolidation for concepts to become embedded in long-term memory.

“In 1993, Ericsson et al published research purporting to show that ‘10,000 hours of deliberate practice’ is what distinguishes ‘experts’ in areas such as music and chess, and that ‘innate ability’ is not relevant. ‘Our theoretical framework can also provide a sufficient account of the major facts about the nature and scarcity of exceptional performance. Our account does not depend on any scarcity of innate ability (talent)… [I]ndividual differences in ultimate performance can largely be accounted for by differential amounts of past and current levels of practice.’ This view had a large influence on the media and various books (e.g. Gladwell’s ‘Outliers’, Shenk’s ‘The Genius in All of Us’, Syed’s ‘Bounce’, and Brooks’ ‘The Social Animal’) have promoted the idea that people require the same amount of practice regardless of talent and that ‘10,000 hours’ is a sort of magic number – put in those hours and you too can be great, don’t worry about your genes.

That may be, but the notion that practice makes perfect remains a good one for educationists to espouse. I remember basing a school assembly on this 10,000 hours theory. For young children the notion that you become good at something by Doing It again and again is certainly one that is worth emphasising, notwithstanding the following:

“Recent analysis of these studies has confirmed that such claims are greatly exaggerated: ‘deliberate practice’ does not account for most of the variance in performance (only about a third of the variance in chess and music), and ‘some people require much less deliberate practice than other people to reach an elite level of performance’.

That in itself is something that is demonstrably true to children in their own experience and can become demoralising when they realise they require more practice than others.

Cummings looks to a future when advances in genetic and scientific research combine with the power of computers to result in a truly personalised education.

But personalised education will produce inequalities that do not sit comfortably with current educational “closing the gap” dogma.

“It is reasonable to hope that the combination of 1) finding the genes responsible for cognitive abilities, 2) scientific research on teaching methods, and 3) the power of computers to personalise learning will bring dramatic improvements to education – but this will not remove genetic influence over the variation in outcomes or ‘close the gap between rich and poor’. ‘The good school … does not diminish individual differences; it increases them. It raises the mean and increases the variance’ (Elliot Eisner, Stanford). Good schools, in the sense of ‘teaching children of different natural abilities as well as possible’, will not ‘eliminate gaps’ – they will actually increase gaps between those of different abilities, but they will also raise floors and averages and give all children the opportunity to make the most of their genetic inheritance (personality as well as IQ)

Cummings advocates an exclusive educational roadmap for the top 2% IQ, especially at university level. “We should give this ~2% a specialist education as per Eton or Kolmogorov, including deep problem-solving skills in maths and physics.”

Presumably based on what he perceives as a ‘bell-curve’ of human aptitudes, he has this to say about ambitions to make all teachers brilliant teachers.

“Only a small proportion of the small number of ‘significantly more than averagely talented people’ are, without huge cultural changes, going to want to be teachers (at least in the conventional career sense).

This is a breathtaking conflation of brilliance in teaching with brilliance in IQ. In other words, teachers are not going to be IQ high-fliers, and so they are not going to be brilliant teachers. That is manifestly nonsense, but it leads Cummings to conclude that the best that can be done is weed out the bad:

“We should give heads power to remove who they consider to be poor teachers; we should let schools experiment with rigorous methods of judging teacher effectiveness

And be a little more realistic about what can be achieved:

“Perhaps we should also stop discussing schools as if we are going to have a quarter of a million ‘brilliant’ teachers and instead think about what to do with tens of thousands of roughly averagely talented people… While some children will always be blessed by a brilliant teacher, by definition this is not a scaleable solution to our problems: real talent is rare, mediocrity ubiquitous.

When he talks about “mediocrity” it becomes clear from the above that he is not making a (scathing) judgement on current teaching quality, but making the philosophical point that at any one period of time the vast majority of teachers will be average (satisfactory = mediocre) rather than above average (outstanding = brilliant).

This acknowledgement/acceptance has repercussions for curriculum design and pedagogy.

“Whilst heads need to be flexible enough to allow talented people to experiment, we also need schools in chains that spread proven approaches (and ‘90% solutions’) without relying on innovation, inspiration, and talent. ‘Direct Instruction’ (DI), in which teachers follow tightly constrained rules to deliver lessons, is generally ignored in English education debates despite its proven effectiveness in randomised control trials around the world. However, standards might improve substantially if thousands of roughly averagely talented teachers simply used DI instead of reinventing square wheels.

Isn’t that, in large part, what we have had already, in terms of the National Curriculum, the Literacy and Numeracy hours, the Phonics Programme etc. Teachers have for a long time been working to a prescribed script not far off from Direct Instruction.

“Added value’ was introduced in 2002 on the basis that ‘raw’ attainment figures are ‘largely an index of the calibre of pupil intake rather than any indication of the school’s performance’. Although it was intended to correct for natural ability, analysis of twin studies shows that ‘about half of the variance of corrected-school achievement is due to genetic differences between children’ – that is, even stripping out g or previous attainment does not strip out all genetic influence in general; ‘achievement independent of ability may be just as heritable as achievement including ability because achievement is as much a function of genetically-driven appetites [such as motivation] as of aptitudes’.

I think that’s a very helpful observation, and goes part way to explaining why some schools are so often in and out of ‘special measures’ based on raw data.

‘Instead of thinking about education as a way of countering genetic differences among children, the field of education might profit from accepting that children differ genetically in how and how much they learn. This way of thinking is compatible with the current trend towards personalizing education by optimizing children’s learning, which is increasingly possible through the use of interactive information technology

I don’t have a problem with that. The use of the word ‘genetically’ WILL be a problem for some. So let’s repeat the essence of what Cummings says in that last quote, simply removing the word ‘genetically’: “The field of education might profit from accepting that children differ in how and how much they learn.”

In schools where the overarching educational effort of ‘management’ is for all children to make the same amount of progress each year (4 National Curriculum points, whatever) and to achieve the same outcomes this is a message that needs to be frequently raised by those who are actually doing the teaching.

I don’t think Cummings has any idea what makes a teacher ‘brilliant’. There are many more brilliant teachers amongst the mediocre hordes than he is willing to credit.

I hope some of them will be prepared to engage with the ideas in this provocative paper.

http://www.theguardian.com/politics/2013/oct/11/gove-adviser-thesis-inflammatory-ideas-education