Genetics and Education: Examining the arguments


17th January 2018

A frightening proposition

Four years ago I sat in the House of Commons being quizzed by the Education Select Committee. Sat next to me was Professor Robert Plomin of Kings College London. Much of what he said about the relationship between genes and educational attainment seemed to threaten my deeply held beliefs about education – if not humanity. I’ve always believed that differences between pupils are largely down to opportunities rather than anything innate. It’s a belief that I think most teachers, parents and people working in the sector share, yet I knew so little about the science of the matter that there was little I could say in response.

Three and a half years later, all hell broke loose over similar questions raised by Toby Young: speaking at a Teach First conference he questioned how much schools could do about social mobility given recent research on education and genetics. His comments were published and then deleted by Teach First.

Whilst the whole thing left me feeling queasy, I couldn’t help but think a serious response to the science was needed rather than (or at least in addition to) the blunt deletion.

Fortunately, a few days after the whole sorry debacle, the TES published a response by Dr Kathryn Asbury. Asbury co-authored “G is for Genes: The Impact of Genes on Education and Achievement” with Professor Plomin. She argued that although Young represented the science correctly, his were not the only plausible conclusions. For example, although Toby concedes that in fact, “environmental differences still account for between 30 and 40 per cent of the variance in GCSE results,” he seems to believe our main hope is “smart drugs that actually make you smarter.”

Though he did not mention what he calls ‘progressive eugenics’ in that blog, recent events surrounding his appointment and subsequent resignation from the Office for Students have also brought to light the links he has drawn between genetics and a more sinister agenda .

In contrast, Asbury explains that:

“In fact, heritability estimates tell us the extent to which behavioural differences between individuals (e.g., where they fall on the IQ distribution) are explained by genetic differences between them. They do not predetermine anything, telling us what is rather than what can be. For example, I can gain a couple of inches by wearing heels or I can correct my vision by wearing glasses, both regardless of my genetic predispositions.”Heritability estimates do not predetermine anything, telling us *what is* not *what can be* Share on X

She therefore argues that education should be better tailored to individual pupils – a recurring theme throughout her and Plomin’s book. In other words, some pupils find things harder than others and need extra help – hardly all that radical after all.

Dubious associations coupled with my desire to believe all are born as blank slates with equal potential, left me wanting to put my fingers in my ears and say “lalala I can’t hear you.” However, in a bid to avoid submitting to a post-fact world – in which only convenient evidence is deemed relevant, I decided to dig deeper.

It quickly became clear that proponents of ‘progressive eugenics’ like Toby Young cannot claim the science makes their conclusions inevitable.Proponents of ‘progressive eugenics’ cannot claim science makes their conclusions inevitable. Share on X However, those who run a mile from the mere idea of genetics research in education should find out more before assuming that behavioural genetics is inevitably toxic. Find out more before assuming behavioural genetics is inevitably toxic. Share on X

More than one possible interpretation

Plomin and Asbury’s research first drew popular attention in light of Michael Gove’s advisor, Dominic Cummings’ interest.

Headlines included “Michael Gove urged to reject ‘chilling views’ of his special adviser” and “Michael Gove held talks with ‘IQ genes’ professor.”

Fellow Govian traditionalist Toby Young’s role in the recent Teach First controversy only seemed to confirm a link between the researchers’ work and a particular political agenda. Links to eugenics, have done nothing to help despite Asbury and Plomin describing this as a “wilful distortion of the science” (p96), a theme also explored in this recent BMJ Journal of Medical Ethics blog.

However, although Asbury’s TES article insists Toby Young’s blog generally “accurately reflected robust, replicated research,” tensions and disagreements seem to be simmering just below the surface and it’s not clear to me why Asbury and Plomin haven’t done more to distance themselves from Young.

For example, whilst Young argues that “what schools cannot do, or haven’t been able to do up to now, is raise the IQs of individual students”, one of Plomin and Asbury’s most explicit recommendations is for schools to introduce “thinking skills lessons focused on IQ and self-confidence” (p169). At the same time, when it comes to social mobility, the authors argue that: 

“We know that environmental factors influence socio-economic status at least as much as genetic traits and that the environment can be used as an agent for change” (p128).

They also note that:  

IQ alone does not predict achievement. For better or worse we are more complex organisms… than Mendel’s peas… The genes for complex traits such as learning ability and IQ are never deterministic. Our aptitude for intelligence is not hard wired and is subject to a panoply of experiences as well as our unique genetic code… equating IQ scores with achievement is akin to believing that regardless of a driver’s experience and skill, the top spot in any motor racing contest will always go to the person driving the car with the biggest or most sophisticated engine.”

For better or worse we are more complex organisms than Mendel's peas Share on X

Given these differing interpretations, perhaps inequality in education is not so immutable after all. Indeed, as the authors acknowledge “just as plants develop differently in different climates, or under the care of different gardeners, so, too, do human beings” (p143). Given this, what can the gardeners of the next generation – teachers, parents and youth workers – learn from genetic research?What can the gardeners of the next generation learn from genetic research? Share on X

The role of early childhood

Confirmation bias is a pervasive problem when reading research. Nowhere is that more acute than when reading about emotive topics where values and principles are at stake. Thus, as I read more about genetics and education, the bits that jumped out were, unsurprisingly, the bits that confirmed my world view. However, one key point gave me considerable pause for thought.

According to Asbury and Plomin, genetic studies raise serious questions about the extent to which “shared environment” accounts for differences in attainment in crucial areas such as reading, writing and numeracy. They cite a twin study in which only 7% of differences in achievement amongst 7 year olds (importantly this is based on teachers’ ratings) were accounted for by shared environment and two-thirds by genes. This seems to clash profoundly with my long held views (based on the conclusions of other research such as the major EPPSE study, on the importance of the Home Learning Environment and Hart and Risley’s seminal work on the 30 million word gap by the age of 3).

However, questions over the role of the home environment have also been highlighted by other researchers. Judith Rich Harris, for example argues in her book “The Nurture Assumption” that:

“The idea that we can make our children turn out any way we want is an illusion. Give it up. Children are not empty canvases on which parents can paint their dreams.”

So could it be that improving the Home Learning Environment, and closing the gap in vocabulary that young children are exposed to, would not in fact close the attainment gap? If so, this would have profound implications for education interventions and what teachers should do when working with parents, particularly in the early years. 

I’m not convinced. Twin studies such as those used in “G is for Genes” have a number of significant limitations, some of which are highlighted in a comprehensive review of the evidence on intelligence in the journal “American Psychologist.” As the review points out, other studies have shown that adoption for example can add 12–18 points to the IQ of unrelated children. Thus the American Psychologist review concludes that “it is almost surely the case that a substantial fraction of the IQ advantage is due to the environments independent of the genes associated with them,” but notes “we have no direct evidence of the impact of any particular environmental factor on IQ”.

Indeed, Asbury and Plomin themselves appear to have some doubts, arguing that even if there are question marks over the long term impact of early years interventions:

“The early years are perhaps best seen as a developmental window in which a positive enriched environment can outweigh the influence of genetic inheritance for a while. During that time children can still catch the learning bug and develop a taste for success before school, cynicism, and DNA, and the interaction between the three, really kick in” (pp.93-94).

They therefore call for “interventions around environmental influences that negate the effects of poverty, reduced stimulation, crowding and chaos”.

Science bucks the trend

Another area in which recent research on twins raises challenges is in relation to science education at primary school. Research has previously shown that enthusiasm for science drops over the course of primary school and in early adolescence. Asbury and Plomin draw links between this and surprising patterns found in twin studies. According to their research, as with other subjects, attainment in science is highly heritable at the age of nine. However this does not seem to be the case by the age of twelve. Thus, unlike in other subjects, non-genetic, shared environmental factors (such as shared family experiences) become have far more influence over science attainment at the age of twelve compared to the age of nine.

Researchers are exploring a number of possible reasons for this, including the possibility that children are being put off science early and therefore not “seeking out scientifically enriching opportunities.” According to the Asbury and Plomin, this might mean we should delay formal science education.

However for now, there is too much uncertainty to justify an immediate response – so rather than telling us “which buttons to press,” genetic research in education leaves us with fascinating insights, but few practical solutions. Rather than telling us “which buttons to press,” genetics research in education leaves us with fascinating insights, but few practical solutions Share on X

So what are the practical implications for teachers?

Once one moves from genetic evidence to practical implications, the research mainly provides an additional rationale or language for things most teachers already recognise. 

Take for example, one of the twin studies highlighted by Asbury and Plomin. This compared identical twins’ experiences in the same classroom and found that these pupils’ experiences can differ considerably. Two pupils might therefore be sat in the same class with the same teacher, but have completely different experiences. It is a striking reminder of the need to pay attention to the minutiae of each individual pupil’s experience and i important to remember. The study is a good way of emphasising it, but the finding is hardly radical. Two pupils might be sat in the same class with the same teacher, but have completely different experiences Share on X 

Similarly, Asbury and Plomin argue that teachers should look out for sudden fluctuations in a pupil’s attainment. As Asbury and Plomin explain there is a strong genetically-based rationale for doing this since continuity and attainment over time tends to have strong genetic associations. In contrast, shorter term fluctuations are driven by environmental factors so teachers and parents can and should respond to the latter. The insight is sensible but few teachers are in any doubt that if a pupils’ performance suddenly changes they need investigate what is going on.

The researchers also demonstrate an apparent genetic element to which pupils find Maths harder and easier, arguing we should therefore provide extra support to pupils who find Maths difficult. But again, whilst the genetic roots of differences in ability may be news, the recognition that the ease with which pupils learn differs is hardly novel, and neither is the need for schools to respond.

Whilst the genetic roots of differences in ability may be news, the recognition that the ease with which pupils learn differs is hardly novel, and neither is the need for schools to respond.    

Perhaps this is no bad thing. One potential benefit of emerging genetic evidence is that it might strengthen the case for education policies which are advisable for ‘not-purely genetic’ reasons. Asbury and Plomin recommend a range of policies including:

  • Providing access to a range of sports for teenagers; allocating support to struggling pupils in a way that does not depend on a diagnosis (p165);
  • Giving all children a key worker – a bit like Scotland’s “Named Person” scheme (p113);
  • Ensuring each school has a substantial team of educational psychologists (p167-168).

All of these are sensible ideas for myriad reasons. We might not need behavioural genetics to recognise that these are a good idea, but the extra source of evidence is welcome.

Tenuous implications

Genetics yields profoundly interesting insights for those keen to understand learning better, and some of these will probably have practical applications in the future. However ultimately, when it comes to the real, meaty questions of who to target, when, and how (in other words, the questions that keep teachers up at night), for now genetics has yet to lend a helping hand. Academics should certainly continue this work and their findings are well worth reading. But, let’s not pretend this emerging field is on the cusp of transforming classroom practice or government policy.

Let’s not pretend this emerging field is on the cusp of transforming classroom practice or government policy.

To do otherwise would be problematic. Plomin used our tête á tête at the Select Committee to argue that ‘learning chips’ could provide a reliable genetic predictor of pupils’ predispositions and difficulties. It was a smart move, and the idea has since turned into complete click-bait. Yet the idea is so blue-sky it’s almost gasping for air. Unfortunately, at the select committee the idea rapidly turned into an unhelpful distraction from the more immediate issues that MPs needed to address. The problem is that attention-grabbing headlines do not necessarily translate into constructive contributions to practice. Attention-grabbing headlines don't always translate into constructive contributions to practice Share on X.

Similarly, Asbury and Plomin’s call for “all teacher training courses (to) include at least one module on the genetics of ability and achievement” (p175) is also misguided. There are a million things that might be worth including in ITT, and most can plausibly be justified, however not all will make the cut. Given the paucity of practical new insights genetics can currently provide, it is surely not one of the limited number of critical things that need to be shoehorned into a year-long, ‘licence to practice’ course.

The question of personalisation is trickier. On one hand, Asbury claims that genetics shows all young people are different and thus education needs to be personalised. On the other, international evidence raises questions about personalisation’s benefits and early tracking (placing different learners on different pathways) in particular has been shown to be particularly damaging. Indeed, a growing number of educationalists are asking whether we need more or less personalisation (Lucy Crehan for example explains the issue well in Principle 3 of Cleverlands – p247).

Asbury and Plomin’s version of hyper-personalisation has philosophical and sociological flaws too. In their ideal “genetically sensitive” secondary school, even the Year 7 compulsory curriculum would be limited to basic literacy, numeracy ICT skills, PE and Science. If they were to have their way this would mean many pupils leaving school without learning basic history, literature or geography – let alone art.

Ultimately, educationalists may have been wrong to dismiss the influence of genetics on children’s education and development, but geneticists would be equally mistaken in drawing conclusions that ignore a vast body of educational experience and evidence.

Missing the mark

I feel sorry for academics. They are stuck between a rock and hard place with three options:

  1. They can choose to say nothing about their research’s practical implications and be criticised for irrelevance.
  2. They can highlight obvious, uncontentious implications and be criticised for teaching grandma to suck eggs.
  3. They can stray into more stormy waters and make more contentious claims.

Although I have criticised Asbury and Plomin for all three, I in fact think their overall contribution to knowledge about education is valuable and it would be a mistake to dismiss their work and field.

Asbury and Plomin may be right that genetics will one day “helps us figure out which buttons to push” (p97). However we are a long way from getting there yet. In the meantime, any claim that the science justifies eugenics should be dismissed. As Ian Brassington, senior lecturer in Bioethics and Medical Law puts it in his blog for the BMJ:

“We can tinker with which embryos are implanted all we like; we could even faff around with particular genes if we wanted. But if we don’t get the broader context right, it’ll all be for very little result.

Being super-intelligent isn’t your route out of poverty if you don’t have the right social context… Science, inasmuch as that it informs public health measures, can help here; but its role is at once much less exciting than embryo selection, and much more important.  Young, I think, has fallen into the trap of thinking that whizzbang science is the solution to all our ills.  And whizzbang science does have its place. But it’s not enough.”