Tag Archives: Sabisky

Nature and Nurture – the Genetics of Education. Why do children differ?

Trying to select which ResearchEd talks to go to was a painful and difficult process. One that I chose was Andrew Sabsiky’s @AndrewSabisky contribution:

Nature and Nurture – the Genetics of Education. Why do children differ?

It’s a hot topic at the moment and I was interested not least because I’ve been rather bemused at some reactions to recent studies for example “If IQ/exam-success are largely heritable then what we do in schools must be largely irrelevant”. So what do I think I learned from Andrew Sabsiky’s ResearchED talk.

Well, firstly, that my brain has been in cognitive decline for the last 23 years! Shattering news, and enough to make one consider retiring to the sofa to watch Doctors, but Andrew has since kindly reassured me that while my fluid IQ skills may have gone-to-pot, I am theoretically still capable of acquiring new vocabulary and knowledge. I therefore aim to test that assertion here by presenting a very simplified, layman’s interpretation of some of what I *think* I learned with some elaborations and diagrams of my own, but please bear my cognitive limitations in mind.

We began with Sir Francis Galton, somewhat infamously associated with eugenics. Love him or loathe him, polymath Galton did and contributed much more than you ever thought he did including inventing methods of scientific inquiry that are still in use today . Particularly relevant to the topic under discussion was Galton’s use of twin and adoption studies.

Very briefly:
• Identical twins share the same genes
• Non-identical twins share 50% of their genes

Children brought up in the same household at the same time will have both:

• a shared environment – same parents, share experiences, shared friends, food etc
• a non-shared environment – different teachers, different friends, unique experiences (eg accidents) etc

By using twin/adoption studies to investigate various traits such as IQ, bullying, conduct-disorder and so on, you can calculate the degree of heritability for each trait studied by calculating and discounting any effects due to both shared and non-shared environments. A meta-analysis of the very many studies in this area indicates that there is a high degree of heritability for a number of traits, such as IQ (50%-80%), that could affect some types of educational attainment.

So what are the implications for education and are some people right to say that what we do in schools doesn’t really matter?

If we accept that at least some measures of educational attainment are closely linked to IQ then in an equitable and meritocratic education system, where all children have the same quality of educational opportunity, if we plotted attainment/IQ we’d expect a normal distribution of results.

fig1

In such an equitable and meritocratic system then the main variable would be IQ.
The corollary of that is if you conducted a ‘heritability of IQ study’ using children brought up and educated within this equitable system that you would conclude that IQ was highly heritable.

In other words, should a study find IQ to be highly heritable, ie that different environments (parents, teacher etc) have a negligible effect, this indicates that the system the children are being educated in is highly equitable.

Does this mean that a child’s environment makes no difference? No, it means that, in an equitable system, each child’s environment makes an equally sized difference to each child.

Neither does it mean that the attainment of a child is limited. Staying with our equitable, meritocratic system, in the diagram below you can see that, due to improvements in the educational environment (Brain Gym and VAK banned, CPD replaced by ResearchEd attendance etc) the whole distribution curve has shifted right .

fig2

As further illustrated below, while each individual’s relative position on the curve would remain the same, their absolute position would shift right.

fig3

 

Theoretically, the scope to improve education for all and shift the curve right is limitless.