Is school performance less heritable in the USA?

CC licensed photo by flickr user Pollbarba. Click for source.A recent twin study looked at educational achievement in the UK and found that genetic factors contribute more than half to the difference in how students perform in their age 16 exams. But this may not apply to other countries.

Twin studies look at the balance between environmental and genetic factors for a given population and a given environment.

They are based on comparing identical and non-identical twins. Identical twins share 100% of their DNA, non-identical twins 50%. They also share a common environment (for example, the family home) and some unique experiences.

By knowing that differences in what you’re measuring in identical twins is likely to be ‘twice as genetic’ or ‘twice as heritable’ in non-identical twins you can work out the likely effect of environment using something called the ACE model.

This relies on various assumptions, for example, that identical twins and non-identical twins will not systematically attract different sorts of experiences, which are not watertight. But as a broad estimate, twin studies work out.

Here’s what the latest study concluded:

In a national twin sample of 11,117 16-year-olds, heritability was substantial for overall GCSE performance for compulsory core subjects (58%) as well as for each of them individually: English (52%), mathematics (55%) and science (58%). In contrast, the overall effects of shared environment, which includes all family and school influences shared by members of twin pairs growing up in the same family and attending the same school, accounts for about 36% of the variance of mean GCSE scores. The significance of these findings is that individual differences in educational achievement at the end of compulsory education are not primarily an index of the quality of teachers or schools: much more of the variance of GCSE scores can be attributed to genetics than to school or family environment.

In other words, the study concluded that over half of the difference in exam results was down to genetic factors.

The most important thing to consider, however, is how well the conclusions apply outside the population and environment being tested.

Because the results give an estimate of the balance between environment and genetic heritability that contribute to the final outcome, the more fixed the environment, the more any differences will be due to genetics and vice versa.

If that’s a bit difficult to get your head round try this example: ask yourself – is difference in height mostly due to genetics or the environment? Most people say genetics – tall parents tend to have tall offspring – but that only applies where everybody has adequate nutrition (i.e. the environmental contribution is fixed to maximum benefit).

In situations where malnutrition is a problem, difference in height is mostly explained by the environment. People who have adequate nutrition during childhood are taller than people who suffered malnutrition. In this situation, genetic factors are a minor player in terms of explaining height differences.

So let’s go back to our education example and think about how genetic and environmental factors balance out.

One of the interesting things about the UK is that it has a National Curriculum where schools have to teach set subjects in a set way.

In other words, the government has fixed part of the environment meaning that differences in exam performance in the UK are that bit more likely to be due to genetic heritability than places where there is no set education programme.

In fact, the same research group speculated in 2007 in a research monograph (pdf, p116) in a similar analysis, that school performance would be less genetically heritable in the USA, because the school environment is more variable.

The U.K. National Curriculum provides similar curricula to all students, thus diminishing a potentially important source of environmental variation across schools, to the extent that the curriculum actually provides a potent source of environmental variation.

In contrast, the educational system in the United States is one of the most decentralized national systems in the world. To the extent that these differences in educational policy affect children’s academic performance, we would expect greater heritability and lower shared environment in the United Kingdom than in the United States.

In other words, all other things being equal, greater equality in educational opportunity should lead to greater heritability.

School performance may be less influenced by genetic heritability in the USA because the educational environment is more variable and therefore accounts for more difference.

Whereas in the UK, the educational environment is more fixed so a greater proportion of the difference in performance is down to genetic heritability.

It’s worth noting that this hasn’t, to my knowledge, been confirmed yet, but it’s a reasonable assumption and demonstrates exactly the question we need to bear in mind when considering studies that estimate heritability – for whom and in what environment?

Link to twin study on school performance in PLOS One.
pdf of research monograph on learning and genetics.


  1. Posted December 19, 2013 at 7:49 am | Permalink

    Can’t you use this data in the opposite way? Assume that heritable variance in learning ability is the same for any human population. Then you can look at the inverse of the heritability as an index of how even the quality of education is across subpopulaitons for different societies.

    Or would that be pushing it well beyond its statistical breaking point?

  2. drkashik
    Posted December 19, 2013 at 2:07 pm | Permalink

    “Identical twins share 100% of their DNA, non-identical twins 50%”

    Okay see that’s hella false. I have a lot of issues with twin studies, including the environment assumption issue mentioned in the post, but this is one of the bigger issues. Any given pair of humans shares something like 99% of our DNA because otherwise most of our proteins wouldn’t work properly. Even out of the, let’s call it, unconserved fraction, any given pair of siblings shares something between 0 and 100% of their DNA. Theoretically, child 1 could get all of fraction A from mom and all of fraction X from dad, and child 2 could get B and Y (except mitochondria, which is all mom).

    So, given a large enough sample size, the normal distribution would center on 50%, sure, but you’ve got some pretty big and non-negligible tails on either end there. And that’s not even considering epigenetics, which affects gene activity in both heritable and non-heritable ways, and usually has environmental causes, meaning identical twins are not actually identical.

    And CERTAINLY one can’t discount the inherent effect on a person’s life caused by having an identical twin at all! A study buddy, a confidant, that’s not only the same sex as you, but nearly you genetically. Surely that shared experience would cause their test scores to be more similar!

    • Alexander Stanislaw
      Posted January 2, 2014 at 3:37 am | Permalink

      Identical twins are much more genetically similar than non-identical twins and this can be quantified. This is sufficient to establish the usefulness of twin studies.

      Also there have been studies of identical twins raised apart – turns out that they end up similar in terms of all the usual measurements.

  3. Posted December 19, 2013 at 2:10 pm | Permalink

    Reblogged this on The Echo Chamber.

  4. Posted December 19, 2013 at 5:02 pm | Permalink

    Umm…shouldn’t the title of the post be “Is school performance less heritable in the USA?” or “more heritable in the UK?”

Post a Comment

Required fields are marked *


Get every new post delivered to your Inbox.

Join 2,572 other followers