Researchers identified 1271 genetic variants called single nucleotide polymorphisms (SNPs) that are linked to how long an individual spends time at school. Each SNP has a minute effect on its own, but together they can account for 11 percent of the variation in how long people remain in education.
'We can explain education as well with saliva samples as with demographics,' said lead author Dr Daniel Benjamin, a behavioural economist at the University of South California in the US. While 11 percent may seem small, it is greater than the effect of household income, but smaller than parents' educational background. A large proportion of variation is still not accounted for.
The majority of the newly discovered SNPs, which were found to be highly expressed both early in development and later in life, were strongly associated with neuronal communication, such as the activation of ion channels and neurotransmitter secretion.
In the study, published in Nature Genetics, US researchers analysed the DNA of both men and women of European descent, whose DNA profiles were collected from sources including the UK BioBank. The SNPs were combined into a Polygenic Score for each person and this score was compared to data on how many years of schooling each individual had completed – a good measure of educational attainment.
'If we can identify genetic risks for low cognitive ability, there is an argument for putting extra provisions in place to mitigate against that disadvantage,' Dr Kathryn Asbury, a genetics researcher at University of York who was not involved in the study, told The Atlantic.
'We already do that for environmental disadvantages, such as through additional funding or free school meals. I think we need to seriously consider the implications of biological risk factors and how a fair society should respond to them.'
However, researchers stress that the scoring does not determine an outcome for any individual. Based on this information, study authors propose that any practical advice or policy lessons based on the current research is 'extremely premature and unsupported by the science'.
'Until the score is better and we understand the causal factors underlying it, I am pretty uncomfortable using it to predict individual outcomes,' said Dr Benjamin. 'There is a lot more work to be done before we even have a conversation about using it that way.'