Better schools can offset genetic disadvantages in learning

By Dr. Liji Thomas, MDReviewed by Lauren HardakerNov 3 2025

New genetic research reveals that when students attend better schools, their reading success depends less on their DNA, showing that quality education can truly level the playing field.

Study: The genetic lottery goes to school: Better schools compensate for the effects of students’ genetic differences. Image credit: Ground Picture/Shutterstock.com

Education’s vital role in shaping lives makes fair and equal access to quality schooling a central policy goal around the world. A recent study published in PNAS highlights the role of schooling in mitigating the effects of poor genetic and social contexts determined by birth.  

Can schools beat genetics?

Schools were initially established to address the inherent differences in background among students, including socioeconomic status, race, and sex. The current study aimed to investigate the impact of schooling on genetically mediated differences.

Such genetic differences significantly influence skill development, as affirmed by twin studies, which have shown that school performance has a 50% heritability. These genetic effects are closely intertwined with environmental influences. Understanding this interaction is essential for effective social policy, as it shows how environments can affect individuals differently based on their genetic makeup.

Twin studies have been limited by the fact that twins usually attend the same school, restricting variation in school-gene interactions. Even so, prior research suggests that teacher quality improved the genetic variance in reading skills and overall academic skills. This suggests that the effect of genetic differences is amplified rather than compensated for by the quality of schooling.

Such studies are improved by the use of polygenic indices (PGIs), which account for a few genetic effects rather than total heritability. These effects are linked to genes that have been variably associated with the trait of interest in genome-wide association studies (GWASs) and are weighted accordingly.

Despite their narrower scope, PGIs are based on more representative samples. When considering PGIs based on GWAS associations with educational outcomes, the final effects are found within a specific social and academic background and do not signal intrinsic genetic weaknesses. In other words, given diverse contexts, lower educational status may be associated with different variants.

PGI-based studies exploring gene-environment interactions related to schooling have yielded conflicting results. This could be due to variations in the type of educational outcome examined, the stage of development under review, and the content being studied. Poor study design is another potential contributor, capturing outcome variability due to inherent characteristics in the students rather than the schools’ ability to enhance learning.

It is also important to exclude bias due to the presence of confounding factors, which are associated factors irrelevant to the study but that affect the outcome. Within-family variation in PGIs is a valuable approach to reducing this type of bias, since it acts as a randomizer, a “genetic lottery”, where the effects are independent of other family attributes.

This still leaves room for bias due to the uncertainty surrounding whether social advantages lead to better educational outcomes. In this situation, the key question is whether supportive social environments amplify genetic advantages or help lessen the effects of genetic differences. The latter could be achieved by compensating for poor genetic risk or by inhibiting the outward effects of genetic risk factors.

DNA meets data

The researchers used data from the Norwegian Mother, Father, and Child Cohort Study (MoBa), combined with that from the Norwegian register. The aim was to assess how the predisposition of children interacted with their educational status and the quality of their school. “The Norwegian context and rich educational data provide the ideal setting for validating our causal gene–environment interaction approach.”

MoBa provided genetic data on the mother, father, and offspring in each family set, which served as the basis for calculating polygenic indices (PGI^EA). The family-based datasets helped identify within-family variation, which allowed the researchers to isolate causal genetic effects.

School value-added measures were estimated using the Norwegian register, resulting in estimates of school quality effects. Such measures were designed to capture the relevant difference in student outcomes precisely, that is, numeracy and reading skills. This makes it highly predictive for the student outcome of interest, with minimal bias.

This was based on standardized students’ test scores such that the mean and standard deviation (SD) were zero and one, respectively, just like the standardized PGI^EA itself. The challenge was to ensure that these measures were not confounded by other factors, such as student ability or the parents’ socioeconomic status, which can independently affect school enrollment and, therefore, school outcomes.

Both sets of data were exploited to help uncover the causes of school-based interactions between the child’s genes and environment.

Better schools close gaps

The study showed that PGI^EA and school value-added measures were independently varying factors. This allowed the researchers to separate the effects of genetic and schooling-related factors from each other and from the gene-environment interaction of interest.

PGI^EA and school quality functioned as independent but substitutable factors in predicting reading outcomes, rather than one acting as a stand-in for the other. For schools that were 1 SD above average (by country standards), a 1 SD increase in PGI^EA had a 6% lower-than-expected impact on reading test scores. This compensatory effect was observed within just one school year (grade 8, ages 13–14). Thus, the better the school was at teaching, the greater the impact on the reading outcomes of children who had a low PGI^EA.

This interchangeability of PGI^EA and school quality occurred because students at the lower end of the PGI^EA gained significantly on reading scores. Better schools showed a lower impact of genetic differences on student outcomes due to the compensatory effect of schools on lower PGI^EA. Conversely, students showed a greater decrease in their test scores when their genetic differences were adverse and they attended low-quality schools.

Such findings were absent, however, with numeracy test scores. This null result likely reflects higher persistence of numeracy skills (0.70 vs. 0.46 for reading), meaning new school inputs have less influence on mathematical ability at this developmental stage.

Invest in fairer learning

This suggests that investments in school quality may help reduce educational inequalities arising from genetic differences between students.

Within just one school year, among students aged 13-14 years, higher school quality helped develop children’s skills more fairly. This occurred because good schooling reduced the gap in test reading scores between children with a better genetic predisposition and those with a worse one.

However, the study also suggests that the difference between low- and high-quality schools further increases the skill development gap between those with less and more genetic advantages in education.

Download your PDF copy now!