Genetic Influences on Educational Achievement in Cross-National Perspective

Abstract

There is a growing interest in how social conditions moderate genetic influences on education [gene–environment interactions (GxE)]. Previous research has focused on the family, specifically parents’ social background, and has neglected the institutional environment. To assess the impact of macro-level influences, we compare genetic influences on educational achievement and their social stratification across Germany, Norway, Sweden, and the United States. We combine well-established GxE-conceptualizations with the comparative stratification literature and propose that educational systems and welfare-state regimes affect the realization of genetic potential. We analyse population-representative survey data on twins (Germany and the United States) and twin registers (Norway and Sweden), and estimate genetically sensitive variance decomposition models. Our comparative design yields three main findings. First, Germany stands out with comparatively weak genetic influences on educational achievement suggesting that early tracking limits the realization thereof. Second, in the United States genetic influences are comparatively strong and similar in size compared to the Nordic countries. Third, in Sweden genetic influences are stronger among disadvantaged families supporting the expectation that challenging and uncertain circumstances promote genetic expression. This ideosyncratic finding must be related to features of Swedish social institutions or welfare-state arrangements that are not found in otherwise similar countries.

Social and genetic associations with educational performance in a Scandinavian welfare state

“…Educational outcomes are among the few traits with a substantial shared environmental component in twin studies ( 3 ). Our results support evidence from twin studies that do not find variation in shared environmental components between countries ( 5 – 7 , 35 ). Even though Scandinavian countries have among the highest economic mobility rates in the world, educational mobility is considerably lower than income-mobility ( 36 ).…”

“…Some twin studies have found that the heritability of educational outcomes is higher when societies provide more equal opportunities (16,17). Other comparative studies of the heritability of educational outcomes find no differences between social democratic welfare state countries and liberal welfare state countries (5)(6)(7)35). Our results not only confirm that direct genetic associations are likely to be fairly constant across societies, but they also raise the possibility that social associations of family background do not systematically vary across policy contexts when models are estimated that are robust to differences in assortative mating and gene-environment covariance.…”

How neighborhood effects vary by achievement level

“…As exemplified in panel (b), it may also more or less directly influence the neighborhood effects, which we call the "Causation model" of moderation. For instance, achievement level is intrinsically related to genetic endowment, with some recent studies suggesting that genetic influence account for 50-80% of the variation in achievement levels in western European countries (Baier et al, 2022;Engzell & Tropf, 2019). Following the Scarr-Rowe hypothesis (Scarr-Salapatek, 1971), we may expect that children with high academic proneness flourish in advantageous neighborhoods with good learning opportunities.…”

“…Yet, there is considerable variation in achievement scores within such characteristics, as highlighted by Figure 1 in the introduction. For example, variance decomposition models suggest that shared environmental factors (including neighborhood effects) explain 10-40% of the variation in children's achievement levels (Baier et al, 2022). Furthermore, family background indicators such as parental education or earnings explain roughly 20% of the variation in children's test scores (von Stumm et al, 2020).…”

A population-wide gene-environment interaction study on how genes, schools, and residential areas shape achievement

“…Many, but not all 9 U.S. twin studies have found that genetic influences are stronger in higher-socioeconomic status families, whereas European and Australian studies have found null or opposite results 10 . A recent twin study found that heritability estimates for achievement were invariant across levels of parental socioeconomic status in Norwegian, German, and U.S. samples, but in a Swedish dataset, heritability estimates were stronger in more disadvantaged families 11 . Genomic studies have also found largely null results.…”