ArticlePDF AvailableLiterature Review


Research has established that genetic differences among people explain a greater or smaller proportion of the variation in life outcomes in different environmental conditions. This review evaluates the results of recent educationally relevant behavioural genetic studies and meta-analyses in the context of recent trends in income and wealth distribution. The pattern of results suggests that inequality and social policies can have profound effects on the heritability of educational attainment and achievement in a population (Gene–Gini interplay). For example, heritability is generally higher at greater equality levels, suggesting that inequality stifles the expression of educationally relevant genetic propensities. The review concludes with a discussion of the mechanisms of Gene–Gini interplay and what the findings mean for efforts to optimize education for all people.
J. Biosoc. Sci., page 1 of 30 © Cambridge University Press, 2018. This is an Open Access article,
distributed under the terms of the Creative Commons Attribution licence (http://creativecommons. org/
licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium,
provided the original work is properly cited.
*Department of Psychology, Tomsk State University, Tomsk, Russia and
Department of Psychology, Goldsmiths, University of London, London, UK
Summary. Research has established that genetic differences among people
explain a greater or smaller proportion of the variation in life outcomes in
different environmental conditions. This review evaluates the results of recent
educationally relevant behavioural genetic studies and meta-analyses in the
context of recent trends in income and wealth distribution. The pattern of
results suggests that inequality and social policies can have profound effects
on the heritability of educational attainment and achievement in a population
(GeneGini interplay). For example, heritability is generally higher at greater
equality levels, suggesting that inequality sties the expression of education-
ally relevant genetic propensities. The review concludes with a discussion of
the mechanisms of GeneGini interplay and what the ndings mean for
efforts to optimize education for all people.
Individual differences in education, health and psychological well-being are a product
of complex processes that involve genetic and socio-demographic factors. These factors
affect life outcomes by interacting with each other, which complicates the search for
specic causal mechanisms underlying variation (Plomin et al., 2016). This review of
recent research evaluates whether the strength of genetic contribution to individual
differences (heritability) depends on the degree of inequality in resource distribution in a
particular population. The rst section introduces the commonly used measure of
inequality the Gini Coefcient (Gini). The second section introduces the fundamental
and commonly misunderstood concept of heritability. The following two sections briey
overview evidence for effects of inequality (Gini) and genes on education, health and
well-being. This is then followed by a review of evidence for the interplay between Gini
and heritability (GeneGini interplay), focusing on educational outcomes in the context
of recent trends in resource distribution in different populations. The GeneGini
Corresponding author. Email:
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
interplay refers to two broad phenomena: (1) the strength of the genetic contribution to
differences among people in a population depends on environmental conditions
(statistical interaction); and (2) environmental conditions may enhance or suppress
genetic effects in an individual (causal mechanism). The paper concludes with a
discussion of what GeneGini interplay means for understanding of the origins of
individual differences and for efforts to improve educational outcomes for all people.
The Gini Coefcient and related concepts
The Gini Coefcient (Gini) was developed in 1912 by Corrado Gini. Gini is the most
widely used single measure of inequality, recognized for its attractive theoretical and
statistical properties and for being technically sound (World Bank Institute, 2005;
De Maio, 2007; Ceriani & Verme, 2012; Cobham, et al., 2015; Gastwirth, 2016; Davies,
2016). It is used to measure inequality of income or wealth distribution within a certain
population or region, or across populations. Gini does not, however, measure the wealth
of a population a wealthy and a poor country can have the same Gini. Wealth of
nations is measured by the Gross Domestic Product (GDP) a monetary measure of the
market value of all nal goods and services produced in a year (or another period). The
GDP per capita is GDP divided by the number of people in the country, which does not
reect differences in the cost of living and the ination rates of the countries. These are
taken into account in the Purchasing Power Parity (PPP) measure. The PPP is an
exchange rate at which the currency of one country is converted into that of the second
country in order to purchase the same volume of goods and services in both countries
(International Monetary Fund; The PPP is expressed in international dollars,
which have the same purchasing power as US dollars have in the United States. The PPP
per capita is GDP converted into international dollars, divided by total population. For
some countries, the per capita GDP and PPP are very similar, whereas for others GDP
and PPP can differ signicantly. Unlike Gini, GDP and PPP measures tell nothing about
how resources are distributed within countries.
Gini ranges from 0 to 1, where 0 means perfect equality (everyone has the same
income/wealth) and 1 means absolute inequality (one person gets the whole income/
wealth and the others get nothing). Gini can also be expressed in percentages (Gini
Coefcient × 100), which is referred to as the Gini Index. However, the two terms
(Coefcient and Index) are commonly used interchangeably. Gini is derived from the
Lorenz curve, which shows the cumulative proportion of the population, sorted from
poorest to richest (or lowest earners to highest earners), on the horizontal axis; and the
cumulative proportion of wealth or income on the vertical axis. So, the point (0.6, 0.2)
would indicate that the bottom 60% of earners of the selected population earn 20% of the
total income. One way to understand the meaning of Gini is as a percentage scale. So, if
income Gini is 0.7, there is a 70% inequality and only 30% equality (equality scoring 30
out of 100). As an example, in the US in 2014, where income Gini was 0.4, 0.1% of the
population earned on average US$6,087,113 each per annum, 1% of the population
earned on average US$1,260,508 each per annum and 90% of the population earned on
average US$33,068 each per annum (Piketty & Saez, 2003; Saez, 2015).
Gini has some limitations. For example, the total Gini of a society is not equal to the
sum of the Gini Coefcients of its subgroups (World Bank Institute, 2005). In addition,
2F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
as Gini depends only on reliable income data, it does not take into account shadow or
informal economies. There is also some controversy on whether Gini is more sensitive to
changes at the centre of the distribution (Atkinson et al., 2011; Cobham et al., 2015) or
to changes in the lower and upper parts of the distribution (Gastwirth, 2016). Despite the
relative strengths and weaknesses of different inequality measures, research has shown
that they are in high agreement. For example, one study (Kawachi & Kennedy, 1997)
found an average correlation of 0.94 among Gini and ve other measures of inequality
(the decile ratio; the proportions of total income earned by the bottom 50%, 60%
and 70% of households; the Robin Hood Index; the Atkinson Index; and Theils
entropy measure).
The currently observed range of inequality across the world is very wide, with
income Ginis in 2014 varying from 0.25 (Norway) to 0.64 (Comoros); and wealth Ginis
varying from 0.42 (Slovakia) to 0.90 (the Russian Federation) (UNDP, 2013; Credit
Suisse, 2014). Wealth Ginis are always more extreme than income Ginis. For example,
the wealth share of the poorest 50% of the population is always less than 5%, whereas
their income share generally falls in the 20% to 30% range (Piketty & Saez, 2014).
In 2014, the wealth and income Ginis were: 0.85 vs 0.4 in the US; 0.73 vs 0.36 in the UK;
and 0.56 vs 0.25 in Norway. Income and wealth inequality are reciprocally linked
and both lead to consumption inequality (Credit Suisse, 2014; Crawford et al., 2016;
OECD, 2017).
It is possible that income and wealth inequality have different impacts on people.
In conditions of high average wealth (e.g. high per capita PPP) and low income Gini, a
high wealth Gini may not signicantly affect peoples lives. This is because most people
would have good access to quality services and resources. Modern Sweden is a good
example of such conditions: the per capita PPP is high (US$46,445), the income Gini is
relatively low (0.27) and the wealth Gini is very high (0.80). In contrast, in conditions of
high PPP coupled with high income Gini and wealth Gini, the impact of inequality on
peoples lives may be very strong, both in terms of lack of resources and negative
psychological experiences. The US exemplies these conditions: the per capita PPP is
high (54,539), income Gini is high (0.41) and wealth Gini is very high (0.86). The
negative impact of inequality can be further inated when average wealth is low because
people who are disadvantaged by inequality have even fewer resources available to them.
For example, in the Russian Federation, the per capita PPP is signicantly lower (US
$25,498), income Gini is high (0.42) and wealth Gini is very high (0.90) (Credit Suisse,
2014; UNDP, 2014; World Bank, 2017a, b).
When reviewing the literature on inequality, one needs to be aware that studies
use different indices and procedures to correct for data limitations. For example,
analyses are often based on what is commonly called Earned Income income
received in return for work. Other studies also take into account Capital Income, Capital
Gains Income and Business Income, which together account for the greater part of
income inequality (Atkinson et al., 2011), but are more difcult measures to ascertain
(Alvaredo et al., 2013). As a result, reports often provide somewhat different estimates
of Gini. For example, wealth Gini estimated for the same year in the Credit Swiss report
was overall higher than that estimated by Allianz (Credit Suisse, 2016; Allianz, 2016).
However, despite these differences across studies and reports, the patterns of results
are consistent.
Genes and Gini 3
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Understanding heritability has become particularly important following recent
scientic advancements in genetic science. Although widely used, the concept of
heritability is still commonly misunderstood (Kovas & Malykh, 2016). For example,
many hold the mistaken view that heritability relates to a specic person. So, a
heritability of 60% is mistakenly taken to mean that 60% of an individuals trait (e.g.
academic achievement) is determined by the persons genes. In reality, it means that
60% of the differences in this trait among people within a particular population is
explained by their genetic differences (see Kovas & Malykh, 2016, for a detailed
explanation). Heritability of a trait within a population can be estimated (quantied)
by using a variety of quantitative genetic methods, such as comparing family members
of different genetic relatedness. One of the most common methods is studying a
sample of monozygotic (MZ) and dizygotic (DZ) twins drawn from a particular
population. Comparing average similarities on a trait of MZ twins (who are
genetically identical) with average similarities of DZ twins (who share on average 50%
of the variable DNA), gives an indication of the relative contribution of genetic and
environmental factors to variation in this trait in this particular population. Another
quantitative genetic method the Genomewide Complex Trait Analysis (GCTA)
estimates heritability directly by comparing DNA and trait similarities of unrelated
individuals (Plomin & Deary, 2015).
Today we know that variation in any trait stems from multiple, possibly hundreds
of DNA variants, and multiple environments, making human traits polyfactorial:
polygenic and polyenvironmental. Heritability reects differences in DNA among all
people within a population. Approximately 0.01% of the human DNA sequence is
different among people, with the rest of the sequence being identical for all humans.
Although 0.01% is a small proportion of the genome sequence, it represents millions of
DNA variants, each potentially contributing to the observed differences in behaviour.
Moreover, recent research suggests that we may be even more different from each other
genetically, with each persons DNA containing many rare or completely new (De novo)
mutations. It is estimated that the DNA sequence of each person contains an average of
8579 novel genetic variants (Telenti et al., 2016). Importantly, these differences do not
refer to race, countries or other population groups, but to differences between any two
individuals, even in the same family.
An important aspect of the denition of heritability is that it refers to genetic
contribution to differences among people within a particular population at a particular
time. This implies that genetic factors can explain a different proportion of individual
differences in different populations (even if the overall amount of trait variability is
similar). For example, children differ greatly in their academic achievement within all
countries. The Programme for International Student Assessment (PISA) study showed
that studentsperformance in mathematics at age 15 varies more within a country than
between countries (OECD, 2013). Within a country, the gap between the highest- and
the lowest-achieving students was over 300 points, the average score in OECD countries
being around 500 points. This difference is enormous, as a performance gap of about
70 score points represents a large difference in skills and knowledge the equivalent of
about 2 years of schooling in a typical OECD country (OECD, 2013). The results also
4F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
suggested that, although most of the variability overlaps across countries, there are also
signicant differences, especially in proportions of students achieving low and high
scores (OECD, 2014).
Between-population variability could arise from differences in allelic frequencies
and differences in environmental provisions across populations. Similarly, within-
population differences are also a product of both genetic and environmental factors.
In fact, genetic variability within a population is so vast that two randomly selected
individuals within a population can be more genetically different from each other
than two randomly selected individuals from two diverse populations (Jorde &
Wooding, 2004; Witherspoon et al., 2007). Environmental provisions also vary greatly
within and between populations. An important question is whether the vast variability
in achievement within each population is explained to a different extent by genetic
and environmental factors in different populations. For example, if two countries
show comparable variances but differ in average performance, variability in one
country may be largely due to environmental disparities, whereas in the other it
may largely be due to genetic variability (free from environmental limitations).
Therefore, interpretations of sources of differences across populations may be
complicated by the GeneGini interplay. This review evaluates whether environmental
provision, and more specically, inequality, moderates the effects of genes on
variation in traits.
Ginis effects on life outcomes
A large number of studies and reports have concluded that inequality, beyond poverty,
has a strong impact on peoples lives, including education, health and well-being (e.g.
Ryan & Adams, 1998; Wilkinson & Pickett, 2006, 2009, 2011; Rowlingson, 2011;
Reardon, 2011; Duncan & Murnane, 2011, 2016; Bischoff & Reardon, 2014; Carter &
Reardon, 2014; Pickett & Wilkinson, 2015; County Health Rankings and Roadmaps,
2015). This section provides a brief summary of evidence for the impact of inequality on
educational and related outcomes.
The impact of inequality on childrens education, health and well-being is partly
mediated by parental psychological and physical well-being. Negative psychobiological
pathways may be activated by inequality-related feelings of injustice, stress, anxiety
about relative social status, loss of social connections and cohesion within communities
and insecurity (Kaplan et al., 1988; Singer & Ryff, 2001; Wilkinson & Marmot, 2003;
Wilkinson & Picket, 2011; County Health Rankings and Roadmaps, 2015). Inequality
may have direct impact on maternal health and stress levels during pregnancy, and on
the overall health and mental health of parents. For example, research suggests that the
most socially disorganized neighbourhoods, rather than the poorest neighbourhoods,
have the highest incidence of schizophrenia (Pinto et al., 2008). Inequality has been
linked to increased prevalence of hallucinations and delusional ideation, depression and
other mental health problems (Ryan & Adams, 1998; Wilkinson & Picket, 2009; Duncan
& Murnane, 2016). These can contribute to parenting problems, such as use of hostile
parenting practices (Johnson et al., 2015).
Inequality gets under peoples skin by activating many biological processes. Research
has shown that relative access to social and material resources is related to differences
Genes and Gini 5
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
among people in integrity of their distributed brain networks. This link is mediated by
multiple processes, including chronic inammation (e.g. Gianaros et al., 2012). Chronic
inammation arises as a result of many factors, including smoking, adiposity and social
integration and support (e.g. Loucks et al., 2006; Ford et al., 2006) which themselves
are associated with socioeconomic status (SES) (e.g. Gianaros et al., 2012). For example,
psychosocial stress can directly provoke pro-inammatory cytokine production (Kiecolt-
Glaser et al., 2010). In addition, exposure to pollution, allergens and lead, as well as
noise-related sleep disruptions more frequently experienced in more disadvantaged
neighbourhoods all trigger long-term biological reactions (Braveman & Gottlieb,
Early environment can be linked to later development through epigenetic regulation
(e.g. Meaney, 2010). For example, several studies have found that negative life events
in early childhood can affect expression of genes involved in immune function (Bick et al.,
2012). Other studies found differences in expression of genes involved in inammation and
anti-inammation between those who grew up in low- and middle-income families (Miller &
Chen, 2007). Socioeconomic disadvantage has been linked to increased expression of
inammatory molecules and proteins, transcriptional changes in genes and intracellular
signalling cascades promoting systemic inammation (Gianoros et al., 2012). The effects can
emerge already pre-birth and affect the whole path of development. Mothersstressand
anxiety during pregnancy can suppress or enhance genetic expression (e.g. reviewed in
Kovas et al., 2016). For example, elevated stress and lack of social support during
pregnancy is linked to increased inammatory activity, affecting immune function, with
increased risk of pregnancy and labour complications (Coussons-Read et al., 2007).
Reviews, based on recent large studies, provide evidence for a causal connection
between inequality and poor health (Pickett & Wilkinson, 2015). A recent large study
ranked the health of 3061 of 3143 counties/county equivalents in the US (County Health
Rankings and Roadmaps, 2015). The results showed that inequality (not just poverty)
contributes to increased risk of mortality and poor health. For example, higher ratio
between high and low earners within a county was associated with lower life expectancy
and higher rates of smoking, cardiovascular disease and obesity. Inequalitys effect was
present even when communities of similar average income and racial composition
were compared (County Health Rankings and Roadmaps, 2015). In a recent book
(Ansell, 2017) inequality is described as a societal diseasethat leads to unequally
dispersed biological disease. For example, an average life expectancy gap of 35 years
has been estimated between the richest and the most deprived US neighbourhoods
(Ansell, 2017). This is consistent with a US study that found substantial correlations
(0.500.66) between mortality and inequality (on six different measures), even after
adjustment for median income and poverty (Kawachi & Kennedy, 1997). Data from the
Ofce for National Statistics in the UK ( also show that inequality,
beyond poverty, affects life expectancy: the gap in life expectancy at birth increased
between the local areas, despite improvements across these areas over the last two
decades. Moreover, people in higher socioeconomic positions live longer than those in
positions slightly lower than themselves (Marmot et al., 1978; Marmot et al., 1984;
Rowlingson, 2011). It has also been shown that rates of infant mortality can be reduced
with improvements in Gini, including in wealthy countries (Hales et al., 1999). For
example, income Ginis of 0.25 (Sweden, Norway, and Finland), 0.35 (UK), and 0.40
6F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
(US) are associated with mortality rates of 2, 4, and 6 per 1,000, respectively
(World Bank, 2015). Taken together, this evidence suggests that, beyond absolute levels
of wealth, peoples experience of inequality, such as feelings of injustice, unfairness
and envy, may lead to elevated stress. In turn, this may trigger cascading negative
biological reactions.
The strength of effects of inequality changes as a function of absolute level of wealth.
Poverty may override the effect of inequality due to its dramatic limiting effects on
development. However, above the poverty line, relative wealth/income may have a
stronger effect. For example, data for 21 regions of Taiwan showed that, as living
standards rose during the course of rapid economic growth (between 1976 and 1995),
income inequality replaced absolute median income as the best predictor of mortality
(Chiang, 1999). In developed countries, people on average have better health in countries
with lower income Gini, than in those with the highest GDP/PPP (Wilkinson, 1996;
Wilkinson & Pickett, 2006). One large international study, with data from 23 high-
income countries, found a strong relationship between income Ginis and the Index of
Health and Social Problems endured by populations, including indicators such as life
expectancy, mental illness, obesity, educational scores, teenage birth rates, social
mobility and education (Wilkinson & Pickett, 2009). Analysis of data from 50 US states
showed very similar results (Kerry et al., 2010).
In developed countries, a clear correlation has also been shown between income
inequality and the UNICEF Index of Child Well-Being (Wilkinson & Picket, 2009).
Children in more unequal societies are more likely to be overweight, experience more
mental health problems, experience more bullying and victimization and become teenage
parents; and less likely to continue education beyond compulsory schooling and have
on average lower achievement (Ryan & Adams, 1998; Pickett & Wilkinson, 2007;
Wilkinson & Picket, 2009; Reardon, 2011; Duncan & Murnane, 2011, 2016; Bischoff &
Reardon, 2014; Pickett & Wilkinson, 2015; UNICEF, 2016).
Several studies suggest that the pattern of increase in inequality and associated
educational investment gap corresponds to the increase in educational achievement gap
(e.g. Magnuson & Waldfogel, 2008; Reardon, 2011; Piketty & Saez, 2003). In the US,
the change in inequality has followed a U-shaped trajectory over the last 100 years: high
in the beginning of the 20th century, falling between 1929 and the late 1970s, and
continuously increasing since then (e.g. 0.40 in 1976 and 0.47 in 2006), with the 2012
Gini being almost as high as in 1929 (Piketty & Saez, 2003; Saez & Zucman, 2014;
Sommeiller & Price, 2015; Atkinson et al., 2011).
This increase in overall inequality is accompanied by an increase in the educational
investment gap. For example, in 1970, a family with school-age children at the 90th
percentile of the income distribution earned ve times as much as a family at the 10th
percentile; in 2013 the high-income families earned eleven times more than the low-
income ones (Reardon, 2013). Consequently, in 1972 high-income families were
spending ve times as much per child as low-income families, and in 2007, they were
spending nine times more (Kornrich & Furstenberg, 2013). One study in the US showed
an increase in income-based gaps in a range of parenting behaviours over a 25-year
period (beginning in 1979), such as teaching children letters or attending museums. For
the most part, these gaps arose from top-income families pulling away from middle- and
low-income families (Kalil et al., 2016).
Genes and Gini 7
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
These increases in income and educational investment gaps have been paralleled by a
growing achievement gap. Test scores of children who were adolescents in the late 1960s
showed an achievement gap of four-fths of a standard deviation (about 80 points on a
SAT-type test), with children from low-income families lagging behind their better-off
peers. Forty years later, this gap was 50% larger, amounting to nearly 125 SAT-type
points. Trends in maths skill gaps were similar (Reardon, 2011; Duncan & Murnane,
2011). Further research similarly showed that the achievement gap between children
from high- and low-income families was roughly 30% to 40% larger among children
born in 2001 than among those born 25 years earlier (Reardon, 2011; Duncan &
Murnane, 2011). Another US study showed similar results: from 1970 to 2013, the
achievement gap between children in families at the 90th percentile and those in families
at the 10th percentile of the income distribution grew by 40% (Reardon, 2013).
Further support for the strong link between the achievement gap and inequality
(beyond poverty) is the evidence that the decrease in inequality between black and white
children after the 1950s and 1960s (when it was very high in the US) was accompanied
by a reduction in the achievement gap between them (Piketty & Saez, 2003; Magnuson &
Waldfogel, 2008). At the same time, increase in income inequality, which reached a
historic high in the 21st century, was accompanied by a signicant increase in the
achievement gap between children from low- and high-earning families. With these
changes, family income became a better predictor of childrens success in school than
race, with the lowhigh income achievement gap being twice as large as the blackwhite
achievement gap (Magnuson & Waldfogel, 2008). Consistent with these ndings, a large
longitudinal Canadian study (NLSCY) showed that, even once children are out of low
income, every incremental increase in household income (in families with low income) is
associated with better child development outcomes; and that there is no point above
which high income ceases to benet childrens development (Phipps & Lethbridge,
Research has also shown that more inequality is associated with less mobility across
generations. For example, among OECD countries (in the 1980s), striking associations
were found when the most equal (Finland, Sweden, Norway and Denmark) were
compared with the least equal (the UK and the US). In Finland, Norway and Denmark,
less than one-fth of economic advantage or disadvantage was passed on from father to
son in adulthood. In the UK and the US, roughly 50% of any advantage or disadvantage
was passed on (Corak, 2013).
The PISA study shows an interesting pattern of results on educational equity and
achievement. The results showed that approximately 15% of the variation in student
performance in mathematics, reading and science was attributed to differences in
studentssocioeconomic status. A more socioeconomically advantaged student scored
39 points higher in mathematics (measured by PISA) the equivalent of nearly one year
of schooling than a less-advantaged student (OECD, 2013). Students attending
socioeconomically advantaged schools outscored those in disadvantaged schools by
more than 104 points in mathematics, on average across OECD countries. In some
countries, small improvements in equity were observed over time (20032012): the extent
to which studentssocioeconomic background predicts their academic performance was
reduced; and the gap in performance between advantaged and disadvantaged students
narrowed. However, in many countries, educational equity decreased over the same
8F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
period. In some cases, the improved equity was accompanied by increased average
performance of a nation. However, in other countries, increase in average performance
was observed in the presence of deterioration of equity (OECD, 2013). As discussed in
this review, national averages do not reect experiences of individual students, just as
GDP and PPP measures do not reect socioeconomic circumstances of individuals.
Focusing on inequality and heritability measures allows further insights to be gained into
the sources of variation in performance within populations.
The trend of growing inequality has been seen in many countries across the world,
with 7 out of 10 people living in a country that has seen a rise in inequality in the last
30 years (Oxfam, 2017b). Over the last 25 years, the top 1% (60 million people) have
gained more income than the bottom 50% put together; and own more wealth than the
rest of the planet (6 billion people). Wealth is consolidating in the hands of very few
people: in 2010, the 388 richest people had as much wealth as the poorest half of the
world (3.6 billion people); in 2017 as few as eight people had as much wealth (Oxfam,
2016, 2017a). In the US, between 2009 and 2012, all income growth happened in the top
1% of earners in eighteen of the states; and in New York and Connecticut, the top 1% of
earners attained average incomes more than 48 times that of the bottom 99%
(Sommeiller & Price, 2015). Over 1989 and 2013 the top 10% owned 87.7% of household
wealth, the top 1% owned 50.8% and the top 0.1% had 20.3% (Saez & Zucman, 2014).
These trends have happened against a background of increases in absolute wealth, with
per capita PPP increasing from US$37,062 in 1990 to US$52,549 in 2015 (UNDP, 2016),
with the US remaining the richest region in the world (Allianz, 2016). At a global level,
it is estimated that over the next 20 years, 500 people will hand over US$2.1 trillion
to their heirs a sum larger than the GDP of India (Oxfam, 2017a). These severe
income inequality trends have been identied by the 2012 survey of the World Economic
Forum as the most likely global risk to manifest in the next 10 years (World Economic
Forum, 2012).
Geneseffects on life outcomes
Decades of behavioural genetic research leave us with little doubt that genes have a
profound effect on differences among people in all aspects of human functioning
(Polderman et al., 2015; Plomin et al., 2016). Recent adequately powered large-scale
studies and meta-analyses have conrmed that the heritability of most traits is
signicant, ranging from moderate to substantial. The most comprehensive meta-
analysis to date of the causes of individual differences in human traits examined 17,804
traits from 2748 publications including 14,558,903 twin pairs and concluded that
average heritability across all traits is 0.49 (Polderman et al., 2015). The most reliable
estimates are available for traits that received much research attention and funding.
In the meta-analysis, average (weighted) heritabilities for some of the most studied traits
were: 0.54 for blood pressure; 0.50 for conduct disorder; 0.40 for depressive episode; 0.62
for general metabolic function; 0.52 for heart function; 0.50 for immune system function;
0.44 for mental and behavioural disorders due to the use of alcohol; 0.50 for mental and
behavioural disorders due to the use of tobacco; 0.44 for other anxiety disorders; and
0.54 for higher-level cognitive functions (Polderman et al., 2015). The Genomewide
Complex Trait Analysis (GCTA) studies, which do not rely on data from family
Genes and Gini 9
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
members, also conrm that common genetic variation among people explains part of
the variance in observed traits. The GCTA heritability estimates are usually lower than
those from family studies (e.g. half for intelligence). This is because GCTA analysis relies
only on genetic effects tagged by common Single Nucleotide Polymorphisms (SNPs) and
does not capture rare SNPs or other types of genetic variation, such as Common
Number Variants (CNVs) (Plomin & Deary, 2015).
Research into individual differences in educational traits also clearly suggests a
substantial role of genetic factors (e.g. reviewed in Krapohl et al., 2014, 2015). For
example, one large-scale UK study demonstrated that state examination performance
at age 16 was highly heritable, with genetic factors explaining 62% of the variation in
examination results (Krapohl et al., 2014; Kovas et al., 2015). There are several reasons
for why genetic effects on educational achievement are strong. First, genetic effects on
achievement have many indirect paths, including through genetic inuences on
cognition, emotion, personality and motivation (Krapohl et al., 2014). Second,
environmental effects, such as SES, are partly confounded by genetic effects (Selzam
et al., 2016). Third, initial genetic predispositions get enhanced through the processes of
active selection and modication of environments (see below for further explanation of
geneenvironment correlations). Finally, a standardized educational set-up (as is the
case of the UK) may lead to more pronounced genetic effects, as students respond
differently to the same circumstances, based on their genetic make-up. This means that
equalized circumstances (same resources, standard teacher training, standard curricula,
standard class sizes) would reduce the proportion of the variance in achievement
explained by environmental disparities, and therefore potentially enhance genetic effects
on learning and achievement (Kovas & Malykh, 2016).
High heritability of education may in turn be reected in the variance of other traits.
For example, a large twin study from the Finnish Twin Cohort found a 0.15 correlation
between educational attainment and BMI (Silventoinen et al., 2004). The correlation
was mainly explained by common genetic factors (affecting both educational attainment
and body weight). This may mean that genetic effects on education contribute to
education-associated disparities in BMI in adulthood, although the direction of effects
in such associations remains poorly understood.
Molecular genetic methods have begun to identify specic DNA variants that add up
to the observed heritability of traits (Plomin & Deary, 2015). Many recent technological
advances enable faster progress in this eld. For example, microarray technology
enables the genotyping of hundreds of thousands of DNA variants simultaneously.
Microarrays can also be used to evaluate epigenetic processes across the genome at
the same time. Whole-genome sequencing is becoming increasingly more accessible. It is
likely that in the near future most DNA variants that drive the heritability of each trait
will be established. Two recent large-scale molecular genetic studies, using DNA and
attainment data from 127,000 and 329,000 participants, have already identied a
number of DNA variants involved in educational attainment and achievement (Rietveld
et al., 2013; Okbay et al., 2016).
Although the effect of each genetic variant is weak, a constellation of a large number
of such genetic factors (polygenic score; Plomin & Deary, 2015) can be used for
prediction. One recent study (Selzam et al., 2016) aggregated polymorphisms implicated
in educational outcomes by previous studies into a polygenic score and examined
10 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
whether this score predicted variance in educational outcomes at ages 7, 12 and 16 in a
sample of 5825 UK individuals. The score predicted up to 9% of the variation in
achievement at age 16 (accounting for 15% of the heritable variance). Continued
progress in genetic research is likely to improve peoples ability to predict learning
outcomes from genetic data, enabling early diagnosis of learning disabilities. Moreover,
this information will be invaluable in further rening of diagnoses and overall better
understanding of the multidimensional nature of traits. Better understanding of the
genetic architecture of complex traits will lead to new research into specic mechanisms
of geneenvironment co-action, ultimately contributing to tailoring learning
environments to every students individual needs.
GeneGini interplay: how genetic effects on life outcomes change as a function of equality
This section examines whether the strength of genetic effects on educational outcomes
differ across environmental conditions. As inequality is a major factor in creating
differences in environments, the term GeneGini interplayis used in this paper to
describe the geneenvironment processes. Heritability may be affected by different socio-
cultural indices, including absolute wealth of the population (e.g. poverty-related
limitations on development such as lack of nutrition, access to education), distribution of
wealth (more or less equal opportunity) and educational policies (e.g. compulsory
education, standardized curricula).
A disadvantaged environment may reduce genetic effects because it lacks food
(educational opportunity, resources, encouragement, choice of activities) for childrens
genetic appetitefor learning. In other words, genetically inuenced potentials for
adaptive functioning may express more in enriched environments (see Tucker-Drob &
Bates, 2016, for more detail on the Scarr-Rowe Hypothesis of Gene by SES interaction).
Equal provision of quality education may buffer the effects of some of the negative
environments and also enable evocative and active geneenvironment correlation
processes. Geneenvironment correlation arises as children are exposed to
environmental effects in a non-random manner depending on their genotypes (Plomin
et al., 1977; Plomin & Bergeman, 1991). In childhood, correlations are mostly passive
and evocative: parents transmit genetic variants to their children and create
environments for their children that partly reect the parentsgenetic make-up;
parents also modify their own behaviours responding to specic (partly genetically
driven) behaviours of their children (e.g. Tucker-Drob & Harden, 2012; Henry et al.,
2016). Active geneenvironment correlations emerge when children actively choose their
educational environment and activities that correspond with their genetically inuenced
abilities and interests. In the presence of geneenvironment correlation, MZ twins
similarity on a trait remains the same across development, as their genetic
predispositions lead to similar environments. In contrast, DZ twins become less
similar with age, as they experience progressively more and more different environments.
Limiting environments, such as lack of quality resources, may suppress gene
environment correlation processes by suppressing expression of the relevant genes.
In addition, as discussed above, socioeconomic inequality may relate to educational
disparities via inammatory and other pathways impacting the structural integrity of
brain networks (Gianaros et al., 2012). These processes may have cascading implications
Genes and Gini 11
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
for heritability. Reduced structural integrity of brain networks may lead an individuals
developmental trajectory diverging from those enabled by their genetic propensities.
For example, genetic predispositions for advanced reading ability may not be expressed
if negative neuro-developmental processes affect brain structure and function,
suppressing reading ability and motivation. In this case, genetic differences among
people would not translate into differences in their reading ability.
Even in the presence of high educational standards and equity, available
environments may be inadequate for some students. For example, in the Netherlands,
average mathematical performance level in primary education is relatively high, largely
because the weakest students are performing better than the weakest students from other
countries participating in PISA and other international studies (Schwabe et al., 2017).
This is probably due to signicant efforts of the Dutch educational system to help
weaker students to achieve adequate educational levels (Schwabe et al., 2017). A recent
behavioural genetic study suggested that the overall heritability of mathematical ability
among Dutch 12-year-olds was high (0.73), suggesting that much of the negative
inuences of home environments on achievement (particularly at the low end) have been
reduced. However, there was an indication that environments explained more variance
for gifted students, potentially because inadequate environments suppressed some of the
genetically driven giftedness (Schwabe et al., 2017). Inadequatein this case refers to a
lack of advanced opportunities that would allow for the expression of the relevant genes.
Good educational provisions may also suppress some of the negative genetic effects.
For example, compulsory secondary education requires that students remain at school
until 16 or even 18 years of age (e.g. in the UK). This may reduce the effects of genetic
propensities on educational attainment (years in education) and associated outcomes.
With genetic contributions to dropping out of school minimized, these effects cannot
explain variance in whether a person progresses to higher education and for how long.
Because different processes unravel simultaneously and may have opposing effects
(enhancing vs suppressing genetic propensities), average heritability differences across
populations may be small and difcult to detect. Nevertheless, several specic
predictions can be formulated in relation to the effects of wealth (e.g. PPP) and
equality (Gini) on heritability of educational traits:
(1) When absolute wealth in a population is low (low PPP), heritability is low, as
deprivation limits genetic expression. (2) When absolute wealth is high (high PPP), but
resources are distributed unequally (high Gini), the overall heritability is lower than in
populations with more equally distributed resources. This is because there are many
more inequality-disadvantaged than inequality-advantaged people, and heritability
within disadvantaged groups may be lower due to reduced opportunity and lack of
geneenvironment processes. For some traits, under certain conditions, heritability may
also be reduced in advantaged environments, as greater education and opportunity may
buffer some negative genetic propensities. (3) When inequality is high (high Gini),
irrespective of absolute wealth, heritability is lower. This is because the experience of
inequality (especially for those disadvantaged) may be an environmental factor that
limits expression of relevant genetic factors. The experience of inequality may activate
detrimental for health and education cascading psychological and biological reactions,
including anxiety, depression and chronic stress. When environmentally triggered, such
effects could contribute to lower heritability of achievement (especially in the
12 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
disadvantaged groups). (4) When absolute wealth is high and more equally distributed
(high PPP and low Gini), a very high proportion of the variance is explained by genetic
The best support for the existence and specic mechanisms of the GeneGini
interplay would come from international studies examining heritability of the same traits
in different populations. However, in practice, such research is extremely difcult to
conduct: large twin studies are needed, representing comparable populations (e.g. school
children of the same age), assessed using the same measures and procedures. The
comparisons are further limited because international samples and birth cohorts differ
on many parameters, such as educational set-ups, nutrition and other socio-cultural
characteristics. This complicates drawing conclusions about the true sources of observed
differences in heritability. There is another problem with testing the GeneGini interplay
hypothesis through international comparisons. To date, most data have been collected
in countries that are relatively similar to each other in terms of educational and cultural
norms. This reduces the power to detect effects of these norms on heritability. In the
most comprehensive meta-analysis to date, twin data were available from only
39 countries, of which most studies were based on US twin samples (34%). Samples
from South America, Africa and Asia were heavily under-represented (5.7%, combined)
(Polderman et al., 2015).
These limitations preclude a robust test of the four predictions. However, available
data allow a preliminary examination of the GeneGini interplay hypothesis. The
following sections review three strands of evidence on whether heritability of educational
traits differs: (1) among population subgroups within a country at a particular time (e.g.
people of lower and higher socioeconomic status); (2) among birth and age cohorts
within a country (e.g. people born before and after social reforms; pre-school and school
children); and (3) across countries, at a particular time.
Subgroups within a country at a particular time
A number of studies have provided evidence for variation in genetic inuences on
intelligence and educational achievement across families, schools and states within
nations, in particular, as a function of socioeconomic status (SES). Several studies have
found that genetic factors have greater effects on IQ and education at higher home SES
levels (see Tucker-Drob et al., 2013, for a review). In addition, a recent study found that
school SES may also moderate genetic inuences on verbal intelligence at the individual
level (Schwartz, 2015). The study used a sample of over 2000 sibling pairs, drawn from
the National Longitudinal Study of Adolescent to Adult Health (AddHealth) in the
United States. The school-level SES measure was based on the average of the household-
level SES measure for each of the students attending each of the 144 schools included in
the study. The results indicated that genetic factors have a greater inuence on verbal IQ
for students who attend schools with higher levels of SES, and shared environmental
factors have a greater inuence on verbal IQ for students who attend schools with lower
levels of SES. However, the differences were signicant only at extremely high and low
levels of SES (within the top or bottom 10th percentiles).
Several studies, conducted outside the US, have failed to replicate the moderating effect
of SES on heritability. For example, a recent study of a representative sample of 1176
Australian adolescent twin pairs found no evidence for higher heritability of intelligence
Genes and Gini 13
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
among those from higher SES backgrounds (Bates et al., 2016). Mean IQ was modestly
higher for higher SES participants, but heritability was equally high (0.83) at all levels of
SES. It is possible that equity of school provision in todays Australia buffers against wealth
and income inequality and allows for greater expression of genetic potential.
Similarly, a study conducted in the UK failed to replicate the moderating effect of SES on
heritability, despite being powered to detect even modest effects (Hanscombe et al., 2012). The
study was based on the data from 8716 twin pairs from the Twins Early Development Study
(TEDS), born between 1994 and 1996. Data on intelligence were collected when the children
were 2, 3, 4, 7, 9, 10, 12 and 14 years of age. The genetic effects on intelligence were similar in
low- and high-SES families, again suggesting that relatively uniform school provision buffers
against wealth and income inequality to some extent. The variance in intelligence was greater
in low-SES families, which was explained by moderation of the environmental effect an
environmentenvironment interaction. Childrens shared environments (rather than
individual specic experiences) appeared to explain the greater variation in intelligence in
lower SES (Hanscombe et al., 2012).
Yet another study, using the UK TEDS sample (Asbury et al., 2005), found that
heritability was greater (not lower) for children from families with a disadvantaged home
atmosphere. The study examined the associations of verbal and non-verbal abilities with ten
measured environments, using a sample of 4-year-old same-sex twins (N=4446 children).
The results showed higher heritability of verbal ability among children whose home was
more chaotic, and whose parents engaged in less communication with their children.
Therefore, heritability can be greater in high-risk environments, with genes providing a
diathesis and environment acting as a trigger for its effect (diathesis-stress model). The lower
heritability in a better home environment might mean that such an environment suppresses
some of the negative genetic effects on early verbal development.
The inconsistent pattern of results across studies indicates that the degree and
direction of the gene by SES interaction differ across populations. A recent meta-
analysis further supported this explanation in relation to intelligence and academic
achievement (Tucker-Drob & Bates, 2016). The study examined differences in
heritability between higher and lower SES groups within the US, Australia, Germany,
the Netherlands, Sweden and the UK. The meta-analytic dataset consisted of 43 effect
sizes from a total of 24,926 pairs of twins and siblings with varying degree of genetic
relatedness from fourteen independent studies. Data were relatively evenly split across
the US and non-US samples: US (eighteen effect sizes, eight studies, 10,831 twin or
sibling pairs); and Western Europe and Australia (25 effect sizes, six studies, 14,095 twin
or sibling pairs). Family SES during childhood was based on parental education, family
income or parental job prestige. Sophisticated methods for estimating effects were used,
accounting for the publication bias and between-study heterogeneity.
The results indicated gene by SES effects in the US: genetic variance in intelligence
was signicantly lower (0.24) at 2 standard deviations (SD) below the mean SES than at
2 SD above the mean SES (0.61). In contrast, zero, or even small reversed effects were
found in studies from Western Europe and Australia. In an unconditional meta-analytic
model, the meta-analytic mean for the heritability by SES interaction was
non-signicant. When a dummy-coded nation variable (0 =non-US, 1 =US) was
added to the meta-regression model, there was a signicant difference between
heritability by SES effect sizes from the US compared with those from Western
14 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Europe and Australia. The SES by heritability interaction found in the US samples was
not a product of methodological differences between the studies, such as childhood age
or how SES, achievement and intelligence were measured. Previous research also
suggests that the observed pattern in the US cannot be explained by racial differences
within the population. For example, heritability by SES effects were still present in a
nationally representative sample of American children when race and its effects on
genetic and environmental components were controlled (Tucker-Drob et al., 2011).
The nding of differences in the strength of heritability between lower and higher
SES in the US, but not in other countries, may be explained by differences in resource
distribution across nations. The Western Europe countries and Australia have more
uniform access to high-quality education, health care, overall social welfare and greater
social mobility than the US (Tucker-Drob & Bates, 2016). As described above, the US
has a higher wealth and income Gini, and a greater proportion of people living below the
line of poverty than Western Europe countries and Australia (Social Policy Research
Centre, 2016; OECD, 2017).
Birth cohorts within and across countries
Another test of GeneGini interplay is examining whether changes in social
provision and equality within a country are accompanied by changes in heritability.
Several studies have provided relevant data for educational attainment (years in
education/degree achieved). Academic attainment is sensitive to many socio-cultural
trends, including general access to education, whether and how much education is
compulsory and whether it is equally available to males and females. Previous research
has found that overall heritability of educational attainment (level of education
achieved) is somewhat lower than that of educational achievement (success of mastering
the curriculum). This should particularly be the case in populations where there are
environmental restrictions on access to education. A recent meta-analysis (Branigan
et al., 2013) examined available data (twin correlations) from ten countries to estimate
heritability of educational attainment in 34 subgroups, split by sex, birth cohort and
nationality. The grand mean heritability of educational attainment for the ten countries
in the meta-analysis was 0.40, consistent with other research. For example, in the UK
TEDS study, heritability was 0.44 for the decision to stay in education for an additional
2 years beyond the compulsory requirement (Rimfeld et al., 2016). In contrast, the
choice of specic subjects (e.g. mathematics vs history) to study for these additional
years was more heritable (0.50.8), and the heritability of grades (success) in different
subjects at the end of the 2 years was on average 0.59. These results suggest that
opportunity/decision to continue education is particularly affected by family and other
socio-cultural factors, such as availability of educational resources (e.g. extra tutoring in
preparation for state exams) or affordability of higher education. In contrast, genetic
factors explain more variance in educational success and choice of subjects, especially in
uniform educational systems like that of the modern UK.
The meta-analysis of heritability of educational attainment in different birth cohorts
provided further evidence for GeneGini interplay (Branigan et al., 2013). Overall,
genetic variation explained 8 percentage points more of the variance in attainment for
those born after 1950 than for those born earlier. The authors suggested that the lower
heritability in earlier cohorts reects greater dependency of educational attainment on
Genes and Gini 15
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
social class for respondents in the earlier cohorts. The trend of increased heritability in
later cohorts was also observed for individual countries in the meta-analysis (although
most differences did not reach statistical signicance). For example, in Norway, for
which three cohorts were available, heritability of educational attainment was: 0.18
(males) and 0.28 (females) for those born in 19151939; 0.68 (males) and 0.34 (females)
for those born in 19401949; and 0.76 (males) and 0.46 (females) for those born in 1950
1960. This increase in heritability happened along with the overall decrease in income
Gini for Norway (Chartbook of Economic Inequality, 2017); increased access to
education aided by government-sponsored student loans and other social changes; and a
dramatic growth in educational attainment for Norwegian adults (from 5.92 years in
1960 to 11.86 years by 2000; Tucker-Drob et al., 2013).
Similar trends were observed in the three Australian samples in the meta-analysis.
Heritability was: 0.34 (males) and 0.44 (females) for those born in 18931950; 0.54
(males) and 0.52 (females) for those born in 19511965; and 0.28 (males) and 0.77
(females) for those born in 19641971. Although most differences were not statistically
signicant, the results overall are consistent with increase in heritability in educational
attainment, with shifts towards universal access to education and other positive social
changes. The income Gini in Australia has decreased from 1950, remaining moderate
and relatively stable in the last three decades (World Bank, 2017a; Chartbook of
Economic Inequality, 2017).
Interestingly, the results for the US cohorts in the meta-analysis did not follow the trend
of increased heritability in later cohorts. For the six available US samples, heritability was:
0.44 (males) for those born in 19171927; 0.46 (males)/0.3 (females) for those born in 1936
1955; 0.44 (males) for those born in 19391957; 0.29 (males) and 0.26 (females) for those
born in 19211969; 0.34 (males) and 0.36 (females) for those born in 19111985; and 0.27
(males) and 0 (females) for those born in 19761984. These results suggest a trend towards a
decrease in heritability in more recent cohorts. This trend is mirrored by the increase of Gini
in the US, starting in the late 1970s until today, as described in the Gini section above.
One country (not included in the meta-analysis), Russia, has gone through similar
decreases in equality of wealth and income distribution as the US. In the 1990s Russia
underwent an enormous change in socioeconomic structure a transition from more
equally distributed resources (in the Soviet Union) to a much greater inequality, with
wealth and income inequality almost doubling during the transition (Galbraith et al.,
2004). For example, income Gini for Russia increased from 0.24 (1988) to 0.48 (1993)
and has remained high (and similar to that of the US) since then (e.g. 0.41 in 2005; 0.42
in 2012) (World Bank, 2017a). The absolute wealth in Russia dropped from a PPP of US
$20,639 in 1990 to that of US$12,813 in 1995, then rose again to US$23,895 in 2015
(UNDP, 2016). Therefore, the mid-1990s was a particularly problematic period, when
inequality emerged across all areas of life, including in educational provision,
accompanied by worsening of funding and educational conditions in schools (Voronin
et al., 2016). Prior to the 1990s, the school system was highly standardized, with all
children receiving free school meals, the same free textbooks and a uniform curriculum.
Today, although the curriculum remains relatively standardized, there is a greater choice
of programmes available to schools; private schools are available at high cost; and
private tuition is widely used by those who can afford it for their children. According to
PISA, the average academic performance in mathematics (at age 15) improved in Russia
16 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
between 2003 and 2012. However, the overall educational equity deteriorated, reected
in the increased percentage of variation in mathematics performance explained by the
PISA index of economic, social and cultural status (OECD, 2013).
Although no large-scale twin studies have been conducted in Russia, the available
data from smaller studies suggest a similar pattern to that observed in the US. For
example, in a study of 160 young twins (1628 years old) who were school children in the
1980s, substantial heritability was found for verbal, performance and full-scale IQ,
measured with the Wechsler Adult Intelligence Scale (WAIS): 0.86, 0.89 and 0.89,
respectively (Malykh et al., 2005). Another small study from the same period assessed
intelligence in 123 pairs of 16-year-old Russian twins (Grigorenko et al., 1992) and
found overall moderate heritability: 0.49 for Performance IQ and 0.29 for Verbal IQ,
assessed by WAIS; and 0.97 for mathematics, 0.57 for spatial and 0.17 for verbal
subtests of the Amthauer Intellectual Test. The two studies differed in age of
participants, which might explain the lower heritability in the younger sample
(Grigorenko et al., 1992). The overall moderate to high heritability found in both
studies could reect the social homogeneity in Russian society at the time of assessment,
eliminating much of the environmental variance between families.
In contrast, low heritability was found after the reforms of the 1990s. One study of
580 twins in Russia examined the heritability of non-verbal ability (measured by
Standard Ravens Progressive Matrices) in adolescents aged 1014 years born in the
mid-1990s (Malykh et al., 2016). Both MZ and DZ twins were highly similar, suggesting
that individual differences in non-verbal ability were explained almost entirely by family
environment (65%) and person-specic environment (29%), with minimal genetic effects
(6%). Another study also found low heritability in the data collected in 19941997 and
19972000 (Voronin et al., 2016). Average heritability was 0.27 (with six out of eleven
subtests showing non-signicant genetic inuences) for intelligence, measured by the
Wechsler Intelligence Scale for Children (WISC-III) in a sample of 201 twin pairs of
13-year-old twins. The same study showed average heritability of 0.27 (with eight out of
eleven subtests showing non-signicant genetic inuences) for intelligence measured with
the WAIS-III in 148 pairs of 16-year-old twins. Overall, heritability of intelligence found
in the two studies was lower than that found in earlier cohorts and in other countries.
For example, heritability of non-verbal intelligence in adolescents born in the mid-1990s
in the UK was moderate (Hanscombe et al., 2012). Explanations for the low heritability,
suggested by the authors, included assortative mating (which masks heritability through
making DZ twins more similar to each other) and features of parenting, such as parental
involvement in childrens education. Consistent with GeneGini interplay, the authors
also discussed as a potential explanation the lower socioeconomic status of families and
associated decrease in geneenvironment processes, as well as inequality-related increase
in environmental variance (Malykh et al., 2016; Voronin et al., 2016).
Spain is another country that has provided an opportunity to test the interaction
between educational equality and heritability of educational attainment. In 1970 a
specic educational policy was introduced in Spain that decreased environmental
variance in the population and increased equality in educational opportunities (Colodro-
Conde et al., 2015). The 1970 General Law on Education marked, among other changes,
an extension of mandatory education to 14 years of age, and the introduction of the
same curricula for boys and girls (Colodro-Conde et al., 2015). A recent study used a
Genes and Gini 17
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
population-based sample of 1271 pairs of adult twins to analyse the effect of this specic
educational policy on heritability. The unique aspect of the study is that it used a very
specic educationally relevant cut-off (introduction of the reform) for comparison
between the cohorts, rather than less-specic cut-offs, such as pre- vs post-war or
collapse of the political system.
Participants were over 40 when they were interviewed, and for the analysis were divided
into two groups: the pre-reform cohort, consisting of twins born before 1961 (65.4%), and
the post-reform cohort, consisting of twins born in 1961 and after (34.6%). The study
provided further support for greater heritability of educational attainment in conditions of
greater equality in educational opportunities. For men, heritability increased from 0.44 in
the pre-reform cohort to 0.67 in the post-reform cohort. Heritability remained in the same
range for women (0.40 vs 0.34%). The increase in heritability (for men) was probably due to
the implementation of more egalitarian educational policies: as the inuence of family
socioeconomic status decreased, genetic differences between individuals made a greater
impact on the variation of educational attainment.
There is no denitive explanation for why heritability did not increase in the post-
reform cohort of women. The total variance in educational attainment was reduced in
the post-reform cohort, with both men and women born after 1960 more likely to have
completed medium and high education levels than those born pre-reforms. However, on
average, fewer women achieved a high educational level regardless of cohort. In men, the
reduction in individual differences was almost exclusively accounted for by shared
environmental factors. In women, the reduction in individual differences occurred due to
reduction of both shared environmental and genetic variance. This could be because
genetic factors, which increase the risk of girls of dropping out of school early, were now
suppressed by compulsory education. In other words, environmental constraints may
have depressed negativeheritability by compelling girls to attain higher levels than they
otherwise would (Tucker-Drob et al., 2013). The absence of an increase in heritability
for women is consistent with studies in other countries that found overall lower
heritability of educational attainment and achievement for women (e.g. Branigan et al.,
2013). It is possible that socio-educational changes unravel more slowly for females, with
benets of reforms reaching women later than men. Consistent with this explanation, a
number of more recent studies reported minor or no differences between men and
women in the weight of genetic and environmental factors on variation in academic
achievement (e.g. Shakeshaft et al., 2013). This could be the result of the substantial
recent efforts to provide equal gender opportunities and the overall positive change in
attitudes towards gender equality in many countries.
Age cohorts within and across countries
Support for the GeneGini interplay also comes from ndings of increased
heritability with age for some traits. For example, the pattern of results that emerged
from cross-cultural longitudinal research on literacy is consistent with the hypothesis of
greater equality leading to greater heritability (e.g. Kovas et al., 2005; Samuelsson et al.,
2008; Hayiou-Thomas, 2008). The research demonstrated that heritability of reading
skills varies as a function of studentsage and educational set-up, including the extent to
which the formal education system is able to buffer the effects of socioeconomic
inequality. One study (Samuelsson et al., 2008) assessed the literacy (reading uency) of
18 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
the same children at the end of kindergarten (around 6 years of age) and at the end of the
1st grade of school (around 7 years of age) in three cohorts: US, Australia and
Scandinavia (Norway and Sweden). At the end of kindergarten, genetic factors
explained different amounts of variability in reading ability in the three populations:
0.84 in Australia; 0.68 in the US; and 0.33 in Scandinavia. The effects of shared
environment were 0.09, 0.25 and 0.52, respectively. However, one year later, by the end
of the rst grade, the differences in genetic and environmental effects between the three
populations disappeared: genetic factors explained most (7983%) of the individual
differences in reading in all three cohorts. The results were very similar for spelling.
The observed differences in pre-school heritability are likely to be explained by
intensity of reading instruction in kindergarten and the home literacy environment. The
Australian pre-school system has a strong emphasis on literacy instruction, with
approximately twice as much instruction as in the US pre-school system. In Scandinavia,
pre-school children rarely receive any literacy instruction in kindergarten; and parents of
pre-school children engage in less shared book reading and letter-based activities with
their children than parents from the US or Australia (Byrne et al., 2006). Although
parentsown reading behaviours are similar across the countries, there is a tradition in
Norway and Sweden not to subject children to formal or informal literacy instruction
in the home during the pre-school years. In the absence of formal instruction,
differences between families in the amount of literacy instruction are more likely to be
reected in childrens literacy skills. However, these differences cannot be the only
explanation for why shared environment explains most of the individual differences in
literacy in Scandinavia. Although home literacy environment was positively correlated
to early literacy skills, it explained only a small proportion of the individual
differences in them. For example, shared book reading, the strongest predictor of
print awareness, explained only 19% of its variance (Byrne et al., 2006). Home
environment is also not independent of genetic effects, as parents shape environments
in line with their own genetic predispositions. Recent molecular genetics research has
provided support for this, nding that DNA variants that contribute to differences in
academic achievement also explain some of the variance in family SES (Selzam et al.,
One explanation for lower heritability of pre-school literacy in Scandinavia and the
US is that the effects of genes that affect early reading and spelling skills cannot be seen
until children participate in intense literacy instruction. The results revealed that many of
the same genetic effects that contribute to pre-school literacy continued to be important
for school literacy. Some new genetic effects also emerged, suggesting that some genes
are specic to further literacy development (Samuelson et al., 2007). These results are
consistent with a large body of behavioural genetic ndings, suggesting that heritability
of pre-school early literacy indicators is on average lower (e.g. Kovas et al., 2005) than
substantial heritability of literacy found throughout formal school education (e.g.
Harlaar et al., 2005; Byrne et al., 2006; Kovas et al., 2013).
The robustly shown pattern of a steady increase in heritability of IQ throughout
development (e.g. Haworth et al., 2010; Polderman et al., 2015; Tucker-Drob & Bates,
2016) also supports the GeneGini interplay. Unlike numeracy and literacy, for which
heritability remains stably high throughout the school system, heritability of IQ is
moderate in primary school and continues to increase throughout education (Kovas
Genes and Gini 19
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
et al., 2013). This may be because there is no formal instruction for intelligence, with
more variance explained by environmental disparity. The gene by age trends for IQ
parallel the gene by SES effects observed in the US (Tucker-Drob & Bates, 2016). It is
possible that both phenomena reect a process of increased and accumulated effects of
geneenvironment interplay (Tucker-Drob & Bates, 2016). Interestingly, similar gene by
age and SES patterns have been reported for variability in BMI. One large study of
Finnish twin pairs born in 19831987 showed that heritability of BMI increased and
shared environment decreased from childhood to adolescence. However, parental
education moderated the genetic and environmental inuences: heritability of BMI was
greater (8587%) for 11- to 12-year-olds with high parental education than those with
lower parental education (6168%) (Lajunen et al., 2012).
The stronger inuence of shared environment in childhood is consistent with
stronger effects of inequality on children (Phipps & Lethbridge, 2008). For example,
inequality-related residential segregation may particularly affect children. Children
generally stay in their neighbourhoods, which differ signicantly in the quality and
quantity of crucial public amenities, such as schools, parks, libraries and recreation
(Bischoff & Reardon, 2014). In the US, children attending public schools spend an
average of only 18% (1000 hours) of their waking hours per year in the school
environment (Kirsch et al., 2016). Therefore, most of a childs development takes place
outside the school environment mainly in the neighbourhood. Moreover, when
disparity in quality across schools is high, the inuence of residential segregation by
income extends to schools. For example, in the US, a child from a low-income family is
two to four times as likely as a child from an afuent family to have classmates with low
skills and with behaviour problems, in both elementary and high school (Duncan &
Murnane, 2016). In the PISA study, students attending socioeconomically advantaged
schools outscored those in disadvantaged schools by more than 104 points in
mathematics (OECD, 2013).
Overall, the pattern of age-related changes in heritability suggests that schools may
buffer at least some of the environmental disparities. Greater equity, afforded by formal
education, leads to less variance in childrens achievement explained by environmental
factors, and more variance explained by genetic factors.
The reviewed research suggests that equality/inequality can have profound effects on the
heritability of educational outcomes in a population. Not all estimates in the reviewed
literature are consistent. This is not surprising as many samples are underpowered and
measures are not always precise. However, the overall pattern of results from recent
large-scale studies and meta-analyses is consistent with the GeneGini hypothesis. The
evidence points to three major conclusions regarding variability in educational
(1) When inequality is high, genetic factors explain more variance among people of
higher SES than among those of low SES. In populations where access to quality
education is the privilege of the rich, differences in academic outcomes are largely due to
socioeconomic disparities.
20 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
(2) When environmental provisions are relatively equally accessible and uniform (e.g.
high-quality unied compulsory education), genetic factors explain most of the
differences among people. Therefore, the strength of heritability can be viewed as a
measure of environmental equality in a population. The observed sex and cohort
differences in the proportion of variance in educational attainment explained by genetic
factors suggest that heritability may also be viewed as an estimate of social mobility:
decrease in heritability and increase in shared environment may reect a decrease in the
opportunity for mobility (Tucker-Drob et al., 2013).
(3) After environmental disparities are reduced, much variability in educational
outcomes remains. This variability is largely driven by genetic differences. Heritability
can also be affected by benign inequality, such as differences across countries in when
children start receiving formal instruction (e.g. in literacy): heritability of a trait (e.g.
reading ability) increases when formal instruction begins.
Reducing the effects of environmental disparity on educational achievement is a challenge
for every society. Another challenge is to address the implications of greater heritability at
greater equality levels. One common misconception is that traits that show substantial
heritability are not modiable through environmental interventions. This is not correct. For
example, heritability of human height is around 80%, yet huge changes have occurred in the
populations average height in recent generations suggesting that genes inuence complex
outcomes through channels that are modiable (Cesarini & Visscher, 2017). Another
common misconception is that high heritability of traits can be used to support eugenic
agenda of manipulating populationsgenetic pools to achieve greater outcomes for societies
(see Afterword for discussion). As this review has shown, genetic effects can be enhanced or
suppressed without any changes in the genetic pool of a population.
With genetic testing allowing for more precise predictions of outcomes, one worry is
that education may adopt gene-based ability tracking/streaming. Ability streaming,
practised in many educational systems, has (to some extent) always been implicitly based
on genetic information, as children are allocated to streams on the basis of their previous
achievement. However, the existing crude system of streaming does not benet
struggling students. In fact, research has shown that early tracking/streaming may
decrease achievement of all students and also increase educational inequality, by leading
to a greater reduction in achievement for low-achieving students than high-achieving
students (Hanushek & Wobmann, 2006). To the extent that the early achievement gap is
associated with differences in family background, track placements may reect these
differences and lead to further disparities in education. Lower achievement tracks may
also not provide challenging or motivational environments to activate genetic
propensities. Indeed, tracking may be another factor that leads to a decrease in
heritability for educational achievement.
Education of the future will have to do much more to address not just wealth
disparities, but also genetic disparities. Children with propensities for learning difculties
will require special and enhanced professional support. Similarly, schools will need to
adapt to the needs of gifted students, so that inadequate environments do not limit their
genetic propensities. The use of educational technologies may enable all children to
optimize their learning, including by studying at different times of the day (suitable for
individual chronotype); and learning material in a different order and pace supporting
individual interest and ability trajectories. Teachers of the future may assume a role of
Genes and Gini 21
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
multi-disciplinary experts who are able to direct and assist individual childrens
trajectories. These efforts will not eliminate the achievement gap, but would be likely to
reduce variability, particularly by enabling greater performance for lower achieving
learners. This narrowing of the achievement gap has already been demonstrated in
populations that have managed to reduce environmental disparity. As demonstrated in
these populations, the remaining variability is likely to still be substantially heritable.
However, by reducing some of the negative effects of genes, personalized education will
lead to better results for everyone.
In isolation, differences in wealth or equality across populations do not directly map
onto the populationsaverage achievement rankings. A simple analysis was conducted
plotting the ranks of twelve countries participating in PISA for three indices: PPP, Gini
and performance on a standardized test of mathematical ability of 15-year-old students
(OECD, 2016, 2017). The countries (presented in decreasing order for PISA results)
were: China, Holland, Canada, Japan, Australia, Denmark, Germany, Sweden, UK,
Norway, Russia and the US. There was no obvious relationship between population
average PISA scores, PPP and Gini. For example, out of the twelve countries, China had
the second worst PPP and the worst Gini, but scored the highest out of the twelve
countries (and out of 34 OECD countries included in PISA 2012). The US had a good
PPP, but high Gini, and scored the lowest of the twelve countries (and below average of
the 34 countries). Norway had the best PPP and low (good) Gini, but about average
performance on PISA. This suggests that many factors contribute to the average
performance of a nation and these factors may differ between countries. As discussed in
this review, GeneGini interplay may be one key factor that affects variance and average
achievement of populations. Further research, with large cross-cultural datasets, should
also test whether an interaction between PPP and Gini may explain more variance in
educational achievement than each indicator alone.
The focus of this review is on outcomes for individuals within populations an
attempt to understand how environmental set-ups, including access to resources,
suppress or enhance the genetic propensities of individuals. Understanding the dynamic
ongoing processes of GeneGini interplay will help educational systems to evolve
further, so that they are adapted to the individual needs of all students.
Afterword (Gini and eugenics)
Corrado Ginis signicance to this paper was limited to that of a statistician who
developed the Gini Coefcient. However, learning about his interest in eugenics added
an unexpected twist to the title of this review and called for a few words on eugenics. For
Corrado Gini, statistics was a methodological lens through which it was possible to
observe populations, economies, social classes, wars and other large-scale phenomena
(Prévost, 2016). The application of statistics to eugenics was his particular passion
(Berlivet, 2016), culminating in his appointment in October 1935 as the inaugural
president of the Latin International Federation of Eugenic Societies. This, together with
his role in various academic and governmental ofces before, during and after the fascist
regime in Italy, makes Corrado Gini a controversial gure (Macuglia, 2014).
Eugenics originated from Francis Galtons theorizing and developed into a popular
movement in the UK in the early 20
century, spreading to many countries.
22 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
The central purpose of the eugenic movement was described as improving a human
population (stock) by controlled breeding to increase the occurrence of desirable
heritable characteristics( Particularly in the Anglo-
Saxon eugenic model, its purpose was formulated as: eliminating from subsequent
generations beings inferior by heredity(negative selection); and increasing the
reproductivity of the best(positive selection). This led to unethical practices in many
countries, including forced sterilization of people deemed unt and marriage
prohibition, and encouraging particularly tindividuals to reproduce (see Kevles,
1985, for a review).
It is necessary to mention that Corrado Gini himself did not support this agenda
for eugenics. According to him, societies were too ignorant of the mechanisms of
genetic and environmental inuences to exercise social control on populations
(Gini, 1932). In fact, he recognized that attempts to reduce the genetic variability
of a population may have catastrophic effects. In his view, the purpose of eugenics was
to study successive generations, isolated populations and migrants and how
populations change over time. He believed that genetics and statistics provided the
foundation of such study. In fact, his proposed programme and motto for eugenics
in the immediate future was facts, facts, facts(Gini, 1932). Altogether, Ginis theorizing
can be viewed as accepting that environment and heredity interact in complex ways
(Berlivet, 2016). It therefore seems more appropriate to describe his interest as
interdisciplinary, including sociology, demography, anthropology, genetics and statistics
with an unfortunate focus on eugenics. It is also worth mentioning that historically, the
term eugenics applied also to benign practices, such as avoidance of inbreeding, or
prenatal care for mothers.
Eugenics cast a long shadow over genetic research, with many people rightly
worrying about the potential unethical applications of genetic knowledge. Apart from
the dangers of selective breeding and sterilization, neo-eugenic risks include forced
prenatal people selectionand gene editing. These and other risks can only be avoided
by appropriate safeguards against misuses of genetic information, such as increasing
genetic literacy (Chapman et al., in press), enacting enforceable laws and providing
equal access to justice (Selita, 2018).
Behavioural genetics is fundamentally different from any form of eugenics.
Eugenics is concerned with factors through which societies rise and decline,
and ways to maximize societiespotential. In contrast, the purpose of behavioural
genetics (and of this review) is to gain insight into the complex processes driving
the development of each individual. This includes understanding the inuences that
societies have on individuals, such as the effects of inequality on peoples genetic
Ethical Approval. The authors assert that all procedures contributing to this work comply with the
ethical standards of the relevant national and institutional committees on human experimentation
and with the Helsinki Declaration of 1975, as revised in 2008.
Conicts of Interest. The authors have no conicts of interest to declare.
Funding. This work was supported by the Tomsk State University Competitiveness Improvement
Program (Grant
Genes and Gini 23
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Allianz (2016) Allianz Global Wealth Report 2016. URL:
media/economic_research/publications/specials/en/AGWR2016e.pdf (accessed 5th May 2017).
Alvaredo, F., Atkinson, A. B., Piketty, T. & Saez, E. (2013) The top 1 percent in international and
historical perspective. Journal of Economic Perspectives 27(3), 320.
Ansell, D. (2017) The Death Gap: How Inequality Kills. University of Chicago Press. ISBN:
Asbury, K., Wachs, T. D. & Plomin, R. (2005) Environmental moderators of genetic inuence on
verbal and nonverbal abilities in early childhood. Intelligence 33, 643661.
Atkinson, A. B., Piketty, T. & Saez, E. (2011) Top incomes in the long run of history. Journal of
Economic Literature 49(1), 371.
Bates, T., Hansell, N. K., Martin, N. G. & Wright, M. J. (2016) When does socioeconomic
status (SES) moderate the heritability of IQ?: no evidence for gx SES interaction for
IQ in a representative sample of 1,176 Australian adolescent twin pairs. Intelligence
Berlivet, L. A. (2016) A laboratory for Latin eugenics: the Italian Committee for the Study of
Population Problems and the international circulation of eugenic knowledge, 1920s1940s.
História, Ciências, Saúde Manguinhos 23(1), 5172.
Bick, J., Naumova, O., Hunter, S., Barbot, B., Lee, M., Luthar, S. S. & Grigorenko, E. L. (2012)
Childhood adversity and DNA methylation of genes involved in the hypothalamus-pituitary-
adrenal axis and immune system: whole-genome and candidate-gene associations. Development
and Psychopathology 24(4), 14171425.
Bischoff, K. & Reardon, S. F. (2014) Residential segregation by income 19702009. In Logan, J.
(ed.) Diversity and Disparities: America Enters a New Century. Russell Sage Foundation,
New York.
Branigan, A. R., McCallum, K. J. & Freese, J. (2013) Variation in the heritability of educational
attainment: an international meta-analysis. Social Forces 92(1), 109140.
Braveman, P. & Gottlieb, L. (2014) The social determinants of health: its time to consider the
causes of the causes. Public Health Reports 129(1) (Supplement 2), 1931.
Byrne, B., Olson, R. K., Samuelsson, S., Wadsworth, S., Corley, R., DeFries, J. C. & Willcutt, E.
(2006) Genetic and environmental inuences on early literacy. Journal of Research in Reading
29(1), 3349.
Carter, P. L. & Reardon, S. F. (2014) Inequality Matters. William T. Grant Foundation URL: (accessed 31st May 2017).
Ceriani, L. & Verme, P. (2012) The origins of the Gini index: extracts from Variabilità e
Mutabilità (1912) by Corrado Gini. Journal of Economic Inequality, doi: 10.1007/s10888-011-
Cesarini, D. & Visscher, P. M. (2017) Genetics and educational attainment. npj Science of
Learning 2(4), doi:10.1038/s41539-017-0005-6.
Chapman, R., Likhanov, M., Selita, F., Zakharov, I., Smith-Woolley, E. & Kovas, Y. (in press)
Genetic Literacy and Attitudes Survey (iGLAS): International Population-Wide Assessment
Instrument. Journal of Community Genetics.
Chartbook of Economic Inequality (2017) Inequality by Country. URL: www.chartbookof (accessed 27th May 2017).
Chiang, T. L. (1999) Economic transition and changing relation between income inequality
and mortality in Taiwan: regression analysis. British Medical Journal 319(7218), 11621165.
Cobham, A., Schlogl, L. & Sumner, A. (2015) Inequality and the tails: the Palma Proposition and
Ratio revisited. United Nations Department of Economics and Social Affairs. Working Paper
No. 143, doi: ST/ESA/2015/DWP/143.
24 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Colodro-Conde, L., Rijsdijk, F., Tornero-Gómez, M. J., Sánchez-Romera, J. F. & Ordoñana, J. R.
(2015) Equality in educational policy and the heritability of educational attainment. PLoS One
10(11), e0143796.
Corak, M. (2013) Income inequality, equality of opportunity and intergenerational mobility.
Journal of Economic Perspectives 27(3), 79102.
County Health Rankings and Roadmaps (2015) Key Findings Report 2015. URL: http://www.rwjf.
org/content/dam/farm/reports/reports/2015/rwjf418649 (accessed 31st May 2017).
Coussons-Read, M., Okun, M. & Nettles, C. (2007) Psychosocial stress increases inammatory
markers and alters cytokine production across pregnancy. Brain, Behavior, and Immunity 21(3),
Crawford, R., Innes, D. & ODea, C. (2016) Household wealth in Great Britain: distribution,
composition and changes 200612. Fiscal Studies 37(1), 3554.
Credit Suisse (2014) Global Wealth Databook 2014. URL:
research/research-institute/global-wealth-report.html (accessed 31st May 2017).
Credit Suisse (2016) Global Wealth Databook 2016. URL:
tasks/render/le/index.cfm?leid=AD6F2B43-B17B-345E-E20A1A254A3E24A5 (accessed 31st
May 2017).
Davies, J. B. (2016) The Gini Coefcient and Personal Inequality Measurement. Department of
Economics Research Report 2016-1. Department of Economics, University of Western Ontario,
London, ON, Canada. URL: (accessed 30th May
De Maio, F. G. (2007) Income inequality measures. Journal of Epidemiology and Community
Health 61(10), 849852.
Duncan, G. & Murnane, R. (eds) (2011) Whither Opportunity? Rising Inequality, Schools, and
Childrens Life Chances. Russell Sage Foundation, New York.
Duncan, G. J. & Murnane, R. J. (2016) Rising inequality in family incomes and childrens
educational outcomes. Russell Sage Foundation Journal of the Social Sciences 2(2), 142158.
Ford, E., Loucks, E. & Berkman, L. (2006) Social integration and concentrations of c-reactive
protein among US adults. Annals of Epidemiology 16(2), 7884.
Galbraith, J. K., Krytynskaia, L. & Wang, Q. (2004) The experience of rising inequality
in Russia and China during the rransition. European Journal of Comparative Economics 1(1),
Gastwirth, J. L. (2016) Is the Gini Index of Inequality Overly Sensitive to Changes in the Middle of
the Income Distribution? SSRN. URL: (accessed 31st May
Gianaros, P., Marsland, A., Sheu, L., Erickson, K. & Verstynen, T. (2012) Inammatory pathways
link socioeconomic inequalities to white matter architecture. Cerebral Cortex 23(9), 20582071.
Gini, C. (1932) Address at the Third International Congress of Eugenics. American Museum of
Natural History, New York City. URL:
Grigorenko, E. L., LaBuda, , Michele, C. & Carter, A. S. (1992) Similarity in general cognitive
ability, creativity, and cognitive style in a sample of adolescent Russian twins. Acta Geneticae
Medicae et Gemellologiae: Twin Research 41(1), 6572.
Hales, S., Howden-Chapman, P., Salmond, C., Woodward, A. & Mackenbach, J. (1999) National
infant mortality rates in relation to gross national product and distribution of income. Lancet
354(9195), 2047.
Hanscombe, K. B., Trzaskowski, M., Haworth, C. M., Davis, O. S., Dale, P. S. & Plomin, R.
(2012) Socioeconomic status (SES) and childrens intelligence (IQ): in a UK-representative
sample SES moderates the environmental, not genetic, effect on IQ. PLoS One 7(2),
Genes and Gini 25
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Hanushek, E. A. & Wobmann, L. (2006) Does educational tracking affect performance and
inequality? Differences-in-differences evidence across countries. Economic Journal 116,
Harlaar, N., Spinath., F. M., Dale, P. S. & Plomin, R. (2005) Genetic inuences on early word
recognition abilities and disabilities: a study of 7-year-old twins. Journal of Child Psychology &
Psychiatry 46(4), 373384.
Haworth, C. M. A., Wright, M. J., Luciano, M., Martin, N. G., de Geus, E. J. C., van Beijsterveldt,
C. E. M. et al.(2010) The heritability of general cognitive ability increases linearly from
childhood to young adulthood. Molecular Psychiatry 15, 11121120.
Hayiou-Thomas, M. E. (2008) Genetic and environmental inuences on early speech, language
and literacy development. Journal of Communication Disorders 41(5), doi: 10.1016/j.
Henry, J., Boivin, M. & Brendgen, M. (2016) Geneenvironment correlations in the context of
parenting and peer relationships. In Kovas, Y. et al. (eds) Behavioural Genetics for Education.
Palgrave Macmillan, pp. 4259.
Johnson, S. L., Wibbels, E. & Wilkinson, R. (2015) Economic inequality is related to cross-national
prevalence of psychotic symptoms. Social Psychiatry and Psychiatric Epidemiology 50(12),
Jorde, L. B. & Wooding, S. P. (2004) Genetic variation, classication and race.Nature Genetics
36, S28S33.
Kevles, D. J. (1985) In the Name of Eugenics: Genetics and the Uses of Human Heredity. University
of California Press.
Kalil, A., Ziol-Guest, K. M., Ryan, R. M. & Markowitz, A. J. (2016) Changes in income-based
gaps in parent activities with young children from 1988 to 2012. AERA Open, doi: 10.1177/
2332858416653732. URL:
Kaplan, G. A., Salonen, J. T., Cohen, R. D., Brand, R. J., Syme, S. L. & Puska, P. (1988) Social
connections and mortality from all causes and from cardiovascular disease: prospective evidence
from eastern Finland. American Journal of Epidemiology 128(2), 370380.
Kawachi, I. & Kennedy, B. P. (1997) The relationship of income inequality to mortality: does the
choice of indicator matter? Social Science & Medicine 45(7), 11211127.
Kerry, B., Pickett, K. E. & Wilkinson, R. (2010) The Spirit Level: Why Greater Equality makes
Societies Stronger. Child Poverty Insights. URL:
(accessed 31st May 2017).
Kiecolt-Glaser, J., Gouin, J. & Hantsoo, L. (2010) Close relationships, inammation, and health.
Neuroscience & Biobehavioral Reviews 35(1), 3338.
Kirsch, I., Braun, H., Lennon, M. L. & Sands, A. (2016) Choosing Our Future: A Story of
Opportunity in America. Educational Testing Service, URL:
opportunity (accessed 31st May 2017).
Kornrich, S. & Furstenberg, F. (2013) Investing in children: changes in parental spending on
children, 19722007. Demography 50(1), 123.
Kovas, Y., Garon-Carrier, G., Boivin, M., Petrill, S. A., Malykh, S. B., Spinath, F. et al.(2015)
Why children differ in motivation to learn: insights from over 13,000 twins from 6 countries.
Personality and Individual Differences 80,5163.
Kovas, Y., Hayiou-Thomas, M. E., Oliver, B., Dale, P. S., Bishop, D. V. M. & Plomin, R. (2005)
Genetic inuences in different aspects of language development: the etiology of language skills
in 4.5 year-old twins. Child Development 76(3), 632651.
Kovas, Y., Henry, J., Feklicheva, I., Soldatova, E. & Malykh, S. (2016) Early environment and
development: epigenetic mechanisms. In Malykh, S. B. et al. (eds) (2016) Behavioural Genomics:
Child Development and Education. Publishing House of Tomsk State University, Tomsk.
Chapter 10, pp. 226237. ISBN: 978-5-94621-585-5.
26 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Kovas, Y. & Malykh, S. (2016) Behavioural genomics and education. In Malykh, S. B. et al. (eds)
Behavioural Genomics: Child Development and Education. Publishing House of Tomsk State
University, Tomsk. pp. 382391. ISBN: 978-5-94621-585-5.
Kovas, Y., Voronin, I., Kaydalov, A., Malykh, S. B., Dale, P. S. & Plomin, R. (2013) Literacy and
numeracy are more heritable than intelligence in primary school. Psychological Science 24(10),
Krapohl, E., Rimfeld, K., Nicholas, G., Shakeshaft, N., Trzaskowski, M., McMillan, A. et al.(2014)
The high heritability of educational achievement reects many genetically inuenced traits, not
just intelligence. PNAS. URL:
Lajunen, H.-R., Kaprio, J., Rose, R. J., Pulkkinen, L. & Silventoinen, K. (2012) Genetic and
environmental inuences on BMI from late childhood to adolescence are modied by parental
education. Obesity 20(3), 583589.
Loucks, E., Berkman, L., Gruenewald, T. & Seeman, T. (2006) Relation of social integration to
inammatory marker concentrations in men and women 70 to 79 years. American Journal of
Cardiology 97(7), 10101016.
Macuglia, D. (2014) Corrado Gini and the scientic basis of fascist racism. Medicina Nei Secoli
26(3), 821855.
Magnuson, K. & Waldfogel, J. (2008) Steady Gains and Stalled Progress: Inequality and the Black
White Test Score Gap. Russell Sage Foundation, New York. URL:
Malykh, S. B., Iskoldsky, N. V. & Gindina, E. D. (2005) Genetic analysis of IQ in young
adulthood: a Russian twin study. Personality and Individual Differences 38, 14751485.
Malykh, S., Voronin, I., Ismatullina, V., Zaharov, I., Belova, A. & Lobaskova, M. (2016) Genetic
and environmental sources of individual differences in non-verbal intelligence in Russian
adolescents. SHS Web of Conferences 29, doi: 10.1051/ 02026.
Marmot, M. G., Rose, G., Shipley, M. J. & Hamilton, P. J. (1978) Employment grade and
coronary heart disease in British civil servants. Journal of Epidemiology and Community Health
32(4), 244249.
Marmot, M. G., Shipley, M. J. & Rose, G. (1984) Inequalities in death-specic explanations of a
general pattern? Lancet 1(8384), 10031006.
Meaney, M. J. (2010) Epigenetics and the biological denition of geneenvironment interactions.
Child Development 81,4179.
Miller, G. & Chen, E. (2007) Unfavorable socioeconomic conditions in early life presage
expression of proinammatory phenotype in adolescence. Psychosomatic Medicine 69(5),
OECD (2013) Excellence through Equity: Giving Every Student the Chance to Succeed. Volume II.
OECD (2014) PISA 2012 Results in Focus: What 15-year-olds now and what they can do with what
they know. URL: (acces-
sed 1st September 2017).
OECD (2016) PISA 2015 Results in Focus: Excellence and equity in education. URL: https://www. (accessed 31st May 2017).
OECD (2017) OECD Income Distribution Database (IDD): Gini, Poverty, Income, Methods and
Concepts. URL: (accessed 24th
May 2017).
OECD (2017) Poverty Rate (Indicator), doi: 10.1787/0fe1315d-en URL:
inequality/poverty-rate.htm (accessed 1st September 2017).
Okbay, A., Beauchamp, J. P., Fontana, M. A., Lee, J. J., Pers, T. H., Rietveld, C. A. et al.(2016)
Genome-wide association study identies 74 loci associated with educational attainment. Nature
533, 539542.
Genes and Gini 27
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Oxfam (2016) An Economy for the 1%. Oxfam Brieng Paper. URL: org/en/research/economy-1
(accessed 27th May 2017).
Oxfam (2017a) An Economy for the 99%. Oxfam Brieng Paper. URL:
economy-99 (accessed 27th May 2017).
Oxfam (2017b) Just 8 men own same wealth as half the world. Press Release. URL:
pressroom/pressreleases/2017-01-16/just-8-men-own-same-wealth-half-world (accessed 31st May
Phipps, S. & Lethbridge, L. (2008) Income and the Outcomes of Children. Analytical Studies
Branch Research Paper Series No. 281. Statistics Canada. URL:
Pickett, K. & Wilkinson, R. G. (2007) Child wellbeing and income inequality in rich societies:
ecological cross sectional study. British Medical Journal 335, 1080.
Pickett, K. E. & Wilkinson, R. G. (2015) Population health: behavioral and social science insights.
Income inequality and health: a causal review. Social Science & Medicine 128, 316326.
Piketty, T. & Saez, E. (2003) Income inequality in the United States, 19131998. Quarterly Journal
of Economics 118(1), 141.
Piketty, T. & Saez, E. (2014) Inequality in the long run. Science 344(6186).
Pinto, R., Ashworth, M. & Jones, R. (2008) Schizophrenia in black Caribbeans living in the UK:
an exploration of underlying causes of the high incidence rate. British Journal of General
Practice 58(551), 429434.
Plomin, R. & Bergeman, C. S. (1991) The nature of nurture: genetic inuence on environmental
measures. Behavioral and Brain Sciences 14, 373427.
Plomin, R. & Deary, I. J. (2015) Genetics and intelligence differences: ve special ndings.
Molecular Psychiatry 20(1), 98108.
Plomin, R., DeFries, J. C., Knopik, V. S. & Neiderhiser, J. M. (2016) Top 10 replicated ndings
from behavioral genetics. Perspectives on Psychological Science 11(1), 323.
Plomin, R., DeFries, J. C. & Loehlin, J. C. (1977) Genotypeenvironment interaction arid
correlation in the analysis of human behavior. Psychological Bulletin 84(2), 309322.
Polderman, T. J. C., Benyamin, B., de Leeuw, C. A., Sullivan, P. F., van Bochoven, A., Visscher, P.
M. & Posthuma, D. (2015) Meta-analysis of the heritability of human traits based on fty years
of twin studies. Nature Genetics 47(7), 702709.
Prévost, J. G. (2016) Statistics, expertise and politics: Corrado Gini and the course of Italian
history. METRON 74, 145165.
Reardon, F. (2013) The widening income achievement gap. Educational Leadership 70(8), 1016.
Reardon, S. F. (2011) The widening academic achievement gap between the rich and the poor: new
evidence and possible explanations. Community Investments 24(2), 1939.
Rietveld, C. A., Medland, S. E., Derringer, J., Yang, J., Esko, T., Martin, N. W. et al.(2013)
GWAS of 126,559 individuals identies genetic variants associated with educational attainment.
Science 340(6139), 14671471.
Rimfeld, K., Ayorech, Z., Dale, P. S., Kovas, Y. & Plomin, R. (2016) Genetics affects choice of
academic subjects as well as achievement. Scientic Reports 6, 26373.
Rowlingson, K. (2011) Does income inequality cause health and social problems? Report. Joseph
Rowntree Foundation. URL:
income-social-problems-full.pdf (accessed 25th May 2017).
Ryan, B. A. & Adams, G. R. (1998) Family relationships and childrens school achievement: data from
the national longitudinal survey of children and youth. Human Resources Development Canada.URL:
Saez, E. (2015) U.S. income inequality persists amid overall growth in 2014. Washington Centre
for Equitable Growth. URL:
persists-amid-overall-growth-2014/ (accessed 24th May 2017).
28 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
Saez, E. & Zucman, G. (2014) Wealth inequality in the United States since 1913: evidence from
capitalized income tax data. NBER Working Paper No. 20625. URL:
Samuelsson, S., Byrne, B., Olson, R. K., Hulslander, J., Wadsworth, S., Corley, R., Willcutt, E. G.
& DeFries, J. C. (2008) Response to early literacy instruction in the United States, Australia,
and Scandinavia: a behavioral-genetic analysis. Learning & Individual Differences 18(3),
Schwabe, I., Boomsma, D. I. & van den Berg, S. M. (2017) Increased environmental sensitivity in
high mathematics performance. Learning and Individual Differences 54, 196201.
Schwartz, J. A. (2015) Socioeconomic status as a moderator of the genetic and shared
environmental inuence on verbal IQ: a multilevel behavioral genetic approach. Intelligence 52,
Selita, F. (2018) Unrepresented Litigants in Modern Courts Ordeal by Combat. Legal Issues
Journal 6(1).
Selzam, S., Krapohl, E., von Stumm, S., OReilly, P. F., Rimfeld, K., Kovas, Y. et al.(2016)
Predicting educational achievement from DNA. Molecular Psychiatry 22, 267272.
Shakeshaft, N. G., Trzaskowski, M., McMillan, A., Rimfeld, K., Krapohl, E., Haworth, C. M. et al.
(2013) Strong genetic inuence on a UK nationwide test of educational achievement at the end
of compulsory education at age 16. PLoS One 8(12), e80341.
Silventoinen, K., Sarlio-Lahteenkorva, S., Koskenvuo, M., Lahelma, E. & Kaprio, J. (2004)
Effect of environmental and genetic factors on education-associated disparities in weight
and weight gain: a study of Finnish adult twins. American Society for Clinical Nutrition 80,
Singer, B. H. & Ryff, C. D. (eds) (2001) New Horizons in Health: An Integrative Approach.
National Research Council (US). National Academies Press, Washington, DC.
Social Policy Research Centre (2016) Poverty in Australia 2016. Australian Council of Social
Service. URL:
pdf (accessed 6th June 2017).
Sommeiller, M. & Price, M. (2015) The Increasingly Unequal States of America: Income Inequality
by State, 1917 to 2012. EARN - Economic Analysis and Research Network. URL: http://www.
Telenti, A., Pierce, L. C. T., Biggs, W. H., DiIulio, J., Wong, E. H. M., Fabani, M. M. et al.(2016)
Deep sequencing of 10,000 human genomes. PNAS 113(42), 1190111906.
Tucker-Drob, E. M. & Bates, T.C. (2016) Large cross-national differences in gene × socio-
economic status interaction on intelligence. Psychological Science 27(2), 138149.
Tucker-Drob, E. M., Briley, A. D. & Harden, K. P. (2013) Genetic and environmental inuences
on cognition across development and context. Current Directions in Psychological Science 22(5),
Tucker-Drob, E. M. & Harden, K. P. (2012) Early childhood cognitive development and parental
cognitive stimulation: evidence for reciprocal geneenvironment transactions. Developmental
Science 15(2), 250259.
Tucker-Drob, E. M., Rhemtulla, M., Harden, K. P., Turkheimer, E. & Fask, D. (2011) Emergence
of a gene × socioeconomic status interaction on infant mental ability between 10 months and
2 years. Psychological Science 22, 125133.
UNDP (2013) UNDP Income Report 2013. URL:
coefcient (accessed 31st May 2017).
UNDP (2014) UNDP Income Report 2014. URL:
coefcient (accessed 31st May 2017).
UNDP (2016) UNDP PPP Report 2016. URL:
(accessed 24
May 2017).
Genes and Gini 29
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
UNICEF (2016) Fairness for Children: A League Table of Inequality in Child Well-Being in
Rich Countries. UNICEF Innocenti Report, Report Card 13. UNICEF Ofce of Research
Innocenti. URL:
Voronin, I., Nijenhuis, J. T. & Malykh, S. (2016) The correlation between gloadings and
heritability in Russia. Journal of Biosocial Science 48, 833843.
Wilkinson, R. G. (1996) Unhealthy Societies: The Afictions of Inequality. Routledge, London,
New York.
Wilkinson, R. & Marmot, M. (eds) (2003) Social Determinants for Health: The Solid Facts. 2nd
Edition. World Health Organization. URL:
Wilkinson, R. G. & Pickett, K. E. (2006) Income inequality and health: a review and explanation
of the evidence. Social Science & Medicine 62, 17681784.
Wilkinson, R. G. & Pickett, K. E. (2009) The Spirit Level: Why Greater Equality Makes Societies
Stronger. Bloomsbury Publishing, New York, London, New Delhi, Sidney.
Wilkinson, R. G. & Pickett, K. E. (2011) Greater equality: the hidden key to better health and
higher scores. American Educator 35(1), 59.
Witherspoon, D. J., Wooding, S., Rogers, A. R., Marchani, E. E., Watkins, W. S., Batzer, M. A. &
Jorde, L. B. (2007) Genetic similarities within and between human populations. Genetics 176(1),
World Bank (2015) The World Bank Infant Mortality Report (per 1,000 live births) URL: http:// (accessed 24th May 2017).
World Bank (2017a) The World Bank GINI Report. URL:
POV.GINI (accessed 27th May 2017).
World Bank (2017b) World Bank PPP Data. URL:
PCAP.PP.CD (accessed 27th May 2017).
World Bank Institute (2005) Introduction to Poverty Analysis, pp. 95105. URL: http:// (accessed 31 May 2017).
World Economic Forum (2012) Global Risks 2012. Seventh Edition. Insight Report. URL: www3. (accessed 31st May 2017).
30 F. Selita and Y. Kovas
available at
Downloaded from Goldsmiths, University of London, on 01 Feb 2018 at 14:39:39, subject to the Cambridge Core terms of use,
... In particular, critical public health and social work scholars note the ways in which intersecting oppressive structures such as capitalism, colonialism, racism, and patriarchy fuel inequality and poverty (Butler 2019;Dennis and Robin 2020;Gonçalves et al. 2019;Prins et al. 2015). These forces in turn shape both people's subjective assessments of life satisfaction as well as material health and well-being outcomes (Pickett and Wilkinson 2015;Selita and Kovas 2019;Wilkinson and Pickett 2011). Recent research also links social and economic policy that exacerbates inequality to the heritability and expression of traits that affect well-being, in what has been called the "Gene-Gini interplay" -shedding new light on how "inequality gets under people's skin" due to the social structures that condition wellness (Selita and Kovas 2019, p. 22). ...
Full-text available
Use of the term “agroecology” has greatly increased over the past few decades, with scholars, civil society actors, and intergovernmental organizations identifying agroecology as a promising pathway for realizing more just and sustainable food systems. Using a community-engaged approach, we explore how diverse agroecological actors in southern Brazil describe and define agroecology. We find that across a range of social differences, agroecological actors come together in describing agroecology as a philosophy of life that promotes well-being, positioning agroecology as a counter-narrative to global discourses on “development” that promote economic growth and productivism as the path to a good life. Our findings suggest that the practice of agroecology can enhance diverse rural peoples’ well-being by providing a pathway to overcoming alienation, commodification, and exploitation, although broader political-economic conditions constrain this potential. As socioecological crises continue to accelerate and undermine human and environmental health and well-being, the capacity for agroecology to promote wellness from various ontological and epistemological positions warrants further attention.
... If this explanation holds true, then heritability estimates will be higher in developed societies than in lessadvanced cultures. Some evidence in this direction has recently emerged (Selita and Kovas, 2019); however, the "WEIRD" gene problem-that nearly all twin studies have been conducted among Western, educated, industrialized, rich, and democratic societies-represents an obstacle for generalizing results and making inferences about cross-cultural heritability differences (Henrich et al., 2010). ...
Full-text available
There is controversy regarding whether gender differences are smaller or larger in societies that promote gender equality highlighting the need for an integrated analysis. This review examines literature correlating, on a national level, gender differences in basic skills-mathematics, science (including attitudes and anxiety), and reading-as well as personality, to gender equality indicators. The aim is to assess the cross-national pattern of these differences when linked to measures of gender equality and explore new explanatory variables that can shed light on this linkage. The review was based on quantitative research relating country-level measures of gender differences to gender equality composite indices and specific indicators. The findings show that the mathematics gender gap from the PISA and TIMMS assessments, is not linked to composite indices and specific indicators, but gender differences are larger in gender-equal countries for reading, mathematics attitudes, and personality (Big Five, HEXACO, Basic Human Values, and Vocational Interests). Research on science and overall scores (mathematics, science, and reading considered together) is inconclusive. It is proposed that the paradox in reading results from the interrelation between basic skills and the attempt to increase girls' mathematics abilities both acting simultaneously while the paradox in mathematics attitudes might be explained by girls being less exposed to mathematics than boys. On the other hand, a more nuanced understanding of the gender equality paradox in personality is advanced, in which a gene-environment-cultural interplay accounts for the phenomenon. Challenges for future cross-national research are discussed.
... WEIRD societies are also roughly aligned on basic environmental factors such as technology, civil infrastructure, healthcare, schooling, and transportation, with typically advanced levels in each of these. Within each of these countries as well, inequality is low compared to non-WEIRD countries, as can be inferred from the comparatively low Gini coefficients of Western countries, with the United States being somewhat of an unequal outlier (Selita & Kovas, 2019). ...
Uchiyama et al. rightly consider how cultural variation may influence estimates of heritability by contributing to environmental sources of variation. We disagree, however, with the idea that generalisable estimates of heritability are ever a plausible aim. Heritability estimates are always context-specific, and to suggest otherwise is to misunderstand what heritability can and cannot tell us.
... WEIRD societies are also roughly aligned on basic environmental factors such as technology, civil infrastructure, healthcare, schooling, and transportation, with typically advanced levels in each of these. Within each of these countries as well, inequality is low compared to non-WEIRD countries, as can be inferred from the comparatively low Gini coefficients of Western countries, with the United States being somewhat of an unequal outlier (Selita & Kovas, 2019). ...
Epigenetics impacts gene–culture coevolution by amplifying phenotypic variation, including clustering, and bridging the difference in timescales between genetic and cultural evolution. The dual inheritance model described by Uchiyama et al. could be modified to provide greater explanatory power by incorporating epigenetic effects.
... WEIRD societies are also roughly aligned on basic environmental factors such as technology, civil infrastructure, healthcare, schooling, and transportation, with typically advanced levels in each of these. Within each of these countries as well, inequality is low compared to non-WEIRD countries, as can be inferred from the comparatively low Gini coefficients of Western countries, with the United States being somewhat of an unequal outlier (Selita & Kovas, 2019). ...
We argue that heritability estimates cannot be used to make informed judgments about the populations from which they are drawn. Furthermore, predicting changes in heritability from population changes is likely impossible, and of limited value. We add that the attempt to separate human environments into cultural and non-cultural components does not advance our understanding of the environmental multiplier effect.
... WEIRD societies are also roughly aligned on basic environmental factors such as technology, civil infrastructure, healthcare, schooling, and transportation, with typically advanced levels in each of these. Within each of these countries as well, inequality is low compared to non-WEIRD countries, as can be inferred from the comparatively low Gini coefficients of Western countries, with the United States being somewhat of an unequal outlier (Selita & Kovas, 2019). ...
We need better understanding of functional differences of behavioral phenotypes across cultures because cultural evolution (e.g., temporal changes in innovation within populations) is less important than culturally molded phenotypes (e.g., differences across populations) for understanding gene effects. Furthermore, changes in one behavioral domain likely have complex downstream effects in other domains, requiring careful parsing of phenotypic variability and functions.
Major changes in the educational distribution of the population and in institutions over the past century have affected the societal barriers to educational attainment. These changes can possibly result in stronger genetic associations. Using genetically informed, population-representative Finnish surveys linked to administrative registers, we investigated the polygenic associations and intergenerational transmission of education for those born between 1925 and 1989. First, we found that a polygenic index (PGI) designed to capture genetic predisposition to education strongly increased the predictiveness of educational attainment in pre-1950s cohorts, particularly among women. When decomposing the total contribution of PGI across different educational transitions, the transition between the basic and academic secondary tracks was the most important. This transition accounted for 60–80% of the total PGI–education association among most cohorts. The transition between academic secondary and higher tertiary levels increased its contribution across cohorts. Second, for cohorts born between 1955 and 1984, we observed that one eighth of the association between parental and one's own education is explained by the PGI. There was also an increase in the intergenerational correlation of education among these cohorts, which was partly explained by an increasing association between family education of origin and the PGI.
Full-text available
This study attempted to determine if the association between genetic admixture and cognitive ability among African, European, and Amerindian descent groups in the USA holds across a large time period. First, we used the large and nationally representative Adolescent Brain Cognitive Development Study (ABCD) sample to examine the association between cognitive ability and parental-reported race, genetically-predicted color, and genetic ancestry among Puerto Ricans, non-Hispanic Whites, Blacks, and American Indians in the 21st century. Second, we use the 1850 to 1930 US censuses to trace ancestry-associated cognitive differences back to the early 19th and early 20th century among African American, American Indian, and Puerto Rican groups by using age-heaping-based numeracy as a proxy for cognitive ability. In the ABCD sample, we found that European ancestry is positively associated with cognitive ability within race/ethnic groups. In the census data, among African Americans and American Indians but not Puerto Ricans, we find that greater apparent European admixture is associated with higher numeracy, and this holds when we subset by age, sex, and literacy status. Therefore, we established that admixture is associated with cognitive ability in the 21st century as it was in the 19th and early 20th century. The implications of these findings are discussed.
The development of the human brain experiences significant growth in the first 2 years of life, laying the foundations for its structural and connective development through childhood, adolescence as well as adulthood. These foundations are dependent upon and shaped by the characteristics as well as the features of the family into which the child is born, but also the characteristics and features of the wider environment in which the family itself exists. In the case of the family, the child may have to compete with siblings for the limited material and psychological resources which the parents may have to offer. For example, the level of the parents’ income may only allow them access to food sources of varying quality and nutritional content. Moreover, the level of parental income will also dictate the frequency with which the children will eat—although the nutritional value of the food that is eaten may have a greater positive impact on brain development than the frequency with which meals are eaten. However, while the nutritional content of food gives the potential for the physiological structures of the brain to develop, it is the infants pre- and post-natal environment which will provide the stimuli to alter the infant’s genes through epigenetic mechanisms to various levels depending on the nature of the stimuli. These altered genes will then determine the quality of the brain structures as well as the quality of neuronal connections within the brain. The altered genes may produce diverse types of proteins, changing the genes themselves, which form the building blocks of all living tissue, including those required to form the structures of the brain. A poor childhood experience may inhibit and constrain gene expression in later life. The result of this may be that the proteins produced later in life may not be of optimal quality and may not be of the correct type. Proteins may also be produced which ‘switch on’ genes which emphasise character weaknesses rather than character strengths. Brain development is optimised when the number of neural connections increases exponentially. This only occurs when the quality of proteins produced is both optimal and high in number. This in turn will affect the level of development of cognitive capacity of the brain development of the infant. It is also at this stage of development that the brain is at its most malleable to change in the external environment; the resulting stimuli allows the infant to acquire new skills and abilities. Furthermore, in late infancy, at the toddler stage, the infant also develops emotional regulation mechanisms because of the amygdala forming connections with other parts of the brain. The development by the infant of mechanisms to emotionally regulate itself is also central to the development of its cognitive ability. However, key to cognitive advancement is the acquisition of language by the infant. As vocabulary expands, the infant’s cognitive capacity develops even further because it can physically construct its thoughts in words through vocal expression. Furthermore, as the infant grows through childhood and adolescence into an adult the parts of the brain may become increasingly specialised in function. It would therefore seem that the first 24 months of life of a human being contribute most towards its lifelong accomplishments. Therefore, the implication is that governments should foster economic and social policies which allow for the development of a conducive environment for an infant to optimally develop its cognitive capacity. Moreover, such economic and social policies should be promoted throughout society on a universal basis regardless of either parental or household income. Such policies may include the provision of either free or subsidised nutritional meals, milk and childcare by the state.
Full-text available
Molecular Psychiatry publishes work aimed at elucidating biological mechanisms underlying psychiatric disorders and their treatment
Full-text available
The Gini index is the most commonly used measure of income inequality. Like any single summary measure of a set of data, it cannot capture all aspects that are of interest to researchers. One of its widely reported flaws is that it is supposed to be overly sensitive to changes in the middle of the distribution. By studying the effect of small transfers between households or an additional increment in income going to one member of the population on the value of the index, this claim is re-examined. It turns out that the difference in the rank order of donor and recipient is usually the most important factor determining the change in the Gini index due to the transfer, which implies that transfers from an upper income household to a low income household receive more weight that transfers involving the middle. Transfers between two middle-income households do affect a higher fraction of the population than other transfers but those transfers do not receive an excessive weight relative to other transfers because the difference in the ranks of donor and recipient is smaller than the corresponding difference in other transfers. Thus, progressive transfers between two households in the middle of the distribution changes the Gini index less than a transfer of the same amount from an upper income household to a lower income household. Similarly, the effect on the Gini index when a household in either tail of the distribution receives an additional increment is larger than when a middle-income household receives it. Contrary to much of the literature, these results indicate that the Gini index is not overly sensitive to changes in the middle of the distribution. Indeed, it is more sensitive to changes in the lower and upper parts of the distribution than in the middle.
The proportion of human genetic variation due to differences between populations is modest, and individuals from different populations can be genetically more similar than individuals from the same population. Yet sufficient genetic data can permit accurate classification of individuals into populations. Both findings can be obtained from the same data set, using the same number of polymorphic loci. This article explains why. Our analysis focuses on the frequency, ω, with which a pair of random individuals from two different populations is genetically more similar than a pair of individuals randomly selected from any single population. We compare ω to the error rates of several classification methods, using data sets that vary in number of loci, average allele frequency, populations sampled, and polymorphism ascertainment strategy. We demonstrate that classification methods achieve higher discriminatory power than ω because of their use of aggregate properties of populations. The number of loci analyzed is the most critical variable: with 100 polymorphisms, accurate classification is possible, but ω remains sizable, even when using populations as distinct as sub-Saharan Africans and Europeans. Phenotypes controlled by a dozen or fewer loci can therefore be expected to show substantial overlap between human populations. This provides empirical justification for caution when using population labels in biomedical settings, with broad implications for personalized medicine, pharmacogenetics, and the meaning of race.
The proportion of human genetic variation due to differences between populations is modest, and individuals from different populations can be genetically more similar than individuals from the same population. Yet sufficient genetic data can permit accurate classification of individuals into populations. Both findings can be obtained from the same data set, using the same number of polymorphic loci. This article explains why. Our analysis focuses on the frequency, ω, with which a pair of random individuals from two different populations is genetically more similar than a pair of individuals randomly selected from any single population. We compare ω to the error rates of several classification methods, using data sets that vary in number of loci, average allele frequency, populations sampled, and polymorphism ascertainment strategy. We demonstrate that classification methods achieve higher discriminatory power than ω because of their use of aggregate properties of populations. The number of loci analyzed is the most critical variable: with 100 polymorphisms, accurate classification is possible, but ω remains sizable, even when using populations as distinct as sub-Saharan Africans and Europeans. Phenotypes controlled by a dozen or fewer loci can therefore be expected to show substantial overlap between human populations. This provides empirical justification for caution when using population labels in biomedical settings, with broad implications for personalized medicine, pharmacogenetics, and the meaning of race.
Authored by David A. Ansell, MD. Kristen Pallok: principle co-contributor for research and editing
Increases in family income inequality in the United States have translated into widening gaps in educational achievement and attainments between children from low- and high-income families. We describe the mechanisms that have produced this disturbing trend. We argue that the three dominant policy approaches states and the federal government have used to improve the education of the disadvantaged have had at best modest success in improving education for disadvantaged children. To conclude, we describe the building blocks for an American solution to the problem of growing inequality of educational outcomes.