Abstract
The use of genetic data has proliferated in social and behavioural genomics research, leading to proposals for genetically-informed social policy. In particular, in the domain of ‘educational genomics’, researchers using ‘polygenic scores’—summary statistics used to calculate social outcomes from genetic data—have generated scientific knowledge claims about the potential to predict educational outcomes from biological samples. These polygenic scores have become the basis for proposed experimental interventions in education policy and practice. In this article analyzing a large corpus of scientific texts, we develop an original conceptualization of five techniques associated with the use of polygenic scores in education. We identify the underlying scientific base for claims of the policy-relevance of educational genomics, and develop a conceptual vocabulary explaining the implications and significance of polygenic scores in education. We conceive of polygenic scores as
Keywords
Introduction
Scientists have analyzed genetic data directly to study the biological bases of social outcomes since around 2010, with a particular emphasis on educational outcomes (Branigan et al., 2013; Silventoinen et al., 2020; Wilding et al., 2024). This has been made possible through the increasing availability of large samples of biological data and analytical technologies associated with advances in the ‘postgenomic’ sciences following the sequencing of the human genome (Kaufmann, 2023; Reardon, 2017). Biomedical genomics methods and technologies have subsequently been adopted by investigators in fields including behaviour genetics and social science genomics, with education a key topic of emerging research, especially in Europe and the US (Malanchini et al., 2020; Mills and Tropf, 2020; Thomas et al., 2015).
Research on the genetic bases of educational outcomes has been described as a ‘genomic revolution for education research and policy’ (Morris et al., 2022: 1), with ‘educational genomics’ characterized as ‘a dynamic field’ (Meaburn, 2025: 49). A complex technoscientific knowledge infrastructure has been assembled for the enactment of educational genomics research, including genetic data banks and bioinformatics technologies (Williamson et al., 2024). Scientists have produced knowledge claims about the genetic heritability of educational attainment, academic achievement, cognitive ability, intelligence, school engagement, non-cognitive skills, and more (Kotouza, 2025). These educational genomics research findings and claims have gained wider public attention through popular science publications (Asbury and Plomin, 2014; Conley and Fletcher, 2017; Harden, 2021; Plomin, 2018), supported by articles in the press and social media (Asbury, 2023; Conley, 2025; Lewis-Krause, 2021). This ‘geneticization of education’ thus promises genetic explanations for matters usually considered the purview of the psychological and social sciences (Matthews, 2024).
Significantly, the findings from educational genomics studies have also been accompanied by concrete proposals for policy and practice (Asbury et al., 2022). The research has circulated widely and been cited in policy reports exploring the potential and risks of incorporating genomic data and methods into education policy and practice, most notably by the UK’s Government Office for Science (2022) and the Department for Education (Hooper et al., 2024). Indeed, educational genomics corresponds with an international educational governance landscape that has increasingly adopted data-centric, ‘scientific’ and ‘medicalized’ methods and knowledge as the basis for policy development and interventions in schools (Moss and Baird, 2025).
What unites these endeavours is the centrality of the ‘polygenic score’ as a scientific instrument that makes genetic data available for use in social policy sectors such as education (Sabatello, 2018). First developed in medical genomics to calculate an individual’s genetic risk of disease or illness, polygenic scores were later taken up by scientists associated with social and behaviour genomics to investigate genetic predispositions to behavioural traits and social outcomes (Harden and Koellinger, 2020). In educational genomics, a polygenic score is a measure of the influence of a vast number of interacting genetic differences on educational outcomes and other educationally relevant traits such as ‘intelligence’, ‘cognitive ability’ and more (Chen et al., 2024; Lin et al., 2025). Described more fully below, a polygenic score is used primarily in educational genomics as a statistical predictor of educational attainment and achievement (as well as other relevant traits), which can be calculated by gathering genetic material from individuals (through, for example, a saliva swab) and then transforming it into statistical data for analysis using advanced software (Burt, 2024).
In this article, our guiding question is how educational genomics researchers and advocates are mobilizing polygenic scores to influence educational policy and practice, and with what implications. This analysis is significant since the use of polygenic scores in education carries major and unresolved social and ethical implications (Parens and Meyer, 2023), including those related to a longer history of racist appropriations of genetic science (Martschenko et al., 2025). Techniques of polygenic scoring are even being enrolled into controversial embryo intelligence testing procedures by commercial biotechnology startup companies, thus resuscitating eugenic ideologies of genetic enhancement through the medium of digital consumer genetics services (Gusev, 2025).
The original contribution of the article, based on in-depth analysis of a large corpus of scientific texts published in the area of educational genomics, is to examine and theorize how polygenic scores function both as scientific instruments in experimental studies and as proposed instruments for policy purposes. We identify the underlying scientific base for claims of the policy-relevance of educational genomics, and develop a conceptual vocabulary explaining the implications and significance of polygenic scores in education. Polygenic scores, we demonstrate, have been positioned to translate the scientific knowledge claims of educational genomics into genetically-informed interventions in education. However, we have discovered that there is no single proposed application of polygenic scores in education, but five approaches that we define and conceptualize as
Significantly, we argue, the emergence of educational genomics marks a transformation in how the core societal purposes of education are framed and what kinds of policy choices are pursued through a renewed confidence in biological authority in policy agenda setting. This has the potential to change what problems governments deem to be worthy of policy action, especially as educational genomics findings and knowledge claims are propelled into public and political discourse and potentially distract from other forms of policy intervention (Meyer et al., 2023a). We next review recent research discussing the use of polygenic scores in education, outline our analytical approach to scientific and policy instruments, describe our methods, and then present our analysis of how polygenic scores have been positioned as policy instruments that could be used to inform educational interventions.
Polygenic scores
Central to proposals to mobilize genomic data and methods for educational purposes is the alleged promise of polygenic scores (sometimes known as ‘polygenic indices’) as measures of the genetic contribution to social outcomes and behavioural traits. Polygenic scores are summary statistics of the thousands of minute molecular genetic differences associated with a particular phenotype, such as a behavioural trait or an outcome like test score achievement (Chabris et al., 2015). For scientists working in molecular genomics fields, polygenic scores ‘represent something of a paradigm shift, moving the focus from estimating net genetic influences in populations to estimating an individual’s genomic likelihood, compared with others in the population, of exhibiting a specific phenotype’ (Meaburn, 2025: 15). It is therefore claimed that ‘[r]ecent advances in genomics make it possible to predict individual differences in education from polygenic scores that are person-specific aggregates of inherited DNA differences’ (Wilding et al., 2024: 1), and that polygenic scores could in the near future be used for predictive purposes and applications in education (Asbury and Wai, 2020).
Polygenic scores are derived from analyzing millions of molecular ‘biomarkers’ known as single nucleotide polymorphisms (SNPs) that correspond with a particular phenotypical trait or propensity, a task requiring enormous quantities of biological data, genomics methods and data-intensive computational infrastructure for data analysis (Mills and Tropf, 2020). The most common polygenic scores are for educational attainment, or years of school and higher education (Domingue et al., 2015). These scores for educational attainment have been calculated through a series of studies since 2013 with escalating samples, statistical power, and effect sizes, the most recent featuring a sample of genotyped data from 3 million individuals (Lee et al., 2018; Okbay et al., 2022). The findings from these studies have catalyzed further efforts to identify what are widely described as ‘genetic propensities’, ‘genetic endowments’ or the ‘genetic influences’ that underpin and predict educational outcomes, including exam grades, or traits such as cognitive ability and intelligence (Burt, 2023).
Despite claims by its practitioners that the research is non-genetically determinist and emphasizes gene-environment interactions (Ghirardi and Bernardi, 2025), it often tends to foreground gene-centric explanations (Bliss, 2018; Burt, 2024). As a result, promissory claims that ‘genomic social policy’ may be possible in the future are now being projected as attainable in the present (Panofsky, 2009, 2015), and enrolled into concrete policy proposals for ‘genomically informed education systems’ (Sabatello, 2018). Recent investments to support educational genomics studies mean that research findings and claims based on polygenic scoring methods are now presented as more ‘policy-relevant’, authoritative and objective ‘direct measures’ of the genetic substrates of educational outcomes and behaviours related to learning (Visscher, 2022).
Claims that findings from genetics studies could be used to inform education policy are not novel in themselves. They stretch back to early 20th-century eugenic attempts to link IQ to its genetic base for differential ability sorting (Lowe, 1998), through twin studies of the heritability of educational attainment (Heath et al., 1985), to more recent proposals for genetically-informed classroom practice (Grigorenko, 2007). What is new is how novel genomic technologies, such as instruments for calculating polygenic scores, are now routinely invoked as transforming the scientific methods, knowledge production, and policy relevance of social and behavioural genetics research (Harden and Koellinger, 2020; Malanchini et al., 2020). On this basis, polygenic scores are positioned by proponents of educational genomics research as providing policy-relevant scientific evidence that ‘DNA may be a powerful predictor of educational success’ (Hughes, 2024: n.p.).
Such claims to novelty, objectivity and salience for policy and practice are highly controversial. There remain significant contests over the policy implications of new educational genomics findings (Martschenko et al., 2019). Educational genomics has also become the subject of intense bioethical and scientific controversy (Martschenko et al., 2025; Nuffield Council on Bioethics, 2025; Parens and Meyer, 2023). Some critics point to their racist and eugenic tendencies (Bliss, 2018; Gillborn, 2016; Gillborn et al., 2022). Others argue the evidence associating polygenic biomarkers with educational outcomes remains too weak and confounded by non-genetic factors to support any form of translation into policy (Burt, 2023; Govindaraju and Goldstein, 2025; Gusev, 2024; Turkheimer, 2025). Given the historical dangers of eugenicist thinking and the revival of ‘race science’ (Martschenko et al., 2025; Panofsky et al., 2024), as well as the rapid commercialization of genetic intelligence testing, polygenic embryo selection and their implications for education (Caplan and Tabery, 2025), there is an urgent need to examine critically the current policy propositions made in educational genomics, and to work through their potential implications and effects. We next outline our analytical approach to the study of scientific and policy instruments.
Scientific instruments and policy instruments
Our conceptual orientation is to focus on the ‘instruments’ through which scientific knowledge is produced and that mediate how policy objectives are achieved. We bring together the study of scientific instruments in science and technology studies with the analysis of policy instruments in education policy analyses. Social scientific analyses of data-intensive biology and the genomic sciences treat scientific instruments themselves as partaking in the production of different conceptions of human biology (Chow-White and García-Sancho, 2012; Stevens, 2013). The design of computer hardware, databases and software all affect how genetic information is analyzed, interpreted, understood and assigned meaning (Kaufmann, 2023). Scientific instruments help to construct the phenomena they are supposed to measure (Pickersgill et al., 2013), because they ‘actively mediate how reality becomes present to—and is treated by—scientists’ (de Boer et al., 2021: 392). The bioinformatics instruments used in contemporary biological research, as well as ‘the quantities of data processed by computers, and the algorithms needed to deal with them, make a qualitatively different kind of knowledge’ with significant effects on how biological phenomena are understood (Stevens, 2016: 172). This produces knowledge claims that then circulate beyond their sites of production and interweave with social, political and economic projects, including being translated into policy proposals (Cruz, 2022).
Scientific instruments are, then, not transparent windows on to actual biological structures and processes, but produce novel configurations of biological reality through their design and functionality (Leonelli, 2016). Bioscientific instruments can also be used for generating the biological knowledge that may be used by authorities in their efforts to govern human lives (Rose, 2007). As products of scientific instruments for the processing of genetic data, polygenic scores produce a particular configuration of human subjects, ‘formatted’ as data, that influences subsequent actions and interventions: One thing that the genetic sciences do with tremendous success is to ‘format’ us as subjects of genetic data. In being formatted, rendered, or organized as genetic data, we come to be defined by that data. Others address us, deal with us, or even impose burdens and benefits on us in terms of the genetic data we have become. (Koopman, 2020: 10)
Choices and constraints on the ‘formatting’ of genetic data matter in terms of how human subjects are known and governed. How these instruments ‘format’ genetic data as policy-relevant results, and their consequences, are particularly germane to our analysis.
Extending this conception of scientific instruments and how they format data, we refer to policy instruments as the technologies and devices that make political programs operational and actionable. This understands policy instruments as material and technical relays of particular policy aspirations and projects—or as ‘technologies of government’ that are both technical and social in their composition and operations (Le Galès, 2016: 510). Already, data-intensive policy instruments such as performance indicators have made it possible to govern sectors such as education through statistical measures that inform associated political actions (Verger et al., 2019). Polygenic scores are now positioned by some as instruments to be used to achieve policy objectives (Asbury and Wai, 2020).
Building on the above approaches to both scientific and policy instruments, here we examine polygenic scores as hybrid instruments combining biological knowledge practices with social policy discourses. In the first instance, a polygenic score is the product of a scientific apparatus of genomic data collection and analysis, and itself constitutes a scientific instrument used in experimental scientific research (Williamson et al., 2024). However, when a polygenic score is fused to social policy aims, as we illuminate later, then it becomes an instrument for the delivery of policy objectives. We thus conceive of polygenic scores as
Methods
Focusing on the scientific instrumentation of polygenic scores, we argue they are not simply the results of objective and neutral scientific methods. Rather, we conceive of polygenic scores as instruments that connect biological knowledge practices to social policy aims and discourses. Deployed as
The article is underpinned by a research project investigating the emergence of new data-intensive biological sciences in education and the circulation of biological research findings and knowledge claims into sites of policy and practice. Previously we detailed the formation of an international research infrastructure of educational genomics, which has enabled scientists to conduct research projects and produce knowledge claims about the genetic substrates of educational outcomes (Williamson et al., 2024; Kotouza, 2025). In this paper we focus on the circulation of educational genomics knowledge through publications and public forums, and specifically on its translation into proposals for policy intervention.
The data collection phase of the project involved the compilation of an archive of documents, consisting of over 100 scientific articles, four full-length popular science books about genetics and education, plus media and social media articles on the topic. We compiled this archive of documents as a Zotero library through searching online academic resources databases and by manually tracing references and citations between publications. This searching was performed mostly manually by searching for publications focused on education in fields such as social science genomics and behaviour genetics, and then screening the publications for relevance to the topic of polygenic scores in education. This was done in an iterative, ongoing process, with several new publications that we analyze here appearing during the duration of the project and being accompanied by significant press coverage that we also collected.
These searches were additionally informed by a mapping of educational genomics research scientists, institutions, centres, networks and associations, which we produced through detailed web searches and compiled as an annotated social network graph with the data visualization tool neo4j. This mapping started with identifying key nodal figures with significant public profiles, then tracing their institutional locations, relations and connections with other individuals and groups (e.g. where they were connected through co-authorship, research grants, shared participation in events, or joint membership of consortia, research networks and associations). The map allowed us to conduct further publication searches using authors’ names, while the recovery of publications enabled us to continue populating the map in an iterative way.
Our analysis of the documentary archive of educational genomics involved attention to how scientific findings reported in such documents are translated into proposals for action—that is, how the results are framed in ways that make certain proposals for the use of polygenic scores in policy or practice seem desirable or attainable. This followed a documentary analysis approach adopted from science and technology studies that treats scientific documents as the products of specific sites of investigation that often frame and represent phenomena as objects of possible intervention, in ways that can shape subsequent actions and behaviours among the audiences addressed by those documents (Asdal and Reinertsen, 2022). From this perspective, scientific documents like those we analyze next can be understood as ‘knowledge tools’, since scientific ‘documents may by the sheer act of textualization make a phenomenon observable and analysable’ and such ‘knowledge tools also often enter into political decision-making’ as ‘tool[s] for governing’ (45).
We view the documents analyzed here likewise as knowledge tools that are infused with the aims of their originating producers to shape perceptions of the usefulness and effectiveness of polygenic scores as instruments to be used for education policy and practice. The corpus of texts has made polygenic scores visible as instruments for governing education, supported by studies which have made the genetic substrates of learning outcomes ‘observable and analysable’, through the five approaches we detail next. Concretely, the analysis involved coding and annotating the individual documents, producing research memos about each one, specifically identifying examples of proposals for the use of polygenic scores in policy and/or practice across the whole corpus, and then thematizing them into a series of clusters.
Through further attention to each of the themes emerging from this process, we finally identified, examined and conceptualized five techniques associated with the use of polygenic scores in educational genomics, and resultant proposals for their use in education, as they are presented in these documents. Each technique generates particular findings about the connections of genetics to outcomes and is used to justify proposals for the use of polygenic scores in genetically-informed education policy and practice. Informed by our conceptual orientation to the generativity of instruments, we sought to foreground and theorize how each of these five techniques is represented in the documents, highlighting how each is constituted by the synthesis of a specific scientific technique, a set of knowledge claims, and existing policy discourses, through which they function as bio-social policy instruments.
Importantly, each of the five variations on polygenic scores as bio-social policy instruments makes particular assumptions about the meaning of learning, the purpose of education and the status of biological knowledge. Each is already embedded in a particular education policy discourse, serving as a mechanism to reinforce and intensify these through the incorporation of biological knowledge claims that we surface in the analysis. As bio-social policy instruments that fuse policy discourses with bioscientific knowledge claims, polygenic scores are therefore being positioned as the basis for genetically-informed education governance, which could—according to advocates—be used to influence how human subjects are governed through new educational interventions.
Polygenic scores as policy instruments
We turn now to fully analyzing how polygenic scores have been positioned to inform educational policy and practice, through five techniques that we designate as
Rating
The first way that polygenic scores have been positioned as bio-social policy instruments is through the ideal of ‘precision education’, or individually-targeted learning interventions based on students’ predicted polygenic scores (Rimfield and Malanchini, 2018). In this approach, polygenic scoring can be understood as a technique for ‘rating’ students on a genetic scale of predicted achievement, as the basis for individually-targeted interventions. A recent systematic meta-analysis of educational genomics studies using ‘DNA-based predictions’, for instance, suggested that ‘[u]nderstanding the origin of people’s differences in educational outcomes is key to improving how we teach and learn’ (Wilding et al., 2024: 1).
Precision education based on polygenic scores is an idea promoted by behaviour geneticists in the UK, notably in the popular sciences books
Some scientists have shifted from the language of polygenic scores to polygenic indices, to signify that a polygenic result is not intended as a hierarchical designation (Becker et al., 2021). However, ‘the shift to index potentially obscures the fact these are “rankings” (i.e., positions on a scale) of genetic associations with socially valued outcomes, whether we call them scores or indices’ (Burt, 2024: 40). Polygenic scores are thus a technology of hierarchical ranking or ordinalization. ‘Ordinalization’ refers to the ways that things may be classified, rated and ranked relative to each other through ‘ordinal judgments’, with these relative valuations occurring through ‘socio-technical channels’ and ‘rank-ordering criterion’ (Fourcade, 2016: 182). Education exemplifies practices of ordinal ranking, as grades ‘aspire to a society of individuals, complete with the egalitarian promise of objectivity’, as ‘every individual could presumably obtain a place on such a scale’ (p. 183).
We conceptualize predictive precision education proposals as a form of
It is in this sense of ordinal rank-ordering by polygenic scoring that ideas about precision education can be understood to reinforce existing policy discourses which emphasize individual measurable achievement levels as proxies of cognitive ability or intelligence, resuscitating ‘hereditarian’ arguments about the biological bases of merit and achievement (Gillborn et al., 2022). Plomin (2018: 180) argues his precision education proposals would better serve students by ‘pinpointing children’ and providing precisely-targeted, personalized support for individual achievement. This mode of bioinformatic ordinalization thus ‘formats’ an ordinal-biological subject as the basis for interventions in policy and practice. What ‘precision education’ proposes is the genetic rating of students via bioinformatics instruments that produce rank-ordered polygenic scores. This technique of rating and ranking positions individuals on a scale of predicted ‘genetic propensities’ for outcomes, with ‘personalized’ interventions to be designed and targeted based on where one falls on the scale. The clear implication is the risk of inserting new forms of genetic discrimination into educational contexts, amplifying medicalized forms of testing and hierarchization of students in schools, while distracting from alternative forms of social and pedagogic intervention.
Sorting
Precision education has been criticized on the grounds that polygenic scores do not have sufficient statistical accuracy for individually-targeted precision education (Harden, 2021). In response, scientists focused on early years genetic screening for group-based differentiation rather than individual personalization (Morris et al., 2020). Such proposals for genetically-based screening and differentiation are here conceptualized as a form of
Recent UK reports on the potential of social and behavioural genomics indicate growing interest in the possibility of using genetic data as the basis for certain forms of policy intervention. The report ‘Genetics and early intervention: Exploring ethical and policy questions’, published by the Early Intervention Foundation (part of the UK government’s ‘What Works Network’), suggests that as genetic science is advancing rapidly, ‘it is increasingly possible to identify at birth children who have an elevated likelihood of outcomes such as struggling at school or being diagnosed with a learning, behaviour or mental health condition’ (Asbury et al., 2021). Polygenic screening through early years genetic testing is thus viewed by some as an attainable form of policy intervention deriving from educational genomics (Shero et al., 2021; Wilding et al., 2024).
Polygenic screening and differentiation emphasize the potential to group children into distinct clusters, identified by similarities in the distribution of SNPs associated with academic achievement or learning difficulty. For example, one UK study assessing the ‘increasing predictive power of polygenic scores for education’ and ‘as a potential tool for genetically informed policy’, concludes that ‘while polygenic scores can be informative for identifying group level differences, they currently have limited use for accurately predicting individual educational performance or for personalised education’ (Morris et al., 2020: 1). A similar study suggests schools could use polygenic scores ‘as an additional tool for universal screening systems’, as ‘a tool for educators in authentic school settings, . . . alongside the progress monitoring tools regularly being used in schools’ (Shero et al., 2021: 2). According to the study, ‘personalizing education at the group level’ requires ‘less precision’; polygenic scores could therefore be used to ‘differentiate based on group needs rather than individual needs’, and ‘in conjunction with other screeners’ reduce the need for continuous ‘waves of progress monitoring’ in schools (pp. 5–6).
Another study indicates how polygenic scores may produce such group differences. In a paper by UK behaviour geneticists titled ‘Exploring the genetic prediction of academic underachievement and overachievement’, authors argue that school achievement can be ‘genomically predicted’ using polygenic scores (Kawakami et al., 2024). Their central claim is that technologies to calculate polygenic scores can operate as ‘early warning systems’ to predict school achievement from infancy, enabling observed academic achievement, as assessed by school test results, to be compared with a polygenic score predicting ‘genomically expected achievement’ from DNA samples collected in childhood (p. 3). Its findings indicate it can identify ‘underachievers’ as having lower education-related ‘genetic propensity’, leading to potential interventions ‘targeting students underachieving genomically’ (p. 7).
Polygenic scores, then, could become additional screening and progress monitoring instruments used in schools for sorting students into groups labelled variously as at risk of low educational attainment or learning disability, or with genetic propensity for high educational attainment. Such uses would also code for the other educational outcomes, such as standardized measurable ‘intelligence’ achievements and interior cognitive processes for which attainment is a proxy (Gusev, 2024). This grouping-based emphasis is proposed to be in contrast with the individualized approach of precision education. It is nested in existing policy discourses of addressing inequality and improving inclusion, particularly for those with additional support needs, albeit with the added apparatus of polygenic scoring as a way to cluster students for support.
We conceive these groupings as
Biosociality can also refer to the ways in which ‘the avalanche of genetic information available about individuals and populations’ has enabled the identification of groups according to genetic markers or indicators that correspond with or predict a biological condition (Hacking, 2006: 89). One does not need to identify as part of a biosocial collective to be identified by others via various techniques as part of a social grouping defined by its distinctive biological characteristics (Rose, 2007). ‘Through the use of computers, individuals sharing certain traits or sets of traits can be grouped together’ through a ‘technocratic administration of differences’ (Rabinow, 2005: 187). As such, genetic signals derived from bioinformatic surveying of the genome have become the basis for algorithmically differentiating individuals into biologically-defined clusters, from which may derive social groupings identified by a biological classifier imposed through bioinformatics techniques and instruments.
The polygenic score has therefore been positioned as a potential policy instrument for formatting and classifying learners by their genetic profiles into distinctive groupings, or polygenic biosocialities. It amplifies policy discourses of inclusion and differentiation but adds genomic analysis as a way of administrating differences and administering interventions. As they are proposed for screening, polygenic scores thus constitute biosocial classification devices for sorting and differentiating individuals into groups judged in terms of their probable genetic propensity for higher or lower educational attainment. The capacity to perform such biosocial sorting through polygenic scoring instruments might pave the way for new genomic screening technologies in education, with the significant consequence of producing new stratifications of genetically-endowed ability and animating subsequent interventions based on polygenic predictions.
Evaluating
The third proposal for using polygenic scores as bio-social policy instruments is to support cost-savings and efficiencies by introducing polygenic scores into policy effectiveness evaluations. Compared to precision education, which seeks ‘systematic’ applications of genetic data, and more ‘targeted’ forms of screening and differentiation, this scientific technique refers to more ‘informational’ applications of genetic knowledge in the education sector (Matthews, 2024). It mobilizes polygenic scores ‘as soft indicators to guide educational support and intervention, but not as deterministic predictors of individual educational outcomes’ (Meaburn, 2025: 53). As ‘soft indicators’, genetic data may be used to indicate whether a policy is cost-effective and to indicate for whom, how and why it works (Asbury and Wai, 2020). In this sense, the potential mobilization of genetic data as soft indicators reinforces existing forms of statistical policymaking and ‘governance by numbers’ that emphasize efficiency, effectiveness and accountability, with various ‘indicators’ used to evaluate policy effects (Grek and Ydesen, 2021).
In the book
Such approaches are based partly on the political argument that ‘genome-blind’ research ‘in the social sciences, designed to identify specific environments that could be targeted with new interventions,
This proposed use of polygenic scores as policy indicators also reinforces the trend of the use of numbers and statistics in education policy and governance, extending this now into the extraction of policy indicators from inside the body. Internationally, education systems are routinely ‘governed by numbers’ through instruments that emphasize performance measurement, efficiency, effectiveness, cost-saving and accountability, with various statistical indicators used to assess whether a policy approach is having the desired effects (Verger et al., 2019).
The positioning of polygenic scores in terms of cost-saving and efficiency reinforces ‘technocratic policy’ models of ‘tweaks and nudges’ that favour low-cost targeted interventions rather than social reforms and political investments in education informed by social science expertise or redistributive aims (Panofsky, 2021: 1449). Matthews (2024: 8), for example, terms Harden’s proposal an ‘informational’ approach to genetically-informed policy, which privileges the ‘practice of altering educational policies and practices in light of findings from the field of genetics’, using ‘genetic findings to influence educational practice and policy’ without necessarily targeting individuals or collectives. In this informational approach, genetic data such as polygenic scores become enrolled into existing statistical infrastructures of policy evaluation and the monitoring and tracking of system performance (Lahtinen et al., 2024).
Such proposals, then, can be conceived as the use of statistically-generated
The polygenic score is interpreted here as a genetic bioindicator instrument. Its use will reinforce the assumption that policy interventions must be steered by the expert use of statistical methods and instruments and thus by quantitative scientific expertise (Bandola-Gill et al., 2022). An informational approach to genetics in education actively dismisses alternative approaches to intervention by characterizing them as ‘genome blind’ and, by implication, unscientific, lacking systematicity, potentially ineffective, and wasteful of public resources. Through such proposals, polygenic scoring technologies are positioned as instruments that can be deployed alongside the battery of existing instruments used to govern education systems through regimes of numbers and indicators. In this sense, a significant potential implication of polygenic scores as indicators is the extension of bio-medicalized diagnostics and interventions into data-driven policy and governance.
Stratifying
The fourth area of policy relevance of polygenic scores is their use for new forms of genetic intelligence or talent testing. Here we foreground consumer genetic IQ tests, specifically how they have been framed as making it possible for parents to pay for genetic intelligence testing of their children, and in extreme cases for pre-implantation embryo selection or even polygenic editing (Meyer et al., 2023b; Visscher et al., 2025). The policy proposal associated with consumer polygenic intelligences tests is that schools should take account of the polygenic scores that parents have purchased to maximize their children’s educational chances, raising significant ethical implications for pedagogic practices in schools (Nuffield Council on Bioethics, 2025). Plomin (2018), for instance, suggests parents may take up the opportunities of purchasing polygenic IQ scores for their children, as a way of informing their choices about education and schooling based on ‘getting a glimpse of their children’s individuality—their strengths and weaknesses, their personalities and their interests’ (p. 178). Another article, co-authored by Plomin, argues that direct-to-consumer DNA testing ‘companies are increasingly marketing their products to encourage parents to test their children . . . for educationally relevant traits’ (Kawakami et al., 2024: 3).
Indeed, a further paper anticipates DNA testing for IQ taking place either at birth or during preimplantation embryo selection, and notes that ‘one not-for-profit company, Impute.me, has begun to compute hundreds of state-of-the-art GPS, including intelligence, for its users’ (Plomin and von Stumm, 2021: 3). The invocation of Impute.me is significant since this not-for-profit enterprise has since been acquired by a commercial company, Nucleus Genomics, that launched a consumer genetic IQ test to the market in 2025 and names Plomin as an advisor (Williamson, 2025). The launch of such a polygenic IQ test thus marks the kind of development that some advocates of educational genomics anticipate as central to the wider use of polygenic intelligence testing in education.
While polygenic IQ testing remains emergent in contexts such as Europe and the US, genetic tests for educationally-relevant phenotypes are more extensively used in the Chinese context. In a study of ‘genetic talent tests’ for children in China, Au (2022: 197) argues that such tests have been designed not only to measure IQ but to capture a wider range of ‘qualities’ in a hyper-competitive context: Genetic talent tests try to inculcate a cosmopolitan disposition for their children that values talent in arts, music, and athletics. By positioning children to only compete in areas where they are reportedly talented, parents identify niches where they believe their children will be competitive. . . . Genetic talent tests claim to enable parents to identify this niche where their investments will yield the highest returns.
Despite the limitations of such genetic talent tests, parents invest in them for their children in combination with intensive regimes of schooling and additional private tuition, using the test results to identify competitive potential and then enrolling their children in educational programs designed to maximize their promise. In this sense, genetic talent testing taps into the model of ‘precision education’ advocated by Plomin and others, albeit in the sphere of the family and under a logic of consumerism and choice, and reinforces the idea that genetic data can be used to construct ordinal hierarchies of abilities and potentials. It also resonates with discourses of ‘school choice’ and the positioning of parents as consumers of quality educational services for their children.
This new form of
The challenge of this biodeterministic stratification, in terms of education policy and practice, is how educational institutions and practitioners should respond when parents who have purchased polygenic intelligence scores request special treatment for their children on the basis of supposed genetic entitlement, exerting their power as consumers to choose what’s best in their children’s future interests. Polygenic IQ tests format children in terms of genetic data signifying their future cognitive abilities, but they remain highly partial and contingent, rooted in the constraints of the tests and the reductivism of the scores. The significant implication of such tests, nonetheless, is the possibility of schools being pressured to make institutional policy decisions concerning individual students despite the absence of credible evidence of the validity of genetic IQ assessments as diagnostic instruments.
Valuing
The final promise of polygenic scores for policy is their potential to be used as the basis for economic prediction of ‘human capital’ (Kotouza, 2025). Here we highlight the techniques of ‘valuing’, emphasizing the role of the emerging interdisciplinary field of ‘genoeconomics’ in making statistical predictions of future economic value from polygenic scores in educational genomics research. This is couched in terms of human capital development of genetic endowments such as educational attainment and cognitive ability. The relationship between genetic differences, education outcomes, and human capital is made explicitly in one systematic review of the contribution of behaviour genetics to education: Educational attainment is a measure of human capital and is indicative of the skills of a population. As countries’ economies gradually shift away from mass production towards becoming knowledge economies, governments are eager to increase the skills and welfare of the population through educational attainment. . . . Higher levels of educational attainment are associated with higher employment rates, better job prospects and higher earnings. (Malanchini et al., 2020: 229)
As Foucault (2008: 227) anticipated, ‘the political problem of the use of genetics arises in terms of the formation, growth, accumulation, and improvement of human capital’. Here, then we can see how polygenic scores can be formatted through the particular disciplinary lens of economics and its concepts as human capital indicators.
Many educational genomics studies are led by economists, or self-described ‘genoeconomists’, who have theorized that genetic data about educational outcomes can be used to understand the genetic basis of other downstream socioeconomic outcomes (Beauchamp et al., 2011; Benjamin et al., 2012). A typical finding from a genoeconomic perspective is that DNA differences affecting brain development lead to different paths of psychological development in the early years, which are then associated with educational achievement and attainment, in turn affecting later occupational, income and socioeconomic status and other ‘success’ measures related to ‘human capital’ (Belsky et al., 2016). The language of economics is prevalent in such studies, with one paper on the genetic bases of educational attainment published in the
As the opening sentence of one typical paper reads, ‘Underachievement in school is costly to society and to the children who fail to maximize their potential’ (Kawakami et al., 2024: 1). It includes, tellingly, a citation to support this point to a paper about the ‘economics of investing in disadvantaged children’ by economist James Heckman of the University of Chicago Center for the Economics of Human Development. Heckman is well known for his work calculating the economic payoffs of investment in early years child development, which is central to the model of ‘human capital development’ he promotes to policymakers. Griffen (2024) has described such programs of human capital development as the ‘economization of early life’, noting especially the influence of Heckman on directing policy attention to early childhood investments in maximizing future human capital gains.
Human capital theory has long focused on both social investments and biological endowments, as ‘economists calculate the potential value of human capital not by estimating market demand but rather by projecting the human capital gains to be made by investment in individuals on the supply side that will inflate the future value of biological selves’ (Griffen, 2024: 179–180). Genoeconomics approaches are thus targeted at the supply-side of human capital formation, seeking to understand how genetic endowments can be maximized through social investments and policy interventions. They are concerned with the genoeconometric measurement of human biocapital and interventions to improve its measurable value in terms of future social and economic outputs and outcomes.
Genoeconomically-oriented educational genomics researchers therefore utilize polygenic scores as predictive econometric instruments to survey for biological signals of future human capital in the genome (Papageorge and Thom, 2020). This can be conceived as
Educational genomics research and advocacy therefore suggests the emergence of genoeconometric education policy, buttressing and fortifying existing econometric tendencies in international education policy with scientific knowledge claims about the genetic substrates of outcomes and their potential economic value. Genoeconomic use of polygenic scores formats human subjects as human biocapital with predicted economic value. Genomic data and genoeconometric measures thus are presented and circulated in ways that are intended to promote genoeconomic solutions—such as early years investments—to address the relationship between children’s outcomes, human capital development and economic prospects. The long-term implications for students are also significant, since commercial organizations involved in the insurance and finance industries could conceivably be interested in accessing an individual’s polygenic score for socioeconomic outcomes when making consequential decisions about providing (or withholding) financial services (Meyer et al., 2024).
Conclusion
Educational genomics is emerging as a source of scientific development with significant implications for social policy and educational practice that are already raising bioethical controversies (Williamson et al., 2024; Nuffield Council on Bioethics, 2025). Mobilizing a methodological approach that treats scientific documents as ‘knowledge tools’ that make scientific phenomena visible as potential objects of intervention and governance (Asdal and Reinertsen, 2022), we have shown how educational genomics research has promoted the polygenic score as an instrument for use in educational policy and practice as well as a scientific instrument of knowledge production. Through five techniques examined in this article, we have documented major efforts across Europe, the US and beyond to bring genomics closer to the persistent problem of social inequalities and learning performance in education. These rely on scientific data and technical practices and promissory knowledge claims which bring forth new proposals for experimental intervention and policy.
It is on the basis of polygenic scores used in educational genomics studies that new pathways have been proposed from saliva samples to social policy. We have thus theorized polygenic scores as constituting five types of
If mobilized in the ways these proposals suggest, these bio-social policy instruments would lead to genomically-informed governance interventions into educational settings and pedagogical practices. With each technique, different formatting arrangements lead to different ways of viewing and doing things to human subjects: rating, sorting, evaluating, stratifying and valuing. These are shaped not only by the available scientific instruments themselves but by their fusion with varied social policy discourses and promissory scientific knowledge claims. The genomic knowledge claims are not neutral but reflect a particular set of evidentiary norms, political rationalities and epistemological commitments that in many cases remain contiguous with the genetic determinism and race science from which the majority of contemporary genomics scholars are at pains to distance themselves.
Genomically-informed interventions in education would take different forms and reflect or reinforce varied existing policy approaches. Some, for instance, build on path-dependent educational discourses which already fit within particular policy commitments. The precision education approach would treat polygenic scores as policy instruments to format, rank and ordinalize students in preparation for individually-targeted ‘personalized learning’ programs. Genomic screening initiatives would arrange or format children into polygenic biosocialities for additional support in terms of assessments of their learning difficulties. Both of these already assume that the purpose of education is to rate and sort individuals, and they can serve to further sediment this assumption.
A more informational approach using polygenic scores as additional measures in policy evaluations would reinforce existing techniques of ‘governing by numbers’ and statistical assessments of ‘what works’ in terms of efficient, cost-effective intervention by formatting genetic data as political indicators. Polygenic IQ testing, by contrast, would involve parents purchasing tests for their children as the basis for making genetically-informed schooling choices and demands. The formatting arrangements of the tests and scores may harden biologically deterministic assumptions about the biological bases of intelligence and other socially-valued qualities. It would also reinforce existing policy discourses of ‘school choice’ that privilege wealthy families with the means to pick and choose educational opportunities with the best future prospects for their children, only now on the basis of purportedly objective biological evidence of their genetic propensities.
Finally, what we have termed genoeconometric biocapitalization would involve genetic data being used to make assessments of children’s human capital potential, formatted in terms of human biocapital calculations, as the basis for genetically-informed educational investments to maximize ‘success’ as conceived through economic reasoning. It would bring biology into the enduring policy preoccupation with economizing education (Kotouza, 2025). By applying logics of evaluation, stratification and econometric capitalization using genetic data at a population level, these three techniques turn persistent context-specific educational divides into biological problems and deficits to be solved through advances in genetic science rather than by socially-oriented policy interventions.
Together, these developments suggest the emergence of a plurality of social policy proposals for genetically-based interventions in education, each of which repositions polygenic scores as bio-social policy instruments. Such instruments fuse scientific findings and associated knowledge claims with political discourses to format genetic data differently, relevant to varied aspirations, ‘each of which carries different consequences for who we, and others, can be’ (Koopman, 2020: 12). They also position genetic differences—made legible at the molecular scale via polygenic analyses—as the legitimate focus of policy attention, and render educationally relevant behaviours and outcomes as phenotypes that can be modified with appropriate interventions. Despite ongoing efforts to translate genetic knowledge into policy and practice, however, educational genomics remains highly contested on both scientific grounds (Govindaraju and Goldstein, 2025) and in terms of its bioethical and sociopolitical implications (Martschenko and Sabatello, 2025). The significant danger here is ‘genetic distractionism’: the risk that attention and resources devoted to genomic research will distract policy attention from other interventions to address social problems, foregrounding biological data over other kinds of expertise (Meyer et al., 2023a: 21).
While there is not one coherent political rationality nor ideology underpinning what is a diverse scientific field, each of the applications of polygenic scores we have described in this paper actively transforms the object and subjects of its potential policy interventions through casting educational and social problems through a ‘genetic lens’ (Asbury and Wai, 2020). Educational genomics is also poised to benefit from existing tendencies in the international educational governance landscape for policy officials to turn to scientific and, increasingly, medicalized explanations for inequalities in educational achievements (Moss and Baird, 2025). This repositions students as biologically legible subjects, rendered in the statistical format of the polygenic score, and as prospective recipients of genetically-informed treatments and interventions.
Such a repositioning of students as genetic data subjects may also affect pedagogic relationships in schools, as teachers are encouraged to consult genetic bioinformation about students to inform their pedagogic practices (Meaburn, 2025; Nuffield Council on Bioethics, 2025). For those who see many national education systems currently failing to address ongoing inequities, injustices and needs, this may have the effect of drowning out other, pluralistic and potentially emancipatory and equity remits for education, while also distracting from non-genetic explanations in other areas of educational research.
The article has traced some of the ongoing developments of educational genomics, focusing specifically on polygenic scoring instruments and the formatting operations they enact on genetic data. Not only has the scientific infrastructure for investigation and knowledge production in educational genomics been assembled and activated over recent years, but now the scientific instrument of the polygenic score has begun undergoing translation into a bio-social policy instrument which may be deployed as an experimental technology of governance. Though the translation of educational genomics into policy or practice remains at this point primarily at the stage of proposals and propositions, and remains highly contested, it is now clear that genomic knowledge is being fused to policy agendas and raising the prospects of data-intensive biological sciences becoming influential in the future governance of education.
Our contribution in this analysis was to establish the conceptual coordinates and empirical foundations necessary for further studies of the concrete development of polygenic scores as bio-social policy instruments in education. While we have provided a detailed examination of the variety of proposals to mobilize polygenic scores in education, scholarship has yet to track the uptake of educational genomics by policy authorities or the real-world deployment of polygenic scores in specific educational contexts and settings. Ongoing research is now urgently required to continue tracing the translation of the genetic sciences into social policy domains such as education, and to critically assess their situated consequences and unfolding effects.
Footnotes
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by a research project grant awarded by The Leverhulme Trust (RPG-2020-395).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.