This chapter uses OECD cross-country evidence to examine how differences in educational attainment, field of study and participation in adult education impact skills development over the life course across demographic groups. Gaps in core 21st-century skills, initially shaped by family background and educational settings, frequently become larger after compulsory education ends. Gender and socio-economic background determine not only for how long individuals study but also what they study and whether they have the chance to continue learning over the life course. Barriers to participation in non-formal adult learning differ depending on adults’ prior education and socio-economic background. While increased educational spending can reduce skills disparities, it alone is insufficient to compensate for initial disadvantage related to parental circumstances. Effective lifelong learning policies are critical for reducing inequalities. The analysis in this chapter underscores the importance of educational quality, accessibility and tailored interventions.
OECD Skills Outlook 2025
Abstract
Disparities in access to learning opportunities and educational trajectories explain a large share of existing gaps in core 21st-century skills between adults with a different socio-economic background, childhood residential context and immigrant background. However, they do not explain gender skills gaps. Socio-economic gaps in skills narrow during compulsory schooling but widen markedly once students leave school, as more advantaged individuals accumulate higher-quality qualifications and access more adult learning.
Gender differences in learning and skills development follow a distinct trajectory: numeracy gaps in favour of men increase from childhood into early adulthood, whereas women’s advantage in literacy peaks in adolescence. Women are more likely than men to attain tertiary qualifications, yet remain under-represented in mathematics-intensive, science, technology, engineering and mathematics (STEM) courses in both vocational and academically oriented programmes. Key findings include:
Formal education and training:
In most countries, schooling provides equalising opportunities for individuals from different socio-economic backgrounds. For example, the difference in literacy proficiency between individuals born around 1990 with and without tertiary-educated parents was 0.57 standard deviations (SD) at age 10, followed by 0.34 SD in adolescence, 0.47 SD in young adulthood and 0.45 SD in adulthood.
Disparities in numeracy in favour of men grow larger over the life course, whereas literacy gender gaps in favour of women are widest during adolescence and narrower in primary school and adulthood. For example, the gender gap in numeracy between men and women born in 1984 was 0.06 SD at age 10, followed by 0.08 SD in adolescence, 0.22 SD in young adulthood and 0.24 SD in mature adulthood.
Individuals with advantaged socio-economic backgrounds are more likely to obtain advanced educational qualifications. Among adults whose parents worked in high-status occupations, 53% attain a bachelor’s degree or higher qualification compared to only 23% of those whose parents worked in low-status occupations. Some 40% of adults who grow up in cities earn a bachelor’s degree or higher qualification compared to only 26% of those who grow up in villages.
Whereas disparities related to socio-economic background lie primarily in how long people study for, disparities related to gender differences also lie in what men and women study. Women are more likely to obtain tertiary educational qualifications but are considerably less likely than men to attend courses with a strong mathematics orientation.
Field-of-study choices during formal education explain around a quarter of gender differences in numeracy skills between men and women, and around half of gender differences in adaptive problem solving. However, gender differences in numeracy in favour of men remain pronounced (0.18 SD) even when comparing men and women with similar levels of educational attainment and fields of study.
Disparities in educational attainment are a key factor in determining socio-economic disparities in skills. For example, around half of disparities in core 21st-century skills between individuals with and without tertiary-educated parents can be explained by differences in educational qualifications between the two groups.
Non-formal education and training:
The very adults who most require upskilling and reskilling to improve their labour market prospects generally participate the least in adult learning. For example, around 61% of tertiary-educated individuals participate in adult learning opportunities compared to only 19% of those without an upper secondary qualification.
Early disparities in skills and educational opportunities shape the likelihood of individuals being able to continue engaging in skills development as adults. Socio-economic disparities in participation in adult learning are not inherent to socio-economic background, but largely arise because of differences in the educational and occupational trajectories of individuals with advantaged and disadvantaged backgrounds.
Disparities pertain not just to the rates of participation of different groups but also to the types of training they engage in. Rather than promote mobility, training may reproduce the intergenerational transmission of disadvantage. Adults from disadvantaged backgrounds are over-represented among those who engage in training that is focused on learning how to operate machinery and equipment or following and maintaining security protocols. By contrast, individuals from advantaged backgrounds are over-represented among those who participate in learning aimed at developing project management or organisational skills, acquiring foreign languages, and developing skills involving numbers and calculations.
Barriers to participation differ across groups. Family obligations are more prominent among younger workers and women. Scarcity of suitable courses, last minute impediments or inconvenient scheduling are especially cited by adults living in rural areas. Women are also especially likely to cite unexpected events as a reason for lack of participation, an indication of the lack of support women have and the fine balance of their time budgets.
Informal learning – volunteering:
Participation in volunteering activities is associated with higher core 21st-century skills. Those who participate in volunteering have 0.11 SD higher literacy and numeracy and 0.10 SD higher proficiency in adaptive problem solving.
On average, 32% of adults reported engaging in some form of volunteering in the previous year, but participation rates vary greatly across countries. Disparities by socio-economic background are large: 38% of adults with tertiary-educated parents engage in volunteering compared to 30% of adults without tertiary-educated parents.
Skills are developed, enhanced and accumulated throughout an individual’s life via the formal educational and training opportunities they participate in and the informal learning that occurs outside these settings. At any given time, the range of skills adults possess – spanning information-processing skills (literacy, numeracy and adaptive problem solving), social and emotional skills (agreeableness, openness to experience, conscientiousness, extraversion and emotional stability), and willingness to delay gratification – is shaped by their access to learning opportunities, the value they assign to life choices and their willingness to engage in skills development opportunities. This developmental trajectory begins in early childhood, progresses through formal schooling and continues into adulthood via formal, non-formal and informal learning contexts including work, leisure and home environments (Cunha and Heckman, 2007[1]).
Formal education has a key role to play in reducing initial disparities arising from circumstances over which individuals have little to no control (such as the educational attainment or the occupational status of their parents, if they are men or women, if they were raised in a city or village, or if their parents decided to leave their country and migrate). Compulsory schooling legislation aims to guarantee a level playing field in skills and educational participation, and support growth. At the same time, there is evidence that the quality of educational opportunities in childhood differs greatly depending on the background of individuals. For example, children with socio-economically advantaged and disadvantaged backgrounds access schools of varying quality, are not equally likely to participate in early learning settings, and, even when they have similar academic results early on, have different rates of participation in further education and training.
Schools offer an environment that is conducive to the development of the information-processing skills critical to thrive in the 21st century (OECD, 2024[2]); however, education systems differ in their capacity to ensure that education standards are of high quality and available to all students, irrespective of their family circumstances. Institutional factors such as tracking and streaming policies, teacher allocation, and curriculum standards are associated with differences in the skills development trajectory of individuals from different backgrounds (Brunello and Checchi, 2007[3]) and gender (van Hek, Buchmann and Kraaykamp, 2019[4]), resulting in disparities in adult skills proficiency.
Students with socio-economically disadvantaged backgrounds, from rural areas or with immigrant backgrounds are frequently channelled into educational paths of comparatively lower quality, limiting their opportunities for skills enhancement (Green and Pensiero, 2016[5]; Silva et al., 2020[6]). Furthermore, parental educational attainment and occupational status profoundly impact children’s skill acquisition trajectories by shaping home learning environments, setting expectations and influencing aspirations (Cunha and Heckman, 2007[1]; Ermisch and Francesconi, 2001[7]). Adult learning opportunities, particularly continuous professional development and informal learning, further amplify existing disparities due to variations in individuals’ employment contexts, residential areas and resources available for ongoing education (Desjardins, 2003[8]). Adults with higher levels of education are often more intrinsically motivated to continue learning as part of their daily lives and to attract support for this from their employers. As a result, disparities in initial educational opportunities can shape lifelong learning trajectories and, potentially, contribute to widening skills disparities over the life course.
Effective lifelong learning systems can promote economic competitiveness by equipping individuals with the foundational knowledge and skills needed to navigate an increasingly dynamic labour market (see Chapter 1). They can also ensure that individuals possess the skills needed to participate effectively in democratic processes, thereby fostering civic engagement and social cohesion. There is a long-standing debate about whether schools function primarily as mechanisms of social mobility or as institutions that reinforce existing disparities, and whether the socio-economic gaps in skills observable in the early years are narrowed or widened through education and training (Cheadle, 2008[9]; Skopek and Passaretta, 2020[10]; Passaretta, Skopek and van Huizen, 2022[11]), with students often sorted into different education programmes along pre-existing class, gender and cultural lines.
This chapter first maps how disparities in core 21st-century skills have evolved over recent decades from childhood into adulthood. It then illustrates disparities in participation in education and training, and the extent to which disparities in learning opportunities exacerbate or alleviate skills disparities between socio-demographic groups. Finally, the chapter considers engagement in adult learning and informal learning, and explores the implications of participation patterns for disparities in core 21st-century skills, offering policy insights on how to broaden access to lifelong learning and mitigate persistent skills disparities. Taken together, the analysis pinpoints whether the key driver of inequality is how long people study for, what they study or factors that lie beyond formal education altogether – which are essential distinctions for designing effective, targeted policy responses.
Public investment in schooling is a key lever for raising average achievement and narrowing the learning gaps between young people with socio-economically advantaged and disadvantaged backgrounds (Jackson, Wigger and Xiong, 2021[12]). Higher per‑pupil spending is associated with smaller class sizes, better‑qualified teachers and richer instructional materials in school, which are factors associated with stronger outcomes and smaller skills gaps. Expenditure can also be directed towards compensatory programmes, such as early childhood education, instructional support, tutoring and needs‑based scholarships that aim to reduce skills disparities and that benefit students from low‑income families.
However, greater resources do not automatically translate into smaller disparities between individuals with different socio-economic backgrounds. More advantaged families generally have more resources to navigate school choice options, lobby for their children to take part in higher quality programmes or supplement public provision with private tutoring. As a result, additional funds may widen, rather than narrow, socio‑economic gaps. Empirical studies of “resource–advantage” interactions show that rises in spending deliver larger skills gains for young people in better-off neighbourhoods unless the allocation formula is explicitly progressive (Lafortune, Rothstein and Schanzenbach, 2018[13]). Against this backdrop, the analysis in this chapter examines whether cross‑national differences in cumulative public expenditure per pupil are reflected in disparities in young adults’ core 21st-century skills.
Socio-economic disparities arise from a combination of individual and social factors. At the individual level, unequal access to economic, cultural and social resources during childhood shapes educational pathways and skills formation. At the national level, broadening access to high‑quality schooling generally entails higher costs, and countries differ markedly in how much they spend. Figure 3.1 illustrates the association between cumulative public expenditure per pupil (from primary to upper secondary level) and gaps in core 21st-century skills among 22‑26 year‑olds with and without tertiary‑educated parents. In countries that invest more, socio‑economic gaps in literacy and numeracy tend to be narrower. The association, however, is weak: similar spending levels are consistent with very different gaps. Chile and Latvia each spend roughly USD 35 000 to 45 000 per pupil, yet Chile’s literacy gap is about 0.5 standard deviations (SD) while Latvia’s approaches 0.8 SD. Denmark, England (United Kingdom) and the Netherlands spend USD 120 000 to 130 000 per pupil, but their literacy gaps range from 0.2 SD in the Netherlands to 0.5 SD in Denmark.
This cross‑section evidence suggests that, on average, higher cumulative investment is accompanied by smaller socio‑economic disparities in core 21st-century skills. However, the large dispersion around the trend line indicates that funding alone is not a sufficient condition to guarantee lower skills disparities. In fact, higher expenditure per student is generally only positively correlated with overall levels of information-processing skills up to a certain threshold: additional investments beyond the cumulative USD 100 000 figure per student do not appear to be associated with higher skills at the country level (OECD, 2024[14]). Structural features such as how resources are targeted, how schools are governed and how families can supplement public provision determine the degree to which additional spending equalises (or not) opportunities (OECD, 2023[15]). Given that how money is spent is at least as important as how much is spent, this chapter details policies and initiatives adopted in OECD countries to reduce disparities in 21st-century skills.
Direct expenditure within educational institutions in equivalent USD converted using purchasing power parity (PPPs) for gross domestic product (GDP)
Note: Adults aged 22-26, restricted to adults whose highest completed level of formal education is upper secondary education. Cumulative expenditure on educational institutions per student (primary to upper secondary) refers to the years 2005-2016. Expenditure data are only available for a limited number of countries that participated in the 2023 Survey of Adult Skills. Data for expenditure for England (United Kingdom) refer to the entire United Kingdom. The dashed line represents the fitted regression line. Parental education (at respondents’ age 14) is based on the International Standard Classification of Education (ISCED) 2011 and distinguishes between adults with at least one tertiary-educated parent (ISCED 2011 5, 6, 7 and 8) and those with no tertiary-educated parent.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html and OECD (2025[17]), Expenditure on educational institutions per full-time equivalent student, (database), Data Explorer, http://data-explorer.oecd.org/s/1cs.
Box 3.1 provides insights into how policymakers can manage budget constraints related to the particular policy challenge of reducing disparities among students with an immigrant background.
Across the OECD, the integration of students with an immigrant background into school systems stands as an increasingly pressing policy challenge. Young people with an immigrant background have lower results in the Programme for International Student Assessment (PISA) than their counterparts without an immigrant background. Similarly, students with an immigrant background report a weaker sense of belonging, with girls who were not born in the country in which they reside as teenagers especially likely to report a low sense of belonging at school.
Language acquisition and inclusive teaching and learning practices can promote positive educational outcomes among students with an immigrant background. Across the OECD, governments have invested in AI infrastructures, programmes and tools that can assist teachers and other stakeholders in their efforts to integrate students with an immigrant background, and their families, into the school community and society more broadly. Table 3.1 provides an overview of AI-powered tools rolled out on a systemic basis in OECD and partner countries to support language teaching in schools.
Government funded – or otherwise supported – AI-powered tools rolled out on a systemic basis in OECD and partner countries
Note: See (Borgonovi et al., 2025[28]) for additional details about the initiatives listed.
Several OECD countries are also funding or otherwise supporting infrastructure for the development of AI tools for public use across low-resource languages (i.e. languages for which a comparatively low number of written texts are available to train AI systems), with the aim of spurring the development of AI-powered educational technology (ed-tech) for language learning. For example, the Lithuanian Ministry of the Economy and Innovation recently announced an investment of EUR 12 million across six projects, including the development or updating of monolingual and multilingual text databases across ten languages, to enable the creation of multilingual chatbots or other tools that incorporate learners’ first languages into the language teaching process (Ministry of the Economy and Innovation of Lithuania, 2024[29]). In Iceland, the government has invested ISK 2.2 billion (Icelandic kronur, ~ EUR 13.6 million) in the “Language Technology for Icelandic” initiative, which includes the development of automatic speech recognition technology for Icelandic. Such technology infrastructure will enable those designing and developing voice-based user interfaces to add Icelandic to machine-translation technology (Language and Voice Laboratory, n.d.[24]; Nikulásdóttir, Guðnason and Steingrímsson, n.d.[30]; Ministry of Culture and Business Affairs, 2024[31]).
Comparative evaluations of skills disparities based on evidence from international large‑scale assessments rarely capture how achievement gaps develop over time. Because most of these studies are cross‑sectional, they reveal disparities at discrete grades or ages, but not trajectories. A growing body of work has addressed this limitation by constructing synthetic cohorts1 to trace changes in information‑processing skills across individuals with different socio-economic backgrounds and between men and women (Borgonovi, Choi and Paccagnella, 2018[32]; Dämmrich and Triventi, 2018[33]; Jerrim and Choi, 2013[34]).
Existing research on the evolution of socio‑economic disparities focuses mainly on the period from early childhood to the end of compulsory schooling. Findings from Australia, Canada, Germany, the United Kingdom and the United States – countries that differ markedly in income inequality and school organisation – indicate that socio‑economic gaps are largely established prior to school entry and widen little, if at all, during the school years (Duncan and Magnuson, 2013[35]; Skopek and Passaretta, 2020[10]). In other words, schools appear to have an equalising effect.
Far less is known about what happens after compulsory schooling, when learning pathways diversify and achievement tests are seldom administered (Schulenberg, Sameroff and Cicchetti, 2004[36]; Schulenberg and Schoon, 2012[37]). Achievement gaps may widen between late adolescence and early adulthood because of initial inequalities and the cumulative nature of advantage (DiPrete and Eirich, 2006[38]). Entry into post‑secondary education depends heavily on success in upper secondary schooling and, unlike earlier stages, participation is voluntary and the quality of provision is more varied (Breen and Jonsson, 2005[39]). Achievement disparities may therefore become more pronounced during this phase.
The literature consistently links socio‑economic gaps in attainment to differences in the resources families invest in their children (Conger and Donnellan, 2007[40]). Individuals with socio-economically advantaged backgrounds are able to provide more material, cultural and social capital; greater involvement in educational decision making; access to better formal learning environments; and richer out‑of‑school experiences (Domina, 2005[41]).
Skills, attitudes and dispositions accumulate throughout life and are transmitted between generations. The effectiveness of new learning depends critically on prior learning, which is in turn influenced by family background and earlier educational experiences. As individuals progress through the life course, past learning increasingly shapes future opportunities. To describe how achievement gaps evolve over time, this section draws on multiple cross‑sectional surveys that sampled individuals from broadly the same birth cohorts at successive ages to construct synthetic cohorts. While this approach does not produce a true longitudinal dataset, it does allow gender and socio‑economic disparities to be compared at key stages of the educational trajectory.
3.3.1. The evolution of gender gaps
Gender gaps in mathematics and numeracy, favouring boys and men, are smallest at age 10 and widen with age, as shown in Figure 3.2. By contrast, gender gaps in reading and literacy, favouring girls and women, peak in mid‑adolescence and then narrow. These patterns appear stable across seven OECD countries for which data are available – Canada, England (United Kingdom), Hungary, the Netherlands, New Zealand, Norway and the United States – and across cohorts born in 1984, 1993 and 1996. For the 1984 cohort, the mathematics gap increased from 0.06 SD at age 10 to 0.08 SD at 16, followed by 0.22 SD at ages 26-34 and 0.24 SD at ages 37-39. The 1993 and 1996 cohorts exhibit similar trends, although the gap does not increase further between age 16 and young adulthood.
Gender gaps in reading and literacy follow a U‑shape: they are largest at age 16 and smaller in primary school and adulthood. This pattern is consistent across the 1990 and 1996 cohorts. Notably, for the 1996 cohort, the relative performance of women improves between age 16 and young adulthood, in contrast to earlier cohorts. These findings corroborate earlier work – e.g. (Borgonovi et al., 2017[42]) and (Borgonovi, Choi and Paccagnella, 2018[32]) – although the magnitude of the gaps varies by country, age and cohort – likely reflecting differences in participation in formal, non‑formal and informal learning.
Difference in standardised mathematics/numeracy and reading/literacy scores between men and women (men minus women), OECD average
Note: Estimates are standardised to an OECD mean of zero and a standard deviation of one in the respective databases. OECD averages include the following countries: Canada, England (United Kingdom), Hungary, the Netherlands, New Zealand, Norway and the United States. Country-specific results in addition to detailed information on data availability are provided in Annex Table 3.A.5.
Source: Calculations based on IEA (1995[43]), TIMSS 1995 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_1995_G4; IEA (2001[44]), PIRLS 2001 Database, https://doi.org/10.58150/PIRLS_2001_data; IEA (2003[45]), TIMSS 2003 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_2003_G4; IEA (2007[46]), TIMSS 2007 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_2007_G4; IEA (2006[47]), PIRLS 2006 Database, https://doi.org/10.58150/PIRLS_2006_data; OECD (2017[48]), PIAAC 1st Cycle Database, www.oecd.org/en/data/datasets/PIAAC-1st-Cycle-Database.html; OECD (2000[49]), PISA 2000 Database, www.oecd.org/en/data/datasets/pisa-2000-database.html; OECD (2009[50]), PISA 2009 Database, www.oecd.org/en/data/datasets/pisa-2009-database.html; OECD (2012[51]), PISA 2012 Database, www.oecd.org/en/data/datasets/pisa-2012-database.html; OECD (2022[52]), PISA 2022 Database, www.oecd.org/en/data/datasets/pisa-2022-database.html; and OECD (2024[53]), PIAAC data and methodology, www.oecd.org/en/about/programmes/piaac/piaac-data.html.
Gender gaps evolve in part because early comparative strengths influence subsequent educational choices. Girls who perform well in mathematics often excel even more in reading; when encouraged to pursue their strongest subjects, they are channelled into humanities‑oriented programmes and receive less exposure to advanced mathematics, reducing the likelihood of progressing into STEM pathways. Boys, by contrast, seldom enjoy a relative advantage in reading. During compulsory schooling they must engage with set texts that may hold little personal interest (OECD, 2015[54]), but outside school they are free to explore an array of reading materials that align with their preferences. This voluntary reading can narrow the literacy gap after adolescence, while their sustained engagement with mathematics maintains or enlarges the numeracy advantage. Gendered expectations, differential guidance and a paucity of visible female role models in STEM further reinforce these trajectories. In adulthood, women are more likely than men to engage in unpaid care activities, which may limit their ability to engage in activities that promote strengthened literacy and numeracy skills. Box 3.2 illustrates between-country differences in how gender disparities in reading and mathematics evolve between primary and secondary school, and Box 3.3 provides examples of policies and initiatives aimed at reducing gender disparities over the life course.
Figure 3.2 suggests that for cohorts of children born between 1984 and 1996, gender gaps in favour of girls and gender gaps in favour of boys grew wider between age 10 and age 15-16.
Figure 3.3 complements this evidence with estimates of gender gaps in mathematics and reading for cohorts of children born in 2004-05 across a wider set of OECD countries to highlight the evolution of gender gaps during compulsory schooling. Estimates are aligned with those illustrated in Figure 3.2: during primary and secondary school gender gaps become more pronounced, although countries differ with respect to the size of gender gaps at age 10, as well as the evolution of such gaps as young people age. For example, the gender gap in favour of boys in mathematics at age 10 is widest in Italy (0.27 SD), whereas in Finland, girls outperform boys in mathematics at age 10 by 0.13 SD. Although in Italy the gender gap in favour of boys is reduced between age 10 and age 15-16 (it is 0.23 SD at age 15-16), it remains the largest among countries with available data. Boys’ disadvantage in Finland narrows, and by age 15-16 the gender gap in favour of girls in mathematics is 0.05 SD. The gender gap in favour of boys in mathematics grows the most in Chile where it is 0.02 SD at age 10 and 0.18 SD at age 15-16.
Similarly, the gender gap in favour of girls in reading is large at age 10 and in most countries, it grows larger by age 15-16. For example, in Finland, girls outperform boys in reading by 0.3 SD at age 10, a gap that grows as large as 0.44 SD by age 15-16. In Portugal, the gender gap in favour of girls in reading is only 0.02 SD at age 10 but grows to 0.20 SD by age 15-16. Comparing the evolution of gender gaps in mathematics and reading in
Figure 3.3 suggests that gaps in mathematics in favour of boys tend to be smaller in countries in which gender gaps in favour of girls are wider and, conversely, gender gaps in mathematics in favour of boys are widest in countries in which gender gaps in reading in favour of girls are smaller. Furthermore, countries in which gender gaps in mathematics in favour of boys grow the most between age 10 and age 15-16 tend to be those in which gender gaps in reading in favour of girls grow the least.
Note: Estimates are standardised to an OECD mean of zero and a standard deviation of one in the respective databases. Norway assessed its fifth grade at the age of 10 to obtain better comparisons with other northern European countries.
Countries are ranked in ascending order based on the score-point difference of boys and girls in mathematics (Panel A) and reading (Panel B).
Source: Calculations based on IEA (2015[55]), TIMSS 2015 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_2015_G4; IEA (2016[56]), Progress in International Reading Literacy Study (PIRLS) 2016 Database, https://doi.org/10.58150/PIRLS_2016_data; and OECD (2022[52]), PISA 2022 Database, www.oecd.org/en/data/datasets/pisa-2022-database.html.
Addressing gender disparities in skills requires a comprehensive, life-course approach – from early childhood education to adulthood – with targeted policies and initiatives that challenge stereotypes, broaden opportunities and support the participation of men and women in all sectors. Beginning in early childhood education and care (ECEC), interventions can address how children are socialised and extend all the way to industry-specific programmes that lower barriers for women accessing labour market opportunities in men-majority fields.
Early childhood education and care
Gender equality initiatives in Sweden’s preschools aim to eliminate gender stereotypes in early childhood education by fostering inclusive environments that challenge traditional gender roles. These schools use gender-neutral language, avoid gendered expectations and offer diverse learning activities – from creative arts to problem solving – ensuring all children engage equally. Classrooms are structured to avoid gender-based divisions, promoting equal participation in all areas of play and learning. A key component is teacher training, equipping educators with strategies to recognise and counteract unconscious bias. The approach aligns with Sweden’s broader commitment to gender equality, embedding values of fairness and inclusion from an early age (Shutts et al., 2017[57]; Brussino and McBrien, 2022[58]).
Schooling and young adulthood
Ireland’s Facts, Faces, Futures campaign, launched in 2023 by the National Apprenticeship Office, promotes the participation of women in apprenticeships. Featuring female role models and distributed to over 130 all-girls schools, it has received positive feedback, with visits continuing into 2024. By the end of 2023, over 100 schools had received apprenticeship information. In 2024, the campaign focused more on craft apprenticeship opportunities and collected feedback to enhance impact (Education Magazine, 2023[59]). The campaign is central to the Action Plan for Apprenticeship 2021-2025; the Minister for Further and Higher Education, Research, Innovation and Science credited Facts, Faces, Futures for promoting women’s participation in apprenticeships in Ireland and for the increase in the share of apprentices who are women (approximately 9% of the overall apprentice population are women, a 70% increase between the end of 2021 and February 2025) (House of the Oireachtas, 2025[60]).
In the Slovak Republic the You too in IT (Aj Ty v IT) initiative encourages and supports girls and women to pursue careers in information and communication technology (ICT). It offers coding workshops for secondary school girls, upskilling for career changers and teacher training. Activities are often delivered in all-girl groups to create a supportive environment, enhancing engagement and confidence. The programme links participants to tech employers, supporting diversity and ensuring technology benefits from a broader range of perspectives (Aj Ty v IT, 2025[61]).
In France, the government has launched the Girls and Maths (Filles et Maths) plan, introducing measures to encourage girls to choose more mathematics-intensive study pathways throughout their formal schooling, and ultimately enrol in mathematics-related higher education pathways. The gap between boys’ and girls’ mathematics achievement and enrolment begins in primary school and widens with age – the plan therefore aims to tackle the persistent gender stereotypes in society and the classroom which lead to girls opting out of studying mathematics. The plan includes additional training for national education staff, setting higher targets for girls’ enrolment in specialised mathematics courses throughout secondary school, and fostering networks and promoting role models to inspire young girls to choose mathematics-related career paths. Beyond mathematics, the plan ultimately aims to guide more girls towards scientific and technological careers, areas often referred to as future-oriented and offering greater economic opportunities (French Ministry of National Education, 2025[62]; 2025[63]).
Adulthood
Women continue to face barriers in entering and progressing within men-majority fields such as construction and technology, which are sectors crucial to the green and digital transition and that are experiencing acute labour shortages. In Denmark, the Boss Ladies project (Boss Ladies, 2025[64]) seeks to shift perceptions and attract more women into construction, where they currently represent just 9% of the workforce. The project also works to reshape narratives through media outreach and social media campaigns featuring female role models. The initiative includes:
Boss ladies ambassadors: Over 300 female builders visiting schools as role models.
Learning labs: Collaboration with vocational schools and counsellors to recruit more women.
Talent development: Camps, workshops and mentoring focused on communication and career planning.
Girl boss garage: Providing practical building experiences for girls without industry connections.
Apprenticeship match: One-on-one support to help women secure apprenticeships.
3.3.2. The evolution of gaps related to parental education
At every age, people from tertiary‑educated families outperform their peers with no tertiary-educated parent, with the gap widening between age 16 and early adulthood in all cohorts. Panel A in Figure 3.4 presents the average standardised difference in mathematics and numeracy between students with at least one tertiary‑educated parent and those with no tertiary-educated parent. As shown, the gap for the 1993 cohort widens from 0.45 SD at age 16 to 0.52 SD at the age of 17-25. The disparity flattens for the 1984 cohort between age 27-34 and 37-39, increases for the 1993 cohort (from 0.52 SD at age 17-25 to 0.63 SD at age 28-30) and drops slightly for the 1996 cohort (from 0.41 SD at age 16 to 0.53 SD at age 25‑27). Although country‑specific patterns diverge, the broad tendency across the seven OECD countries used for this analysis is one of increasing socio‑economic disparity.
In reading and literacy (Figure 3.4, Panel B), the parental education gap is likewise present at every age. It contracts between primary and lower‑secondary schooling, then expands and plateaus through early adulthood. For instance, the disparity for the 1990 cohort starts at 0.57 SD at age 10, drops to 0.34 SD at age 16, increases back to 0.47 SD at age 20-28 and then drops slightly to 0.46 SD at age 31-33. Overall, socio‑economic gaps are wider than those linked to gender, but national trajectories vary.
These results highlight that socio-economic disparities in literacy and numeracy observed at a young age not only persist but often widen into young adulthood in most countries. Previous work attributes this widening to unequal access to post‑secondary education and training, higher dropout rates among students with socio-economically disadvantaged backgrounds, and their increased likelihood of unemployment or employment in occupations that do not foster skills development. Compulsory schooling appears to act as an equalising force, but afterwards divergent educational and labour market experiences frequently amplify socio-economic gaps in information-processing skills, particularly numeracy skills.
Box 3.4 presents evidence on country specific disparities in mathematics achievement at age 10 and ages 15-16 for two cohorts: one that participated in the Trends in International Mathematics and Science Study (TIMSS) in 2011 and PISA in 2018, and one that participated in TIMSS in 2015 and PISA in 2022. Box 3.5 provides examples of initiatives that promote access to compulsory schooling for children and the uptake in formal skills development opportunities throughout the life course.
Differences in standardised mathematics/numeracy and reading/literacy scores between individuals whose parents have tertiary education and those whose parents do not, OECD average
Note: Estimates are standardised to an OECD mean of zero and a standard deviation of one in the respective databases. OECD averages include the following countries: Canada, England (United Kingdom), Hungary, the Netherlands, New Zealand, Norway and the United States. See the note for Figure 3.1 for a description of parental educational attainment. Country-specific results in addition to detailed information on data availability are provided in Annex Table 3.A.5.
Source: Calculations based on IEA (1995[43]), TIMSS 1995 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_1995_G4; IEA (2001[44]), PIRLS 2001 Database, https://doi.org/10.58150/PIRLS_2001_data; IEA (2003[45]), TIMSS 2003 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_2003_G4; IEA (2007[46]), TIMSS 2007 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_2007_G4; IEA (2006[47]), PIRLS 2006 Database, https://doi.org/10.58150/PIRLS_2006_data; OECD (2017[48]), PIAAC 1st Cycle Database, www.oecd.org/en/data/datasets/PIAAC-1st-Cycle-Database.html; OECD (2000[49]), PISA 2000 Database, www.oecd.org/en/data/datasets/pisa-2000-database.html; OECD (2009[50]), PISA 2009 Database, www.oecd.org/en/data/datasets/pisa-2009-database.html; OECD (2012[51]), PISA 2012 Database, www.oecd.org/en/data/datasets/pisa-2012-database.html; OECD (2022[52]), PISA 2022 Database, www.oecd.org/en/data/datasets/pisa-2022-database.html; and OECD (2024[53]), PIAAC data and methodology, www.oecd.org/en/about/programmes/piaac/piaac-data.html.
Although it is possible to map the evolution of socio-economic disparities for selected cohorts, as shown in Figure 3.4, there is a lack of data on parental education for fourth graders in early editions of the TIMSS study, which prevents examining early trajectories for numeracy. Figure 3.5 shows that disparities related to parental education are generally wider at age 10 than at ages 15-16. On average across OECD countries, the difference in mathematics achievement for the 2004/2005 cohort decreased from 0.66 SD at age 10 to 0.41 SD at ages 15-16.
Across countries with available data, score-point differences in mathematics between young people whose parents had and did not have a tertiary qualification were largest in Hungary and the Slovak Republic (1.06 SD and 0.84 SD) at age 10, and in the Czech Republic (hereafter ‘Czechia’), Poland and the Flemish Region (Belgium) (0.60 SD, 0.58 SD and 0.52 SD) at age 15-16 (Figure 3.5, Panel A). Socio-economic disparities in mathematics narrowed the most between age 10 and age 15‑16 in Hungary (from 1.06 to 0.47), whereas they remained relatively stable in the Flemish Region (Belgium) (0.56 SD and 0.52 SD, respectively).
In reading, socio-economic gaps narrowed the most in the Slovak Republic (from 0.83 to 0.32) and did not widen in any country with available data between age 10 and age 15-16 (Figure 3.5, Panel B). Israel was the country with the widest socio-economic gap in reading at age 10 (0.88 SD) but by 15-16 the gap narrowed to 0.52 SD. Similarly, in Hungary, the socio-economic gap in reading at age 10 was 0.87 SD but by 15-16 the gap narrowed to 0.43 SD.
Note: Estimates are standardised to an OECD mean of zero and a standard deviation of one in the respective databases. Norway assessed its fifth grade at the age of 10 to obtain better comparisons with other northern European countries. Parental education is based on students’ responses. Information on their mothers’ and fathers’ education was used to distinguish between young people with at least one tertiary-educated parent (ISCED 2011 levels 5, 6, 7 and 8) and those with no tertiary-educated parent.
Countries are ranked in ascending order based on the score-point difference of adults with tertiary and non-tertiary educated parents in mathematics (Panel A) and reading (Panel B).
Sources: Calculations based on IEA (2015[55]), TIMSS 2015 Grade 4 Database, https://doi.org/10.58150/IEA_TIMSS_2015_G4; IEA (2016[56]), Progress in International Reading Literacy Study (PIRLS) 2016 Database, https://doi.org/10.58150/PIRLS_2016_data; and OECD (2022[52]), PISA 2022 Database, www.oecd.org/en/data/datasets/pisa-2022-database.html.
Access to quality education throughout the life course can reduce disparities in the short term and prevent them from accumulating over time. Policies may range from investment in ECEC infrastructure that considers the needs of families with socio-economically disadvantaged backgrounds, programmes promoting access to higher education, and interventions in adulthood that mitigate disparities in early years and prevent the accumulation of disadvantages resulting from disrupted schooling or low levels of skills.
Early childhood education and care
Sure Start Children’s Centres in the United Kingdom offer integrated support for families with children under five, particularly in disadvantaged areas. Services include health, parenting, financial advice and employment support for parents, and early education. Evaluations show positive outcomes: improved early development, physical health, home environments and early identification of special educational needs. The programme is considered cost-effective and impactful in reducing early childhood inequalities and has achieved significant educational gains for young children, with benefits lasting into secondary school (Carneiro et al., 2025[65]; UK Department of Education, 2010[66]; OECD, 2025[67]). At its peak in 2009-2010, Sure Start had around 3 600 centres operating in England. Following funding cuts, local authorities scaled back or closed most of the centres by 2018. The programme cost GPB 2.7 billion a year in 2023-24 prices, delivering a return of around GBP 2 for every GBP 1 invested (Carneiro et al., 2025[65]).
Finland’s Programme for Equality and Non-Discrimination in Education and Training 2025 is a ten-point action plan covering every level of the education system and the stakeholders involved in children’s educational and developmental journeys. Related to ECEC, the programme introduced new quality indicators that promote inclusivity and cultural sensitivity, giving teachers the tools to evaluate their pedagogical practices accordingly. Professional development for teachers and school staff to recognise unconscious bias, identify discriminatory behaviour and create inclusive classroom environments is also envisioned as part of this programme, including in initial teacher training. The programme builds upon the Right to Learn initiative implemented between 2020 and 2022, which gave schools additional resources and teacher training to identify and support students requiring additional help facing barriers related to socio-economic or linguistic background or learning obstacles. Both programmes in-built family and community involvement to ensure effectiveness in reducing achievement gaps, particularly among students with an immigrant background (OECD, 2022[68]; Ministry of Education and Culture, Finland, 2025[69]).
Head Start programmes across the United States promote school readiness for low-income families with children up to age five. In addition to early learning, the programmes support child health, parent engagement and access to public services. Programmes are federally funded and locally delivered through schools and non-profits, aiming to support both children’s development and family well-being (Bierman et al., 2013[70]; Bierman et al., 2008[71]; OECD, 2025[67]).
Schooling and young adulthood
The Irish Higher Education Access Route (HEAR) eases entry into higher education for students from disadvantaged backgrounds. Applicants must meet financial, social and cultural criteria, with household income the most important, and can apply for HEAR at the same time as university. Each participating higher education institution has a reserved number of places to offer eligible HEAR applicants at lower or reduced Leaving Certificate (the national end of high school exam) points. Like all other applicants, HEAR prospective beneficiaries need to meet the minimum entry requirements (at the institutional level) and any specific programme requirements before being considered for a HEAR reduced points offer (HEAR, 2025[72]). In 2022, 11% of university applicants applied through HEAR, with an average 84% acceptance rate (Irish Universities Association, n.d.[73]).
HEAR also offers support such as orientation, tutoring, mentoring, study skills help and financial advice, and participants report a strong sense of identity and belonging (Byrne et al., 2013[74]). HEAR students are more likely to transition successfully to higher education (Byrne et al., 2013[74]), achieve honours degrees and perform better academically than peers with similar backgrounds (Byrne et al., 2013[74]; Denny et al., 2014[75]). The programme is funded as a part of wider efforts by the Irish government to widen access to higher education, promote innovation in higher education and support national development.
Adulthood
Luxembourg’s FutureSkills Initiative targets unemployed individuals with secondary education, offering three months of training followed by a six-month public sector internship. Training focuses on soft skills, digital competencies and office tasks. The aim is to improve employability and open access to public sector roles, which may otherwise be out of reach for individuals from disadvantaged backgrounds (Agence pour le développement de l’emploi, 2025[76]; OECD, 2023[77]).
Norway’s Skills Plus Work programme provides workplace-based basic skills training for employed adults or individuals engaged in volunteering who have low formal qualifications. Training focuses on literacy, numeracy, digital and communication skills using job-relevant content. Employers or voluntary organisation collaborate with training providers to design tailored programmes. Funded by the Ministry of Education and Research, the programme targets learners unlikely to participate in formal training (OECD, 2020[78]; Directorate of Higher Education and Competence, Norway, 2025[79]).
Spain’s Vives Emplea Saludable programme aims to enhance employability by integrating physical, emotional and social well-being into employment support programmes among unemployed individuals. The programme combines job-readiness training with interventions focused on mental health, physical activity, sleep and nutrition. Emotional resilience, healthy behaviours and job placement rates were the key metrics through which programme success was evaluated. The initiative is delivered through labour-market programmes led by non-governmental organisations (European Commission, 2023[80]).
3.4.1. Disparities in educational attainment
The educational attainment of 30-65 year-olds varies markedly across OECD countries and is unevenly distributed across demographic groups: younger adults (30-49 year-olds) “out‑credential” their older peers; women match or exceed men in tertiary attainment; and individuals with socio-economically advantaged backgrounds or who grew up in urban settings on average have higher levels of educational qualifications than less advantaged adults or adults who grew up in rural areas. Programme orientation further stratifies opportunities, with vocational-focused upper secondary pathways typically facilitating smoother transitions into work but enabling less exposure to advanced academic content than general tracks. At the tertiary level, professionally‑oriented short‑cycle programmes offer a rapid route to skilled employment. Because measured proficiency reflects both the quantity and the nature of formal learning, disparities in educational attainment are a primary driver of skills gaps. This section quantifies the extent to which differences in educational attainment level and field of study between different groups of individuals explain disparities in 21st-century skills.
Figure 3.6 shows the highest educational qualification of 30-65 year-olds who participated in the OECD Survey of Adult Skills. On average across OECD countries, 14% have not completed upper secondary education, 28% hold an upper or post‑secondary credential with a vocational orientation, 15% possess an upper or post‑secondary certificate with an academic orientation, 9% have completed a short‑cycle tertiary programme, and 34% have a bachelor’s degree or higher.
National profiles vary sharply. The proportion of adults without an upper secondary credential ranges from 48% in Portugal to 4% in Czechia. Vocational upper secondary and post‑secondary qualifications are the modal attainment in Czechia (66%) yet account for just 8% of attainment in Portugal. Academically oriented upper secondary programmes are most common in the United States (31%) and least common in Germany (2%). Participation in short‑cycle tertiary education is 21% in Japan but remains below 1% in Czechia, Germany, Poland and the Slovak Republic. Finally, the share of adults with a bachelor’s degree or above ranges between 48% in the Flemish Region (Belgium) and 17% in Italy.
Share of 30-65 year-olds, by highest educational attainment
Note: Adults aged 30-65. Respondents' educational attainment is based on the International Standard Classification of Education (ISCED) 2011, grouped into below upper secondary (ISCED 0, 1, 2), upper secondary and post-secondary general (gen. ISCED 3 short, gen. ISCED 3 access 3, gen. ISCED 3 access 3/4, gen. ISCED 3 access 5/6/7), upper secondary and post-secondary vocational (voc. ISCED 3 short, voc. ISCED 3 access 3, voc. ISCED 3 access 3/4, voc. ISCED 3 access 5/6/7), short-cycle tertiary education (ISCED 5 nfs [not further specified], gen ISCED 5, voc ISCED 5) and bachelor’s or equivalent or above (ISCED 6, 7, 8).
Countries are ranked in descending order based on the percentage of adults with below upper secondary education.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
When disaggregating data on educational attainment by age, gender, socio-economic background, childhood residential context and own occupation, there are clear patterns and differences, as shown in Figure 3.7. Educational expansion is evident: 40% of 30‑49 year‑olds hold a bachelor’s degree or higher, compared with 26% of 50‑65 year‑olds. Moreover, 16% of 50‑65 year‑olds have not completed an upper secondary qualification compared to only 11% of 30-49 year-olds. Women outpace men in tertiary educational attainment (37% vs. 31%, respectively) but are less likely to hold vocationally oriented upper secondary or post-secondary degrees (26% vs. 31%, respectively). Figure 3.7 also indicates that socio‑economic background is a powerful determinant of educational attainment. Among adults with tertiary‑educated parents, 61% hold at least a bachelor’s degree, compared with 26% of those with no tertiary-educated parent. A similar pattern emerges by parental occupation: 53% of adults from high‑status occupational backgrounds hold a bachelor’s degree or higher, compared to 23% of those from low‑status backgrounds. As many as 40% of adults who grew up in cities earn a bachelor’s degree or higher compared to 26% of those who grew up in villages. By contrast, differences by immigrant background are small.
Share of adults, by socio-demographic characteristic and highest educational attainment, OECD average
Note: Adults aged 30-65. Respondents' occupation and parental occupation (at respondents’ age 14) is based on the International Classification of Occupations (ISCO) and grouped into high-status: managers, professionals, and technicians and associate professionals (ISCO 1-3); and low-status: clerical support workers; service and sales workers; skilled agricultural, forestry and fishery workers; craft and related trades workers; plant and machine operators, and assemblers; and elementary occupations (ISCO 4-9). Childhood residential context (at respondents’ age 14) refers to whether the respondent grew up in villages or cities. Groups by immigrant background distinguish between children of immigrants, immigrants and non-immigrants. Children of immigrants were born in the country in which they currently reside, but their parents were not, or they were born in a different country and moved to their current country of residence before the age of 18. Immigrants are defined as those who migrated to their current country of residence at age 18 or older. Non-immigrants were born in their current country of residence, as were their parents. See the note for Figure 3.1 for a description of parental educational attainment. See the note for Figure 3.6 for a description of respondents’ educational attainment. Numbers in the figure reflect percentages associated with each educational level for each subgroup. Country-level results are provided in Annex Table 3.A.1.
* Next to percentages in the figure indicate that disparities across groups (30-49 vs. 50-65, women vs. men, non-tertiary vs. tertiary, low-status vs. high-status, children of immigrants vs. non-immigrant, village vs. city) are significant at the 5% level.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
3.4.2. Association between disparities in educational attainment and disparities in skills
Higher levels of educational attainment are, on average, significantly and positively associated with core 21st-century skills, with each successive step up the “qualification ladder” associated with progressively stronger skills, as shown in Table 3.2. For example, relative to those who left education before completing upper secondary education, adults who hold upper or post‑secondary vocational qualifications have higher literacy (0.59 SD difference), numeracy (0.64 SD difference) and adaptive problem solving (0.51 SD difference) skills. Adults who possess an upper or post‑secondary certificate with a general orientation have even higher literacy (0.73 SD difference), numeracy (0.78 SD difference) and adaptive problem solving (0.64 SD difference) skills than individuals who did not complete upper secondary qualifications. Adults who have completed a short‑cycle tertiary programme have higher literacy (0.87 SD difference), numeracy (0.93 SD difference) and adaptive problem solving (0.77 SD difference) skills than individuals who did not complete upper secondary qualifications. Finally, adults who have completed a bachelor’s degree or higher have higher literacy (1.22 SD difference), numeracy (1.30 SD difference) and adaptive problem solving (1.08 SD difference) skills than individuals who did not complete upper secondary qualifications.
When looking at skills differences by age, on average, around one-third of the difference in numeracy skills between 50-65 year-olds and 30-49 year-olds can be explained by differences in educational attainment between the two groups (Table 3.2). The corresponding shares are about one-quarter for literacy and one-fifth for adaptive problem solving.
Gender differences in key 21st-century skills point to possible gender-specific experiences both within and beyond formal education. As noted, women have higher literacy skills but score lower in numeracy and adaptive problem solving. These patterns persist after adjusting for other socio-demographic characteristics (age, parental education and occupation, childhood residential context) (Table 3.2). However, additionally adjusting for educational attainment widens gender gaps in numeracy (from 0.18 SD to 0.24 SD, a 33% increase) and adaptive problem solving. By contrast, literacy is similar among men and women with similar levels of educational attainment. These factors are further explored in Sections 3.5 (field of study) and 3.6 (adult education and training).
The numeracy returns of educational qualifications differ across countries (Figure 3.8). For example, in New Zealand, after taking into account differences in socio-demographic characteristics, individuals who have obtained an upper and post-secondary education with a vocational orientation have 0.97 SD higher numeracy proficiency than individuals who did not obtain an upper secondary education qualification. In Israel, the difference is considerably smaller at 0.27 SD. Similarly, in Finland, individuals who have obtained an upper and post-secondary education with a general orientation have 1.25 SD higher numeracy proficiency compared to individuals who did not obtain an upper secondary education qualification; in Israel, this difference is 0.32 SD. The numeracy returns associated with having obtained a short-cycle tertiary education rather than no upper secondary qualification range between 1.45 SD in Singapore and 0.48 SD in Italy; for those who have obtained a bachelor’s degree or above, the numeracy returns range between 1.82 SD in Singapore and 0.70 SD in Croatia.
Adults with at least one tertiary-educated parent score higher on core 21st-century skills than their peers with no tertiary-educated parent (Table 3.2). For example, the difference in literacy between individuals with and without tertiary-educated parents is 0.30 SD when adjusting for other socio-demographic characteristics and 0.13 SD when additionally adjusting for differences in educational attainment between the two groups, a 57% decline. Similarly, the difference in numeracy between individuals with and without tertiary-educated parents is 0.29 SD and 0.12 SD, respectively, after adjusting for differences in educational attainment, a 59% decline. Even when comparing individuals from different backgrounds but with similar educational trajectories, differences remain, reflecting the varied quality of education experiences available to different groups and the differences in learning that occur outside of formal education settings over the life course. However, results suggest that a key reason adults with socio-economically advantaged backgrounds have considerably higher levels of information-processing skills is their greater opportunity to obtain higher levels of educational attainment.
Regression coefficients before and after adjusting for respondents’ educational attainment, OECD average
Note: Adults aged 30-65. Coefficients in bold are statistically significant at the 5% level. Results in columns (1), (3) and (5) explain the respective skill while adjusting for differences in gender, age, parental education, parental occupation, childhood residential context and immigrant background. Results in columns (2), (4) and (6) additionally adjust for differences in respondents’ educational attainment. See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Standard errors are provided in Annex Table 3.A.1.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Adults with parents who worked in high-status occupations score higher on core 21st‑century skills than their peers whose parents worked in low-status occupations. On average, differences in educational attainment explain around half of the observed gap in numeracy between adults whose parents did or did not work in high-status occupations, but the role of differences in educational opportunities in shaping skills disparities differs across countries. Figure 3.9 illustrates the variation across countries in how differences in educational attainment relate to disparities in numeracy proficiency between individuals with and without at least one parent who worked as a manager or professional (high-status occupations) when the respondent was 14 years old. Differences between individuals with and without parents working in high-status occupations are highest in the United States (0.23 SD difference) and lowest in Korea (0.03 SD difference). Such differences can be explained, to a large extent, by differences in the educational opportunities different groups benefit from. In the United States, differences in educational attainment explain around 52% of the difference in numeracy skills between adults with and without parents who worked in high-status occupations when they were growing up, compared to 27% in Korea.
Change in numeracy (standardised difference) associated with respondents’ educational level (ref. below upper secondary) after accounting for socio-demographic characteristics
Note: Adults aged 30-65. Socio-demographic characteristics comprise gender, age, parental education, parental occupation, childhood residential context and immigrant background. See the note Figure 3.7 for a description of respondents’ educational attainment.
Countries are ranked in descending order based on the effect of parents working in high-status occupations on numeracy proficiency.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Change in numeracy (standardised difference) associated with parents working in high-status occupations (ref. low-status) before and after accounting for respondents’ education
Note: Adults aged 30-65. Other socio-demographic characteristics comprise gender, age, parental education, parental occupation, childhood residential context and immigrant background.
Countries are ranked in descending order based on the effect of parents working in high-status occupations on numeracy proficiency.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Adults who grew up in cities outperform peers from towns and villages in core 21st-century skills, but adjusting for educational attainment narrows differences (Table 3.2). Individuals who grew up in cities may have had better access to quality education, resources and supportive learning environments, which in turn could have led to higher educational attainment and subsequently better skills. In contrast, those from villages may have faced more limited educational opportunities, contributing to lower attainment and, consequently, weaker skills outcomes. These results suggest that urban–rural disparities in the quality and breadth of educational opportunities explain part – but not all of the skills divergence between the two groups.
Adults who are the children of immigrants have lower information-processing skills than their peers without an immigrant background, even after accounting for other background factors (Table 3.2). Contrary to other characteristics, differences in educational attainment do not explain any of the observed differences in skills between the two groups, with barriers outside of formal education settings appearing to be the key drivers of the immigrant skills gap.
3.4.3. Skills returns to educational qualifications
While education is a key factor in shaping skills acquisition, its effects can vary considerably depending on age, gender, socio-economic background, childhood residential context and immigrant background. This section considers if the skills returns to education – i.e. the skills that individuals are more likely to possess after completing certain levels of education – differ across different socio-demographic groups. Table 3.3 illustrates differences in the skills returns to educational qualifications among men and women, age groups, and socio-economic background of the family of origin. The skills returns to educational qualifications are broadly homogeneous with respect to childhood residential context and immigrant background, with detailed results reported in Annex Table 3.A.1.
Educational attainment is generally less strongly associated with information-processing skills among individuals with socio-economically advantaged backgrounds, i.e. individuals whose parents worked in high-status occupations or obtained tertiary qualifications. For example, the literacy return to having obtained a bachelor’s degree or more advanced qualifications rather than not having obtained an upper secondary qualification is 0.17 SD higher for adults from high-status rather than low-status households.
When family resources are scarce, formal education becomes the primary channel for developing and signalling competence. By contrast, when resources are abundant, additional credentials add relatively little. Advantaged families can provide physical and social environments for their children that foster the development of information‑processing skills outside formal education settings. For disadvantaged youth, classrooms remain the main place where such skills can be cultivated. The lower skills returns among adults with socio-economically advantaged backgrounds may also reflect selection processes. Young people with socio-economically disadvantaged backgrounds are less likely to pursue advanced educational qualifications because of economic and logistical barriers, lack of role models, and lower perceived returns to education. Therefore, those who do pursue such paths are generally highly motivated and especially well-suited to acquire the set of skills that formal education is designed to promote, including literacy, numeracy and adaptive problem solving. Box 3.6 presents examples of policies and interventions aimed at addressing the risk factors for school dropout.
Gender differences in information-processing skills associated with having obtained a bachelor’s degree or more advanced qualifications rather than not having completed an upper secondary qualification favour men. For example, the numeracy return to having obtained a bachelor’s degree or more advanced qualifications rather than not having obtained an upper secondary qualification is 0.13 SD higher for men than women (Table 3.3). Differences in the skills returns to advanced qualifications between men and women could reflect gender differences in field of study, which is examined in the next section.
In contrast, the skills returns to education tend to be relatively homogeneous across age groups (Table 3.3). For example, the literacy returns to having obtained a bachelor’s degree or more advanced qualifications are 0.1 SD higher for those aged 50-65 compared to those aged 30-50, a difference that is not statistically significant at any conventional significance level (Table 3.3).
Each skill is adjusted for socio-demographic characteristics, and the interaction is then shown between educational attainment and the characteristic of interest (age, gender, parental education or parental occupation), OECD average
Note: Adults aged 30-65. Coefficients in bold are statistically significant at 5% level. See the note for Figure 3.7 for a description of respondents’ educational attainment. Results explain the skill while adjusting for differences in gender, age, parental education, parental occupation, childhood residential context, immigrant background, educational attainment and an interaction term between educational attainment and the characteristic of interest (i.e. depending on the row these are age, gender, parental education or parental occupation). See the note for Figure 3.7 for the definitions of groups based on parental education and parental occupation. Standard errors are provided in Annex Table 3.A.1.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
In their Shoes (En Sus Zapatos) is a Spanish emotional education programme tackling bullying and promoting positive school environments. It targets students aged 4-17, as well as teachers, families and non-teaching staff, using a cascade learning model. Trained teachers pass on emotional literacy techniques to peers, parents and students, with older students mentoring younger ones. The programme is designed to build emotional literacy, empathy and conflict resolution skills to reduce bullying and foster peaceful coexistence between different students. Students learn to identify and manage emotions and resolve conflicts constructively. Evaluations by the Spanish Ministry of Education found that 92% of students could better recognise emotions, 90% of teachers reported increased empathy, and parents noted improved relationships with their children after participation. Over 120 000 individuals have been reached, with strong institutional backing and media visibility. Funding for the programme comes from partnerships with public institutions, including the Community of Madrid, which supports its widespread implementation across schools (UNESCO, 2024[81]; OECD, 2024[2]; Teatro de Conciencia, 2025[82]).
In Finland, KiVa is a research-based anti-bullying programme developed by the University of Turku and funded by the Ministry of Education and Culture. It includes three core components: prevention, intervention and monitoring. Preventive measures are embedded in the curriculum through lessons and online games; intervention strategies provide schools with structured, solution-focused tools to manage incidents; and annual surveys track progress and highlight areas for improvement. KiVa has demonstrated strong results, significantly reducing bullying and victimisation in over 200 Finnish schools within its first year. Benefits include improved school climate, student motivation and academic performance. The programme has also shown positive effects internationally, including in the Netherlands and Italy. Teachers are supported with ready-to-use resources such as lesson plans, guides and digital tools to ensure effective implementation (KiVa, 2025[83]).
In France, the Roped party of Success (Cordées de la Réussite) programme boosts academic ambition and combats self-censorship among disadvantaged secondary school students. It connects higher education institutions with secondary schools, especially in marginalised areas, offering mentorship and exposure to academic and career pathways. Support is tailored to individual needs and fosters long-term ambition – see Box 4.8 in (OECD, 2023[84]).
Australia’s Be You is a national initiative supporting mental health in early learning and schools. It equips educators with tools to build mentally healthy environments through professional learning modules, covering areas such as family partnerships, resilience, early support and crisis response. Resources include action plans, tools and expert guidance. The whole-community approach promotes collaboration across schools and families and aligns with national education policy (McBrien, 2022[85]; Hoare et al., 2020[86]).
The data presented in this section are drawn from the Survey of Adult Skills, which reflects some key characteristics that shape access to education opportunities. However, within countries, other characteristics also play an important role. For example, Box 3.7 provides insights into how some OECD countries have succeeded in addressing the complex and intersectional needs of the Roma and Irish traveller population, and in promoting the uptake of tertiary education and skills development more broadly.
Roma are the EU’s largest ethnic minority, numbering around six million; they often have a socio-economically disadvantaged background and face exclusion and discrimination (EPRS, 2025[87]). Only 44% of Roma children aged 3-7 attend ECEC, compared to the EU average of 93%, and Roma employment rates remain low at 43%, compared to 70% for the EU population overall. Despite EU strategies for inclusion (e.g. the EU Roma Strategy 2020-2030), progress remains slow, especially in combating educational segregation (European Commission, 2023[88]; European Parliament, 2023[89]).
EU-led initiatives have sought to improve educational inclusion and outcomes. The Inclusive Schools: Making a Difference for Roma Children (INSCHOOL) project 2017-2021 involved 25 schools in Czechia, Hungary, Romania, the Slovak Republic and the United Kingdom, implementing teacher training, school governance reforms and community engagement. Over 4 000 stakeholders participated, and evaluations show improvements in inclusive practices and school conditions, inspiring further national policy actions (European Union/Council of Europe, 2025[90]).
In the Slovak Republic, where Roma comprise 8% of the population, there have been persistent challenges in including Roma children in the education system (EPRS, 2025[87]; Kahanec et al., 2020[91]; Council of Europe, 2012[92]). With the support of the EU, the Slovak Republic developed the Roma Equality, Inclusion and Participation Strategy 2030 (Office of the Government of the Slovak Republic, 2022[93]). Recent interventions have prioritised ECEC accessibility, notably through:
Schools Open to All: Employing teaching assistants and promoting informal learning, reaching over 130 schools and benefiting 416 children and families (Koreň, 2018[94]).
Inclusion in Maternity Schools: Employing Roma parental assistants and additional teaching staff and collaborating with social workers to improve kindergarten attendance among Roma children (European Commission, 2024[95]).
Hungary’s Roma population, which comprises about 7% of the country and is geographically unevenly distributed, faces persistent school segregation and low educational attainment, with just 24% completing secondary education (Council of Europe, 2012[92]). The Hungarian National Social Inclusion Strategy 2020-2030 (HNSIS) recognises these challenges and prioritises inclusive ECEC initiatives and targeted support (European Commission, 2022[96]). One notable example is the Gandhi High School in Pécs – although it was opened with Roma children in mind, making it largely segregated, it provides mentoring, mental health support, family engagement and academic tutoring and is explicitly designed to mitigate educational disadvantages and promote progression to higher education, which are interventions and resources usually inaccessible to Roma students within the broader education system (Együtt fejlődünk, 2025[97]; Pályázati Portál, 2025[98]; van Driel, 2006[99]).
Ireland’s Traveller and Roma communities similarly experience significant disadvantage, including barriers to education and employment. Only 31.4% of Travellers completed the Leaving Certificate in 2022 (Irish Government, 2024[100]), markedly below the national rate of 70.8% in 2020 (OECD, 2023[101]). To tackle these gaps, Ireland has implemented targeted policies and a national strategy:
National Traveller and Roma Inclusion Strategy 2024-2028 (NTRIS III): The goal of the strategy is to build a safe, fair and inclusive Ireland where Travellers and Roma are supported to lead inclusive, healthy and fulfilling lives. In its education pillar, the Strategy allocates EUR 1.25 million in 2024 to establish community link workers to support educational participation among Travellers and Roma at risk of exclusion. The Irish government is also looking towards developing a Traveller and Roma Education Strategy, incorporating targeted measures to incremental progress towards the norm in access, retention and progression (Eurydice, 2024[102]; Irish Government, 2024[100]).
Delivering Equality of Opportunity in Schools (DEIS): Targets schools within disadvantaged communities, providing resources such as additional learning support to facilitate smaller class sizes in primary schools, priority access to psychological services and free school meals in secondary schools. DEIS includes the Home School Community Liaison scheme, which provides for a teacher to be released from teaching duties in order to work intensively with and support parents or guardians through home visits, school-based classes for parents or guardians, and connecting them with other relevant resources and community initiatives (Tusla, 2025[103]). DEIS aims to raise educational attainment and school completion among Traveller and Roma students by undertaking the following: collaborating with the Child and Family Agency (Tusla) and Traveller Representative Groups on measures to improve Traveller engagement with education; re-evaluating current Traveller-specific resources in the context of outcomes and experiences; and developing best practices and innovative measures to support traveller attendance, participation and retention in a pilot context of the School Excellence Fund (Irish Government, 2017[104]). Some evaluations, even though not causal in inference, indicate narrowing achievement gaps between DEIS and non-DEIS schools. (OECD, 2024[105]).
Social Inclusion and Community Activation Programme 2024-2028 (SICAP): Delivered locally through partnerships with disadvantaged groups, SICAP specifically targets Travellers, Roma, migrants and refugees. It supports initiatives such as employment projects within the Traveller economy and language skills for Roma students, supporting their integration into education and the labour market (Pobal, 2025[106]; 2021[107]).
Prior sections have shown that when leaving formal education, there are differences in the qualifications obtained between men and women and between adults with socio-economically advantaged and disadvantaged backgrounds. In particular, differences in educational trajectories are reflected in large disparities in information-processing skills. However, educational qualifications do not fully explain socio-economic disparities in skills.
This section considers whether differences in fields of study nurture distinct 21st-century skills and offer varying opportunities for individuals to practice them in the long term, for example by facilitating entrance into different labour-market opportunities. It also considers if people with different skills have a different propensity to choose different fields. The section first documents the field-of-study profile of adults aged 30-65 by educational programme, where differentiation is prevalent (upper and post-secondary qualification with a vocational orientation, short-cycle tertiary education, and bachelor’s degree or equivalent and above). Results are also reported by gender and parental educational attainment. Results relating to field of study by parental occupation, immigrant status and childhood residential context are available in Annex Table 3.A.2. The section then considers to what extent patterns of field of study account for the disparities in skills between different socio-demographic groups, after accounting for the differences in educational attainment.
3.5.1. Field-of-study profile of adults by educational programme
Fields of study differ for adults who have obtained an upper or post-secondary degree with vocational orientation, short-cycle tertiary education, and those with a bachelor’s degree or more advanced qualifications (Figure 3.10). In particular, findings from the 2023 Survey of Adult Skills show that technical fields of study such as engineering and manufacturing; construction; agriculture, forestry, fisheries and environment; and personal and community services tend to be especially popular fields among adults who have completed vocationally oriented upper secondary qualifications. By contrast, among adults who have completed advanced academic qualifications, professional fields such as law; social and behavioural sciences; education and teacher training; humanities, languages and arts; and natural sciences, mathematics and statistics are especially popular (OECD, 2024[16]).
Among vocational graduates, the largest shares pursued qualifications in economic, business and administration, and engineering and manufacturing (20% each), followed by personal and community services (14%), construction (11%), and health-related programmes (8%). Among those with short-cycle tertiary education, the largest shares pursue economics, business and administration (25%), engineering and manufacturing (13%) and health (12%). At the bachelor’s level or equivalent and above, economics, business and administration is the single largest field of study (21%), followed by engineering and manufacturing (10%), and education and teacher training (12%), and health (11%). Up to 36% of adults with a vocational qualification pursued STEM2 qualifications, compared to 17% of students with a short-cycle tertiary education and 26% of adults with a bachelor’s degree or equivalent and above (OECD, 2024[16]).
Percentage of adults by field of study and highest educational attainment, OECD average
Note: Adults aged 30-65. See the note for Figure 3.7 for a description of respondents’ educational attainment. Country-level results are provided in Annex Table 3.A.2.
Fields of study are ranked in descending order based on the percentage of adults with upper and post-secondary (vocational orientation). Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Disparities related to gender relate not only to what level people study, but also what they study. Although women are more likely to obtain tertiary educational qualifications, they are considerably less likely than men to attend courses with a strong mathematics orientation. For example, among those who have obtained bachelor’s degrees and above, women are over-represented in pursuing degrees with a focus on education (10 percentage point difference, Panel C in Figure 3.11), whereas men are over-represented in pursuing degrees with a focus on engineering (15 percentage point difference, Panel C in Figure 3.11.
Among those who have obtained an upper or post-secondary degree with a vocational orientation, women are especially likely to have completed qualifications in economics, business and administration (31%); personal and community services (22%); and health (15%) (Panel A in Figure 3.11). By contrast, men are especially likely to have completed qualifications in engineering and manufacturing (35%); construction (19%); and economics, business and administration (10%). Among those who have obtained a short-cycle tertiary education, women are especially likely to have qualifications in economics, business and administration (29%); health (19%); and education and teacher training (11%). For men, engineering and manufacturing (25%); economics, business and administration (19%); and ICT (12%) are the most common fields of study (Panel B in Figure 3.11).
Among those who have obtained bachelor’s degrees and above, women are most heavily represented in economics, business and administration (20%); education and teacher training (17%); and health (15%); whereas men are especially likely to have obtained degrees in economics, business and administration (21%); engineering and manufacturing (19%); and ICT (11%) (Panel C in Figure 3.11). Box 3.8 provides insights into how some OECD countries have succeeded in promoting the participation of men and women in fields in which they are underrepresented.
Germany
Germany addresses stereotypes in early childhood education through teacher training and grants to men entering the field. It also supports STEM-based extracurricular activities and campaigns to raise girls’ awareness of STEM. For example, Girls’ Day – Girls’ Future Day takes place once a year and is sponsored by the Federal Ministry for Family Affairs, Senior Citizens, Women and Youth and the Federal Ministry of Education and Research. On this nationwide day of action, schoolgirls from fifth grade (age 10/11) onwards are given an insight into professions and courses of study in which women have thus far been under-represented. Several ministries also support Klischeefrei, which is a coalition of over 600 members from various sectors advocating for gender-neutral career and study choices. The coalition provide resources, networking and support through a service centre and portal. Other ministries contribute through a range of policies, such as grants to encourage women to enter trades (Ministry of Labour), investments in childcare in universities (Education and Research), grants to support later life transitions to in-demand occupations (Labour), investments in campus safety (Justice), grants to increase innovation in gender-specific safety equipment and standards (Labour), measures to increase women’s representation in academic leadership positions in STEM (Education and Research), interventions with employers to ensure women have better access to STEM and innovation careers (Education and Research), and programmes to support women’s networks in men-majority fields (Culture) (OECD, 2025[108]).
Spain
Universitat Politècnica de Catalunya (UPC) is a public research and higher education institution in engineering, architecture, science and technology. UPC has a low percentage of women students (less than 30% of bachelor’s and master’s students), with women also poorly represented across the academic career, falling to the lowest percentage in the full professor category. In 2010, UPC established a plan to reduce structural obstacles for women in the evaluation of professors qualifying for full professorship according to National Agency for Quality Assessment and Accreditation, the public body responsible for carrying out evaluation, certification and accreditation activities of the Spanish university system.
In 2016, women accounted for only 8.6% of full professors. In response, UPC approved an affirmative measure to correct the inequalities and structural obstacles driving women’s under-representation. The Full Professor Programme applies a gender coefficient, which is a correction coefficient in the final scores of women candidates when applying to a professorial positions; if a female and male candidate end on the same final score, the female candidate earns the professorship. The gender coefficient is calculated on the basis of the percentage of women professors who are not full professors, and the goal is to reach the same percentage of women with full professorship.
In 2017, a coefficient of 1.15 was applied, which has since been maintained or increased, further supporting women’s pathways to full professorship. In 2021, 5% of the full professor positions opened were reserved for women candidates, and the coefficient applied was 1.25. This measure has increased the numbers of women promoted to full professor, with 12.9% of women in this category in 2021. The measure also promotes a greater presence of women in decision-making positions by having more women in the higher categories. Finally, it can impact organisational culture by showcasing the effects of transformative and structural measures to promote gender equality (European Institute for Gender Equality, 2025[109]; Oxford Research AB, 2021[110]).
Norway
In Norway, men accounted for 10% of teachers in early childhood educational development and pre-primary education in 2022 (OECD, 2024[14]). While this still reflects a significant gender gap in the teaching profession, in international comparisons Norway ranks among the countries with the highest share of male teachers. In 1997, Norway began increasing efforts to raise the number of men working in childcare, for example through positive discrimination policies (Norwegian Government, 2009[111]; Nordic Information on Gender, 2018[112]). According to regulations, men may receive preferential treatment when filling positions in teaching or childcare – provided they are equally qualified as female applicants. The right for affirmative action for men was a form of special measure aimed at increasing male participation in ECEC, schools and child protection services, and to achieve gender equality in the long term. Further efforts have been made to attract young men to the childcare profession, with municipalities encouraged to invite boys in lower secondary school to gain experience working in childcare (Nordic Information on Gender, 2018[112]; Lundgaardsløkka, 2025[113]; Norsk Rikskringkasting, 2014[114]).
Germany
In Germany, the federal government supports initiatives to attract more men into the nursing profession. Between 2020 and 2023, the share of male trainees in nursing professions with newly concluded training contracts increased from 24% to 27% (Statistisches Bundesamt, 2025[115]). The nationwide “Boys’ Day” is a school-based career orientation project that allows boys from fifth grade onwards to explore professions in which men are the minority during a one-day visit to companies or institutions in the social, educational and caregiving sectors (German Government, 2024[116]; Boys' Day, 2025[117]). Between 2021 and 2023, the project “Modern Men Do Care” (funded by the Federal Ministry of Health [Bundesministerium für Gesundheit]), aimed to understand the underlying reasons for the low share of men in the healthcare sector and develop practical measures for institutions to attract more men to the nursing profession (Pflege-Netzwerk Deutschland, 2023[118]; Gesundheitswirtschaft Nordwest, 2023[119]; n.d.[120]).
United Kingdom
The National Health Service (NHS) in England launched a GBP 8 million campaign in 2018 to recruit nurses (NHS England, 2018[121]). The campaign highlighted the vast range of opportunities available in the NHS for potential new recruits through advertising across TV, radio, posters, digital and social media. In addition, the NHS worked with nursing and midwifery ambassadors who helped change the perceptions of these professions to help parents, teachers and young people see nursing and midwifery as careers of choice. The research accompanying the campaign found that only four in ten parents said they would be proud of their son becoming a nurse. The year following the campaign's launch not only saw a surge in male applicants but also a reversal of the overall trend of declining nursing applications observed in recent years (NHS England, 2019[122]).
Australia
In 2020, the Australian College of Nursing (ACN) strengthened efforts to recruit and retain more men in the nursing profession. As part of their campaign, the Men in Nursing ebook was created, featuring a powerful collection of stories of 28 men who outline their experience in nursing (Australian College of Nursing, 2020[123]). As part of their efforts, the ACN established The Australian College of Nursing Men in Nursing Working Party, a group dedicated to improving the nursing workforce to allow for the greater retention and recruitment of men (Australian College of Nursing, 2020[123]).
Differences in field-of-study choices related to whether parents are tertiary educated are less pronounced, suggesting that disparities related to socio-economic background lie primarily in to what level people study, not what they study (Figure 3.12). For those who have obtained a short-cycle tertiary, adults without tertiary-educated parents are marginally over-represented among those who have obtained a degree in economics, business and administration, and engineering and manufacturing (around 3.6 percentage point difference in both cases), and are marginally under-represented among those who have obtained a degree oriented towards for example humanities, languages and arts (3.5 percentage point difference), personal and community services (1.8 percentage point difference), education and teacher training (1.0 percentage point difference), and ICT (0.9 percentage point difference). As will be explored in Chapter 4, there are different returns to working in STEM-oriented and humanities-education fields, meaning that these results suggest that individuals with socio-economically disadvantaged backgrounds are marginally more likely to pursue degrees with tangible individual economic returns than those from more advantaged backgrounds.
Percentage of women and men, by field of study and highest educational attainment, OECD average
Note: Adults aged 30-65. See the note for Figure 3.7 for a description of respondents’ educational attainment. * Next to field of study indicate that disparities across groups (women vs. men) are significant at the 5% level. Country-level results are provided in Annex Table 3.A.2.
Field of study for all panels are ranked in descending order based on the percentage of women with upper and post-secondary (vocational orientation).
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Percentage of respondents with tertiary and non-tertiary educated parents, by field of study and highest educational attainment, OECD average
Note: Adults aged 30-65. See the note for Figure 3.7 for a description of respondents’ educational attainment. See Note 6 in Chapter 1 for the definition of parental education groups. Country-level results are provided in Annex Table 3.A.2.
Field of study for all panels are ranked in descending order based on the percentage of respondents with non-tertiary educated parents with upper and post-secondary (vocational orientation) in Panel A.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
3.5.2. Linking field-of-study patterns with skills differences across socio-demographic groups
Field‑of‑study choices explain around a quarter of differences in numeracy skills between men and women and around half of gender differences in adaptive problem solving, both of which are in favour of men. At the same time, gender differences in numeracy in favour of men remain pronounced even when comparing men and women with similar levels of educational attainment and similar fields of study (Table 3.4).
By contrast, field-of-study choices explain little of the differences in 21st-century skills between different groups with similar levels of educational qualifications. This is consistent with the fact that there are few differences in field of study between young and mature adults, between adults with and without parents who obtained a tertiary educational qualification, between adults with and without parents who worked in high-status occupations, between individuals with and without an immigrant background, and between individuals who grew up in cities, towns or villages.
However, adults who have completed similar educational qualifications but in different fields do have different levels of skills proficiency (Table 3.4). Adults who have completed STEM courses have higher numeracy and higher literacy and adaptive problem solving skills than those who have completed education and teacher training courses, health courses, or courses in other fields. For example, adults who have completed STEM courses have levels of numeracy proficiency that are significantly higher than those of adults who have completed education and teacher training qualifications (0.26 SD difference) and health qualifications (0.20 SD difference) (Table 3.4). Although differences are largest for numeracy, adults who have completed STEM courses also have levels of adaptive problem-solving proficiency that are significantly higher than those of adults who have completed education and teacher training qualifications (0.21 SD difference) and health qualifications (0.16 SD difference). Similarly, they have levels of literacy proficiency that are significantly higher than those of adults who have completed education and teacher training qualifications (0.13 SD difference) and health qualifications (0.10 SD difference). These differences could reflect either the content of different field-of-study programmes, differences in the occupational trajectories of individuals who studied different fields, or differences in selection into different educational programmes. These patterns are consistent with evidence that extended exposure to mathematics and scientific training can sharpen general analytic reasoning (Attridge and Inglis, 2013[124]), and that reading technical texts or engaging in scientific reasoning can transfer beyond the narrow quantitative content of STEM courses.
Regression coefficients before and after adjusting for respondents’ educational attainment and field of study, OECD average
Note: Adults aged 30-65. The field-of-study “other” category contains the fields: economics, business and administration; law; welfare; social and behavioural sciences; journalism and information; agriculture, forestry, fisheries and environmental studies; personal and community services; security and transport; humanities, languages and arts; and no main area of study or emphasis. Coefficients in bold are statistically significant at the 5% level. Results in columns (1), (4) and (7) explain the respective skill while adjusting for differences in gender, age, parental education, parental occupation, childhood residential context and immigrant background. Results in columns (2), (5) and (8) additionally adjust for differences in respondents’ educational attainment. Results in columns (3), (6) and (9) further adjust for differences in respondents’ field of study. Estimation results in columns (1) to (9) are restricted to the same number of observations. See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Standard errors are provided in Annex Table 3.A.2. STEM fields comprise information and communication technologies, natural sciences, mathematics and statistics, engineering and manufacturing, and construction.
*Differences by immigrant background vary greatly across countries because of differences in the size of immigrant groups as well as composition and context. Readers are therefore encouraged to consult country-specific results in Annex Table 3.A.2.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
The association between socio-demographic characteristics and numeracy skills before and after accounting for adults’ educational attainment and field of study differs across countries. Figure 3.13 shows the OECD average and highlights the countries with the largest and smallest disparity in the basic model that does not control for educational attainment and field of study. For example, in Switzerland the gender gap in numeracy is large compared to the OECD average (0.32 SD), and approximately half of this gap can be explained by differences in educational attainment and field of study between men and women (after accounting for these factors the gap is 0.17 SD). By contrast, in New Zealand, the gender gap in numeracy skills is aligned with the OECD average and corresponds to 0.19 SD before and after controlling for educational attainment and field of study.
In New Zealand, the difference in numeracy skills between adults with and without tertiary-educated parents is 0.47 SD before and 0.22 SD after accounting for differences between the two groups in educational attainment and field of study, suggesting that around 53% of the parental education gap is mediated by educational attainment and field of study. Similarly, in Chile, the difference in numeracy skills between adults who grew up in cities rather than villages is 0.50 SD before and 0.23 SD after accounting for differences between the two groups in educational attainment and field of study, suggesting that around 54% of the urban–rural numeracy gap is mediated by educational attainment and field of study. By contrast, in Germany, the difference in numeracy skills between adults who grew up in cities and those who grew up in villages is -0.05 SD before and ‑0.05 SD after accounting for differences between the two groups in educational attainment and field of study, suggesting that educational attainment and field of study do not mediate urban–rural skills disparities.
Results show that fields of study are associated with skills proficiency, which underpins the importance of adults’ educational choices. Box 3.9 highlights differences in career guidance and counselling services between students of different socio-economic backgrounds as well as initiatives to address these shortcomings.
Change in numeracy (std.) by socio-demographic group: basic and fully adjusted country-specific regression coefficients
Note: Adults aged 30-65. The figure shows the difference in regression coefficients for numeracy between the basic adjusted model (Table 3.4 Model (4)) which accounts for differences in socio-demographic characteristics (gender, age, parental education, parental occupation, childhood residential context, immigrant background) and the fully adjusted model (Table 3.4, Model (6)), which additionally accounts for differences in respondents’ educational attainment and field of study. See the note for Figure 3.7 for the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. For immigrant background, the figure only indicates a country if at least 200 adults for each group are part of the final PIAAC sample.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Field of study may have some impact on the development of information-processing skills, such as numeracy and problem solving, throughout people’s lives. Beginning at the secondary school level, the types of qualifications, teaching and learning methodologies, and extracurricular opportunities associated with different fields of study diverge. Even if students are studying within the same educational system and towards a similar qualification, their field of study may impact their entire educational experience – including the skills and proficiency levels with which they enter the labour market.
The effects of field-of-study choices add up over time, administrative barriers notwithstanding, individuals may find it difficult to choose or change field of study as a result of the skills they have, or have not, built up over time. Career guidance and counselling services throughout formal education and beyond can help individuals make decisions about their field of study and understand the skills and opportunities associated with different pathways.
Students with low socio-economic status (SES) backgrounds in the formal education system
Occupational stratification by SES backgrounds may be worsened by the availability of career guidance services in the formal education system. The 2024 OECD report Challenging Social Inequality Through Career Guidance: Insights from International Data and Practice (OECD, 2024[125]) highlights a range of issues related to career guidance and field-of-study selection for students with low SES backgrounds:
Across the OECD, students in schools with a higher proportion of disadvantaged students have less access to career guidance counsellors than students in schools with higher proportions of advantaged students. However, disadvantaged students rely more on schools for career guidance than their advantaged peers.
Students with low SES backgrounds are less likely to participate in any career development activity than their peers with high SES backgrounds, including both in and out of school.
Students with low SES backgrounds often express misaligned career and education goals
Higher-performing students with low SES backgrounds are less likely to expect to work in high-skilled jobs than their peers with high SES backgrounds; they are also less likely to expect to complete tertiary education.
The report highlights a range of initiatives that have been implemented across OECD countries to address these shortcomings in the provision of career guidance and counselling. These include:
Preferential funding for schools with high concentrations of students with low SES background, such as the Delivering Equality of Opportunity in Schools in Ireland, which provides extra funding for eligible schools to provide career guidance activities.
Leveraging the existing social capital of institutions, such as the UK’s Inspiring the Future and Speakers for Schools programme that connects schools with volunteers from a range of jobs and backgrounds who go into schools to share their story with children. The programme works exclusively with public institutions to try to democratise access to hard-to-reach networks, information and, ultimately, career pathways.
Adulthood
Challenges to accessing career guidance services persist into adulthood, particularly for individuals with low SES backgrounds, as services that enable them to gain information about reskilling or career changing opportunities can be non-existent, exclusive to certain workplaces, or private and expensive.
Since 2021, the Australian state of Victoria’s Jobs Victoria Career Counsellors Service (JVCCS) has been offering free, personalised career guidance to all adults, regardless of employment status. The services are delivered by the Career Education Association of Victoria (CEAV) and staffed by around 35 professionally endorsed practitioners through the Career Industry Council of Australia (CICA), including specialists for Aboriginal communities and people with disabilities. The service is accessible via face-to-face, phone or video sessions. CEAV counsellors are highly qualified and must maintain professional standards through ongoing development. They are also based in Skills and Jobs Centres within Technical and Further Education (TAFE) institutions, which offer integrated support on training, employment, welfare referrals, financial advice, skills recognition and labour-market trends, while engaging with local industries (Victoria State Government, 2025[126]).
Against a backdrop of rapid technological change, demographic shifts and evolving patterns of work, adult education has become a critical lever for sustaining employability and inclusive growth. The skills that adults draw upon in the labour market – information‑processing skills such as literacy, numeracy and adaptive problem solving, alongside social and emotional skills – are shaped not only by initial schooling but also by individuals’ life‑course experiences and opportunities to learn. Skills disparities, therefore, reflect not only disparities in educational pathways but also disparities that occur in learning outside the formal education system. For example, occupational placement determines the access adults have to learning‑rich environments at work and the barriers they face in engaging with learning opportunities.
For most adults, returning to formal education is constrained by tuition costs, foregone earnings and practical considerations such as childcare or shift‑work schedules. As a result, non‑formal learning – short, job‑related courses delivered in the workplace, online or by specialised providers – often constitutes the most viable route for a person to acquire new skills or update skills needed for their current position or for a transition to a new role. Such provision is typically demand‑driven, narrowly targeted and employer‑sponsored, aiming to increase task‑specific know‑how rather than to strengthen broad foundation or social and emotional skills. The extent to which different groups engage in, benefit from or encounter barriers to this training therefore warrants close scrutiny.
This section considers non-formal learning opportunities, which include intentional, institutionalised learning of short duration that is not formally recognised by relevant authorities (e.g. short courses or workshops) but that may award “alternative credentials” such as digital badges, micro-credentials, and professional or industrial certificates3 (OECD, 2025[127]). Drawing on data from the 2023 Survey of Adult Skills, the analysis in this section reflects the types of non‑formal learning in which men and women and adults from varied backgrounds participate, identifying disparities in participation, learning goals and stated reasons for attendance. It further examines self‑reported obstacles – time, cost, lack of employer support – to determine whether barriers are group‑specific and whether the training undertaken genuinely equips disadvantaged adults for future labour‑market requirements or merely consolidates existing roles.
By identifying who participates in training activities, what they study, why they enrol, where demand remains unmet and which obstacles prevent them from participating or from participating more, this section pinpoints the levers – such as financial incentives, targeted guidance, flexible scheduling and tailored programme design – through which policymakers can change uneven engagement in learning into more inclusive and effective skills development opportunities for adults.
3.6.1. Disparities in the uptake of non‑formal education and training activities
On average across OECD countries, 43% of adults report having participated in some form of non‑formal education and training in the 12 months preceding the interview (including both job-related and non-job-related) (Annex Table 3.A.3). However, there are large disparities in non-formal training participation across socio‑demographic groups, with non‑formal learning opportunities unevenly distributed, and the very adults often considered in need of upskilling and reskilling to improve their labour‑market prospects highly under‑represented (Figure 3.14).
Participation in non-formal learning is higher among younger adults, those with socio‑economically advantaged backgrounds, adults with higher educational attainment and individuals employed in high‑status occupations and with permanent employment contracts. Participation gaps between men and women and between adults who grew up in cities and those who grew up in villages are modest, and individuals with and without an immigrant background also report similar participation levels. Adults in high-status occupations report the highest incidence of training (62%), whereas adults with below upper secondary education report the lowest (19%). Educational attainment exhibits the largest gap, with 61% of those holding a bachelor’s degree or higher reporting participation in training in the past 12 months compared to 19% of those with below upper secondary education, a difference of around 42 percentage points. Adults who have an indefinite (permanent) work are also considerably more likely to participate non-formal learning compared to those with fixed term and seasonal contracts, temporary contracts with an agency, zero hour contracts, or contractor/freelance or consultant contracts (52% vs. 45%).
Participation differs by age, with 49% of adults aged 30-49 reporting to have participated in non-formal learning compared with 35% of those aged 50-65. More than half of adults whose parents were tertiary educated or employed in high‑status occupations participate in training (55% and 52%, respectively). The corresponding shares fall to 40% and 38% among adults whose parents lacked tertiary education or had low‑status jobs. Adults who grew up in cities are more likely to participate than those raised in towns and villages (45%, 44% and 40%, respectively). Gender differences are small, with a 2 percentage point advantage for women (44% vs. 42%).
Share of adults participating in non-formal adult education and training in the past 12 months, OECD average
Note: Adults aged 30-65. Survey question used to measure non-formal adult education and training participation: “During the last 12 months, that is since [Interview date], have you participated in any training activity? Include any training activity even if it lasted for only one hour. Please also include training activities that are still ongoing.” Darker colours denote statistically significant differences (at the 5% level) in the percentage of non-formal adult education and training activities between groups: 30-49 vs. 50‑65, men vs. women, non-tertiary vs. tertiary (parental education), low-status vs. high-status (parental occupation), children of immigrants vs. non-immigrant, village vs. city, bachelor's or equivalent and above vs. below upper secondary (respondents’ education), low-status vs. high-status (respondents’ occupation), permanent and non-permanent contract (only for those who are employed). See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Differences by immigrant background vary greatly across countries because of differences in the size of immigrant groups as well as composition and context. Readers are therefore encouraged to consult country-specific results in Annex Table 3.A.3.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Box 3.10 highlights initiatives and interventions aimed at promoting engagement in different forms of learning throughout the life course, with a particular focus on opportunities and infrastructures that enable individuals to acquire and exercise their skills.
Lifelong learning through different modes and in different settings can promote the uptake of core 21st‑century skills, as well as a range of social and emotional skills that can help individuals achieve better educational and labour market outcomes. From psychosocial support in ECEC, providing spaces and platforms for learning in young adulthood, to funding access to digital tools, policymakers can pull a range of levers to promote learning throughout the life course.
From childhood to young adulthood
In Chile, the Life Skills Program (Programa Habilidades para la Vida, HPV) supports students from pre-kindergarten to 12th grade by fostering emotional, social and cognitive development. Delivered in public and subsidised private schools, it targets psychosocial risks across three stages: 1) early years (pre-kindergarten to 4th grade), with a focus on emotional regulation and adaptation; 2) middle school (5th to 8th grade), with a focus on school coexistence and psychosocial well-being; and 3) upper grades (7th to 12th grade), with targeted support for students at risk of psychosocial and academic challenges. In crisis contexts, “Temporary Psychosocial Support Devices” provide tailored actions to restore school operations and create spaces that foster the socio-emotional well-being of students affected (Chile Atiende, 2025[128]).
In Spain, the Programme for Educational Guidance, Advancement and Enrichment (Programa para la orientación, avance y enriquecimiento educativo, PROA+) initiative aims to reduce school dropout and low performance by focusing on the digitalisation and modernisation of educational infrastructure from early childhood education to higher education, including the provision of digital tools and complementary teacher training. Schools commit to responsive service provision through agreements with educational authorities that outline the shared objectives and define specific actions and resources needed to achieve the objectives. Key goals may include strengthening personalised learning, improving educational outcomes and preventing dropout among students facing personal or social obstacles, particularly in rural areas or those with a significant number of students in a situation of educational vulnerability. The initiative is co-funded by the EU and the Spanish government; between 2021 and 2024, PROA+ was allocated EUR 360 million and covered 1 million children and young people across 3 600 schools and colleges (Spanish Government, 2025[129]; Spanish Government, 2025[130]; European Commission, 2025[131]; Comunidad de Madrid, 2025[132]; Spanish Government, 2024[133]).
Young adulthood and beyond
In France, the Personal Training Account (compte personnel de formation, CPF) allows individuals aged 16+ to access career training throughout their working life. Primarily accessible via a secure online platform, CPF funds a range of certified courses, including upskilling, entrepreneurship and driving licences, with support available through employers or local employment offices if needed. The programme remains valid until retirement and ensures flexibility across job changes (Perez and Vourc’h, 2020[134]).
In Mexico, The UTOPÍAS initiative in Iztapalapa, Mexico City, aims to transform neglected urban spaces into free-access centres offering educational, cultural, health and recreational services. Designed with community input, the initiative aims to foster social inclusion and youth engagement in order to tackle a range of social issues, including socio-economic disparities. With 12 sites developed in four years, the programme has the ability to reach 100 000 weekly users (Urban Sustainability Exchange, 2025[135]).
Seniors and individuals with disabilities
In the Slovak Republic, the Digital Seniors project provides digital training to over 100 000 seniors and people with disabilities. Participants learn to use IT tools and e-services, helping address digital divides worsened by the COVID-19 pandemic. Graduates receive tablets and data plans to support ongoing learning. The project is part of the Slovak Republic’s Recovery and Resilience Plan (Ministry of Investments, Regional Development and Informatization of the Slovak Republic, 2025[136]; OECD, 2025[137]).
The association between participation in non‑formal adult education and training and a range of socio‑demographic characteristics before and after adjusting for adults’ education, occupation and social and emotional skills differs across countries, as highlighted in Figure 3.15. Prior to adjustment, on average across OECD countries, men are 6 percentage points less likely than women to participate in non-formal learning. Adults aged 50-65 have a participation rate that is 8 percentage points lower than 30‑49 year‑olds. Children of immigrants have a participation rate that is 0.7 percentage points lower than adults with native‑born parents. By contrast, adults who grew up in cities are 2 percentage points more likely to engage with non-formal learning than those raised in villages, and individuals with socio‑economically advantaged backgrounds enjoy a 7 percentage point advantage in non-formal learning participation over those whose parents did not have tertiary education or who worked in low‑status occupations.
Once adults’ own qualifications and job status are taken into account, many of these gaps narrow markedly, with the age‑related shortfall decreasing from 8 percentage points to 5 percentage points, and the socio‑economic advantage shrinking to 1 percentage points for adults with tertiary‑educated parents and for those from high‑status occupational backgrounds. The gender difference also narrows slightly, with the participation deficit for men decreasing from 6 percentage points to 3 percentage points. There are no differences by immigrant background and childhood residential context. Further adjusting for differences in social and emotional skills leads to a small reduction in the explained disparities related to gender and socio-economic background, while the difference remains constant for age and immigrant background. The participation deficit of those who grew up in villages slightly increases from 0.3 to 0.6 percentage points.
These findings indicate that most observed disparities in adult learning participation are not inherent to socio‑economic background but largely arise through the educational and occupational channels a person enters, which are in turn shaped by their background. Individuals from less advantaged families, rural areas or immigrant origins enter adulthood with lower qualification levels and are concentrated in lower‑status jobs. These two factors together almost fully account for their lower subsequent engagement in non‑formal training, and hence for reduced opportunities to upgrade skills and progress later in life. Conversely, once education and job status are held constant, background‑related gaps shrink to negligible levels, underscoring that early disparities in attainment set in motion a cumulative process: initial credentials determine occupational placement, and, together, education and occupation shape access to further learning that could support mid‑career mobility. Remaining differences in participation could reflect institutional and workplace factors that shape access to training, such as employment stability as detailed in Figure 3.14. These factors can create unequal opportunities for skill development among workers with similar educational and occupational profiles.
Figure 3.15. highlights the countries for which educational attainment and occupation mediate the most and the least in terms of disparities in participation in non-formal adult education and training activities. For example, in Lithuania the gender gap in participation is large compared to the OECD average (18 percentage points), and almost half of this gap can be explained by differences in educational attainment and field of study between men and women (after accounting for these factors the gap is 10 percentage points). By contrast, in the Israel, the gender gap in participation is aligned with the OECD average and corresponds to 6 percentage points before controlling for educational attainment and occupation.
In Italy, the difference in participation between adults with and without tertiary-educated parents is 21 percentage points before and 10 percentage points after accounting for differences between the two groups in educational attainment and occupation, suggesting that around 52% of the parental education gap is mediated by educational attainment and occupation. By contrast, in Sweden, the difference in participation between adults with and without tertiary-educated parents is small and not explained by differences between the two groups in educational attainment and occupation.
In Chile, the difference in participation between adults who grew up in cities rather than villages is 11 percentage points before and 2 percentage points after accounting for differences between the two groups in educational attainment and occupation, suggesting that 82% of the urban–rural numeracy gap is mediated by educational attainment and occupation.
Participation in adult education and training is positively associated with literacy, numeracy and adaptive problem solving skills. Adults who participated in education and training have significantly higher levels of literacy than adults who have not participated in education and training (0.23 SD), with the same pattern observed for numeracy (0.21 SD) and adaptive problem solving (0.23 SD) (Table 3.5). However, participation in adult education and training does not explain differences in 21st-century skills between different groups. For example, differences in numeracy between adults who grew up in cities and villages are 0.11 SD after accounting for other socio-demographic characteristics and 0.04 SD after additionally accounting for educational attainment. However, the difference of 0.04 SD remains stable after additionally accounting for participation in adult training. This pattern is similar for the remaining socio-demographic characteristics.
Change in probability of participation in non-formal adult education by socio-demographic group: basic and fully adjusted country-specific regression coefficients
Note: Adults aged 30-65. The figure shows the difference in regression coefficients for participation in non-formal adult education between the basic adjusted model, which accounts for differences in socio-demographic characteristics (gender, age, parental education, parental occupation, childhood residential context, immigrant background), and the fully adjusted model, which additionally accounts for differences in respondents’ educational attainment, occupation. Survey question used to measure non-formal adult education and training participation: “During the last 12 months, that is since [Interview date], have you participated in any training activity? Include any training activity even if it lasted for only one hour. Please also include training activities that are still ongoing.” See the note for Figure 3.7 for the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. For immigrant background, the figure only indicates a country if in the country at least 200 adults for each group are part of the final PIAAC sample. For immigrant background, the figure only indicates a country if at least 200 adults for each group are part of the final PIAAC sample.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Regression coefficients before and after adjusting for respondents’ educational attainment and participation in adult education and training, OECD average
Note: Adults aged 30-65. Coefficients in bold are statistically significant at the 5% level. Results in columns (1), (4) and (7) explain the respective skill while adjusting for differences in gender, age, parental education, parental occupation, childhood residential context and immigrant background. Results in columns (2), (5) and (8) additionally adjust for differences in respondents’ educational attainment. Results in columns (3), (6) and (9) further adjust for differences in respondents’ participation in adult education and training. Estimation results in columns (1) to (9) are restricted to the same number of observations. See the note for Figure 3.7 for the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Standard errors are provided in Annex Table 3.A.3.
*Differences by immigrant background vary greatly across countries because of differences in the size of immigrant groups as well as composition and context. Readers are therefore encouraged to consult country-specific results in Annex Table 3.A.3.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
3.6.2. Disparities in the content of the training activities adults enrol in
The focus of the training activities participants enrol in vary depending on socio-demographic characteristic. On average, the most frequently cited principal focus areas for participants in non-formal learning were “security” (17%); “computer or software skills” (13%); “handling customers, clients, patients or students” (9%); “team‑working or leadership skills” (8%); and “operating machinery or equipment” (7%). Smaller proportions reported include “foreign language”, and “project‑management or organisational skills” (6% each), “communication and presentation skills”, and “creative or musical skills” (4% each), and “sports” (3%). Training devoted to core information‑processing skills – “skills involving numbers, calculating skills” (1.5%) and “reading and writing skills” (0.9%) – was rarer, while one in five adults selected “other focus” Annex Table 3.A.3.
Figure 3.16 illustrates relative representation ratios by socio‑demographic characteristic, highlighting where specific groups are over‑ or under‑represented in each training category. Age‑related disparities in participation are limited, with 30-49 year-olds slightly over‑represented in courses on “skills involving numbers, calculating skills” and “foreign language”; age differences are negligible for most other topics. Gender disparities are more pronounced, with women over‑represented in training on “creative or musical skills”, “reading and writing skills”, “handling customers, clients, patients or students”, “communication and presentation skills” and “foreign language”. Men are over‑represented in “operating machinery or equipment”, “security”, “computer or software skills” and “project‑management or organisational skills”.
Adults with socio-economically disadvantaged backgrounds are over‑represented among those who indicate they engaged in training focusing on “operating machinery or equipment” and “security”, whereas those with socio-economically advantaged backgrounds gravitate towards “project‑management or organisational skills”, “foreign language” and “skills involving numbers, calculating skills”. Children of immigrants are over‑represented in “foreign language”, and adults who grew up in villages are more likely to pursue “operating machinery or equipment” and “security”, while their urban counterparts favour “foreign language”. Adults’ own occupational status and educational attainment reinforce these tendencies, with adults in low‑status occupations or who have not studied beyond upper secondary qualifications over‑represented in “operating machinery or equipment” and “security”. Conversely, individuals in high‑status occupations or with tertiary education are disproportionately enrolled in “project‑management or organisational skills”, “team‑working or leadership skills”, “communication and presentation skills”, “reading and writing skills”, “skills involving numbers, calculating skills” and “sports”.
For each socio-demographic characteristic, the relative representation ratio (RR) is presented. RR > 1 (< 1) indicates a higher (lower) likelihood that members of the underlined group cite a specific focus, OECD average
Note: Adults aged 30-65. The relative representation ratios refer only to participants in non-formal adult education and training activities in the past 12 months. Survey question used to measure training focus: “What was the main focus of this training activity? Please name only one.” See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Country-level results are provided in Annex Table 3.A.3.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Disparities in the focus of training risk entrenching existing disparities in occupational placement rather than promoting upward labour-market mobility through upskilling and reskilling. Beyond differences in participation, qualitative aspects of training (such as its relevance, content and duration) also matter, as disadvantaged groups are often offered narrowly focused on compliance-oriented learning that does little to enhance mobility. Further examining this segmentation of learning opportunities through complementary qualitative evidence would help shed light on how training quality and quantity shape disparities in skill development. Courses focused on security tasks or the operation of machinery tend to enhance task‑specific know‑how that is applicable to occupations dominated by men, residents of rural areas, and workers with lower educational qualifications and working in low-status occupations. These workers are often the children of low-educated parents who worked themselves in low-status occupations, reflecting the intergenerational occupational transmission of disadvantage. By contrast, training for skills that frequently serve as gateways to higher‑status, technology‑intensive and internationally oriented roles (for example project management, advanced digital competencies, training involving numbers and calculations, training involving writing, training to promote digital skills, or foreign language acquisition) is concentrated among adults with a socio-economically advantaged background. Without corrective action, this pattern may curtail upward mobility for disadvantaged groups and perpetuate gendered and socio‑economic disparities in access to the occupations projected to expand most rapidly in light of the digital and green transitions and demographic shifts.
3.6.3. Disparities in participation in job-related non‑formal education and training
Understanding who undertakes job‑related (as opposed to interest-driven) courses and why they do so is central to designing incentives that raise productivity and foster career mobility. Job‑related training tends to be sponsored or required by employers and therefore exposes where firms invest. Interest‑driven learning, by contrast, signals intrinsic demand that may remain untapped if finance or information are lacking.
Across OECD countries, 84% of adults who participated in non‑formal education and training did so primarily for job‑related reasons, while only 16% reported non‑work motivations Annex Table 3.A.3. Figure 3.17 presents the distribution of work‑related training across socio‑demographic groups. Younger adults, men, individuals whose parents worked in low‑status occupations, individuals without an immigrant background, those who grew up in villages, adults with upper secondary or post‑secondary vocational qualifications, and workers in high‑status occupations are marginally more likely to pursue training for job‑related purposes. However, between‑group differences are modest. For example, 85% of 30-49 year‑olds undertook training for job‑related reasons compared with 82% of 50‑65 year-olds. Similarly, 84% of individuals without an immigrant background, 83% of the children of immigrants and 81% of immigrant adults reported participating in training for job-related reasons, and 84% of adults who grew up in villages or towns engaged in work‑oriented learning compared to 83% of those who grew up in cities. The most pronounced gap is the difference between men and women: 87% of men cited work‑related reasons for participation in training compared to 80% of women – a difference of 7 percentage points.
Adults’ own educational attainment and occupational status show similarly small yet consistent disparities. Among adults with a bachelor’s degree or higher and those with upper secondary or post‑secondary vocational qualifications, 84% and 85%, respectively, pursued training for work‑related reasons, compared with 83% of those with short‑cycle tertiary credentials, 82% of adults with general upper secondary or post‑secondary education, and 80% of those without an upper secondary qualification. Among workers in high‑status occupations, 88% undertook job‑related training compared to 85% of those in low‑status roles.
Share of adults participating in non-formal adult education and training for job-related reasons, OECD average
Note: Adults aged 30-65. Percentages presented in this figure refer only to adults who participated in non-formal adult education and training activities in the past 12 months. Survey question to measure whether or not the adult education and training was job-related: “Was this training activity mainly job-related? ‘Job-related’ can refer to your specific job, but also to improving career and employment chances in general.” Darker colours denote that differences in the percentage of non-formal adult education and training activities between groups (16-29 vs. 50-65, men vs. women, non-tertiary vs. tertiary, low-status vs. high-status, immigrant vs. non-immigrant, village and city) are statistically significant at the 5% level. Villages, towns and cities refer to childhood residential context at the age of 14. See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Country-level results are provided in Annex Table 3.A.3. Differences by immigrant background vary greatly across countries because of differences in the size of immigrant groups as well as composition and context. Readers are therefore encouraged to consult country-specific results in Annex Table 3.A.3.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
The association between participation in job-related non-formal adult education and training activities and a range of socio‑demographic characteristics before and after adjusting for adults’ education and occupation differs across countries. Figure 3.18 shows that men are 7 percentage points more likely than women to pursue job‑related non-formal training courses, while 50‑65 year-olds are 3 percentage points less likely than those aged 30-49 to engage in job‑related non-formal training courses. When comparing individuals with similar levels of educational attainment and working in similar occupations, the gender gap in participation decreases to 5 percentage points, and no quantitatively relevant gaps are observed between other groups. This may be because men‑majority occupations (e.g. skilled trades, security, transport) carry more mandatory or employer‑financed training requirements, whereas women are concentrated in sectors where continuous learning is more often self‑initiated.
Figure 3.18 shows the countries for which educational attainment and occupation mediate the most and least disparities in participation in job-related non-formal adult education and training activities. For example, in Japan, the gender gap in participation is large compared to the OECD average (23 percentage points), and approximately half of this gap can be explained by differences in educational attainment and occupation between men and women (after accounting for these factors, the gap is 15 percentage points). By contrast, in the Slovak Republic, the gender gap in participation is aligned with the OECD average and corresponds to 2 SD before and after controlling for educational attainment and occupation.
In Korea, the difference in job-related education and training participation between adults with and without tertiary-educated parents is -9 percentage points before and -5 percentage points after accounting for differences between the two groups in educational attainment and occupation, suggesting that around 44% of the parental education gap is mediated by educational attainment and field of study. By contrast, in Czechia, differences in participation between adults with and without tertiary-educated parents are negligible. In the United States, the difference in participation between adults who grew up in cities rather than villages is 3 percentage points before and -2 percentage points after accounting for differences between the two groups in educational attainment and occupation.
Change in probability of participating in job-related non-formal adult education by socio-demographic group: basic and fully adjusted country-specific regression coefficients
Note: Adults aged 30-65. The figure shows the difference in regression coefficients for participation in job-related non-formal adult education between the basic adjusted model, which accounts for differences in socio-demographic characteristics (gender, age, parental education, parental occupation, childhood residential context, immigrant background), and the fully adjusted model, which additionally accounts for differences in respondents’ educational attainment and occupation. Survey question to measure whether or not the adult education and training was job-related: “Was this training activity mainly job-related? ‘Job-related’ can refer to your specific job, but also to improving career and employment chances in general. See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. For immigrant background, the figure only indicates a country if at least 200 adults for each group are part of the final PIAAC sample.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
3.6.4. Group‑specific motivations for engagement in job‑related non‑formal learning
Reasons for participating in job-related non-formal training vary depending on socio-demographic characteristic. Among all those who participated in job-related training, the most frequently mentioned reasons were: “to improve my knowledge or skills on a subject that interests me” (23%), “to improve my job or career opportunities” (19%), “to better carry out my regular work tasks” (20%), “I was obliged to participate” (17%), “to better deal with new or changing work tasks” (11%) and to “to obtain or to renew a certificate” (8%) Annex Table 3.A.3.
When breaking these down by socio-demographic group, Figure 3.19 suggests that younger adults prioritise career advancement (1.5 RR), while older adults more often say they were obliged to attend (0.8 RR) – reflecting age‑biased promotion opportunities and mandatory refresher courses. Women are more likely to pursue personal‑interest learning (1.2 RR), whereas men are more likely to focus on certificates (0.7 RR), which could reflect differences in the occupations men and women are employed in, and related differences in occupational licensing regimes, or men’s greater awareness of the importance of obtaining certifications to effectively navigate skills-based hiring markets (OECD, 2025[138]). Adults with socio-economically disadvantaged backgrounds are more likely to participate in job-related non-formal training because they are required to or because it will allow them to gain a certificate, although adults with socio-economically advantaged backgrounds are more likely to participate in non-formal education in general (Figure 3.14) . For example, adults with non-tertiary educated parents are more likely to participate in training to gain a certificate or because they were obliged to participate (1.3 RR each). Adults working in low-status occupations or who did not obtain tertiary education are more likely to participate in training because they were obliged to (1.5 RR and 1.33 RR, respectively) or to obtain or renew a certificate (1.6 RR each), suggesting fewer voluntary opportunities and a need to prove the skills they possess through formal qualifications among prospective employers.
For each socio-demographic characteristic, the relative representation ratio (RR) is presented. A RR > 1 (< 1) indicates a higher (lower) likelihood that members of the underlined group cite a specific reason for participation, OECD average
Note: Adults aged 30-65. Survey question to measure adults’ main reason for participating in adult education and training: “Could you please specify your main reason for participating in this training activity?”. See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Country-level results are provided in Annex Table 3.A.3.
Source: Calculations based on OECD (2017[139]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
3.6.5. Satisfied and unmet demand for non‑formal adult education and training: Overall prevalence and group differences
Adults participating in the 2023 Survey of Adult Skills were asked to report if there were courses they wanted to attend but did not. A “no” from a participant implies satisfied demand, whereas a “no” from a non‑participant may indicate either lack of interest or a low perceived value of training given the existing supply of training opportunities. In other words, adults might have been willing to participate in training but not in the training available, hence their potential demand was unmet.
On average, 68% of participants felt their training needs were met, while 84 % of non‑participants declared no interest or unmet demand, as indicated in Annex Table 3.A.3. Older adults are over-represented among participants who have a satisfied demand, with 71% of those aged 50-65 vs. 66% of those aged 30‑49 who have participated in training indicating not wanting to attend other courses (Figure 3.20). Similarly, 88% of 50-65 year-olds and 80% of 30-49 year-olds who have not participated in training indicated not wanting to attend other courses. Adults with parents who worked in low-status occupations are over-represented among participants with a satisfied demand and an unmet demand. Among those who have participated in training, 71% of adults whose parents worked in low-status occupations and 64% of adults whose parents worked in high-status occupations indicated not wanting to attend other available courses. Similarly, for those who have not participated in training, 85% of adults whose parents worked in low-status occupations and 81% of adults whose parents worked in high-status occupations indicated not wanting to attend other available courses.
Overall, these numbers reflect that lack of interest in available training opportunities or unmet demand is high. Increasing enrolment in adult education and training therefore requires not only addressing potential lack of motivation to participate in non-formal learning, but also providing information about the benefits of training as well as easily accessible opportunities that meet the current training needs of adults.
The association between the extent to which individuals report lacking training activities of interest and a range of socio‑demographic characteristics before and after adjusting for adults’ education and occupation differs across countries. Figure 3.21 indicates that differences in educational attainment and occupational status explain a large part of the variation in whether individuals with different socio-economic backgrounds report lacking training activities of interest; however, they do not explain much of the differences by gender and age. On average across participating OECD countries, and before accounting for adults’ educational attainment and occupation, men have an 8 percentage point higher probability of reporting no training activities of interest, and older adults (aged 50-65) have a 7 percentage point higher probability. Adults with socio-economically advantaged backgrounds have a lower probability (6 percentage points for tertiary-educated parents and 4 percentage points for parents who worked in high-status occupations), children of immigrants have a 3 percentage point lower probability, and those who grew up in cities have a 2 percentage point lower probability.
Some of these differences are explained by adults’ educational attainment and occupation. On average across participating OECD countries, the higher probability of older adults reporting no training activities slightly reduces from 7 to 5 percentage points after accounting for adults’ educational attainment and occupation. For adults with tertiary-educated parents, the lower probability changes from 6 to 3 percentage points, and for adults with parents in high-status occupations, it changes from 4 to 1 percentage point. Adjusting for adults’ educational attainment and occupation also affects differences by gender, immigrant status and childhood residential context.
Figure 3.21 highlights the countries for which educational attainment and occupation mediate disparities in participation in job-related non-formal adult education and training activities the most and the least. For example, in Italy half of the gap in whether men and women report lacking training activities of interest can be explained by differences in educational attainment and occupation between men and women. By contrast, in England (United Kingdom), educational attainment and occupation do not explain the small gender gap in lack of training activities of interest.
In Portugal the difference in whether adults with and without tertiary-educated parents report lacking training activities of interest is -3 percentage points before and 2 percentage points after accounting for differences between the two groups in educational attainment and occupation. By contrast, in France, this difference is ‑5 percentage points before and ‑6 percentage points after accounting for differences between the two groups in educational attainment and occupation. In Latvia, the difference in participation between adults who grew up in cities rather than villages is -1 percentage points before and 4 percentage points after accounting for differences between the two groups in educational attainment and occupation. By contrast, in the United States, the difference in participation between adults who grew up in cities and those who grew up in villages is -2 percentage points before and after accounting for differences between the two groups in educational attainment and occupation, suggesting that educational attainment and occupation do not mediate the small urban–rural disparities in lack of training activities of interest.
Share of training participants (e.g. those with satisfied demand) and share of non-training participants (e.g. those with unmet demand) reporting that there were no (other) courses they wanted to participate in, OECD average
Note: Adults aged 30-65. Survey question: “In the last 12 months, were there any training activities you wanted to participate in but did not?” See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Differences by immigrant background vary greatly across countries because of differences in the size of immigrant groups as well as composition and context. Readers are therefore encouraged to consult country-specific results in Annex Table 3.A.3.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Change in the probability that individuals will have “no training activities that I wanted to participate in”: basic and fully adjusted country-specific regression coefficients
Note: Adults aged 30-65. The figure shows the difference in unmet demand for participation in training activities between the basic adjusted model, which accounts for differences in socio-demographic characteristics (gender, age, parental education, parental occupation, childhood residential context, immigrant background), and the fully adjusted model, which additionally accounts for differences in respondents’ educational attainment and occupation. Survey question: “In the last 12 months, were there any training activities you wanted to participate in but did not?” See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. For immigrant background, the figure only indicates a country if at least 200 adults for each group are part of the final PIAAC sample.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
3.6.6. Disparities in barriers to participation in job-related non‑formal education and training
Barriers to participating in training vary markedly by gender, age, place of residence and socio‑economic background. Each barrier filters out a share of potential learners, limiting aggregate upskilling and perpetuating inequality. Failing to recognise the range of reasons that hinder participation risks misdirecting resources. By identifying precisely which hurdles deter initial participation and which constrain additional or diversified learning, policymakers can design interventions that convert latent demand into effective engagement and progression. The design of effective skills strategies requires identifying the people facing obstacles. This is more than just those who take no training at all, but also involves those who engage in training yet would have preferred to participate more often or in different activities.
For adults participating in the 2023 Survey of Adult Skills who answered “yes” to the question “In the last 12 months, were there any training activities you wanted to participate in but did not?” – signalling interest curtailed by barriers – a follow‑up question was asked about the reasons for non‑participation. Figure 3.22 shows how these self‑reported reasons vary across socio‑demographic groups, highlighting those over‑represented (relative‑representation ratios > 1). The figure distinguishes between constrained participants (individuals who took at least one course but missed another) and non‑participants (those who took none), allowing a direct comparison of the two forms of unmet demand.
There is a wide range of obstacles to participation in training, and the mix differs between constrained participants and non-participants and by socio-demographic characteristics. Family obligations, for example, are prominent for younger adults and women regardless of whether they were already engaged in training, whereas supply‑side barriers – scarcity of suitable courses, last‑minute impediments or inconvenient scheduling – feature more strongly among adults who grew up in rural areas, especially those who report not having been engaged in any training activity in the past 12 months. By contrast, both constrained participants and non-participants frequently mention the absence of appropriate courses and unmet prerequisites, pointing to gaps in provision and entry restrictions. Lack of employer support is particularly salient among non‑participants but less prominent among constrained participants, suggesting that employer backing may be decisive in crossing the threshold into any training.
Obstacles to participation in training differ by age. Lack of time due to family responsibilities and the cost of training are more important barriers to participation among constrained participants and non-participant 30-49 year-olds (1.7 RR and 1.8 RR, respectively) than among constrained participants and non-participant 50-65 year-olds. By contrast, 50‑65 year-olds are more likely to mention lack of suitable courses (0.7 RR for constrained participants and 0.6 RR for non-participants), unexpected events (0.7 RR and 0.6 RR) or cancellations (0.8 RR and 0.7 RR) as reasons for not participating. Several factors could explain these patterns: adults between the ages of 30 and 49 often have family responsibilities, and training must be woven around schedules constrained by childcare and sometimes elder‑care. As a result, they are acutely sensitive to the time cost of evening or weekend courses and to direct financial costs. By contrast, adults aged 50-65 typically face fewer childcare demands but may confront a different hurdle: a scarcity of courses tailored to late‑career upskilling or transitions towards retirement.
Furthermore, because older workers may have very specific training needs – as the heterogeneity of life experiences is wider for older workers – providers may cancel or postpone courses aimed at older workers more frequently due to low take-up or difficulty in organising relevant curricula. Finally, health fluctuations such as medical appointments can prevent participation at short notice for older adults. Taken together, these supply‑side limitations and age‑specific life events shift the barrier profile from time and cost among mid‑career adults to course availability and unpredictability among those approaching retirement.
There are also gender differences in the obstacles to participation reported. Among constrained participants, men are over‑represented in reporting that they lack prerequisites (0.7 RR), and among non‑participants, men are over‑represented in citing a lack of employer support (0.5 RR). By contrast, women are considerably more likely than men to emphasise family responsibilities and cost as reasons for an unfulfilled willingness to participate in training. Family responsibilities are an especially strong barrier to the participation of non-participant women (2.0 RR), and cost considerations are an especially strong barrier to additional participation among constrained participants (1.7 RR). Family responsibilities are currently not equally shared between men and women, with women’s larger share of unpaid care and mid‑career adults’ dual pressures raising the opportunity cost of engagement among women.
Women’s responses also highlight an overlooked form of vulnerability in the form of “unexpected events”. In principle, such events should strike men and women at similar rates, yet women are markedly more likely to cite them as a key barrier to participation in training (1.2 RR among non-training participants). This does not suggest that women face more emergencies but indicates that they have far less spare capacity to absorb emergencies that occur. Because women still shoulder the larger share of unpaid care, a sudden illness in the family, a school closure or even a public transport strike could derail their learning plans. Whether at home or in the workplace, women are often the de facto back‑up system for others (whether as mothers or partners at home or as administrative support staff in the workplace). What these results imply is that women can attend training only when nothing goes wrong, whereas men are better insulated by partner support, social expectations and workplace flexibility. That so many women report “unexpected events” as the reason for not participating reveals how finely balanced their time budgets are – and how easily the goal of upskilling is sacrificed when shocks occur.
Among adults with socio‑economically disadvantaged backgrounds, training participants are over‑represented in indicating that a lack of suitable courses or prerequisites constrained their participation, whereas non‑participants emphasise lack of employer support and cost. For example, participants whose parents were not tertiary educated are more likely to report that they did not find suitable training activities (1.4 RR) and did not meet prerequisites (1.2 RR), and their non-participating counterparts are more likely to report a lack of employer support (1.3 RR) or cancellation or postponements of training activities (1.3 RR). Adults with socio-economically disadvantaged backgrounds may live far from their place of work or from the location of in-person courses, making it harder for them to participate. These workers are also often employed with temporary contracts, making their employers less likely to be willing to accommodate their training needs (Albert, Garcia-Serrano and Hernanz, 2005[140]). Lack of information about the benefits of participation and expectations about how training could translate into improved opportunities may lower participation among adults from disadvantaged backgrounds. They may undervalue further rounds of training, be less familiar with subsidies available to them or be deterred by unclear entry criteria. Furthermore, employers in low‑productivity sectors, in which individuals with socio-economically disadvantaged backgrounds are over-represented, may doubt the payoff of helping their staff engage in training opportunities.
For the children of immigrants, both constrained participants and non-participants are over‑represented in reporting that courses were too expensive (1.2 RR each), with non‑participating adults also emphasising a lack of prerequisites as a key barrier to participation (2.0 RR). Among constrained participants, the children of immigrants are considerably under-represented in the group that cites lack of prerequisites as a reason for lack of participation (0.5 RR), among those who cite lack of suitable training (0.7 RR), and among constrained non-participants who cite that training takes place at an inconvenient time or location (0.6 RR).
For childhood residential context, adults raised in villages report different barriers to training participation than those raised in cities. Adults raised in villages who participate in training are over‑represented among those affected by cancellations (1.5 RR), with village non‑participants more often citing unsuitable courses (1.3 RR), lack of employer support (1.2 RR) and inconvenient scheduling or location (1.3 RR). This may be due to structural factors such as the geographic concentration of providers, digital‑connectivity gaps and sectoral training cultures, alongside differences in occupation, which may make it harder for rural residents to find suitable training.
Overall, constrained participants are more likely to report that available courses do not match their upskilling needs, whereas non‑participants are more likely to emphasise cost or time barriers. Ensuring that workers have the time and financial security needed to participate, for example through paid training leave and study allowances can reduce disparities. Equally important is adapting training content and delivery to better reflect workers’ needs, job contexts and personal circumstances, thereby strengthening both participation and learning outcomes. Even when practical barriers related to cost and scheduling are removed there still needs to be investments in making curricula relevant to workers. Unless provision is aligned with workers’ needs and aspirations, they will continue to under-invest in upskilling and reskilling.
For each socio-demographic characteristic, the relative representation ratio (RR) > 1 (< 1) indicates a higher (lower) likelihood of underlined groups being represented across different reasons for non-participation, OECD average
Note: Adults aged 30-65. The figure provides responses to the survey question: “Which of the following reasons prevented you from participating in these training activities? Please indicate the most important reason.” This survey question is a follow-up question to respondents answering “yes” to: “In the last 12 months, were there any training activities you wanted to participate in but did not?” See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Country-level results are provided in Annex Table 3.A.3.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
International evidence suggests that informal learning is a key means through which individuals build their skills in adulthood (OECD, 2025[127]). Acquiring knowledge through everyday problem solving, peer exchange, community participation and self‑directed practice allows individuals to develop transversal and technical skills in ways that formal curricula and structured workplace training rarely match. Informal learning is a particularly useful skills development path for adults who have had negative experiences in formal education settings or who may feel ill at ease with classroom‑style provision.
However, the very spontaneity that makes informal learning effective also renders it hard to promote through policy action. Governments can provide information alongside financial or human resources to support formal and non-formal learning, but they cannot legislate exchanges between co-workers. One exception to this is volunteering, which already sits within the reach of policy levers, with programmes to encourage civic participation, from national volunteer schemes to employer‑supported volunteering leave, well established across OECD countries (see Box 3.11). Although primarily framed in terms of social cohesion and service delivery, volunteering remains an under‑explored vehicle for skills development, offering a socially valued route for individuals to engage in skills development.
Voluntary work is more than civic altruism: its collaborative, problem‑focused nature creates the opportunity to develop both information‑processing skills and the ability to work with others, communicate effectively and self‑regulate (OECD, 2015[141]). Unlike engagement in formal or non‑formal learning activities, and in line with work-based learning and apprenticeships, volunteering embeds learning in real‑world contexts, providing the opportunity to practice skills and receive immediate feedback – conditions that reinforce skills acquisition and promote employment prospects (Spera et al., 2015[142]). Adults with socio‑economically disadvantaged backgrounds – who, as indicated in previous sections, generally hold lower formal qualifications and are less likely to participate in organised training activities – may view classrooms with scepticism, either because of earlier negative experiences or because fees, scheduling and entry requirements represent important barriers. Volunteering is therefore a potentially less stigmatising and more flexible pathway to skill development for these groups.
The share of adults who volunteer varies widely across countries. Figure 3.23 indicates that on average across OECD countries, 32% of adults reported having engaged in some form of volunteering in the previous year Annex Table 3.A.4. Around 50% of adults in Norway volunteer compared to less than 20% in Croatia, Korea, Lithuania and Spain. Frequency of participation also differs, with around 10% or slightly more of adults in Austria, Denmark, Ireland, the Netherlands, New Zealand and Norway volunteering weekly (Figure 3.23). In Canada, Denmark, Finland, Israel, the Netherlands, New Zealand, Norway, Switzerland and the United States, at least 10% of adults volunteer monthly. The shares of sporadic volunteers, i.e. adults who volunteer less than once a month, are highest in Denmark, Finland, Hungary, Norway and the United States, ranging between 20% and 27%. Disparities in volunteering by gender, age, childhood residential context and immigrant background are negligible, whereas adults with tertiary‑educated or high‑status occupation parents are more likely than their less advantaged counterparts to volunteer (a difference of between 7 and 9 percentage points) (as indicated in Figure 3.24).
Share of 16-65 year-olds engaging in volunteering activities, by country
Note: Volunteering activities are measured based on the following survey question: “In the last 12 months, how often, if at all, did you do voluntary work, including unpaid work for a charity, political party, trade union or other non-profit organisation?” With answer options being: “Never”, “Less than once a month”, “Less than once a week but at least once a month”, “At least once a week but not every day” or “Every day”.
* Indicates that differences between the share of adults who volunteer in a given country and those who volunteer on average across OECD countries and economies differ statistically significant at the 5% level.
Countries are ranked in descending order based on the percentage of adults who volunteer at least once a month.
Source: Calculations based on OECD (2024[53]), PIAAC data and methodology, www.oecd.org/en/about/programmes/piaac/piaac-data.html.
After adjusting for adults’ educational attainment level, further adjusting for volunteering contributes very little to explaining the differences in 21st-century skills between different groups. However, the results suggest that individuals who participate in volunteering activities have higher levels of proficiency in literacy and numeracy (0.11 SD each), and in adaptive problem solving (0.10 SD) (Table 3.6). These results suggest that volunteering can be an effective path for skills development. Box 3.11 provides examples of government policy measures, targeted programmes and strategic support to encourage volunteering.
Share of adults, by socio-demographic characteristic and volunteering activities, OECD average
Note: Volunteering activities are measured based on the following survey question: “In the last 12 months, how often, if at all, did you do voluntary work, including unpaid work for a charity, political party, trade union or other non-profit organisation?” With answer options being: “Never”, “Less than once a month”, “Less than once a week but at least once a month”, “At least once a week but not every day” or “Every day”. See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context. Country-level results are provided in Annex Table 3.A.4. Differences by immigrant background vary greatly across countries because of differences in the size of immigrant groups as well as composition and context. Readers are therefore encouraged to consult country-specific results in Annex Table 3.A.4.
* Indicates that shares in volunteering across groups (16-29 vs. 50-65, women vs. men, non-tertiary vs. tertiary, low-status vs. high-status, immigrant vs. non-immigrant, village vs. city) are significant at the 5% level.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Regression coefficients before and after adjusting for respondents’ educational attainment and volunteering activities, OECD average
Note: Adults aged 30-65. For this report, any volunteering activity in the past year is categorised as a positive outcome. Coefficients in bold are statistically significant at the 5% level. Results in columns (1), (4) and (7) explain the respective skill while adjusting for differences in gender, age, parental education, parental occupation, childhood residential context and immigrant background. Results in columns (2), (5) and (8) additionally adjust for differences in respondents’ educational attainment. Results in columns (3), (6) and (9) further adjust for differences in volunteering participation. Estimation results in columns (1) to (9) are restricted to the same number of observations. See the note for Figure 3.7 for a description of respondents’ educational attainment and the definitions of groups based on parental education, parental occupation, immigrant background and childhood residential context.
*Differences by immigrant background vary greatly across countries because of differences in the size of immigrant groups as well as composition and context. Readers are therefore encouraged to consult country-specific results in Annex Table 3.A.4.
Standard errors are provided in Annex Table 3.A.4.
Source: Calculations based on OECD (2024[16]), Survey of Adult Skills (PIAAC) 2nd cycle database, www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html.
Volunteering can enhance social cohesion and community well-being at the same time as benefiting volunteers themselves (Gagliardi, Pérez-Raynaud and Robinson, 2024[143]). Governments can promote volunteering through laws and strategies, policy measures, targeted programmes and strategic support, as also highlighted by the OECD Recommendation on Creating Better Opportunities for Young People.
Encouraging volunteering from an early age
The inclination to volunteer often develops early in life, influenced by family, community and educational environments. Volunteering at a young age fosters the development of pro-social values and social and emotional skills (Wilson, 2012[144]). Governments can nurture volunteering habits among young people through:
Mandatory service programmes: Some OECD countries integrate mandatory service into education. For example, Ontario, Canada, requires secondary students to complete 40 hours of community service, while Washington D.C. mandates 75 hours for graduation. National civic service programmes in Israel offer alternatives to compulsory military service. Such programmes have been shown to positively influence educational attainment and future earnings (Kim and Morgül, 2017[145]; Gagliardi, Pérez-Raynaud and Robinson, 2024[143]; Kol Zhekut, 2025[146]).
National youth volunteering and civic service programmes: At least 16 OECD countries implement voluntary, national, government-led youth volunteering and civic service programmes to foster young people’s civic participation and help them gain skills to enter the labour market. Most of these programmes offer training opportunities for young volunteers to promote civic engagement as well as personal and professional skills. Some programmes provide young participants with skills certifications at the end of their volunteering placement (Gagliardi, Pérez-Raynaud and Robinson, 2024[143]).
Financial support: Governments can fund volunteering directly or via organisations, thus helping to address the financial barriers reported by youth organisations (Gagliardi, Pérez-Raynaud and Robinson, 2024[143]; OECD, 2025[147]; 2020[148]). Assistance includes micro-grants, tax-breaks, stipends, language and skills training, and transport discounts.
Leveraging mass volunteering events
Governments can use large-scale events, such as Olympic Games or national disaster responses, to increase volunteering among adults. These events create widespread momentum and are often gateways to regular volunteering (Holmes et al., 2024[149]). Governments can sustain momentum by offering continuous training opportunities and support networks, ensuring volunteers develop relevant and transferrable skills (Benson et al., 2013[150]). The French government's role in the 2024 Paris Olympics exemplifies this by aligning volunteering roles with individual aspirations, promoting sustained involvement in local sports and community activities (Petit, 2024[151]). By supporting organisers and training delivered as part of volunteering, governments can ensure that training is tailored towards addressing skills shortages and labour market demands. Well-designed initiatives that are responsive to labour market needs are an avenue for adults to engage in meaningful upskilling opportunities that will enable them to gain skills in high demand; this may be particularly relevant for older adults, who have limited mobility in the job market (Lancee and Radl, 2012[152]).
Ensuring inclusive and accessible volunteering opportunities
Individuals with socio-economically backgrounds face significant barriers to volunteering, including limited access and exclusionary contexts. Governments can improve outcomes for these groups by ensuring accessible volunteering opportunities with inclusive designs and quality training (Southby, South and Bagnall, 2019[153]). Structural barriers and unnecessary eligibility criteria – such as education levels – often deter individuals who would otherwise benefit significantly. Simplified eligibility processes and clearly defined volunteering frameworks help engage marginalised groups, including migrants and refugees. For instance, the Civic Service for Youth in the Netherlands includes non-citizens through municipal databases. In France, the Civic Service assesses young applicants based on motivation, work ethic and interests, rather than specific skills or previous experiences (Gagliardi, Pérez-Raynaud and Robinson, 2024[143]). Furthermore, exploring alternative informal volunteering channels can bridge gaps for migrants typically excluded from formal opportunities (Wilson, 2012[144]). By integrating informal and formal volunteering systems, governments can widen participation, ensuring incentives and resources are accessible to all who contribute positively to their communities.
References
[76] Agence pour le développement de l’emploi (2025), FutureSkills Initiative, https://www.lifelong-learning.lu/orientation-et-reconversion/future-skills-initiative/en (accessed on 15 July 2025).
[61] Aj Ty v IT (2025), My sme Aj Ty v IT, https://ajtyvit.sk/o-nas/ (accessed on 15 July 2025).
[140] Albert, C., C. Garcia-Serrano and V. Hernanz (2005), “Firm-provided training and temporary contracts”, Spanish Economic Review, Vol. 7/1, pp. 67-88, https://doi.org/10.1007/s10108-004-0087-1.
[154] Australian College of Nursing (2020), Men share powerful videos of their experiences in the nursing profession, https://www.acn.edu.au/nurseclick/men-share-powerful-videos-of-their-experiences-in-the-nursing-profession (accessed on 18 June 2025).
[123] Australian College of Nursing (2020), Perspectives on men who care, https://www.acn.edu.au/nurseclick/perspectives-on-men-who-care (accessed on 18 June 2025).
[150] Benson, A. et al. (2013), “Training of Vancouver 2010 volunteers: A legacy opportunity?”, Contemporary Social Science, Vol. 9/2, pp. 210-226, https://doi.org/10.1080/21582041.2013.838296.
[71] Bierman, K. et al. (2008), “Promoting Academic and Social‐Emotional School Readiness: The Head Start REDI Program”, Child Development, Vol. 79/6, pp. 1802-1817, https://doi.org/10.1111/j.1467-8624.2008.01227.x.
[70] Bierman, K. et al. (2013), “Effects of Head Start REDI on Children’s Outcomes 1 Year Later in Different Kindergarten Contexts”, Child Development, Vol. 85/1, pp. 140-159, https://doi.org/10.1111/cdev.12117.
[28] Borgonovi, F. et al. (2025), “AI adoption in the education system: International insights and policy considerations for Italy”, OECD Artificial Intelligence Papers, No. 52, OECD Publishing, Paris/Fondazione Agnelli, Turin, https://doi.org/10.1787/69bd0a4a-en.
[32] Borgonovi, F., Á. Choi and M. Paccagnella (2018), “The evolution of gender gaps in numeracy and literacy between childhood and adulthood”, OECD Education Working Papers, No. 184, OECD Publishing, Paris, https://doi.org/10.1787/0ff7ae72-en.
[42] Borgonovi, F. et al. (2017), “Youth in Transition: How Do Some of The Cohorts Participating in PISA Fare in PIAAC?”, OECD Education Working Papers, No. 155, OECD Publishing, Paris, https://doi.org/10.1787/51479ec2-en.
[64] Boss Ladies (2025), About Boss Ladies, https://www.boss-ladies.dk/new-page-78 (accessed on 15 July 2026).
[117] Boys’ Day (2025), Homepage, https://www.boys-day.de/ (accessed on 18 June 2025).
[39] Breen, R. and J. Jonsson (2005), “Inequality of opportunity in comparative perspective: Recent research on educational attainment and social mobility”, Annual Review of Sociology, Vol. 31/1, pp. 223-243, https://doi.org/10.1146/annurev.soc.31.041304.122232.
[3] Brunello, G. and D. Checchi (2007), “Does school tracking affect equality of opportunity? New international evidence”, Economic Policy, Vol. 22/52, pp. 782-861, https://doi.org/10.1111/j.1468-0327.2007.00189.x.
[58] Brussino, O. and J. McBrien (2022), “Gender stereotypes in education: Policies and practices to address gender stereotyping across OECD education systems”, OECD Education Working Papers, No. 271, OECD Publishing, Paris, https://doi.org/10.1787/a46ae056-en.
[74] Byrne, D. et al. (2013), An Evaluation of the HEAR and DARE Supplementary Admission Routes to Higher Education, DARE/HEAR Strategic Development Group, https://mural.maynoothuniversity.ie/id/eprint/8969/1/DB-Evaluation-2014.pdf.
[65] Carneiro, P. et al. (2025), The short- and medium-term effects of Sure Start on children’s outcomes, The Institute for Fiscal Studies, https://ifs.org.uk/sites/default/files/2025-05/IFS%20Report.%20The%20short-%20and%20medium-term%20effects%20of%20Sure%20Start%20on%20children%E2%80%99s%20outcomes.pdf (accessed on 17 July 2025).
[9] Cheadle, J. (2008), “Educational investment, family context, and children’s math and reading growth from kindergarten through the third grade”, Sociology of Education, Vol. 81/1, pp. 1-31, https://doi.org/10.1177/003804070808100101.
[128] Chile Atiende (2025), Programa Habilidades para la Vida (HPV) [Life Skills Program (HPV)], https://www.chileatiende.gob.cl/fichas/2080-programa-habilidades-para-la-vida-hpv (accessed on 2 July 2025).
[132] Comunidad de Madrid (2025), Programa de Cooperación Territorial PROA+ [PROA+ Territorial Cooperation Program], https://www.comunidad.madrid/servicios/educacion/programa-cooperacion-territorial-proa (accessed on 2 July 2025).
[40] Conger, R. and M. Donnellan (2007), “An interactionist perspective on the socioeconomic context of human development”, Annual Review of Psychology, Vol. 58/1, pp. 175-199, https://doi.org/10.1146/annurev.psych.58.110405.085551.
[92] Council of Europe (2012), Inclusive Education for Roma Children as Opposed to Special Schools, https://rm.coe.int/16800890d5?utm (accessed on 15 July 2025).
[1] Cunha, F. and J. Heckman (2007), “The technology of skill formation”, American Economic Review, Vol. 97/2, pp. 31-47, https://doi.org/10.1257/aer.97.2.31.
[33] Dämmrich, J. and M. Triventi (2018), “The dynamics of social inequalities in cognitive-related competencies along the early life course – A comparative study”, International Journal of Educational Research, Vol. 88, pp. 73-84, https://doi.org/10.1016/j.ijer.2018.01.006.
[75] Denny, K. et al. (2014), “Money, mentoring and making friends: The impact of a multidimensional access program on student performance”, Economics of Education Review, Vol. 40, pp. 167–182, https://doi.org/10.1016/j.econedurev.2014.03.001.
[8] Desjardins, R. (2003), “Determinants of literacy proficiency: A lifelong-lifewide learning perspective”, International Journal of Educational Research, Vol. 39/3, pp. 205-245, https://doi.org/10.1016/j.ijer.2004.04.004.
[38] DiPrete, T. and G. Eirich (2006), “Cumulative advantage as a mechanism for inequality: A review of theoretical and empirical developments”, Annual Review of Sociology, Vol. 32/1, pp. 271-297, https://doi.org/10.1146/annurev.soc.32.061604.123127.
[79] Directorate of Higher Education and Competence, Norway (2025), Kompetansepluss [Competence Plus], https://hkdir.no/programmer-og-tilskuddsordninger/kompetansepluss (accessed on 12 August 2025).
[41] Domina, T. (2005), “Leveling the home advantage: Assessing the effectiveness of parental involvement in elementary school”, Sociology of Education, Vol. 78/3, pp. 233-249, https://doi.org/10.1177/003804070507800303.
[35] Duncan, G. and K. Magnuson (2013), “Investing in preschool programs”, Journal of Economic Perspectives, Vol. 27/2, pp. 109-132, https://doi.org/10.1257/jep.27.2.109.
[59] Education Magazine (2023), ‘Fact’s, Faces, Futures’, https://educationmagazine.ie/2023/04/12/facts-faces-futures/ (accessed on 15 July 2025).
[97] Együtt fejlődünk (2025), From disadvantage to advantage: A unique development program at Gandhi High School, https://egyuttfejlodunk.hu/en/from-disadvantage-to-advantage-a-unique-development-program-at-gandhi-high-school/ (accessed on 2 July 2025).
[87] EPRS (2025), Understanding EU action on Roma Inclusion, European Parliamentary Research Service, https://www.europarl.europa.eu/RegData/etudes/BRIE/2021/690629/EPRS_BRI(2021)690629_EN.pdf.
[7] Ermisch, J. and M. Francesconi (2001), “Family matters: Impacts of family background on educational attainments”, Economica, Vol. 68/270, pp. 137-156, https://doi.org/10.1111/1468-0335.00239.
[131] European Commission (2025), Program for orientation, progress and educational enrichment (“PROA+”), https://commission.europa.eu/projects/program-orientation-progress-and-educational-enrichment-proa_en (accessed on 2 July 2025).
[95] European Commission (2024), Transforming early education for marginalised Roma communities in Slovakia, European Social Fund Plus, https://european-social-fund-plus.ec.europa.eu/en/projects/transforming-early-education-marginalised-roma-communities-slovakia.
[80] European Commission (2023), SIM Case Study: Vives Emplea Saludable, https://european-social-fund-plus.ec.europa.eu/en/social-innovation-match/case-study/vives-emplea-saludable (accessed on 1 July 2025).
[88] European Commission (2023), The European Commission decides to refer SLOVAKIA to the Court of Justice of the European Union for not sufficiently addressing discrimination against Roma children at school, Press Release, 19 April 2023, https://ec.europa.eu/commission/presscorner/detail/pt/ip_23_2249.
[96] European Commission (2022), Civil society monitoring report on the quality of the national strategic framework for Roma equality, inclusion, and participation In Hungary, Directorate-General for Justice and Consumers, https://romacivilmonitoring.eu/wp-content/uploads/2023/07/RCM2-2022-C1-Hungary-FINAL-PUBLISHED-CATALOGUE.pdf.
[109] European Institute for Gender Equality (2025), Gender Equality in Academia and Research - GEAR tool, https://eige.europa.eu/gender-mainstreaming/toolkits/gear/legislative-policy-backgrounds/spain?language_content_entity=en (accessed on 18 July 2025).
[89] European Parliament (2023), European Parliament resolution on the segregation and discrimination of Roma children in education, https://www.europarl.europa.eu/doceo/document/B-9-2023-0394_EN.html.
[90] European Union/Council of Europe (2025), INSCHOOL2, Inclusive schools: Making a difference for Roma children, https://pjp-eu.coe.int/en/web/inclusive-education-for-roma-children/inschool2 (accessed on 2 July 2025).
[102] Eurydice (2024), Ireland: Supporting Traveller and Roma students and those at risk of educational disadvantage, European Commission, https://eurydice.eacea.ec.europa.eu/news/ireland-supporting-traveller-and-roma-students-and-those-risk-educational-disadvantage (accessed on 2 July 2025).
[62] French Ministry of National Education (2025), Communiqué de presse: Plan Filles et Maths [Press release: Girls and maths], https://www.education.gouv.fr/communique-de-presse-plan-filles-et-maths-450370 (accessed on 1 July 2025).
[63] French Ministry of National Education (2025), Filles et mathématiques: Lutter contre les stéréotypes, ouvrir le champ des possibles [Girls and mathematics: fighting stereotypes, opening up the field of possibilities], https://www.education.gouv.fr/filles-et-mathematiques-lutter-contre-les-stereotypes-ouvrir-le-champ-des-possibles-416773.
[143] Gagliardi, P., O. Pérez-Raynaud and A. Robinson (2024), “Promoting youth volunteering and civic service engagement: A stocktake of national programmes across OECD countries”, OECD Working Papers on Public Governance, No. 77, OECD Publishing, Paris, https://doi.org/10.1787/39659e6a-en.
[116] German Government (2024), Zahl der Eintritte in die Pflegeausbildung 2023 deutlich gestiegen [Number of entries into nursing training in 2023 increased significantly], Federal Ministry for Education, Family Affairs, Senior Citizens, Women and Youth, http://www.bmfsfj.de/bmfsfj/aktuelles/alle-meldungen/zahl-der-eintritte-in-die-pflegeausbildung-2023-deutlich-gestiegen-237880#:~:text=Laut%20Statistischem%20Bundesamt%20haben%20im,eine%20Ausbildung%20in%20der%20Pflege.&text=Das%20Statistische%20Bundesam (accessed on 18 June 2025).
[119] Gesundheitswirtschaft Nordwest (2023), So gelingt Diversifizierung in der beruflichen Pflege [How to achieve diversification in professional nursing], https://www.gesundheitswirtschaft-nordwest.de/downloads/HCM_3-2023_Gesundheitswirtschaft_Nordwest.pdf (accessed on 18 June 2025).
[120] Gesundheitswirtschaft Nordwest (n.d.), Modern Men do Care: Mehr Männer Für Die Pflege Von Morgen [More Men for the Care of Tomorrow], https://pflegenetzwerk-deutschland.de/fileadmin/files/Bilder/Wissenschaft/momedocare_Broschuere_08-03-2023_RZ_ONLINE-VERSION.pdf (accessed on 18 June 2025).
[18] Government of Canada (2023), The Government of Canada invests in artificial intelligence and teaching French as a second language from early childhood, Department of Canadian Heritage, http://www.canada.ca/en/canadian-heritage/news/2023/05/thegovernment-of-canada-invests-in-artificial-intelligence-and-teaching-french-as-a-second-language-from-early-childhood.html (accessed on 30 June 2025).
[5] Green, A. and N. Pensiero (2016), “The effects of upper‐secondary education and training systems on skills inequality. A quasi‐cohort analysis using PISA 2000 and the OECD survey of adult skills”, British Educational Research Journal, Vol. 42/5, pp. 756-779, https://doi.org/10.1002/berj.3236.
[72] HEAR (2025), What is HEAR?, Higher Education Access Route, https://accesscollege.ie/hear/what-is-hear/ (accessed on 1 July 2025).
[19] Hector Institute for Empirical Educational Research (2024), POLKE: AI in Education, https://uni-tuebingen.de/de/247993 (accessed on 31 October 2024).
[86] Hoare, E. et al. (2020), “Be You: A national education initiative to support the mental health of Australian children and young people”, Australian & New Zealand Journal of Psychiatry, Vol. 54/11, pp. 1061-1066, https://doi.org/10.1177/0004867420946840.
[149] Holmes, K. et al. (2024), “Volunteering legacies from the Olympic Games: Missed opportunities”, VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, Vol. 35/4, pp. 768-779, https://doi.org/10.1007/s11266-024-00643-w.
[60] House of the Oireachtas (2025), Apprenticeship Programmes - Dáil Éireann Debate, Tuesday - 25 March 2025, https://www.oireachtas.ie/en/debates/question/2025-03-25/875/#:~:text=As%20of%20end%2DFebruary%202025,Apprenticeship%202021%2D2025%20was%20launched. (accessed on 21 August 2025).
[56] IEA (2016), Progress in International Reading Literacy Study (PIRLS) 2016 Database, https://doi.org/10.58150/PIRLS_2016_data (accessed on 15 October 2025).
[55] IEA (2015), Trends in International Mathematics and Science Study (TIMSS) 2015 Grade 4 Database, International Association for the Evaluation of Educational Achievement, https://doi.org/10.58150/IEA_TIMSS_2015_G4 (accessed on 15 October 2025).
[46] IEA (2007), Trends in International Mathematics and Science Study (TIMSS) 2007 Grade 4 Database, International Association for the Evaluation of Educational Achievement, https://doi.org/10.58150/IEA_TIMSS_2007_G4 (accessed on 15 October 2025).
[47] IEA (2006), Progress in International Reading Literacy Study (PIRLS) 2006 Database, International Association for the Evaluation of Educational Achievement, https://doi.org/10.58150/PIRLS_2006_data (accessed on 8 July 2025).
[45] IEA (2003), Trends in International Mathematics and Science Study (TIMSS) 2003 Grade 4 Database, International Association for the Evaluation of Educational Achievement, https://doi.org/10.58150/IEA_TIMSS_2003_G4 (accessed on 15 October 2025).
[44] IEA (2001), Progress in International Reading Literacy Study (PIRLS) 2001 Database, International Association for the Evaluation of Educational Achievement, https://doi.org/10.58150/PIRLS_2001_data (accessed on 8 July 2025).
[43] IEA (1995), Trends in International Mathematics and Science Study (TIMSS) 1995 Grade 4 Database, International Association for the Evaluation of Educational Achievement, https://doi.org/10.58150/IEA_TIMSS_1995_G4 (accessed on 8 July 2025).
[100] Irish Government (2024), National Traveller and Roma Inclusion Strategy II 2024-2028, Department of Children, Disability and Equality, https://www.gov.ie/en/department-of-children-disability-and-equality/publications/national-traveller-and-roma-inclusion-strategy-ii-2024-2028/.
[104] Irish Government (2017), DEIS Delivering Equality of Opportunity In Schools, Department of Education, https://www.gov.ie/en/policy-information/4018ea-deis-delivering-equality-of-opportunity-in-schools/#deis-schools-2024-2025 (accessed on 8 July 2025).
[73] Irish Universities Association (n.d.), DARE HEAR Facts and Figures 2018-2022: Executive Summary, Disability Access Route to Education (DARE)/Higher Education Access Route (HEAR), https://accesscollege.ie/wp-content/uploads/2024/07/DARE-HEAR-Facts-and-Figures-Executive-Summary-2018-2022-s.pdf.
[12] Jackson, C., C. Wigger and H. Xiong (2021), “Do school spending cuts matter? Evidence from the Great Recession”, American Economic Journal: Economic Policy, Vol. 13/2, pp. 304-335, https://doi.org/10.1257/pol.20180674.
[34] Jerrim, J. and Á. Choi (2013), “The mathematics skills of school children: How does England compare to the high-performing East Asian jurisdictions?”, Journal of Education Policy, Vol. 29/3, pp. 349-376, https://doi.org/10.1080/02680939.2013.831950.
[91] Kahanec, M. et al. (2020), The Social and Employment Situation of Roma Communities in Slovakia, Policy Department for Economic, Scientific and Quality of Life Policies, European Parliament, https://www.sav.sk/uploads/monography/31/368/fulltext/03141433The_social_employment_situation_of_Roma.pdf.
[27] Khonthai Foundation (2021), Annual Report 2021, http://khonthaifoundation.org/wp-content/uploads/2022/06/2021-annual-report-KTF-final-for-web-EN.pdf (accessed on 7 November 2024).
[145] Kim, J. and K. Morgül (2017), “Long-term consequences of youth volunteering: Voluntary versus involuntary service”, Social Science Research, Vol. 67, pp. 160-175, https://doi.org/10.1016/j.ssresearch.2017.05.002.
[83] KiVa (2025), What is KiVa?, https://newzealand.kivaprogram.net/what-is-kiva/ (accessed on 2 July 2025).
[146] Kol Zhekut (2025), National-civilian service (in Arabic), http://www.kolzchut.org.il/he/%D7%A9%D7%99%D7%A8%D7%95%D7%AA_%D7%9C%D7%90%D7%95%D7%9E%D7%99-%D7%90%D7%96%D7%A8%D7%97%D7%99#:~:text=%D7%94%D7%A9%D7%99%D7%A8%D7%95%D7%AA%20%D7%94%D7%9C%D7%90%D7%95%D7%9E%D7%99-%D7%90%D7%96%D7%A8%D7%97%D7%99%2C%20%D7%94%D7%A (accessed on 10 March 2025).
[94] Koreň, M. (2018), Schools open to all? Slovakia uses EU funds to include Roma children, Euractiv, https://www.euractiv.com/section/economy-jobs/news/schools-open-to-all-slovakia-uses-eu-funds-to-achieve-inclusive-education/ (accessed on 2 July 2025).
[13] Lafortune, J., J. Rothstein and D. Schanzenbach (2018), “School finance reform and the distribution of student achievement”, American Economic Journal: Applied Economics, Vol. 10/2, pp. 1-26, https://doi.org/10.1257/app.20160567.
[152] Lancee, B. and J. Radl (2012), “Social connectedness and the transition from work to retirement”, The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, Vol. 67/4, pp. 481-490, https://doi.org/10.1093/geronb/gbs049.
[24] Language and Voice Laboratory (n.d.), Language Technology for Icelandic 2018-2022, https://rafhladan.is/bitstream/handle/10802/20054/mlt-en.pdf?sequence=1 (accessed on 4 November 2024).
[113] Lundgaardsløkka (2025), Ungdomsskolegutter i barnehagen [Middle school boys in kindergarten], https://lundgaardslokka.barnehage.no/innhold/side/20747 (accessed on 23 July 2025).
[85] McBrien, J. (2022), “Social and emotional learning (SEL) of newcomer and refugee students: Beliefs, practices and implications for policies across OECD countries”, OECD Education Working Papers, No. 266, OECD Publishing, Paris, https://doi.org/10.1787/a4a0f635-en.
[31] Ministry of Culture and Business Affairs (2024), Language Technology Programme for Icelandic 2024-2026, The Ministry of Culture and Business Affairs, http://www.government.is/library/01-Ministries/Ministry-of-culture-and-business-affairs/Language-technology-programme-for-icelandic-2024-2026-web.pdf (accessed on 5 November 2024).
[69] Ministry of Education and Culture, Finland (2025), Programme for equity and non-discrimination in education and training, http://urn.fi/URN:NBN:fi-fe2025033122398 (accessed on 11 August 2025).
[136] Ministry of Investments, Regional Development and Informatization of the Slovak Republic (2025), O projekte Digitálni seniori [About the Digital Seniors project], https://www.digitalniseniori.gov.sk/o-projekte/ (accessed on 15 July 2025).
[29] Ministry of the Economy and Innovation of Lithuania (2024), €12 million for AI solutions for the Lithuanian language, https://eimin.lrv.lt/en/structure-and-contacts/news-1/eimin-12-million-for-ai-solutions-for-the-lithuanian-language/ (accessed on 30 October 2024).
[122] NHS England (2019), Young male nursing applicants surge after ‘We are the NHS’ recruitment campaign, News, National Health Service England, https://www.england.nhs.uk/2019/02/young-male-nursing-applicants-surge-after-we-are-the-nhs-recruitment-campaign/#:~:text=Casualty%20actor%20Charles%20Venn%2C%2045%2C,change%20misconceptions%20of%20the%20profession (accessed on 18 June 2025).
[121] NHS England (2018), NHS launches multi million pound TV advertising campaign to recruit thousands of nurses in landmark 70th year, News, National Health Service England, https://www.england.nhs.uk/2018/07/nhs-launches-multi-million-pound-tv-advertising-campaign-to-recruit-thousands-of-nurses-in-landmark-70th-year/ (accessed on 18 June 2025).
[30] Nikulásdóttir, A., J. Guðnason and S. Steingrímsson (n.d.), Language Technology for Iceland 2018-2022, https://www.clarin.is/media/uploads/mlt-en.pdf (accessed on 5 November 2024).
[112] Nordic Information on Gender (2018), Subsidised Childcare for All: The Nordic Gender Effect at Work, https://nikk.no/wp-content/uploads/2019/10/2018-Subsidised-childcare-for-all.pdf.
[114] Norsk Rikskringkasting (2014), Desse gutane jobbar i barnehage etter skuletid – målet er å rekruttere fleire menn [These boys work in a kindergarten after school - the goal is to recruit more men], https://www.nrk.no/vestland/unge-gutar-i-barnehage-1.11691066 (accessed on 18 June 2025).
[111] Norwegian Government (2009), Men, Male Roles and Gender Equality, Ministry of Children and Equality, https://www.regjeringen.no/contentassets/ed3e28fc824c44118c17b63de7362aae/en-gb/pdfs/stm200820090008000en_pdfs.pdf.
[138] OECD (2025), Empowering the Workforce in the Context of a Skills-First Approach, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/345b6528-en.
[137] OECD (2025), Enhancing the Digital Skills of Seniors in the Slovak Republic, OECD, Paris, https://www.oecd.org/en/about/programmes/sg-reform/enhancing-the-digital-skills-of-seniors-in-the-slovak-republic.html (accessed on 11 August 2025).
[17] OECD (2025), Expenditure on educational institutions per full-time equivalent student, (database), Data Explorer, http://data-explorer.oecd.org/s/1cs (accessed on 4 March 2025).
[108] OECD (2025), Gender Equality in a Changing World: Taking Stock and Moving Forward, Gender Equality at Work, OECD Publishing, Paris, https://doi.org/10.1787/e808086f-en.
[67] OECD (2025), Reducing Inequalities by Investing in Early Childhood Education and Care, Starting Strong, OECD Publishing, Paris, https://doi.org/10.1787/b78f8b25-en.
[127] OECD (2025), Trends in Adult Learning: New Data from the 2023 Survey of Adult Skills, Getting Skills Right, OECD Publishing, Paris, https://doi.org/10.1787/ec0624a6-en.
[147] OECD (2025), “Voluntary work in the community: A guide to delivering an effective career development activity”, OECD Education Policy Perspectives, No. 115, OECD Publishing, Paris, https://doi.org/10.1787/ae8726c8-en.
[125] OECD (2024), Challenging Social Inequality Through Career Guidance: Insights from International Data and Practice, OECD Publishing, Paris, https://doi.org/10.1787/619667e2-en.
[14] OECD (2024), Education at a Glance 2024: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/c00cad36-en.
[105] OECD (2024), OECD Review of Resourcing Schools to Address Educational Disadvantage in Ireland, Reviews of National Policies for Education, OECD Publishing, Paris, https://doi.org/10.1787/3433784c-en.
[53] OECD (2024), PIAAC data and methodology, OECD, Paris, https://www.oecd.org/en/about/programmes/piaac/piaac-data.html (accessed on 10 December 2024).
[2] OECD (2024), Social and Emotional Skills for Better Lives: Findings from the OECD Survey on Social and Emotional Skills 2023, OECD Publishing, Paris, https://doi.org/10.1787/35ca7b7c-en.
[16] OECD (2024), Survey of Adult Skills (PIAAC) 2nd cycle database, OECD, Paris, http://www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html (accessed on 10 December 2024).
[84] OECD (2023), Building Future-Ready Vocational Education and Training Systems, OECD Reviews of Vocational Education and Training, OECD Publishing, Paris, https://doi.org/10.1787/28551a79-en.
[101] OECD (2023), “Implementation of Ireland’s Leaving Certificate 2020-2021: Lessons from the COVID-19 Pandemic”, OECD Education Policy Perspectives, No. 73, OECD Publishing, Paris, https://doi.org/10.1787/e36a10b8-en.
[77] OECD (2023), OECD Skills Strategy Luxembourg: Assessment and Recommendations, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/92d891a4-en.
[15] OECD (2023), PISA 2022 Results (Volume II): Learning During – and From – Disruption, PISA, OECD Publishing, Paris, https://doi.org/10.1787/a97db61c-en.
[68] OECD (2022), “Finland’s Right to Learn Programme: Achieving equity and quality in education”, OECD Education Policy Perspectives, No. 61, OECD Publishing, Paris, https://doi.org/10.1787/65eff23e-en.
[52] OECD (2022), PISA 2022 Database, OECD, Paris, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 5 November 2024).
[78] OECD (2020), Continuous Learning in Working Life in Finland, Getting Skills Right, OECD Publishing, Paris, https://doi.org/10.1787/2ffcffe6-en.
[148] OECD (2020), Governance for Youth, Trust and Intergenerational Justice: Fit for All Generations?, OECD Public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/c3e5cb8a-en.
[48] OECD (2017), PIAAC 1st Cycle Database, OECD, Paris, https://www.oecd.org/en/data/datasets/piaac-1st-cycle-database.html (accessed on 14 November 2024).
[139] OECD (2017), Trends Shaping Education, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/publications/reports/2017/04/country-roads_3e5f7ba4/ea43a39d-en.pdf.
[141] OECD (2015), How’s Life? 2015: Measuring Well-being, OECD Publishing, Paris, https://doi.org/10.1787/how_life-2015-en.
[54] OECD (2015), The ABC of Gender Equality in Education: Aptitude, Behaviour, Confidence, PISA, OECD Publishing, Paris, https://doi.org/10.1787/9789264229945-en.
[51] OECD (2012), PISA 2012 Database, OECD, Paris, https://www.oecd.org/en/data/datasets/pisa-2012-database.html (accessed on 5 November 2024).
[50] OECD (2009), PISA 2009 Database, OECD, Paris, https://www.oecd.org/en/data/datasets/pisa-2009-database.html (accessed on 14 November 2024).
[49] OECD (2000), PISA 2000 Database, OECD, Paris, https://www.oecd.org/en/data/datasets/pisa-2000-database.html (accessed on 14 November 2024).
[93] Office of the Government of the Slovak Republic (2022), Action Plans: To the Strategy of Equality, Inclusion and Participation of Roma Until 2030 for 2022-2024, https://www.romovia.vlada.gov.sk/site/assets/files/1526/action_plans_2022_2024_of_the_strategy_of_equality_inclusion_and_participation_of_roma_until_2030-1.pdf.
[110] Oxford Research AB (2021), Equality and diversity in the Spanish higher education institutes: Case study for KOTAMO-project, https://okm.fi/documents/1410845/113231925/KOTAMO+Espanja,+Equality+and+diversity+in+the+Spanish+higher+education+institutes.pdf/a2952232-78b2-d800-b317-42922975ff7e/KOTAMO+Espanja,+Equality+and+diversity+in+the+Spanish+higher+education+institutes.pdf (accessed on 5 August 2025).
[98] Pályázati Portál (2025), Továbbtanulást erősítő kezdeményezések a Gandhi Gimnázium Közhasznú Nonprofit Kft. által [Initiatives to strengthen further education by the Gandhi Gymnasium Public Benefit Nonprofit Ltd.], https://www.palyazat.gov.hu/eredmenyek/tamogatott-projektek/1474630201 (accessed on 2 July 2025).
[11] Passaretta, G., J. Skopek and T. van Huizen (2022), “Is social inequality in school-age achievement generated before or during schooling? A European perspective”, European Sociological Review, Vol. 38/6, pp. 849-865, https://doi.org/10.1093/esr/jcac005.
[134] Perez, C. and A. Vourc’h (2020), “Individualising training access schemes: France – the Compte Personnel de Formation (Personal Training Account – CPF)”, OECD Social, Employment and Migration Working Papers, No. 245, OECD Publishing, Paris, https://doi.org/10.1787/301041f1-en.
[151] Petit, M. (2024), Hidden heroes: The valor of volunteers at the Paris 2024 Olympics and Paralympics, OECD, Paris, https://oecdcogito.blog/2024/06/20/hidden-heroes-the-valor-of-volunteers-at-the-paris-2024-olympics-and-paralympics/.
[118] Pflege-Netzwerk Deutschland (2023), Mehr Männer für die Pflege gewinnen [Attract more men to nursing], https://pflegenetzwerk-deutschland.de/mehr-maenner-fuer-die-pflege-gewinnen-1 (accessed on 18 June 2025).
[106] Pobal (2025), Social Inclusion and Community Activation Programme (SICAP) 2024 – 2028, https://www.pobal.ie/programmes/social-inclusion-and-community-activation-programme-sicap-2024-2028/ (accessed on 15 July 2025).
[107] Pobal (2021), The Role of SICAP in Supporting New Communities, https://www.pobal.ie/wp-content/uploads/2021/03/The-Role-of-SICAP-in-Supporting-New-Communities_23.03.2021.pdf (accessed on 15 July 2025).
[22] Radboud University (2025), About the National Education Lab AI, https://www.ru.nl/en/nolai/about-nolai (accessed on 15 July 2025).
[23] Richter, C. et al. (2022), DComputer-assisted pronunciation training in Icelandic (CAPTinI): Developing a method for quantifying mispronunciation in L2 speech, Research-publishing.net, https://doi.org/10.14705/rpnet.2022.61.1480.
[124] Rustichini, A. (ed.) (2013), “Advanced mathematical study and the development of conditional reasoning skills”, PLoS ONE, Vol. 8/7, p. e69399, https://doi.org/10.1371/journal.pone.0069399.
[36] Schulenberg, J., A. Sameroff and D. Cicchetti (2004), “The transition to adulthood as a critical juncture in the course of psychopathology and mental health”, Development and Psychopathology, Vol. 16/04, https://doi.org/10.1017/s0954579404040015.
[37] Schulenberg, J. and I. Schoon (2012), “The transition to adulthood across time and space: overview of Special Section”, Longitudinal and Life Course Studies, Vol. 3/2, https://doi.org/10.14301/llcs.v3i2.194.
[57] Shutts, K. et al. (2017), “Early preschool environments and gender: Effects of gender pedagogy in Sweden”, Journal of Experimental Child Psychology, Vol. 162, pp. 1-17, https://doi.org/10.1016/j.jecp.2017.04.014.
[6] Silva, P. et al. (2020), “Student selection and performance in higher education: Admission exams vs. high school scores”, Education Economics, Vol. 28/5, pp. 437-454, https://doi.org/10.1080/09645292.2020.1782846.
[10] Skopek, J. and G. Passaretta (2020), “Socioeconomic inequality in children’s achievement from infancy to adolescence: The case of Germany”, Social Forces, Vol. 100/1, pp. 86-112, https://doi.org/10.1093/sf/soaa093.
[25] Smart Nation Singapore (2019), National Artificial Intelligence Strategy, http://www.smartnation.gov.sg/files/publications/national-ai-strategy.pdf (accessed on 30 October 2024).
[153] Southby, K., J. South and A. Bagnall (2019), “A rapid review of barriers to volunteering for potentially disadvantaged groups and implications for health inequalities”, VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, Vol. 30/5, pp. 907-920, https://doi.org/10.1007/s11266-019-00119-2.
[130] Spanish Government (2025), Componente 21: Modernización y digitalización del sistema educativo, incluida la educación temprana de 0 a 3 años [Component 21: Modernization and digitalization of the education system, including early education from 0 to 3 years old], https://planderecuperacion.gob.es/politicas_y_componentes/componente-21-modernizacion-y-digitalizacion-del-sistema-educativo-incluida-la-educacion-temprana (accessed on 2 July 2025).
[129] Spanish Government (2025), Programa para la orientación, avance y enriquecimiento educativo PROA+ [Program for Guidance, Advancement and Educational Enrichment PROA+], https://www.educacionfpydeportes.gob.es/mc/sgctie/cooperacion-territorial/programas-cooperacion/proa.html (accessed on 2 July 2025).
[133] Spanish Government (2024), Spain. Biennial progress report on the implementation of the European Child Guarantee, https://www.juventudeinfancia.gob.es/sites/default/files/infancia/garantia/2025-01-14/Spain.%20Biennial%20progress%20report%20ECG.pdf.
[142] Spera, C. et al. (2015), “Out of work? Volunteers have higher odds of getting back to work”, Nonprofit and Voluntary Sector Quarterly, Vol. 44/5, pp. 886-907, https://doi.org/10.1177/0899764015605928.
[115] Statistisches Bundesamt (2025), Pflegeausbildungsstatistik: Auszubildende in Pflegeberufen mit neu abgeschlossenem Ausbildungsvertrag [Trainees in nursing professions with newly concluded training contracts: Germany, reference date, gender], http://www-genesis.destatis.de/datenbank/online/statistic/21241/table/21241-0002/search/s/cGZsZWdlYXVzYmlsZHVuZ3NzdGF0aXN0aWs= (accessed on 18 June 2025).
[82] Teatro de Conciencia (2025), En Sus Zapatos: Un Espacio de Empatía Activa (In Their Shoes: A Space for Active Empathy), Teatro de Conciencia (Theatre of Conscience), https://programaensuszapatos.org/ (accessed on 2 July 2025).
[20] Tübingen Center for Digital Education (2024), Pedagogically oriented extraction of linguistic knowledge and generation of natural language with controllable readability (POLKE), https://uni-tuebingen.de/en/research/centers-and-institutes/tuebingen-center-for-digital-education/projects-1/polke/ (accessed on 31 October 2024).
[21] Tübingen Center for Digital Education (2024), Use of AI for thematic personalization of grammar learning, https://uni-tuebingen.de/forschung/zentren-und-institute/tuebingen-center-for-digital-education/projekte/ki-fuer-grammatik/ (accessed on 31 October 2024).
[103] Tusla (2025), HSCL - Home School Community Liaison, https://www.tusla.ie/services/educational-welfare-services/hscl/ (accessed on 11 August 2025).
[66] UK Department of Education (2010), The quality of group childcare settings used by 3-4 year old children in Sure Start Local Programme areas and the relationship with child outcomes, https://assets.publishing.service.gov.uk/media/5a7ae26aed915d670dd7f442/DFE-RR068.pdf.
[81] UNESCO (2024), Mainstreaming Social and Emotional Learning in Education Systems: Policy Guide, United Nations Educational, Scientific and Cultural Organization, https://doi.org/10.54675/ORWD6913.
[135] Urban Sustainability Exchange (2025), UTOPÍAS, https://use.metropolis.org/case-studies/utopias#casestudydetail (accessed on 2 July 2025).
[99] van Driel, B. (2006), “The Gandhi Secondary School: An experiment in Roma education”, European Journal of Intercultural studies, pp. 173-182, https://doi.org/10.1080/0952391990100205.
[4] van Hek, M., C. Buchmann and G. Kraaykamp (2019), “Educational Systems and Gender Differences in Reading: A Comparative Multilevel Analysis”, European Sociological Review, Vol. 35/2, pp. 169-186, https://doi.org/10.1093/esr/jcy054.
[26] Vathanalaoha, K. (2022), “Effects of Gamification in English Language Learning: The Implementation of Winner English in Secondary Education in Thailand”, LEARN Journal: Language Education and Acquisition Research Network, Vol. 15/2, pp. 830-857, https://so04.tci-thaijo.org/index.php/LEARN/article/view/259953.
[126] Victoria State Government (2025), Employment and small business, Jobs, Skills, Industry and Regions, https://djsir.vic.gov.au/jobs-victoria/how-we-help (accessed on 30 June 2025).
[144] Wilson, J. (2012), “Volunteerism research: A review essay”, Nonprofit and Voluntary Sector Quarterly, Vol. 41/2, pp. 176-212, https://doi.org/10.1177/0899764011434558.
← 1. A synthetic cohort is a group that analysts create by combining data from different people who are at similar stages of life, rather than following the same people over time as would typically be done in longitudinal cohort studies.
← 2. STEM fields comprise information and communication technologies; natural sciences, mathematics and statistics; engineering and manufacturing; or construction.
← 3. The 2023 Survey of Adult Skills defines adult education and training activities as organised learning activities, namely training activities such as courses, webinars, workshops, lectures or private lessons. These activities can be job-related or for personal interest. Examples of training activities include foreign language course, computer or software course, job-related training, hobby course (e.g. drawing, swimming, guitar or lecture on a specific topic), communication training (e.g. workshop on public speaking), or health and safety training (e.g. first aid course).