The most important motivation for adults to participate in adult learning is to improve their job performance or open new job opportunities. Adults who do participate in adult learning report that it was useful, and participation is associated with tangible benefits such as higher wages. Employer financial support for adult learning is commonplace and is a strong predictor of participation rates overall. Despite this, nearly a quarter of adults report barriers that prevent them from participating in adult learning; these barriers are more prevalent among women, younger adults, and more highly educated adults. The most commonly cited barrier is a lack of time, due to work and family responsibilities.
Trends in Adult Learning
New Data from the 2023 Survey of Adult Skills
Abstract
Most adults participate in learning to improve their job performance and prospects. This is the primary motivation for half of adult learners, and adults’ motivations for participation have not changed significantly over the past 10 years.
Most adults find this learning to be valuable. On average, 46% of adults find their non-formal job-related learning to be “very useful”, and another 31% report it was “moderately useful”. The rates at which people find learning to be useful have not changed significantly over the past decade.
Adults who participate in adult learning earn more. Participation in learning is associated with 7% higher wages, even after controlling for age, gender, educational attainment, and occupational characteristics. This may represent both workers improved skills and knowledge, as well as the fact that more motivated and productive adults are more likely to participate in adult learning in the first place. Adult learning is not significantly associated with higher life satisfaction.
Around half of adults neither take part in nor show interest in participation in adult learning. In addition to motivational barriers, this may reflect structural ones – such as limited opportunities, past exclusion, or learning opportunities that are misaligned with job needs.
Among those who do participate or want to, one in two face barriers to this participation. The most common barriers are lack of time either due to family or work responsibilities (48%), the lack of suitable training opportunities (14%) and costs (13%).
Some groups are more likely than others to face barriers than others. Women, younger adults, and more highly educated adults are most likely to report that they wanted to but faced barriers.
Employers have an important role to play in supporting adult learning. The majority of adult learning takes place at the workplace, during working hours. Strong financial support from employers is one of the best predictors of participation in adult learning. Co‑operation between policymakers and leaders within industry to provide high-quality, relevant adult learning activities may help ensure that employers see value in investing in these activities for their employees.
Adult learning can help employees to adapt to changing workplaces. A slight majority of adults who report changes in their workplace over the past three years reported that their employers paid for training to help adapt to the change. Employer-supported training was most common for those experiencing changes related to information and communication technology (66%, on average).
Adult learning plays a vital role in helping individuals navigate a world of rapid technological change, shifting labour markets, and evolving personal aspirations. Adults participate in learning activities for a multitude of reasons, and no single motivation appears to clearly account for differences in participation rates between countries. Nevertheless, adults who do participate in learning consistently report their training to be useful to them. Moreover, when adults do participate, the benefits are tangible – particularly in terms of wages. Despite this, nearly a quarter of adults across OECD countries report having encountered a barrier that prevented them from participating in a desired training activity over the past 12 months. These barriers are not distributed evenly, with gender, age, and educational attainment being most notably associated with the propensity to encounter barriers. All of this leaves an important role for employers in supporting adult learning. The majority of adult learning activities take place at the workplace, during work hours, and receives at least some financial support from employers. Indeed, employer financial support is one of the strongest predictors of a country’s overall adult learning participation rate.
This chapter explores the motivations behind adults’ decision to participate in adult learning, the value they find in this learning, and the resulting benefits in terms of earnings and life satisfaction overall. It then describes the prevalence of different barriers that some adults face in accessing adult learning and analyses which socio-demographic groups are most impacted by these barriers. Lastly, it explores the ways in which employers can support adult learning for their employees.
Adults participate in adult learning for a range of reasons, including to achieve professional objectives, to renew or update their credentials, or simply to fulfil personal goals. In doing so, they can improve their skills, making them more productive employees and improving their earning potential. Participation in adult learning is associated with higher wages and with higher life satisfaction overall; however, it is likely that these effects are mediated through effects on educational attainment and occupation itself. This section explores the reasons adults cite for participating in adult learning and trends between cycles. It also provides evidence of the benefits of participating in adult learning, including economic and social rewards. It discusses to what extent adults apply their learnt skills directly at work or in new roles, demonstrating the practical value of learning.
Most adults participate in learning to improve their career prospects
To improve one’s job performance or to exploit new job opportunities was most frequently cited as the most important reason for participation in adult learning. On average across OECD countries and economies, about half of adults who participated in non-formal job-related adult learning cited this reason, and it was the most-cited reason in every country (Figure 3.1). The second most common reason was to satisfy a personal interest or curiosity. Given that these respondents also indicated that their learning activity was job-related, this would imply that these adults are following a personal interest in learning new skills that do have some importance for their job, even though they do not have a specific professional goal in mind.
The primary motivation for undertaking non-formal job-related adult learning has not changed significantly for most countries over the past ten years (Figure 3.2). The most pronounced overall change has been a five-percentage point increase in the likelihood of adults reporting obtaining or renewing a certificate as their primary motivation for undertaking adult learning. Other notable changes in individual countries include Norway, where improving job performance has become far more prevalent at the same time as pursuing personal interests has become less common, and Japan, where the share of adults citing training to improve job performance has declined by 17 percentage points. It is not obvious what drives these changes; they may be due to changes in individuals’ own preferences, or changes in the types of trainings that are most likely to be subsidised or supported by employers.
Share of adults reporting a given primary reason for participating in learning, percentage
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Percentage point change in the share of adults reporting a given primary reason for participating in learning
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. Results that are not statistically significant at the 5% level (p > 0.05) are shown in striped bars. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2012, 2015, 2018, 2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Adult learning is a tool to help workers adapt to changing workplaces
Most adults complete formal training at a young age, before entering the workforce. Nevertheless, the skill and technical requirements of most workplaces will continue to evolve throughout people’s working lives. Non-formal adult learning can be a useful means to help workers adapt to changes in their workplace by giving them familiarity with new tools and technologies and teaching them the skills to thrive and advance amid these shifts. One would expect that workplaces undergoing changes would rely on adult learning for precisely these reasons.
Nevertheless, there is a great deal of variation in the extent to which employers will pay for training to support employees who have undergone changes in their workplace. Among those who reported changes in the use of information and communication technologies over the past three years, for example, 85% or more of workers in Korea and Poland reported that their employer had paid for training to help adapt to some or all of the changes (Table 3.1). For Denmark, France, and Hungary, the share was less than half. On average across OECD countries and economies, 66% of workers who experienced changes in this domain were supported by training provided by their employer. Over the six types of workplace changes presented, between 54% of workers (Amount of contact with clients or customers) and 66% of workers (Information and communication technologies) received accompanying training activities, on average.
Share of adults experiencing workplace changes who received employer-financed training, by type of change
Note: Adults aged 25‑65. Table reports the share, in percent, of adults whose employer paid for training activities for some or all of the workplace changes, by type of change, among those adults who responded positively to the question: “In the last three years, has your working environment significantly changed in any of the following areas?”. Standard errors are reported in parentheses. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
As technological advancements reshape industries, occupations, and tasks, the demand for digital competencies – from basic digital literacy to more advanced ICT skills – has surged. Improving workers’ ability to use digital tools and navigate digital processes at work via digital upskilling is a crucial component in helping workers adapt to the changes in their work environments. Many roles will increasingly involve at least some component of digital literacy, or the ability to manage and understand the workflow of other employees who have a more specialised digital skill set. There is already evidence of an increase in demand for technicians, data entry clerks, human resources managers, and marketing specialists with whose requisite skills are highly digital: for example, data management, online marketing, web analytics, and quality assurance (OECD, 2022[2]).
On average, just under half of non-formal job-related adult learning activities include some element wherein participants learn how to better use computerised equipment or navigate digitally supported processes. Singapore has the highest share of such activities, with 60% of trainings involving the use of digital equipment and knowledge. On the other hand, France stands out as the only country where less than 30% of trainings involve such content. There is a loose negative correlation between overall participation rates (Figure 1.1) and the share of trainings making use of digital equipment.
With workplaces continuing to evolve, it will be increasingly important to ensure that training is available to help ensure workers can develop the necessary facility with new digital technologies. This will in turn help workers to slot into the new and changing roles that will emerge as digital technologies become increasingly integrated into workplaces.
Share of trainings involving learning how to better use computerised equipment or digitally supported processes, percentage
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Most adult learners find their training activities to be valuable
In order for participation in adult learning to lead to positive outcomes, participants must first and foremost find their training useful. On average across participating OECD countries and economies, 46% of adults who participated in non-formal job-related learning activities reported that their training was “very useful” (Figure 3.4) and another 31% reported that the training was “moderately” useful to them. Conversely, less than 10% of adults reported that their training was “not useful at all”. Adults in Denmark, Chile and Croatia were most likely (60% or more) to report their training as very useful, whereas 25% of adults or less found this to be the case in Japan and Korea. In Hungary, Israel, and Korea, 15% of adults or more found their training to be not at all useful.
Share of non-formal job-related learning activities by usefulness to participant, percentage
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
For the OECD on average, there has been comparatively little change since Cycle 1 in the relative usefulness of non-formal job-related adult learning, with the share of adults who reported their training to be “very useful” falling by just 2 percentage points (Figure 3.5). However, this conceals the fact that for four countries (Chile, Estonia, Israel and Poland), there were significant increases in this share of adults, whereas for ten countries the share of adults actually decreased. For Denmark, Hungary and Spain, the decrease was greater than 10 percentage points.
Percentage point change in share of adults who reported non-formal job-related adult learning to be “very useful”
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. Results that are not statistically significant at the 5% level (p > 0.05) are shown in striped bars. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2012, 2015, 2018, 2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Social and economic returns to training
Adult learning offers adults the opportunity to improve their skills and job prospects, to move into better paid jobs and to become more productive (Albert, Garcia-Serrano and Hernanz, 2010[3]) (Bassanini et al., 2005[4]) (Fialho, Quintini and Vandeweyer, 2019[5]). It also provides an opportunity to satisfy one’s own curiosity by acquiring new knowledge; indeed, 23% of adults participating in non-formal adult learning for work say that they do so to satisfy their own personal interest (Figure 3.1). While there is evidence of a relationship between participation in adult learning and adults’ earnings, adult learning is not associated with greater overall life satisfaction.
Table 3.2 presents the results of a series of regressions of log wages on participation in adult learning and several other variables. At first sight, participation in adult learning is associated with 17% higher wages (column 1); even after controlling for socio-demographic characteristics, educational attainment, sector of employment and job skill requirements, participation is still associated with 7% higher wages (model 6).
While this positive association suggests that individuals who participate in adult learning may be more likely to earn higher wages, causality cannot be assumed. It is possible that adult learning helps individuals to acquire new skills and knowledge that contribute to better paid jobs. However, it could also be the case that individuals who are already in higher paid jobs – or those who are more motivated, career oriented or have stronger cognitive or social skills – are also more likely to participate in adult learning. In other words, the relationship may reflect both the potential benefits of learning and the characteristics of those who are more likely to participate. Other variables such as educational attainment, age, occupational skills and gender also show consistent associations with earnings.
Participation in non-formal job-related adult learning is associated with higher life satisfaction, but this association is largely due to observable factors other than the training itself. In a simple bivariate regression, adults who had participated in adult learning in the previous 12 months were 10% more likely to report high life satisfaction than those who had not (Table 3.3). However, this effect disappears when other socio-demographic variables are included in the regression model (models 2 and 3). The inclusion of additional control variables, such as whether or not one is employed or how much one earns, eliminates much of the association between education and life satisfaction. The most consistent predictors of high life satisfaction, on the other hand, are whether one lives with a partner or alone, whether one is employed or not, and how much one earns. The initial association between training and satisfaction can be explained by the fact that adults who live with a partner, are employed and have higher earnings are more likely to participate in adult learning in the first place.
Regression estimates for the effect of adult learning on log wages
Note: Table summarises the statistical effect of each variable on log wages; each estimate therefore can be understood as the percentage change in wages associated with each variable. Participation in adult learning refers to non-formal job-related adult learning over the past 12 months. Estimates are for the average of participating OECD countries. P values are reported in parentheses. Stars denote statistical significance at the 1% (**) and 5% (*) level.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html.
Regression estimates for the effect of adult learning on the probability of reporting high life satisfaction
Note: Table summarises the statistical effect of each variable on the probability, in percent, of reporting high life satisfaction. High life satisfaction is defined as responding 7 or higher to the question, “All things considered, how satisfied are you with your life as a whole these days?”, with 0 defined as “extremely dissatisfied” and 10 defined as “extremely satisfied”. Participation refers to non-formal job-related adult learning over the past 12 months, using a linear probability model (OLS). Estimates are for the average of participating OECD countries. P values are reported in parentheses. Stars denote statistical significance at the 1% (**) and 5% (*) level.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html.
Approximately half of all adults neither participate in nor express interest in participation in adult learning. While this may appear to reflect a lack of motivation, such disengagement often stems from deeper structural factors. Limited availability of relevant learning opportunities, past exclusion from education systems, and training perceived as disconnected from labour market needs can undermine both interest and confidence. These barriers reduce not only access but also expectations and the perceived value of participation.
One in four adults say that they would have liked to participate in (more) learning, but have been prevented from participating by some barrier or another. These barriers vary in nature; they may emerge because adult learning takes place at an unfavourable time or location, because learning may be too expensive, or adults may have to prioritise other responsibilities – to family or to work – that leave them little time for adult learning. Moreover, these barriers are not distributed uniformly throughout the population; some socio-demographic groups face higher incidences of these barriers than others. This section examines the reasons adults cite for not participating in adult learning, and how these have evolved over time. It also analyses to what extent these reasons differ between socio-demographic groups.
Owing to the design of the questions on barriers to learning in the 2023 Survey of Adult Skills, the sample is restricted to participation in non-formal (job-related) adult learning.
One in four adults report face barriers to adult learning
A sizeable minority of adults would like to participate in (more) adult learning but are unable to do so. Figure 3.6 shows the proportion of adults who experienced barriers to participation in adult learning. On average, 60% of adults have not participated in non-formal job-related learning activities over the past 12 months, of whom 11% reported that they would have liked to participate in a training activity (light grey bar). This implies that one in two adults did not participate in learning, neither did they have the desire to.
Furthermore, of the remaining 40% who did participate in training, 13% of these adults would have liked to participate in additional adult learning activities. Altogether, this means that 24% of adults encountered some barrier to participate in (more) adult learning.
Share of adults reporting barriers to participation in adult learning by participation status, percentage
Note: Adults aged 25‑65. Adults who reported barriers are plotted under: “wanted to participate” and “wanted to participate further”. Adults who “did not want to participate” did not report any barriers, nor did adults plotted under “participated”. OECD is an unweighted average of all participating member countries.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Increasing participation in adult learning requires an understanding of the nature of the barriers to learning faced by adults. By far the most common barrier to participation in non-formal job-related learning is lack of time (Figure 3.7). For almost half of adults (48%) who experience barriers to participation in adult learning, time constraints due to work or family responsibilities are the most important barrier to training, on average across the participating OECD countries and economies. Cost is the third most common barrier (13%). Reasons related to the availability of relevant training activities, such as suitability of training, necessary prerequisites, or taking place at an accessible time or place, are cited by a total of 14% of adults. Unexpected barriers or the cancellation or postponement of training accounts for a further 10% of barriers.
Average share of adults citing the following barrier as their main reason for not participating in learning, percentage
Note: Adults aged 25‑65; Figure plots the average share of adults citing each given barrier to participation among those who reported being unable to participate in a desired training over the past 12 months. Respondents were asked to choose the single most important reason that prevented them from participating from a pre‑defined list of potential barriers. OECD is an unweighted average of all participating member countries.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Figure 3.8 again shows that lack of time due to work or family responsibilities is by far the most commonly cited barrier to adult learning in OECD countries, accounting for the majority of responses in almost every country. This issue is particularly pronounced in Italy, Japan, Korea and Singapore, where 60% of adults or more cite time constraints as a major barrier. Other barriers – such as high costs, lack of employer support and lack of suitable training opportunities – vary significantly between countries. For example, financial barriers are most prominent in Italy, Hungary, Lithuania and Portugal. Meanwhile, a lack of suitable training is more frequently cited in countries such as Estonia, Latvia, Poland or the Slovak Republic. These patterns underline the need for broad-based policy responses – from extending training leave and childcare availability to improving course availability and financial support – to address country-specific challenges and increase participation rates in lifelong learning.
The most common barriers to adult learning have not changed over the past decade (Figure 3.9). Despite differences in the estimated shares of adults who flag time constraints, costs, or other barriers to adult learning between Cycle 1 and Cycle 2, for most countries there are no statistically significant changes in the relative importance of certain barriers. This raises questions about the effectiveness of policies, which have aimed to address these barriers in the past decade.
On average across OECD countries, the tendency has been for time constraints or unexpected impediments – such as last-minute cancellations – to become relatively more prevalent, but these changes are not very severe. Notable exceptions are in Chile and France, where there is statistically significant evidence that time constraints became relatively more prevalent than other types of barriers.
Average share of adults who reported each given barrier to training, percentage
Note: Adults aged 25‑65. Figure plots the average share of adults citing each given barrier to participation, among those who reported being unable to participate in a desired training over the past 12 months. Respondents were asked to list the single most important reason why they were prevented from participating. Some categories combine multiple responses: “No time due to work or family” includes adults who responded yes to either “I had no time due to family responsibilities” or “I had no time due to work-related reasons”. “No suitable training” combines “I did not find any suitable training activity”, “I did not have the prerequisites”, and “Training activity took place at an inconvenient time or location”. “Other” combines “Training activity was cancelled or postponed”, “Something unexpected came up that prevented me from participating”, and “Other reason”. Figure 3.7 plots all possible responses. OECD is an unweighted average of all participating member countries.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2012, 2015, 2018, 2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Percentage point change in share of adults reporting they were unable to participate in training, by reason
Note: Adults aged 25‑65. Figure plots the average share of adults citing each given barrier to participation, among those who reported being unable to participate in a desired training over the past 12 months. Respondents were asked to list the single most important reason why they were prevented from participating. Some categories combine multiple responses: “No time due to work or family” includes adults who responded yes to either “I had no time due to family responsibilities” or “I had no time due to work-related reasons”. “No suitable training” combines “I did not find any suitable training activity”, “I did not have the prerequisites”, and “Training activity took place at an inconvenient time or location”. “Other” combines “Training activity was cancelled or postponed”, “Something unexpected came up that prevented me from participating”, and “Other reason”. Figure 3.7 plots all possible responses. Results that are not statistically significant at the 5% level (p > 0.05) are shown in striped bars. OECD is an unweighted average of all participating member countries.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2012, 2015, 2018, 2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Barriers to adult learning follow similar patterns to participation overall
Barriers to adult learning are recorded for people who reported that they wanted to participate but were unable to do so. This is an important qualification: around one in two adults seem to have no desire participate in adult learning. These individuals are fundamentally different from those who wanted to learn but encountered barriers – whether logistical, financial or related to time. Consequently, the reported barriers largely reflect the experiences of a self-selecting group of motivated learners, and therefore largely follow the overall participation patterns. While addressing their barriers is essential to making adult learning more equitable and accessible to motivated learners, wider participation gaps are also shaped by the motivation or perceived need to engage in learning in the first place.
Still, even among those who do want to participate, certain groups are more likely to encounter obstacles. Figure 3.10 shows raw gaps in reported barriers by gender, education, and age across countries, while regression results in Table 4.4 provide a more detailed regression analysis that accounts for additional characteristics and provides a more nuanced picture of the prevalence of barriers across various groups.
On average, women are 7 percentage points more likely than men to report barriers to learning, with 27% of women reporting barriers, compared to 20% of men (Figure 3.10, Panel A). Indeed, this gap is only partly mitigated by other factors such as educational attainment or one’s sector of employment (see Table 3.4). Even when accounting for occupational, educational, and socio-demographic characteristics, women are about 5 percentage points more likely than men to encounter barriers to participation. The greatest gaps are in Estonia, Finland, Korea, and Lithuania, where gaps are 12 percentage points or higher.
There are pronounced gaps in the average likelihood of reporting barriers to participation between adults with different educational attainment (Figure 3.10, Panel B). Those with a tertiary degree are the most likely, with 30% on average reporting barriers. The rates for those without tertiary education are closer, at 20% for those with upper secondary and 15% for those who have not completed upper secondary. Because adults with below upper secondary education are half as likely as tertiary-educated adults to report barriers, but they are only a third as likely to participate in training, this means that, in fact, the burden of barriers falls disproportionately upon the least-educated adults.
Finally, young adults are most likely to report barriers to non-formal job-related learning (Figure 3.10, Panel C). Though participation and reported barriers are both highest, on average, for those aged 25‑34, adults aged 25‑54 are nearly as likely to report that they encounter barriers to learning (28% and 25%, respectively). This is likely due to the constraints faced by early-career workers who may also be in the process of raising young children.
Age, gender, and education are the best predictors of barriers to learning
Table 3.4 summarises the results of various regression models that regress the probability of reporting a barrier to participating in non-formal job-related adult learning on various socio-demographic, educational, and occupational characteristics. The leftmost column (1) summarises the results of bivariate regressions. This is the simple association between a given variable and the likelihood of reporting barriers to adult learning, without any other control variables. Columns (2) through (7) present the results of different model specifications. As noted above, because an individual must have the desire to participate in an adult learning activity in order to report a barrier in the first place, the patterns of barriers to learning resemble the overall participation patterns to some extent (see Chapter 2, Table 2.1).
Age and gender are consistently associated with the probability of encountering barriers to participation in adult learning. Older adults are less likely to report such barriers, and women are more likely to do so. Though, intuitively, the strongest effects appear in bivariate regressions, the effects are quite consistent even as additional control variables are added to the model. Adults in the oldest age group are consistently between 8% and 9% less likely than adults in the youngest group to report barriers to adult learning, and similarly women are between 5% and 7% more likely than men. Additional socio-demographic variables – whether one has children, lives alone or with a partner, is native‑born – do not have strong associations with the probability of reporting barriers to adult learning, even before the inclusion of control variables. Once control variables are included in the regression model, motherhood (an interaction variable) is no longer statistically significantly associated with the likelihood of reporting barriers either (model 3). Because of this, these variables are excluded from models 4‑7.
Parental education is an interesting case, in that its raw effect on the probability of encountering barriers is quite strong, at 13% (column 1), but this effect diminishes as successive variables are included in the model. Once these characteristics are also included, the size of the estimated effect of parental education on the probability of encountering barriers to adult learning diminishes substantially. This suggests that a large part of the effect parental education is, in fact, mediated through other variables such as educational attainment and where one works.
Adults who have completed tertiary education are between 8% and 12% more likely than adults who have not completed upper secondary to report barriers to adult learning, even with the inclusion of various control variables (2‑7). The estimated effect is smallest when literacy proficiency is also included in the model; this is not surprising, given that educational attainment and skill proficiency are closely correlated. Despite this correlation, model 5 shows that attainment and literacy proficiency each have independent and statistically significant associations with a higher likelihood of encountering barriers. Nevertheless, due to their similarity and because including both variables does not substantially affect the magnitude of other estimates, models 6 and 7 include only educational attainment.
Wages and occupational skill requirements are associated with barriers to adult learning when all other variables are excluded: the magnitude of the effect and its statistical significance erodes in models 6 and 7. However, including controls for wages, occupational skill, and industry does not change the estimated effects for other variables, with the exception of parental education (as noted above).
Overall, educational attainment has consistently the strongest effect on the likelihood of encountering barriers when controlling for other variables, followed by age and gender. Most other variables are not clearly associated with encountering barriers. It should also be understood that the R squared statistic on all models presented above does not exceed 0.06, meaning that the majority of variance in whether adults encounter barriers to adult learning is attributable to factors other than the broad categories presented in Table 3.4.
Share of adults reporting having encountered a barrier to participation over the past 12 months, percentage
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. OECD is an unweighted average of all participating member countries.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Estimated effect on probability of reporting barriers to adult learning
Note: Table summarises the statistical effect of each variable on the probability, in percent, of reporting barriers to participation in non-formal job-related adult learning over the past 12 months, using a linear probability model (OLS). Estimates are for the average of participating OECD countries. P values are reported in parentheses. Stars denote statistical significance at the 1% (**) and 5% (*) level.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html.
There are costs associated with the provision of and participation in adult learning. These costs may be borne by, or shared between, different actors: employers, trade unions, civil society organisations, public employment services, or individuals themselves. Indeed, cost is the third most frequently cited barrier to participation in adult learning. Together with a lack of support from employers, these barriers prevented 18% of adults from participating in a desired adult learning activity in the last 12 months (Figure 3.7). This section analyses the prevalence and trends in employer-sponsored training.
On average across the participating OECD countries and economies, 65% of adults participating in non-formal work-based adult learning reported that they received at least partial financial support from their employer (Figure 3.11). This share rises to almost 80% in the Netherlands, while in Korea only 35% of training was supported in this way. Strong financial support from employers is consistently associated with higher overall participation in adult learning (Figure 3.12).1 Indeed, employer financial support for adult learning is one of the strongest predictors of a country’s overall participation rate: a ten percentage point increase in the share of adults with employer financial support is associated with a seven percentage point increase in the country’s overall participation rate. It should be noted that adults may not always be conscious of the cost of employer provided training, in particular when it is provided in the workplace.
The lesson for policymakers is that while lack of employer support may not be the most commonly cited barrier to participation by adults themselves, there is strong evidence that employer support can have a powerful effect in increasing participation. Collaboration between policymakers and employers to provide high quality, relevant adult learning can help ensure that employers see the value in investing in these activities for their employees.
Share of non-formal job-related training where participant receives financial support from employer, percentage
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. Employer financial support is defined as the employer covering some or all of the expenses for learning, including tuition fees, expenses for books and travel costs. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Relationship between employer financial support for adult learning and overall participation rates
Note: Adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
Most adult learning takes place while adults are at work
The workplace is the main place where non-formal job-related learning takes place. Across OECD countries, 61% of non-formal job-related learning activities take place in the normal working environment and 74% during paid working hours (Figure 3.13). These figures suggest that the workplace plays a central role in facilitating opportunities for adults to improve their skills, although this reliance on workplace learning also raises important questions about the accessibility of non-formal learning to those who are not in employment.
There are notable differences in the role of the workplace in non-formal job-related learning across countries. In Poland and England (United Kingdom), for example, the workplace is the most common setting for non-formal learning, accounting for 83% and 75% of activities respectively. At the other end of the spectrum, Korea stands out as the only country where less than half of non-formal learners report taking part in such activities in their normal working environment. Similar patterns emerge for training during paid working hours (Figure 3.14, Panel A), with Korea again reporting the lowest levels, followed by Israel and Spain, where just over half of non-formal learning takes place during paid working hours. In contrast, countries such as France and Denmark integrate non-formal learning almost entirely into paid working time, with shares of 92% and 88% respectively, suggesting a systemic commitment to skills development in the workplace.
Share of adults reporting non-formal adult learning occurring in a normal working environment, percentage
Note: Employed adults aged 25‑65; non-formal job-related adult learning in the 12 months prior to the survey. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) (2023) Database, www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
On average across OECD countries and economies, there has been a six percentage point increase in the share of non-formal job-related learning that takes place only or mostly during working hours, with implications for people who are not in employment or whose workplace does not provide learning opportunities (Figure 3.14, Panel B). These developments take on added significance when considered in the context of wider trends in non-formal job-related adult learning. Overall participation rates in non-formal learning have declined in many countries (see Chapter 1), meaning that the observed increase in workplace learning reflects a redistribution of learning activities rather than a net increase in learning opportunities. The increasing concentration of non-formal learning in the workplace raises important issues of accessibility and inclusivity. As non-work learning opportunities decline in many countries, those outside the labour force or in precarious employment may face additional barriers to participating in skills development. As work-based learning is often tailored to specific job roles and needs, its increasing prevalence may limit access to broader skills development opportunities that promote adaptability in a rapidly changing labour market.
Note: Employed adults aged 25‑65, non-formal job-related adult learning in the 12 months prior to the survey. Results that are not statistically significant at the 5% level (p > 0.05) are shown in striped bars. See Box 1.3 for a note on comparison of data between Cycle 1 and Cycle 2. OECD is an unweighted average of all participating member countries. See Box 1.1 for a precise definition of job-related adult learning.
Source: OECD (n.d.[1]), Survey of Adult Skills (PIAAC) Database (2023), www.oecd.org/en/about/programmes/piaac/piaac-data.html. Underlying data are reported the accompanying Appendix Tables, available at www.oecd.org/en/publications/trends-in-adult-learning_ec0624a6-en/support-materials.html.
References
[3] Albert, C., C. Garcia-Serrano and V. Hernanz (2010), “On‐the‐job training in Europe: Determinants and wage returns”, International Labour Review, Vol. 149/3, pp. 315-341, https://doi.org/10.1111/j.1564-913x.2010.00089.x.
[4] Bassanini, A. et al. (2005), “Workplace training in Europe”, IZA Discussion Paper, Vol. 1640, http://ftp.iza.org/dp1640.pdf.
[5] Fialho, P., G. Quintini and M. Vandeweyer (2019), “Returns to different forms of job related training: Factoring in informal learning”, OECD Social, Employment and Migration Working Papers, No. 231, OECD Publishing, Paris, https://doi.org/10.1787/b21807e9-en.
[2] OECD (2022), Skills for the Digital Transition: Assessing Recent Trends Using Big Data, OECD Publishing, Paris, https://doi.org/10.1787/38c36777-en.
[1] OECD (n.d.), Survey of Adult Skills (PIAAC) Database, http://www.oecd.org/en/about/programmes/piaac/piaac-data.html (accessed on 11 June 2025).
← 1. This relationship holds even if the outlier Korea is excluded from the analysis.