Research · for u3a groups, adult-learning groups and older people's charities

Learning in Later Life: Who Takes Part, What Stops Them and Its Links With Wellbeing

Published 2026-11-15

Across the European Union in 2022, 35.4% of people aged 55 to 64 had taken part in some education or training in the past year, against 56.5% of people aged 25 to 34[1], and the official adult learning surveys of both Europe and Australia stop asking at 69 and 74 respectively. Studies of older learners find that learning for interest, especially in groups, goes with better wellbeing and, in one large Japanese survey, lower odds of loneliness. None of them can yet show that learning causes either.

Learning in later life takes many forms: a language class at the local college, a history group in a church hall, an online course in the evening, a craft taught by one neighbour to another. For some the subject is the point; for others it is the reason to leave the house, the people they now know by name. This study looks at how many older people take part in different countries, what stops those who would like to, and what research has found about learning, wellbeing, loneliness and memory. Our u3a study covers the University of the Third Age in detail.

35.4%of people aged 55 to 64 in the EU took part in formal or non-formal education or training in the 12 months before the 2022 Adult Education Survey[1]
21.9%of people aged 55 to 69 in the EU who wanted to learn but did not gave health or age as a reason, against 8.3% of 25 to 34 year olds, in the same survey[2]

In Europe, about a third of people aged 55 to 64 took part in learning in a year, and the surveys stop at 69

The best international data on adult learning come from the European Union’s Adult Education Survey. In its 2022 round, the share of people who had taken part in any formal or non-formal education or training in the previous 12 months fell steadily with age, from 56.5% of 25 to 34 year olds to 46.2% of 45 to 54 year olds and 35.4% of 55 to 64 year olds. Across the wider band of 55 to 69, it was 29.4%[1].

Share of adults who took part in education or training in the last 12 months, by age, EU, 2022Adult Education Survey 2022, European Union (27 countries), formal or non-formal education and training in the previous 12 months: aged 25 to 34, 56.5%; 35 to 44, 49.9%; 45 to 54, 46.2%; 55 to 64, 35.4%.0%15%30%45%60%56.5%25-3449.9%35-4446.2%45-5435.4%55-64Age group
Formal and non-formal learning combined, much of it job-related; informal learning (reading, self-teaching, learning from friends) is not counted. The survey interviews people aged 18 to 69 in private households, so it says nothing about people over 69. Source: Eurostat, Adult Education Survey, table trng_aes_101.

Those figures need two warnings. First, the survey counts courses, classes, workshops and job training; much of what it captures at working ages is training for work, so the fall with age partly reflects retirement rather than a loss of interest. Informal learning, such as reading, teaching yourself or learning from a friend, is not in these figures. Second, the survey interviews people aged 18 to 69 only: its published tables have no age band above 69[1]. For people in their seventies, eighties and nineties, the most widely used European measure of adult learning is silent.

Where someone lives makes a large difference. Among people aged 55 to 64 in EU countries, participation ranged from 63.8% in Sweden and 53.8% in the Netherlands to 11.6% in Poland and 7.8% in Greece[1]. The same countries lead and trail at younger ages too: among 25 to 34 year olds, participation was 77.3% in Sweden and 72.5% in the Netherlands, against 31.5% in Poland and 27.3% in Greece[1]. That pattern is consistent with differences in the supply of courses, workplace training and public provision in each country, not with anything about older people themselves.

Share of people aged 55 to 64 who took part in education or training in the last 12 months, selected European countries, 2022Adult Education Survey 2022, people aged 55 to 64, formal or non-formal education and training in the previous 12 months: Sweden 63.8%; Netherlands 53.8%; Germany 51.5%; Hungary 50.6%; Denmark 46.0%; Ireland 40.3%; France 35.5%; EU average 35.4%; Spain 35.1%; Italy 26.7%; Poland 11.6%; Greece 7.8%.0%17.5%35%52.5%70%Sweden63.8%Netherlands53.8%Germany51.5%Hungary50.6%Denmark46.0%Ireland40.3%France35.5%EU average35.4%Spain35.1%Italy26.7%Poland11.6%Greece7.8%
Formal and non-formal education and training, including job-related courses. Small differences between countries should not be over-read. Source: Eurostat, Adult Education Survey, table trng_aes_101.

Australia’s official survey goes to age 74, but its 2024–25 round did not report learning for personal interest separately. On what it did measure, 6.0% of people aged 65 to 74 were studying for a qualification or doing work-related training, against 34.3% of all 15 to 74 year olds, and 5% of 65 to 74 year olds had done work-related training in the previous year[3]. Among the few aged 65 to 74 who were studying for a qualification, 46.2% gave personal development as the main reason, a category that includes enjoyment, interest, confidence and meeting people[3]. In our search we found no national statistics, in any country, that count older people’s informal and community learning in the way these surveys count courses and training.

Health, age and timing stand between older people and the learning they want

Not taking part is not the same as not wanting to. The Adult Education Survey asks people who wanted to learn but did not what stopped them, and the answers change with age. Among 25 to 34 year olds, the commonest reasons were the timing of courses (43.6%) and cost (35.0%). Among 55 to 69 year olds, timing was still the most common reason (28.4%), but health or age came next (21.9%), ahead of cost (19.5%) and family reasons (18.4%)[2].

Why people who wanted to learn did not, by age, EU, 2022Adult Education Survey 2022, European Union, people who wanted to take part in education or training but did not, percentage giving each reason (more than one allowed), aged 55 to 69 against 25 to 34: schedule 28.4% against 43.6%; health or age 21.9% against 8.3%; cost 19.5% against 35.0%; no suitable offer 16.3% against 14.2%; distance 10.2% against 11.6%.Aged 55-69Aged 25-340%12.5%25%37.5%50%Schedule (timing)28.4%43.6%Health or age21.9%8.3%Cost19.5%35.0%No suitable course onoffer16.3%14.2%Distance10.2%11.6%
Base: people who wanted to take part in education or training but did not (Eurostat’s definition). People could give more than one reason. Source: Eurostat, Adult Education Survey, table trng_aes_177.

Two answers deserve attention from anyone who runs or funds a learning group. One in six of the older people who wanted to learn, 16.3%, said there was no suitable course on offer, a slightly higher share than among younger adults (14.2%), and 10.2% named distance[2]. These are things a provider or local government can change, and “health or age” partly is too: a course without step-free access, or in a room where it is hard to hear, turns a health condition into a barrier.

The survey does not ask anyone over 69, so it cannot tell us what keeps people in their seventies and eighties away. Our companion study of the u3a sets out who joins that movement and who is missing from it.

Learning for interest, especially with others, goes with better wellbeing

The strongest evidence on learning and wellbeing in later life comes from the English Longitudinal Study of Ageing (ELSA), which follows the same people aged 50 and over in England over time. Andrew Jenkins and Tarek Mostafa used four waves of ELSA and fixed-effects models, which compare each person with themselves at different times and so remove the influence of stable traits such as personality or earlier schooling. They measured wellbeing with CASP-19, a quality-of-life scale covering control, autonomy, pleasure and self-realisation. “Learning was associated with higher wellbeing after controlling for a range of other factors,” they found, and “more informal types of learning were associated with higher wellbeing”. But “there was no evidence that formal education/training courses were associated with higher wellbeing”[4]. Their informal categories included education, music or arts groups and evening classes, and sports clubs, gyms and exercise classes[4].

That is a careful design but not a trial: it cannot rule out changes happening at the same time, such as better health making both joining a class and feeling well more likely. The contrast with formal courses may suggest that what goes with wellbeing is learning that is chosen, enjoyed and often shared, not gaining a qualification.

A Canadian study points the same way, with a weaker design. Miya Narushima and colleagues surveyed 416 people aged 60 and over enrolled in non-credit, general-interest courses in a public continuing education programme. Compared with those who had been learning for the shortest time, those who had kept attending longest had lower odds of being in the “distressed” category of a psychological wellbeing measure (odds ratio 0.4, 95% confidence interval 0.2 to 0.9), after adjusting for age, gender and an index of vulnerabilities such as poor health, low income and living alone[5]. Nearly one-third of the longest-attending group fell into the study’s high-risk category for vulnerability, so long-term learners were not simply the healthiest people[5]. But everyone surveyed was a current learner, people who felt worse may have left earlier or not answered, and the authors state that, given the cross-sectional design, “we were unable to determine causation”[5].

What learners themselves say is consistent with this. In a 2015 survey of US adults of all ages by the Pew Research Center, 74% had done some learning for personal interest in the previous year. Of those personal learners, 87% said it had helped them feel more capable and well rounded, 64% that it had helped them make new friends and 58% that it made them feel more connected to their local community[6]. These are answers from adults of every age and describe what people value, not a measured effect.

The links with loneliness are modest, and come mostly from one-off surveys

Few studies of learning in later life measure loneliness directly; most measure wellbeing, quality of life or mood. The largest we found is from Japan. Researchers analysed 5,395 people aged 65 and over living in the community in the Tokai region, grouping 16 kinds of social participation into types and measuring loneliness with the UCLA Loneliness Scale (version 3), a widely used research questionnaire. The “hobbies and learning activities” type was associated with lower odds of loneliness (adjusted odds ratio 0.900, 95% confidence interval 0.848 to 0.953)[7]. Work, leisure and outings, community contribution and support and care were also associated with lower odds, while taking part in local events showed no significant association[7]. The authors are clear that “causal inferences cannot be drawn from the present findings”[7]: people who feel less lonely may simply find it easier to join in.

The pandemic moved many learning groups online. When two social and educational programmes for older adults in South Florida switched to video classes in March 2020, researchers worked with their members to survey participants, and repeated the survey a year later; respondents included 211 older adults with an average age of 75.5. More frequent online class attendance was strongly associated with feeling more connected, in better spirits and less socially isolated[8]. It used no standard loneliness scale, and the people most cut off may not have been online to answer. Our u3a study and our study of digital inclusion say more on learning online.

Intervention studies are rarer still. A small study gave a three-month English-language course to 19 people with current or past depression, average age 69.7, and to a comparison group of 15 older people without depression. Those with a history of depression showed significant falls in apathy, social loneliness and everyday cognitive slips, but the authors judged that the course “had limited benefits for cognition and psychosocial well-being” overall, and called for “larger samples and a no-training control sample”[9]. Our study of what reduces loneliness sets learning alongside other group programmes.

Learning something new may matter for the mind as well as for company

Is it the learning or the company that counts? One US experiment, the Synapse Project, tried to separate them. Older participants spent an average of 16.51 hours a week for three months learning to quilt, learning digital photography, or both. Others met as a social group for activities that were not intellectually demanding, or did undemanding tasks with no social contact. Episodic memory improved more in the groups learning demanding new skills, and the researchers reported, “somewhat surprisingly”, “limited cognitive benefits of sustained engagement in social activities”[10]. The trial was short and measured memory, not loneliness, and does not show that learning prevents dementia; it does suggest that the challenge of learning something new matters, not only the company. Our study of the Lancet Commission’s dementia risk factors covers education and social isolation as risk factors across the life course.

Better long-term evidence is on its way. Sweden’s UGU-LIFE study is following people born in 1948 and 1953, whose records go back to their school years, to study what shapes learning across life. Its authors note one of the field’s recurring problems: only 26.2% of the original cohort took part in the latest follow-up, “with selective participation particularly related to childhood cognitive ability and educational level”[11]. The people research loses are often the people learning programmes most want to reach.

What the evidence asks of us

The World Health Organization counts the ability to “learn, grow and make decisions” as part of functional ability, alongside the ability to “build and maintain relationships”[12]. The evidence above points to several things for providers, funders, councils and statistics agencies.

  • Count learning at every age. The EU’s survey stops at 69 and Australia’s at 74, and Australia’s latest round does not report learning for personal interest separately[1][3]. Without counts for people in their seventies and beyond, policy cannot see who is left out.
  • Remove the barriers people name. Among 55 to 69 year olds in the EU who wanted to learn but did not, timing, health or age, cost and family reasons were the commonest, and one in six found no suitable course[2]. Daytime sessions, step-free and well-lit venues, hearing loops, low fees and courses designed with older people are places to start.
  • Value informal, shared learning. In ELSA it was informal learning, not formal courses, that went with higher wellbeing[4]. Funding rules that reward qualifications and job outcomes can miss the groups that matter most to older learners.
  • Keep face-to-face options. In one US survey, attending online classes more often went with feeling more connected during the pandemic[8], but such classes reach only people who are online, and a shared room offers things a screen may not.
  • Measure loneliness properly. Most studies here are one-off surveys of current learners. Evaluations that use a validated loneliness scale, follow people who leave as well as those who stay, and compare learners with similar people who do not take part would tell groups and funders far more.

For the wider picture, see the loneliness research library.

Language: this piece says "older people" and treats loneliness as a common human experience that anyone can have, not a personal failing or an illness. It keeps apart three things that are often mixed up: loneliness (how people feel about their relationships), social isolation (how few social contacts they have) and living alone (a household arrangement). Where a study uses another term, such as "seniors" or "older adults", we keep it when quoting that study.

Why we're publishing this

Learning groups for older people, from u3a interest groups to college classes and community workshops, are mostly run on goodwill, small fees and volunteer time. They are increasingly asked by funders and health services to show what difference they make to loneliness and wellbeing. This study brings together the official participation figures and the research so that groups can make their case with evidence that stands up, and can see honestly where that evidence is still thin.

How to read this data

Participation figures come from household surveys: the EU Adult Education Survey (2022, ages 18 to 69) and Australia’s adult learning survey (2024–25, ages 15 to 74). They count courses and training, much of it job-related, leave out informal learning and are not directly comparable with each other. On wellbeing, the ELSA analysis follows the same people over time (CASP-19 quality of life); the Canadian study is a single survey of current learners. On loneliness, the Japanese study used the UCLA Loneliness Scale (version 3); its abstract, which is what we could read, does not give the cut-off used. The Florida survey asked about feelings of isolation and connectedness, not a loneliness scale. All of these are associations: people who are healthier, more confident and better connected are more likely to take up learning in the first place. We searched Europe PMC, Eurostat and national statistics agencies in November 2026; we found no randomised trial of general adult learning that used loneliness as its main outcome.

Use this data

Free to cite with attribution to Understanding Aging. u3a groups, men's sheds, senior centres, lunch clubs, befriending services, charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Suggested citation: Understanding Aging (2026). Learning in Later Life: Who Takes Part, What Stops Them and Its Links With Wellbeing. https://understandingaging.org/research/lifelong-learning-later-life

Important: This article summarises published research and statistics for general information. It is not medical or care advice and cannot be used to diagnose, treat or predict anything about an individual. If loneliness, low mood or isolation is affecting your health or someone else’s, a doctor or another health professional can help, and local services for older people can often put you in touch with groups and support nearby.

Sources

  1. Eurostat. Participation rate in education and training by age (Adult Education Survey), table trng_aes_101. Eurostat, 2022 survey data. ec.europa.eu/eurostat/databrowser/view/trng_aes_101
  2. Eurostat. Persons who want to participate in education and training but do not by reason and age (Adult Education Survey), table trng_aes_177. Eurostat, 2022 survey data. ec.europa.eu/eurostat/databrowser/view/trng_aes_177
  3. Australian Bureau of Statistics. Work-Related Training and Adult Learning, Australia, 2024–25. ABS, 2026. abs.gov.au/statistics/people/education/work-related-training-and-adult-learning
  4. Jenkins A, Mostafa T. The effects of learning on wellbeing for older adults in England. Ageing & Society 35(10):2053–2070, 2015. cambridge.org/core/journals/ageing-and-society
  5. Narushima M, Liu J, Diestelkamp N. Lifelong learning in active ageing discourse: its conserving effect on wellbeing, health and vulnerability. Ageing & Society 38(4):651–675, 2018. europepmc.org/article/PMC/PMC5848758
  6. Pew Research Center. Lifelong Learning and Technology. Pew Research Center, 2016. pewresearch.org/internet/2016/03/22/lifelong-learning-and-technology
  7. Shimoda T, Katayama O, Yamaguchi R, et al. Classification of social participation and its association with loneliness in older adults: a cross-sectional study of the National Center for Geriatrics and Gerontology–Study of Geriatric Syndromes. Geriatrics & Gerontology International, 2026. europepmc.org/article/MED/42297725
  8. Derynda B, Siegel J, Maurice L, Cook N. Virtual lifelong learning among older adults: usage and impact during the COVID-19 pandemic. Cureus, 2022. europepmc.org/article/MED/35651387
  9. Brouwer J, van den Berg F, Knooihuizen R, Loerts H, Keijzer M. Language learning as a non-pharmacological intervention in older adults with (past) depression. Brain Sciences, 2025. europepmc.org/article/MED/41008351
  10. Park DC, Lodi-Smith J, Drew L, et al. The impact of sustained engagement on cognitive function in older adults: the Synapse Project. Psychological Science, 2014. europepmc.org/article/MED/24214244
  11. Hansson I, Berg AI, Bjälkebring P, et al. Cohort profile: the Swedish Evaluation Through Follow-up Study of Learning Later in Life (UGU-LIFE). BMJ Open, 2026. europepmc.org/article/PMC/PMC13691816
  12. World Health Organization. Healthy ageing and functional ability (questions and answers). WHO, accessed November 2026. who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability