Adults in the UK are tracked over 12 years to reveal who keeps learning, who stops, and who never starts. The project uses longitudinal data from the Understanding Society study to move beyond one-off snapshots of adult education participation. Current cross-sectional surveys cannot distinguish between a lifelong learner and someone who took a single course a decade ago, nor can they show how training benefits health, career progression, or community involvement over time. By applying statistical matching and panel regressions to 12 years of data, the researchers will identify patterns of frequent, occasional, and non-participation, and simulate whether the benefits of training persist or fade. The findings are intended for policymakers across education, labour markets, and health—sectors that currently rely on weaker evidence. If successful, this work could reshape how the UK funds and targets adult retraining programmes, particularly in response to automation, an ageing workforce, and economic shocks such as Brexit or COVID-19. The project also includes a one-day conference and partnership with the Learning & Work Institute to ensure results reach decision-makers directly.
View original technical description
The proposed project aims to investigate patterns of participation in education and training among adults and their related potential benefits in the domains of (1) health, (2) work and careers, and (3) the social community. Structured according to three objectives, the project will utilize data from the UK’s Understanding Society Household Longitudinal Study (UKHLS). We will apply Propensity Score Matching techniques and multivariate longitudinal data analyses. The argument for this proposal is grounded in the findings of the first Work Package of our ongoing ESRC Standard Grant ‘A UK-Ireland investigation into the statistical evidence-base underpinning adult learning and education policy-making’ (ES/X000826/1), resulting in the need to supplement cross-sectional analyses on participation in adult education and training with longitudinal data. Engagement with learning opportunities beyond initial education is important to cope with fast changing economies and societies. This includes the rise of Artificial Intelligence and automation, an ageing population and the need to adapt to changing circumstances caused by critical junctures such as an economic crisis, Brexit, COVID-19 lockdowns and pressures to transform into green and digital economies. Supplementing more sophisticated analyses to the current knowledge base will increase the quality of evidence which we will share and discuss with policymakers and stakeholders across social policy fields such as education and training, the economy, labour market and health. We will further nourish our existing partnership with the Learning & Work Institute and organise a one-day conference at the end of the project. Academic contributions will take the format of journal articles and conference presentations. The project will be carried out towards the fulfilment of three objectives. Objective 1: The first objective is to generate more sophisticated insights into the patterns and frequency of participation in adult education and training over time (12 years). Cross-sectional data do not give us insights in who the ‘frequent’ versus ‘occasional’ learners are, or who ‘never’ engages over an expanded period. Using data from Understanding Society, notable the ‘participation in any training’ variable, we will run descriptive statistics and longitudinal logistic regression models to test variations in participation at the individual level over time. Analyses will control for the core determinants of participation, which are age, gender, educational attainment and employment characteristics. Objective 2: The second objective is to zoom in on frequent and occasional learners and to investigate how their training characteristics vary over time. This is done through investigating their potential shifts in learning purposes, defined in Understanding Society through seven categories including ‘learning to increase skills at work’ and ‘for hobby or leisure’ reasons. Outputs of sequence analyses will visualise potential fluctuations in participation purposes and be supplemented by panel regressions, controlling for time variant variables such as job changes and age. Objective 3: The third objective is to reinvigorate the research agenda on the benefits of participation in adult education and training. In line with previous work, we will analyse potential benefits in three public policy domains: health, work and careers, and social community. Based on our cleaned dataset through Propensity Score Matching (PSM), making groups of learners versus non-learners more comparable, we will run panel regressions to investigate these potential benefits and their possible strength in relation to participation in training. As part of these analyses, we will simulate potential benefits in the short versus longer term.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know