It is challenging to investigate wider benefits of adult learning, especially in later life, due to limited data on educational activities and non-monetary returns in large, longitudinal surveys. Statistical matching provides an approach to exploit the potential of existing data by combining data sources with complementary features based on shared information. The paper describes the matching of two data sources (German Ageing Survey and Study of Educational Attainment and Interests of Older People) in order to examine the effects of educational participation on well-being in later life. We emphasize the matching procedure and how to identify the best-matched dataset. Based on matched data, effects of educational activities on life satisfaction are examined in later life. The