Group singing may be an optimal intervention strategy to promote active ageing and well-being; however, evidence with experimental validity is scarce. This study aims to fill this gap by analysing the effects of a 34-session singing group programme (SGP) on participants' subjective and social well-being and the mediating roles of social identification with the singing group and of self-esteem. An RCT with intervention (n = 89) and active waiting-list control (n = 60) conditions was conducted, and a mixed method quantitative and qualitative data collection and analysis were performed. Participants were mostly elderly day-care centre users (M = 76.66 years old; SD = 8.79) with low average levels of education and income. Structured measures of life satisfaction, positive and negative affect, self-esteem, loneliness, social identification and social well-being were collected, as well as interviews on the perceived benefits of participating in the SGP. Results showed significant effects of the SGP on the positive affect, social well-being and marginally on the self-esteem of the participants. The observed effects were sustained at the follow-up. Qualitative analysis corroborated the quantitative results. Mediation analysis showed indirect effects of social identification with the singing group on loneliness and social identification with the social care institution group; and of self-esteem on positive and negative affect.