The work in this thesis explores the relationship between singing and well-being. In particular, I focus on investigating the potential for group singing to promote feelings of hedonic, eudaimonic and social well-being. The theoretical evidence suggests that such is possible because music activities are structured to engage mind and body in overcoming a challenge, generating feelings of achievement and pleasure in the process. However, the empirical evidence on such effects is less conclusive, reflecting the novelty of the field. I start by developing a model capable of generating hypotheses and framing the empirical work on the relationship between group singing and well-being. In this model, I integrate a conceptualization of well-being as a multidimensional concept with a conceptualization of group singing as a challenging multifaceted activity. The model suggests that well-being is promoted by group singing through the effects this activity has on its psychological, bodily and social correlates. However, the model also accounts for possible moderating effects of situational factors and individual differences. My empirical work focus first on clarifying that group singing involves the types of experiences that can generate feelings of hedonic, eudaimonic and social well-being. I then show that singing in a group relates to long-term measures of well-being, particularly social and eudaimonic. In the short-term, group singing is also consistently related to correlates of well-being. It is shown to increase positive affect, decrease negative affect, increase feelings of achievement and connection to others. Situational factors such as the social context in which the singing occurs are shown to have a moderating effect on endocrine effects of group singing. However, gender differences do not emerge. I also show that group singing can have effects over and above those of a similar, though non-musical, group activity. Overall, the evidence suggests a meaningful relationship between group singing and well-being.