This study explores the spontaneous singing of three- and four-year-old children at home, with emphasis on how young children use singing in their everyday lives. Spontaneous singing pervades the everyday lives of young children and can provide insights into a child's musical and extra-musical experience at home. Although several studies have examined spontaneous singing in educational settings, young children's musical lives at home are rarely studied in detail. The home is a difficult space to access, and data collection methods often rely on parental reporting. As a result, some types of singing have been overlooked. Located within the sociocultural theoretical tradition, this thesis draws on and develops theories of musical agency to explore how children act musically to engage with others and manage their own experience. Audio data were collected using LENA all-day recording technology supplemented by semi-structured parental interviews. Over 183 hours of audio recording were collected from 15 children (7 boys, 8 girls), aged between 3:0 and 4:10 years (average age 3:8). The children were recorded for continuous periods during their normal everyday routines. The recordings contained more than nine hours of spontaneous singing in total. The data were analysed using qualitative thematic analysis with an element of embedded numerical analysis. Interpretive analysis indicated that the children sang to act on themselves and manage social interactions. Spontaneous singing was used as a tool through which the children could realise personal and social agency and influence themselves and others. The children used different modes of singing in social and solitary contexts, demonstrating knowledge of culturally meaningful ways of singing. The home musical environment, and particularly parental singing, appeared to influence the way young children use singing in their everyday lives. This research used an innovative methodology to access young children’s singing in the home. The findings contribute to a greater understanding of young children’s musical behaviours and the home musical lives of young children. Further, the thesis provides an original contribution to the understanding of how young children use spontaneous singing as musical agents acting in and on the world around them. This research has educational implications relating to the way young children’s musicality is understood and encouraged and the importance of music in young children’s lives.