This dissertation examines the mechanisms of formal and informal sources of social diffusion and its impact on fertility attitudes and outcomes in southern Ghana. More specifically, theories of structural modernization and diffusion of innovations are tested by examining the effects of ideational and social interaction mechanisms on fertility preferences and outcomes. Drawing from theories of structural modernization, structural economic changes and the role of the mass media, the determinants of fertility preferences are explored. The structure of core discussion networks of men and women and its implications for fertility are also investigated. Finally, a novel analytical model incorporates the complimentary effects of discussions of matters of childbearing within men's and women's respective informal social networks on one hand and the effects of spousal communication and spousal approval of family planning, on the other, on changes in couples' joint fertility intentions and outcomes.

Data are drawn from longitudinal surveys collected in six communities in southern Ghana. Reproductive-age women and their male partners were interviewed in three rounds over a three-year period. Ordinary least square regression, generalized linear and logistic regression models as well as controls for unmeasured variables through fixed and random-effects models are used to determine the impact of social diffusion on fertility.

Estimates from the analysis indicate that; (1) forces of modernization and mass media are associated with fertility preferences of men and women; (2) the content of information exchanged in personal social networks are positively associated with fertility attitudes and experiences, even after other factors are controlled; (3) men's social networks have a significant impact on couples joint fertility intentions both directly and indirectly through spousal communication about family and spousal approval of family, while that of women only affects couples joint fertility intentions indirectly. These effects are significant even after unobserved factors are controlled. The limitations of the analysis and implications for population policy are discussed.

Theories of social diffusion in general and informal discussions in social networks in particular are important determinants of fertility change in transitional societies. Associated with this change is the relevance of gendered social interaction and its mechanisms.