In this thesis I set out to further our understanding of two broad questions, 1) why it is that sexual selection favours the evolution of particular traits and 2) why do the patterns of sexual selection on such traits differ between groups and populations? Specifically, I focused on the role that variation in intrasexual competitive structure, the non-random distribution of socio-competitive environments across individuals, plays in shaping variation in the operation of sexual selection both within and across populations. I explore the roles of three main determinants of competitive structure, namely; population structure, polyandry and non-random variation in the distribution of the intensity of competition. To achieve this, I used a combination of empirical and theoretical tools, using the model system Red junglefowl, Gallus gallus. Throughout this thesis I both develop and employ network quantitative tools as a framework to describe variation in intrasexual competitive structure. Overall, this thesis demonstrates a complex relationship between competitive structure and the operation of sexual selection. This structure can modify the strength and direction of sexual selection operating on phenotypic traits, obscure the operation of selection at the population level and influence the relative roles of pre- and post-copulatory sexual selection. Furthermore, this thesis explores how differences in local ecology can shape competitive structure itself and in turn shape sexual selection. In doing so, this thesis sheds some light on the role that variation in competitive structure may play in shaping the operation of sexual selection both within and between populations and generating the great diversity in sexually-selected traits and behaviours between populations.