Collective behavior is widespread in nature and examples include schools of fish and nest building in social insects. Although collective behavior and other group-level phenotypes are assumed to be shaped by selection, we do not know to what degree they are heritable and how selection acts on them. Furthermore, we have identified relatively few genes underlying variation in group-level phenotypes, hindering our understanding of the molecular mechanisms by which genes influence these traits and how they evolve. Elucidating the genetic architecture underlying group-level phenotypes is especially diffuclt because it depends on the genotypes of multiple interacting individuals. In this thesis, we use a new pharaoh ant (Monomorium pharaonis) laboratory mapping population to investigate the genetic architecture underlying a number of group-level phenotypes. These group-level phenotypes include collective behaviors (foraging, aggression, and exploration) and cuticular hydrocarbons, which play a vital role in chemical communication within social insect colonies. We demomonstrate that these phenotypes are heritable and have fitness consequences – the two prerequites for evolution via natural selection. Next, we perform genome-wide association studies to identify many interesting candidate genes associated with variation in group-level phenotypes, including genes associated with variation in collective behavior that have been implicated in neurological disorders or in the development of the visual system. Next, we explore how the genetic makeup of groups affects collective behavior and find that the specific combinations of genotypes within a group influence group-level outputs. Finally, we focus on the important social interactions between nurse workers and larvae. We first explore the evolutionary origin of sibling care and find that it likely shares a genetic basis with maternal care. Next, we demonstare that some nurse workers are behaviorally specicialized to care for larvae of different development stages and identify genes differentially expressed between nurses caring for different larval types. These specialized nurse workers likely play a large role in regulating divion of labor within social insect colonies. Overall, this work begins to identify the genetic architecture underlying group-level phenotypes, highlights the importance of within-group genetic composition on group-level output, and demonstrates the important role of nurse workers in modulating group-level phenotypes.