Causal inference in any science is about differentiating among alternative explanations. To discern among these explanations one needs good theory, good data, and models that can support scientific discourse about the mechanisms of an effect. This chapter delineates how this general process can be contextualized to make causal inferences for social network analysis. Specifically, the chapter focuses on two fundamental processes associated with networks: the influence of network members on one’s belief or behavior, and the selection of network members with whom to establish a tie. The chapter discusses how these two different yet intertwined processes create challenges for causal inferences, and possible approaches and techniques that can be undertaken to mitigate potential bias. Finally, given the almost inevitable debates about causal inferences for social network analysis, the chapter implores network scientists to use transparent scientific models and sensitivity analysis to inform the discourse for sense-making by research communities and practitioners.