Social networks and their antecedents and consequences are phenomena located at multiple, intertwined levels. While past research has pointed out that different units of analysis are involved when doing social network research, mono-method studies have struggled to manage this complexity. Here, we discuss this problem from the perspective of mixed methods research. Specifically, we argue that qualitative and quantitative approaches to doing network research have their strengths for researching particular units of analysis; hence, the complexity of these multiple, intertwined levels may be better grasped using mixed methods designs.