2022

Hünermund, Paul; Louw, Beyers; Rönkkö, Mikko

Control variables have a central role when empirical data are used to support causal claims in management research. The current literature has been intransparent in so far as to how control variables should be chosen, how many control variables should be chosen and whether a potential control variable should be included. Causal diagrams provide a transparent framework on how to select control variables for causal identification. This article delineates how causal graphs can inform researchers in leadership and management in finding the correct set of control variables and possible solutions in the case that causal identification is not possible or when causal identification requires unobserved variables.