Abstract
Unfortunately, most researchers in leadership studies (and management in general) ignore one key design and estimation problem rendering parameter estimates uninterpretable: endogeneity. This chapter discusses the problem of endogeneity in depth and explains conditions that engender it. It shows how consistent causal estimates can be derived from the randomized experiment, in which endogeneity is eliminated by experimental design. It then reviews the reasons why estimates may become biased (i.e., inconsistent) in nonexperimental designs and presents a number of useful remedies for examining causal relations with nonexperimental data. This chapter is written in intuitive terms using nontechnical language to make it accessible to a large audience, irrespective of discipline.