2024/12/02 by Paul Hünermund, Beyers Louw, Mikko Rönkkö · 1 voice · 21 citations
Computer Science · Mathematics · #Advanced Causal Inference Techniques #Artificial intelligence #Bayesian Modeling and Causal Inference #Cognitive Science and Mapping #Computer science #Control (management) #Econometrics #Economics #Management science #Mathematics
paper · doi:10.1016/j.leaqua.2024.101845
published in The Leadership Quarterly 36(2), 101845 (Elsevier BV)
openalex publication_date 2024/12/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/15
The Leadership Quarterly and the management community more broadly prioritize identifying causal relationships to inform effective leadership practices. Despite the availability of more refined causal identification strategies, such as instrumental variables or natural experiments, control variables remain a common strategy in leadership research. The current literature generally agrees that control variables should be chosen based on theory and that these choices should be reported transparently. However, the literature provides little guidance on how specifically potential controls can be identified, how many control variables should be used, and whether a potential control variable should be included. Consequently, the current empirical literature is not fully transparent on how controls are selected and may be contaminated with bad controls that compromise causal inference. Causal diagrams provide a transparent framework to address these issues. This article introduces causal diagrams for leadership and management researchers and presents a workflow for finding an appropriate set of control variables.