On the Nuisance of Control Variables in Causal Regression Analysis

Presentor Biography

Paul Hünermund is an Assistant Professor of Strategy and Innovation at Copenhagen Business School. In his research, Dr. Hünermund studies how firms can leverage new technologies in machine learning and artificial intelligence for value creation and competitive advantage. His research has been published in the Journal of Management Studies, the Econometrics Journal, Research Policy, Journal of Product Innovation Management, International Journal of Industrial Organization, MIT Sloan Management Review, and Harvard Business Review, among others. He serves on the editorial board of the Journal of Causal Inference and the executive team of the Technology and Innovation Management (TIM) division at the Academy of Management. Dr. Hünermund pursued his studies in economics at the University of Mannheim, HEC Lausanne, and NYU Stern School of Business. He holds a Ph.D. in business economics from KU Leuven in Belgium.

Abstract

Control variables are included in regression analyses to estimate the causal effect of a treatment on an outcome. In this article, we argue that the estimated effect sizes of controls are unlikely to have a causal interpretation themselves, though. This is because even valid controls are possibly endogenous and represent a combination of several different causal mechanisms operating jointly on the outcome, which is hard to interpret theoretically. Therefore, we recommend refraining from interpreting the marginal effects of controls and focusing on the main variables of interest, for which a plausible identification argument can be established. To prevent erroneous managerial or policy implications, coefficients of control variables should be clearly marked as not having a causal interpretation or omitted from regression tables altogether. Moreover, we advise against using control variable estimates for subsequent theory building and meta-analyses.

Quick Chat

How to Access Webcast Lecture

Access to Live Online Webcast Lectures is available for Institutional Premium, Institutional Basic, and Affiliate Members. To access the live broadcast, click on the Access Live Events tab in your CARMA User Area.

If you are not a current member, please visit our Institutional Membership page or Affiliate page to learn more about how to gain access to our CARMA content.