The age old problem of disentangling cause and effect from mere association remains a central concern for quantitative researchers. Indeed, most estimation procedures we rely on to test theory rest on the cornerstone assumption that predictors should be exogenous—in other words, on the assumption that any observed covariation between an independent and dependent variable should not be due to reverse causality, simultaneous relationships, or to a third (omitted) variable. The cost of violating this assumption is high as it leads to biased estimates rendering the interpretation of results meaningless. Outside of certain fields (e.g., economics), this problem of endogeneity has yet to be properly addressed. Even in context of experiments—the gold standard for testing causality—endogeneity is still be a concern in the context of mediation analysis. In the course of this workshop, we will look at the nature of this problem and review some of the methods and designs which can allow to deal with endogeneity—either through statistical correction (e.g., using instrumental variable regression), or through creating equivalent comparison groups (e.g., with difference-in-differences estimation). We will look at how such designs and methods have been applied and go through practical examples (using Stata) in terms of how to conduct such analyses. Ideally, participants should have a solid background in regression analysis.
Philippe is currently an Associate Professor at Emlyon Business School, in France, and on sabbatical at ESADE, in Spain, for the current academic year. Philippe's research interests focus on the cognitive processes underlying the leadership process. His current research focuses on understanding the evaluation and selection of leaders, the motivation to lead, charisma and its effects, gender stereotypes, as well as on methodological topics. In particular, he is interested in how quasi-experimental methods can be leveraged to understand causal mechanisms. He is an Associate Editor and Method Advisor at The Leadership Quarterly. His research has been published in top refereed journals, such as the Academy of Management Journal, the Annual Review of Organizational Psychology and Organizational Behavior, and The Leadership Quarterly among others.
9:00-10:15 session 1
10:30-11:45 session 2
11:45-14:00 long lunch break
14:00-15:15 session 3
15:30-16:45 session 4
More about Prof. Dr. Philippe Jacquart: here