Welcome Prof. Urbig to the Brown Bag Seminar at the Schumpeter School of Business and Economics.
Last week, we were very pleased to welcome Prof. Diemo Urbig to our Brown Bag Seminar at the Schumpeter School of Business and Economics.
For Prof. Urbig, the visit was a return to a familiar place: before taking up his current position, he spent several years as assistant professor at our research center. It was therefore especially nice to welcome him back and to hear more about his current work.
His talk was based on an in-press paper in the Journal of Management and focused on nonlinear and interaction effects in empirical management research.
The topic may sound technical, but the underlying issue is highly relevant for many quantitative studies. Management scholars often study relationships that are not simply linear: effects may become stronger or weaker at different levels, or they may depend on other variables. In such cases, the way a model is specified can strongly affect the conclusions researchers draw.
Prof. Urbig and his co-authors show that this is not a minor issue. In a review of 548 quantitative articles published between 2021 and 2023 in Academy of Management Journal, Journal of Management, and Strategic Management Journal, they found that around 73% tested nonlinear or interaction effects, while only 3% included the corresponding nonlinear or interactive control variables. Their work shows that omitting such controls can distort statistical tests, effect-size estimates, and, in some cases, even reverse substantive conclusions.
Against this background, the paper introduces a five-step guide for identifying and integrating relevant nonlinear and interactive controls in a systematic and theory-driven way. The broader message of the seminar was clear: careful model specification is not just a technical detail, but central to robust empirical findings and theory development.
The presentation led to an engaged discussion on causal inference, model specification, and the interpretation of complex empirical models.
Many thanks to Prof. Urbig for the visit and the insightful exchange.
