Towards Stronger Causal Claims in Management Research: Causal Triangulation Instead of Causal Identification

Abstract

This article addresses widespread concerns about the reliability and strength of many causal claims made in management research. We first critically review the three prevalent forms of theorizing used to identify causal relationships in this field, i.e., propositional, configurational, and process approaches to causation. Highlighting the strengths and limitations of these approaches, we show that while no single approach is sufficient by itself as the basis for robust causal claims, researchers can nonetheless enhance and strengthen claims significantly by combining approaches and thus subjecting them to multiple criteria for drawing robust inferences. We emphasize the risks of continuing with narrow monolithic approaches, using examples of weak claims to show how these could have been strengthened (or abandoned) if the researchers had followed our proposed model of causal triangulation. Finally, we elucidate the practical benefits for management researchers and stakeholders in society of adopting this theoretically pluralistic approach to causation.

Panacea or Dangerous Practice: A Counterpoint to Hanisch's Argument for Prescriptive Theorizing

“…Above we have argued that elaborations of the distinction between description and prescription veils a deeper‐seated homology between both, rooted in an instrumental logic that reduces complexity to isolated factors which, for the sake of analysis, are taken to be real, and then set into causal relations (Cornelissen, 2023; Cornelissen and Kaandorp, 2023). But this logic, and the prescriptions derived from them, begins to idle in face of the unknowability of future events.…”

Unlocking the power of diversity for supply chain knowledge: Is pluralism in theorizing styles the key?

“…First of all, a pluralistic system based on multiple styles of theorizing leads to better knowledge by joining up efforts—across styles—in pursuit of a common epistemic goal where that is shared between them. To this end, Cornelissen (2024) and Cornelissen and Kaandorp (2023) recently argued that aligning propositional, process, and configurational theorizing styles (all styles that are based on the epistemic goal of explanation) allows researchers to form drastically more probable causal explanations of phenomena. Hence, for the epistemic aim of explanation and prediction, such “triangulation” offsets the threats to validity and inferential limits that come with relying on one of these styles only, such as the strong likelihood that by following the propositional style in isolation, it generates false positives.…”