Why theory matters for causal inference? Rethinking endogeneity in entrepreneurship research
指出创业研究中的内生性不仅是统计问题，更是概念问题，强调理论在明确构念、机制和边界条件中的核心作用，并结合因果模型提升识别能力，对创业学者和管理者均有启发。
Abstract Endogeneity in entrepreneurship research is often treated as a statistical complication addressable through advanced econometric tools. This commentary argues that such an approach overlooks a deeper issue: endogeneity is conceptual before it is statistical. Because entrepreneurial phenomena involve reciprocal relationships, evolving mechanisms, and context‐dependent processes, biased estimates frequently stem from underspecified constructs and unclear causal logic. I contend that theory, sufficiently precise to specify constructs, articulate mechanisms, and establish temporal ordering and boundary conditions, is the primary tool for reducing endogeneity in empirical estimation. Integrating theory with structural causal modeling and rigorous empirical design strengthens identification while enhancing explanatory value. I conclude with practical recommendations for scholars, emphasizing theory's central role in producing credible, cumulative knowledge in entrepreneurship research. Managerial Summary Entrepreneurs and managers often rely on data to understand what drives venture success, but data alone can be misleading if the underlying assumptions about cause and effect are unclear. This article explains why strong theory—clear ideas about how and why things work—is essential for drawing reliable conclusions from evidence. Endogeneity, a common problem in business research, occurs when factors influence each other in ways that make results appear stronger or weaker than they are. By using theory to map out mechanisms and likely confounding factors, leaders can make better decisions about which actions truly create value. The article also shows how tools like causal modeling can combine conceptual clarity with rigorous analysis, helping organizations design strategies that are both evidence‐based and trustworthy.