Common Error 1: Using Moderation Analysis Driven by Problematic Motives

A prevalent issue is the use of moderation analysis without adequate theoretical justification (Dawson, 2014; Memon et al., 2019; Becker et al., 2023). A crucial first step is determining if a moderation model is theoretically warranted (Memon et al., 2019). Our review found 29.2% of articles were driven by questionable motives, such as using a “trial and error” approach to find moderators. This post-hoc strategy often produces meaningless conclusions, even with statistically significant results. Instead, moderation should be motivated by research needs and solid theory (Frazier et al., 2004), such as to explain inconsistent relationships or test novel propositions (Andersson et al., 2014). Researchers must explicitly provide a theoretical rationale for proposed moderators (Memon et al., 2019).

常见问题一：调节分析的研究动机不当

许多研究（如Dawson, 2014; Memon et al., 2019; Becker et al., 2023）指出，调节分析常因动机不当而被误用。是否采用调节模型，首先应基于充分的理论依据（Memon et al., 2019）。然而我们发现，仍有 29.2% 的相关研究出于可疑的动机使用这一方法，例如通过“试错”来寻找可能的调节变量。这种缺乏理论支撑的事后策略，即使得出统计上显著的调节效应，也可能导致无意义的结论。调节分析应根植于实际研究需求与扎实的理论框架（Frazier et al., 2004），例如用于解释不同研究中预测变量与结果变量之间意外微弱或不一致的关系，或用于检验新的理论命题（Andersson et al., 2014）。因此，研究者应在提出调节变量时明确说明其理论依据——例如引用先前不一致的发现，或指出相关的背景因素（Memon et al., 2019）。

问题实质：

调节分析若缺乏理论驱动，仅凭“试错”或追求方法复杂度的动机，容易导致分析结果脱离实际研究问题，即使统计显著也缺乏解释力与理论价值。

解决建议：

Ÿ  理论先行：在设计调节分析前，明确该模型是否回应了理论上的真实需求，例如解释不一致的研究结果或拓展某一理论的情境边界。

Ÿ  动机透明化：在论文中清晰阐述为何引入调节变量，避免仅描述为“探索性分析”或“丰富模型”。

Ÿ  拒绝随意尝试：不应为增加分析层次或模仿他人研究而随意加入调节变量，每一步分析都应有相应的理论铺垫。

Reference

• Andersson, U., Cuervo-Cazurra, A., & Nielsen, B. (2014).      From the Editors: Explaining interaction effects within and across levels      of analysis. Journal of International Business Studies, 45(9),      1063-1071.

• Becker, J. M., Cheah, J. H., Gholamzade, R., Ringle, C. M.,      & Sarstedt, M. (2023). PLS-SEM's most wanted guidance. International      Journal of Contemporary Hospitality Management, 35(1), 321-346.

• Dawson, J. F. (2014). Moderation in management research: What,      why, when, and how. Journal of Business and Psychology, 29(1),      1-19.

• Frazier, P. A., Tix, A. P., & Barron, K. E. (2004).      Testing moderator and mediator effects in counseling psychology research. Journal      of Counseling Psychology, 51(1), 115-134.

• Memon, M. A., Cheah, J. H., Ramayah, T., Ting, H., Chuah, F.,      & Cham, T. H. (2019). Moderation analysis: Issues and guidelines. Journal      of Applied Structural Equation Modeling, 3(1), i-xi.

• Xu, Y., & Shiau, W. L. (2026). Moderation analysis in      business and management research: Common issues, solutions, and guidelines      for future research. International Journal of Information      Management, 86, 102995.