Common Error 13: Ignoring Nested Model Difference When Interpreting Main Effect

Early methodological work addressing multicollinearity concerns in moderation analysis proposed omitting the components (X and M) and testing reduced models containing only the product term (Y = XM; Sharma et al., 1981). Some researchers adopted this approach when it appeared to enhance statistical significance, a practice that persists in approximately 26.64% of contemporary studies - an improvement from the 43.9% prevalence reported by Aguinis et al. (2017), yet still methodologically problematic. This specification suffers from fundamental limitations (Becker et al., 2023; Dawson, 2014). The Y-XM correlation inherently depends on the constituent variables' properties (Irwin & McClelland, 2001), potentially conflating main and interaction effects. This confounding becomes particularly acute when main and moderation effects share directional signs (Cohen, 1978), compromising the reliability of moderation tests. Consequently, scholars must include all constituent terms alongside their product in moderation analyses regardless of their statistical significance (Cronbach, 1987), as proper model specification supersedes collinearity concerns.

调节分析:常见问题十三  在解释主效应时忽视完整模型（含交互项）

当调节效应存在时，自变量（X）与因变量（Y）之间的关系是有条件的，它依赖于调节变量（M）的取值。然而，我们的综述发现约26.64%的研究在解释X的主效应时，错误地引用了从不包含交互项的基线模型中得到的系数。这个系数仅代表X对Y的平均效应，它掩盖了X的真实影响在不同M水平下的变化。在调节作用显著的情况下，使用这个平均值可能导致对X效应存在性或强度的错误结论（I类或II类错误）。

问题实质：

在调节分析的语境下，问“X对Y的总体效应是什么？”是一个错误的问题。正确的问题是：“X对Y的效应在M的不同条件下如何变化？”基线模型的主效应系数无法回答后一个问题。

解决建议：

Ÿ  在完整模型中解释主效应：任何关于X效应的实质性解释，都应基于包含了交互项（X*M）的完整模型。

Ÿ  进行简单斜率分析：当交互项显著，或理论上存在调节可能时，必须通过简单斜率分析来揭示并报告X的效应在M的不同关键取值（如均值、均值±1标准差）上是如何具体变化的。

Reference

• Aguinis, H., Edwards, J. R., & Bradley, K. J. (2017).      Improving our understanding of moderation and mediation in strategic management research. Organizational Research Methods, 20(4),      665-685.

• 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.

• 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.