当代中介分析:常见问题 三：是否提出微观中介假设，却用宏观模型检定，或使用宏观中介假设，却用微观模型检定？ by Shiau et al. (2026)
Source from: Shiau, W.-L., Shih, C.-H., Qureshi, I., Yuan, Y., & Teng, C.-I. (2026). Do matched and mismatched null hypothesis significance testing of mediators matter? An example of fintech continuance and guidelines for the future. Decision Support Systems, 211, 114764.
Guideline 3: Ensuring Alignment Between Hypotheses and Testing ModelsResearchers should verify that the mediation hypothesis and the testing model address the same mechanism under the same model scope. Micro hypotheses require corresponding micro tests, and macro hypotheses require corresponding macro tests. A mismatch should be resolved before interpreting the results. Researchers should provide an alignment statement explaining the hypothesis–test correspondence and apply the mediation decision tree only after that correspondence has been established. Revisions made after examining the results should be disclosed as post hoc changes.
指引三：确认假设与检定模型的一致性研究者应在解读结果之前，确认中介假设与检定模型是否相同。依照Shiau等人 (2026) 提出微观假设应对应微观分析，宏观假设应对应宏观分析。若两者不一致，即使取得显著结果，结论也可能未对应提出的理论假设。
研究者应该/不应该做的事
应做事项 | 不应做事项 |
1. If a mismatch occurs, make the hypothesis level and the testing level consistent. 1. 若出现不一致，应使假设层级与检定层级一致。 | 1. Use a macro model to test all mediation effects. 1. 使用宏观模型检定所有中介效果。 |
2. Recommend writing an alignment statement of the NHST consistency. 2. 建议撰写一致性说明，交代 NHST 中假设与检定的一致性。 | 2. Only write test mediation. 2. 仅写出「检定中介效果」。 |
Reference
Shiau, W.-L., Shih, C.-H., Qureshi, I., Yuan, Y., & Teng, C.-I. (2026). Do matched and mismatched null hypothesis significance testing of mediators matter? An example of fintech continuance and guidelines for the future. Decision Support Systems, 211, 114764.
转载本文请联系原作者获取授权，同时请注明本文来自萧文龙科学网博客。
链接地址：https://wap.sciencenet.cn/blog-3444471-1558383.html?mobile=1
收藏
