博文
[Knowledge Sharing] Guideline 6:
||
當代中介分析:常見問題 六：是否未依中介類型解釋理論意涵？by Shiau et al. (2026) and Zhao et al. (2010)
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.
Zhao, X., Lynch, J. G., Jr., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of Consumer Research, 37(2), 197–206. https://doi.org/10.1086/651257
Guideline 6: Linking Mediation Results to Theory Building and Decision MakingResearchers should use the result type to guide theory-building and explain its implications for decision-making. Zhao et al. (2010) identify unexplained direct effects as clues for investigating additional mediators. Complementary and competitive mediation support the proposed indirect mechanism while indicating a remaining direct effect. Direct-only results call for reconsideration of the proposed mediator and investigation of alternative mechanisms. Within Zhao et al.’s framework, indirect-only mediation suggests a lower likelihood of an omitted mediator, but this pattern does not establish that all relevant mechanisms have been identified. No-effect results warrant reassessment of the proposed relationships and study conditions. Theoretical and practical conclusions should remain within the limits of the evidence.
指引六：將中介結果連結至理論建構與決策研究者應以中介分析類型引導理論建構，並說明決策意涵。Zhao等人 (2010) 指出，互補與競爭中介支持提出的間接機制，同時保留顯著的直接效果；僅直接效果需要重新檢視原先提出的中介機制，並探索其他解釋。僅間接效果提示遺漏中介的可能性較低，但結果不足以證明均納入。若直接與間接效果均未達顯著，則應重新評估理論關係與研究條件。理論及實務結論均應符合研究證據支持的範圍。
研究者應該/不應該做的事
應做事項 | 不應做事項 |
1. Use mediation type as the basis for theory-building. 1. 以中介類型作為理論建構。 | 1. Theory-building overreach by bypassing mediation type. 1. 跳過中介類型，對理論建構作出過度推論。 |
2. Link the theory-building to decision-making by stating that an incomplete or incorrect theoretical framework may bias understanding and affect decisions. 2. 將理論建構連結至決策，說明不完整或不正確的理論架構可能使理解產生偏誤，並影響決策。 | 2. Only claim contribution without improving the completeness of the theoretical framework through consistent mediation. 2. 僅宣稱研究貢獻，卻未透過假設與檢定一致的中介分析，改善理論架構的完整性。 |
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.
Zhao, X., Lynch, J. G., Jr., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of Consumer Research, 37(2), 197–206. https://doi.org/10.1086/651257
https://blog.sciencenet.cn/blog-3444471-1559453.html
上一篇：[Knowledge Sharing] Guideline 5:
