中介作用的意涵及其資料分析策略之探究

The Conceptualization and Data Analytical Strategies of Mediation

張再明(Tsay-Ming Chang)；曾迎新(Yin-Hsing Tseng)

中介作用 ； 因果步驟 ； 係數乘積 ； 拔鞋法 ； medation ； causal steps ； product of coefficients ； bootstrapping

家庭教育與諮商學刊

16期（2014 / 06 / 01）

29 - 43

繁體中文

背景：中介作用的概念長久來即見於心理學的論述中，但實證性地探究中介作用則 直至1980年代，Baron與Kenny（1986）、James與Brett（1984）及Judd與Kenny（1981）等幾位學者構建資料分析的策略後才開始普及。目前中介作用的探究已廣見於社會科學、健康學、心理學、教育及社會學的研究中（Wu & Zumbo, 2008）。然遺憾的是，中介作用的因果本質常被忽視或誤解，因而導致許多誤用或錯誤的解釋（Frazier, Tix, & Baron, 2004；MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002）。目的：本文的目的在於釐清中介作用的基本概念，並探究驗證中介作用的資料分析策略，期能有助於研究者進行中介作用的探究。討論焦點：除釐清中介作用的基本概念外，本文著重於討論Kenny的因果步驟、係數乘積考驗法中的Sobel考驗法及拔鞋法（bootstrapping）等三種檢證中介作用效果的方法，並比較其優缺點。結論：檢驗中介作用的方法均有其使用上的限制，進行分析時必須審慎。檢證中介作用效果時，可考慮採行多重的檢驗方法。同時並用數種檢驗方法，其所獲致的結果或 許更為真確。

Background: The methodology of mediation is commonly used in social science, health, psychological, educational, and sociological research (Wu & Zumbo, 2008). Unfortunately, mediation effects have been widely misunderstood and misused as data analytical tools and statistical hypothesis testing with little theoretical and methodological considerations. (Frazier, Tix, & Baron, 2004; MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002). Aims: The current paper aims to clarify the concept of mediation and investigate data analytical strategies for testing mediation effects. Hopefully, the results could assist researchers in designing and conducting their studies when mediation effects are examined. Discussion focus: In addition to clarifying the concept of mediation, this paper compares three analytical approaches for testing mediation effects, namely, Kenny’s causal steps analysis, Sobel test, and bootstrap procedure. Besides data analyzing procedures, strengths and weaknesses of these methods are discussed. Conclusion: These data analytical methods all have their limitations. Researchers should use them with caution. An integrated strategy is suggested when mediation effects are tested. By examining the testing results obtained from each approach simultaneously, it is more likely to gain a valid result.

1. Kenny, D. A. (2009). Mediation Marco. Retrieved from http://davidakenny.net/dtt/mediate.htm
2. Woodworth, R. S. (1928). Dynamic psychology. In C. Murchison (Ed.), Psychology of 1925. Worcester, MA: Clark University Press
3. 4 種中介效應分析對比（2012 年8 月5 日）：4 種中介效應分析對比。經濟學家。取自http://bbs.jjxj.org/thread-1165128-1-1.html
4. Aroian, L. A. (1944). The probability function of the product of two normally distributed variables. Annals of Mathematical Statistics, 18, 265-271
5. Preacher, K. J., & Leonardelli, G. J. (2001). Calculation for the Sobel test: An interactive calculation tool for mediation tests. Retrieved from http://www.quantpsy.org/sobel/sobel.htm
6. Wuensch, K. L. (2003). Statistical Tests of Models That Include Mediating Variables. Retrieved from http://core.ecu.edu/psyc/wuenschk/MV/multReg/MediationModels.doc
7. Arbuckle, J. L.,Wothke, W.(1999).AMOS 4.0 user's guide.Chicago:SPSS.
8. Baron, R. M.,Kenny, D. A.(1986).The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.Journal of Personality & Social Psychology,51,1173-1182.
9. Bollen, K. A.,Stine, R.(1990).Direct and indirect effects: Classical and bootstrap estimates of variability.Sociological Methodology,20,115-140.
10. Collins, L. M.,Graham, J. W.,Flaherty, B. P.(1998).An alternative framework for defining mediation.Multivariate Behavioral Research,33,295-313.
11. Frazier, P. A.,Tix, A. P.,Baron, K. E.(2004).Testing moderator and mediator effects in counseling psychology.Journal of Counselling Psychology,51,115-134.
12. Gilbert, D.(Ed.),Fiske, S.(Ed.),Lindzey, G.(Ed.)(1998).The handbook of social psychology.Boston, MA:McGraw-Hill.
13. Holmbeck, G. N.(1997).Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures.Journal of Consulting & Clinical Psychology,65,599-610.
14. Holmbeck, G. N.(2002).Post-hoc probing of significant moderational and mediational effects in studies of pediatric populations.Journal of Pediatric Psychology,27,87-96.
15. Hoyle, R. H.(Ed.)(1999).Statistical strategies for small sample research.Newbury Park:Sage.
16. James, L. R.,Brett, J. M.(1984).Mediators, moderators, and tests for mediation.Journal of Applied Psychology,69,307-321.
17. Judd, C. M.,Kenny, D. A.(1981).Process analysis: Estimating mediation in treatment evaluations.Evaluation Review,5,602-619.
18. Kraemer, H. C.,Wilson, G. T.,Fairburn, C. G.,Agras, W. S.(2002).Mediators and moderators of treatment effects in randomized clinical trials.Archives of General Psychiatry,59,877-883.
19. Linn, R. L.(Ed.)(1989).Educational measurement.New York:American Council of Education.
20. Lockwood, C. M.,MacKinnon, D. P.(1998).Bootstrapping the standard error of the mediated effect.Proceedings of the 23rd annual meeting of SAS Users Group International,Cary, NC:
21. Long, J. S.(Ed.)(1988).Common problems/proper solutions: avoiding error in quantitative research.Beverly Hill, CA:Sage.
22. MacKinnon, D. P.(2008).Introduction to Statistical Mediation Analysis.New York:Lawrence Erlbaum Associates.
23. MacKinnon, D. P.(1994).Analysis of mediating variables in prevention and intervention research.NIDA Research Monograph,139,127-153.
24. MacKinnon, D. P.,Fairchild, A. J.,Fritz, M. S.(2007).Mediation analysis.Annual Review of Psychology,58,593-614.
25. MacKinnon, D. P.,Krull, J. L.,Lockwood, C. M.(2000).Equivalence of mediation, confounding, and suppression effect.Prevention Sciences,1,173-181.
26. MacKinnon, D. P.,Lockwood, C. M.,Hoffman, J. M.,West, S. G.,Sheets, V.(2002).A comparisons of methods to test mediation and other intervening variable effects.Psychological Methods,7,83-104.
27. MacKinnon, D. P.,Warsi, G.,Dwyer, J. H.(1995).A simulation study of mediated effect measures.Multivariate Behavioral Research,30,41-62.
28. Mallinckrodt, B.,Abraham, W. T.,Wei, M.,Russell, D. W.(2006).Advances in testing the statistical significance of mediation effects.Journal of Counselling Psychology,53,372-378.
29. Muller, D.,Judd, C. M.,Yzerbyt, V. Y.(2005).When moderation is mediated mediation is moderated.Journal of Personality and Social Psychology,89,852-863.
30. Preacher, K. J.,Hayes, A. F.(2004).SPSS and SAS procedures for estimating indirect effects in simple mediation models.Behavior Research Methods, Instruments, & Computers,36(4),717-731.
31. Preacher, K. J.,Hayes, A. F.(2008).Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models.Behavior Research Methods,40,879-891.
32. Reis, H. T.(Ed.),Judd, C. M.(Ed.)(2000).Handbook of research methods in social and personality psychology.New York:Cambridge University Press.
33. Rose, B. M.,Holmbeck, G. N.,Coakley, R. M.,Franks, E. A.(2004).Mediator and moderator effects in developmental and behavioral pediatric research.Developmental and Behavioral Pediatrics,25,58-67.
34. Rozeboom, W. W.(1956).Mediation variables in scientific theory.Psychological Review,63,249-264.
35. Sansone, C.(Ed.),Morf, C.(Ed.),Panter, A. T.(Ed.)(2003).Handbook of methods in social psychology.Thousand Oaks, CA:Sage.
36. Shrout, P. E.,Bolger, N.(2002).Mediation in experimental and nonexperimental studies: New procedures and recommendations.Psychological Methods,7,422-445.
37. Sobel, M. E.(1982).Asymptotic confidence intervals for indirect effects in structural equation models.Sociological methodology,13,290-312.
38. Spencer, S. J.,Zanna, M. P.,Fong, G. T.(2005).Establishing a causal chain: why experiments are often more effective than mediational analysis in examining psychological processes.Journal of Personality and Social Psychology,89,845-851.
39. Stone, C. A.,Sobel, M. E.(1990).The robustness of total indirect effects in covariance structure models estimated with maximum likelihood.Psychometrika,55,337-352.
40. Wu, A. D.,Zumbo, B. D.(2008).Understanding and Using Mediators and Moderators.Social Indicators Research,87(3),367-392.

1. 邱仕凱(2018)。家庭社經地位與學業成就關係之研究－以學業發展自我效能為中介變數。教育學誌，40，59-120。