Science of Emotion and Sensibility (감성과학)

Korean Society for Emotion and Sensibility (한국감성과학회)
권호 표지
DOI QR코드

DOI QR Code

Multilevel Mediation Analysis: Statistical Methods, Analytic Procedure, and a Real Example

다층자료의 매개효과 분석: 통계방법, 분석절차 및 실례

  • Park, Sun-Mi (Department of Education, Chonbuk National University) ;
  • Bak, Byung-Gee (Department of Education, Chonbuk National University)
  • Received : 2016.06.27
  • Accepted : 2016.08.22
  • Published : 2016.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of this study was to propose a proper method for the multilevel mediation analysis, for which the hierarchical method should be utilized, then MLM (multilevel modeling) approach as a hierarchical method has been popularly utilized until MSEM (multilevel structural equation modeling) approach was not proposed. This purpose was covered by three research questions about statistical methods, analytic procedure, and real example. First, MSEM statistical method was preferred to MLM method for its estimation accuracy and analytic flexibility. Second, the four-step procedures of model building, assumption examination, model comparison, and coefficient testing were proposed for the multilevel mediation analysis. Third, the real data of 2695 students of elementary and secondary schools and 89 teachers were analyzed in the multilevel directions of $2{\rightarrow}2{\rightarrow}1$ and $1{\rightarrow}1{\rightarrow}2$. Out of these directions of $2{\rightarrow}2{\rightarrow}1$, and $1{\rightarrow}1{\rightarrow}2$ model, only the coefficient of $2{\rightarrow}2{\rightarrow}1$ model was significant at the 95% CI. Mplus programs used for the real example are attached on the Appendix. Based on the results, significance and limitations of this study, were discussed in detail.

본 연구의 목적은 다층자료 매개효과의 분석 방법을 제안하는 것이다. 연구내용은 다층자료 매개효과의 통계방법 탐색, 분석절차 제안, 그리고 분석의 실례 제시 등 세 가지다. 첫째, MLM (multilevel modeling)과 MSEM (multilevel structural equation modeling) 중에서 어떤 방법이 다층자료의 매개효과 분석에 유용한지 탐색하였다. MSEM은 MLM의 약점을 극복한 것으로서 유용한 다층 매개효과 분석방법이었다. 둘째, 다층자료 매개효과의 분석절차를 연구모형설정, 전제조건 검토, 모형검증, 계수검증의 4단계로 전개하였다. 셋째, 매개효과 분석의 실례에 사용된 자료는 2,695명의 초중등 학생과 88명의 학급교사로 구성되었다. 분석 실례로 2층${\rightarrow}$2층${\rightarrow}$1층과 1층${\rightarrow}$1층${\rightarrow}$2층 두 가지를 제시하였다. 2층${\rightarrow}$2층${\rightarrow}$1층과 1층${\rightarrow}$1층${\rightarrow}$2층 모형은 완전매개모형이 지지되었지만, 2층${\rightarrow}$2층${\rightarrow}$1층 모형의 매개효과 계수만 95% 신뢰구간에서 유의하였다. 분석 실례에 사용된 Mplus 프로그램은 부록에 제시하였다. 연구결과를 기초로 본 연구의 의의와 제한점, 그리고 후속연구의 방향이 논의되었다.

Keywords

References

  1. Bae, B. R. (2011). Structural equation modeling with Amos 19: Principles and practice. Seoul, Korea: Chungram.
  2. Barrett, P. (2007). Structural equation modeling: Adjudging model fit. Personality and Individual Differences, 42(5), 815-824. https://doi.org/10.1016/j.paid.2006.09.018
  3. Bentler, P. M. (1976). Multistructure statistical model applied to factor analysis. Multivariate Behavioral Research, 11(1), 23-25.
  4. Bentler, P. M. & Chou, C. P. (1987). Practical issues in structural modeling. Sociological Methods and Research, 16(1), 78-117. https://doi.org/10.1177/0049124187016001004
  5. Braun, S., Peus, C., Weiswiler, S., & Frey, D. (2013). Transformational leadership, job satisfaction and team performance: A multilevel mediation model of trust. The Leadership Quarterly, 24(1), 270-283. https://doi.org/10.1016/j.leaqua.2012.11.006
  6. Burnham, K, P. & Anderson, D. R. (2002). Model Selection and Multimodel Inference. New York: Springer.
  7. Burnham, K. P. & Anderson, D. R.(2004). Multimodel inference: Understanding AIC and BIC in model selection. Sociological Methods Research, 33(2), 261-304. https://doi.org/10.1177/0049124104268644
  8. Chen, F., Bollen, K. A., Paxton, P., Curran, P., & Kirby, J. (2001). Improper solutions in structural equation modeling: Causes, consequences and strategies. Sociological Methods Research, 29(4), 468-508. https://doi.org/10.1177/0049124101029004003
  9. Cruz, K. S. & Pil, F. K. (2011). Team design and stress: A multilevel analysis. Human Relations, 64(10), 1265-1289. https://doi.org/10.1177/0018726711409264
  10. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. National Communication Association, 76(4), 408-420.
  11. Hong, S. H. (2000). The criteria for selecting appropriate fit indices in structural equation modeling and their rationales. Korean Journal of Clinical Psychology, 19(1), 161-177.
  12. Hwang, J. & Hopkins, K. (2012). Organizational inclusion, commitment and turnover among child welfare workers: A multilevel mediation analysis. Administration in Social Work, 36, 23-39. https://doi.org/10.1080/03643107.2010.537439
  13. Joo, M. J. & Lee, J. S. (2014). The relationship between older driver's BIS/BAS and life satisfaction: Mediating effects of mobility and affectivity. Korean Journal of the Science of Emotion & Sensibility, 17(4), 29-40.
  14. Kim, K. (2011). Comparisons Between the Longitudinal Multilevel Models and the Latent Growth Models and the Realization of Piecewise Growth Modeling: Longitudinal Changes and Influential Factors in Anxiety in Korean Adolescence. Unpublished master thesis, Korea University, Korea.
  15. Kolenikov, S. & Bollen, K. A. (2012). Is a Heywood case a symptom of misspecification? Sociological Methods Research, 41(1), 124-167. https://doi.org/10.1177/0049124112442138
  16. Krull, J. L. & MacKinnon, D. P. (2001). Multilevel modeling of individual and group level mediated effects. Multivariate Behavioral Research, 36(2), 249-277. https://doi.org/10.1207/S15327906MBR3602_06
  17. Lee, C. H., Sohn, Y. W., & Rim, H. B. (2015). The influence of cultural similarity and empathy on helping intention: Testing the moderated mediating effect of cosmopolitanism. Korean Journal of the Science of Emotion & Sensibility, 18(4), 35-46. https://doi.org/10.14695/KJSOS.2015.18.4.35
  18. Lee, H. A. & Lee, K. H. (2009). The Relationship among job stress, job satisfaction, and burnout of the professional school counselors. The Korean Journal of School Psychology, 6(1), 83-102. https://doi.org/10.16983/kjsp.2009.6.1.83
  19. Lee, S. K. & Park, J. H. (1996). A study on the relationship between autonomy and group cohesiveness perceived by nurses and their job satisfaction, organizational commitment, motivation and intend to stay on jobs. The Journal of Korean Academy of Nursing Administration, 2(1), 5-15.
  20. MacKinnon, D. P., Warsi, G., & Dwyer, J. H. (1995). A simulation study of mediated effect measures. Multivariate Behavioral Research, 30(1), 41-62. https://doi.org/10.1207/s15327906mbr3001_3
  21. Matsumura, L. C., Garnier, H. E., & Spybrook, J. (2013). Literacy coaching to improve student reading achievement: A multi-level mediation model literacy coaching to improve student reading achievement. Learning and Instruction, 25, 35-48. https://doi.org/10.1016/j.learninstruc.2012.11.001
  22. Mulaik, S. A., James, L. R., Alstine, J. V., Bennett, N., Lind, S., & Stilwell, C. D. (1989). Evaluation of goodness-of-fit indices for structural equation modeling. Psychological Bulletin, 105(3), 430-445. https://doi.org/10.1037/0033-2909.105.3.430
  23. Muthen, B. O. (2013). Advances in latent variable using Mplus version7. Retrieved on June 4, 2014 from the http://www.modeling.uconn.edu/m3c/aGSSeJST/File
  24. Muthen, B. O. & Asparouhov, T. (2008). Growth mixture modeling: Analysis with non-Gaussian random effects. In G. Fitzmaurice, M. Davidian, G. Verbeke, & G. Molenberghs (Eds.), Longitudinal data analysis (pp. 143-165). Boca Raton, FL: Chapman & Hall/CRC.
  25. Muthen, L. K. & Muthen, B. O. (2012). Mplus User's Guide. Los Angeles, CA: Muthen & Muthen.
  26. Na, K. J., Lee, Y. G., & Yook, H. Y. (2010). The mediating effects of design values on the relationship between consumers' design evaluative elements and brand Loyalty. Korean Journal of the Science of Emotion & Sensibility, 13(3), 511-522.
  27. Park, D. Y. (2011). A tendency analysis on the School effectiveness using Multilevel Structural Equation Modeling. The Korean Journal of Educational Evaluation, 24(2), 345-376.
  28. Park, S. M. & Bak, B-G. (2014). A trend analysis of the mediation effect researches and a suggestion for the new direction of them. The Korean Journal of Educational Psychology, 28(1), 41-70.
  29. Park, S. M. (2015). Mediator analysis methods for the longitudinal, multilevel, and categorical data. Unpublished doctoral dissertation, Chonbuk University, Korea.
  30. Pituch, K. A., Murphy, D. L., & Tate, R. L. (2010). Three-level models for indirect effects in school-and class-randomized experiments in education. The Journal of Experimental Education, 78(1), 60-95. https://doi.org/10.1080/00220970903224685
  31. Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework for assessing multilevel mediation. Psychological Methods, 15(3), 209-233. https://doi.org/10.1037/a0020141
  32. Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2011). Alternative methods for assessing mediation in multilevel data: The advantages of multilevel SEM. Structural Equation Modeling, 18(2), 161-182. https://doi.org/10.1080/10705511.2011.557329
  33. Raftery, A. E. (1995). Bayesian model selection in social research. Sociological Methodology, 25(1), 111-163. https://doi.org/10.2307/271063
  34. Song, J. H. (2011). Exploring the Explanatory Factors of Student Happiness by Applying the Multi-level Model. Unpublished doctoral dissertation, Chonbuk University, Korea.
  35. Zhang, Z., Zyphur, M. J., & Preacher, K. J. (2009). Testing multilevel mediation using hierarchical linear models: Problems and solutions. Organizational Research Methods, 12(4), 695-719. https://doi.org/10.1177/1094428108327450