The Principle of Multilevel Structural Equation Modeling Analysis by Using Optimal Sample Size and Estimation Methods
Article Sidebar
Main Article Content
Abstract
Multilevel structural equation model (MSEM) is an advanced statistical analysis technique that is developed by integrated the analysis of the concept of the Structural equation model (SEM) and Hierarchical linear model (HLM). To study the effect of multilevel predictive variables on the dependent variables and to study cross-level interactions, cause to comprehensive and profound than traditional analysis statistics and can help improve limitations in the analysis HLM and SEM analysis. The sample size that is considered first about appropriate with macro-level or group level. The sample size of the highest level of analysis should be more than 30 groups. For the sample size, the micro-level or individual level should be 10-20 times the number of parameters. Parameter estimation by using methods of the Maximum likelihood (ML) and the Full information maximum likelihood (FIML) suitable for balanced group sizes and data has a normal distribution. For methods of Muthen and Muthen’s Quasi-maximum likelihood (MUML), Partial maximum likelihood (PML) and the maximum likelihood with robust standard errors and chi-square (MLR) using for unbalanced group sizes and data have a non-normal distribution. However, if the sample size is large, parameter estimation by using ML and MUML methods have similar results.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
ข้อความและบทความในวารสารการวัด ประเมินผล สถิติ และการวิจัยทางสังคมศาสตร์ เป็นแนวคิดของผู้เขียน มิใช่ความคิดเห็นของกองบรรณาธิการวารสาร จึงมิใช่ความรับผิดชอบของวารสารการวัด ประเมินผล สถิติ และการวิจัยทางสังคมศาสตร์ บทความในวารสารต้องไม่เคยตีพิมพ์ที่ใดมาก่อน และสงวนสิทธิ์ตามกฎหมายไทย การจะนำไปเผยแพร่ ต้องได้รับอนุญาตเป็นลายลักษณ์อักษรจากกองบรรณาธิการ
References
Anderson, J. & Gerbing, D. W. (1984). The effects of sampling errors on convergence, improper solution and goodness of fit indices for maximum likelihood confirmatory factor analysis. Psychometrika, 49, 155-173.
Bentler, P. M. & Yuan, K. H. (1999). Structural equation modelling with small samples: Test statistics. Multivariate Behavioral Research, 34(2), 181–197.
Browne, W., Goldstein, H., Rashbash, J., Plewis, I., Draper, D., & Yang, M. (1998). User's guide to MLwiN. London: Institute of Education.
Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models newbury park. Calif: Sage Publications.
Diamantopoulos, A. & Siguaw, A. D. (2000). Introducing LISREL: A guide for the uninitiated. London : Sage Publications.
Goldstein, H. (1991). Non-Linear Multilevel Models, with an Application to Discrete Response Data. Biometrika, 78, 45-51.
Goldstein, H. (1995). Multilevel statistical models. London : Edward Arnold.
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (1998). Multivariate data analysis, 5(3), 207-219.
Hair, J. F., Black, W.C., Babin, B.J. & Anderson, R. E. (2010). Multivariate data analysis (7th ed.). New York: Pearson.
Heck, R. H. & Thomas, S.L. (2000).An introduction to multilevel modeling techniques. Mahwah, NJ:
Lawrence Erlbaum.
Hox, J. J. & C. J. M. Maas. (2001). The accuracy of multilevel structural equation modeling with pseudobalanced groups and small samples. Structural Equation Modeling, 8(2), 157-174.
Hox, J. J. & C. J. M. Maas. (2004). Robustness issues in multilevel regression analysis. Statistica Neerlandica, 18(2), 127–137.
Hox, J. J. & C. J. M. Maas. (2005). Sufficient sample sizes for multilevel modeling. Methodology, 1(3), 86–92.
Hox, J. J. (2002). Multilevel analysis: Techniques and applications. Mahwah, NJ: Erlbaum.
Kanjanawasee, Sirichai. (2007). Multi-level analysis (4th ed.). Bangkok: Chulalongkorn University Printing House. (in Thai)
Kanjanawasee, Sirichai. (2011). Multi-level analysis (5th ed.). Bangkok: Chulalongkorn University Printing House. (in Thai)
Lindeman, R. H., Merenda, P. F. & Gold, R. Z. (1980). Introduction to bivariate and multivariate analysis. Glenview, Illinois: Scott, Foresman and Company.
Meuleman, B. & Billiet, J. (2009). A Monte Carlo sample size study: how many countries are needed for accurate multilevel SEM?. Survey Research Methods, 3(1), 45-58.
Muthen, B. O. (1994). Multilevel Covariance Structure Analysis. Sociological Methods and Research, 22(1), 376-398.
Muthen, B. O. (2011). Applications of causally defined direct and indirect effects in mediation analysis using SEM in Mplus. Retrieved from statmodel.com/download/causalmediation.pdf
Muthen, L. K. & Muthen, B. O. (1998). Mplus technical appendices. Los Angeles, CA: Muthen and Muthen.
Muthen, L. K. & Muthen, B. O. (2004). Mplus user’s guide (3rd ed). Los Angeles, CA: Muthen and Muthen.