Reliability and Stability on the Numerical Analysis in Structural Equation Modeling

Structural equation modeling has been widely used in various research studies and has been incorporated into many software programs such as Mplus, CALIS, EQS, LISREL, Amos and so on. At the meantime, the diversity, instability of the analysis results or no solutions has been found in some analysis models. Several problems on the numerical analysis such as the constraints of residual variance, initial values and the different solutions by software in use have also been found in SEM analysis. This paper discusses the reliability and stability on the numerical analysis in structural equation modeling by using four kinds of SEM programs and three kinds of sample data. The study will introduce an application of the optimized calculation of genetic algorithms (GA) in structural equation modeling in order to see elaborately what is going on with these issues and also to examine the goodness-of-fit, validity, stability and reliability of structural model. Furthermore, the empirical analysis is presented to discuss the above issue in the questionnaire survey data about drunken driving behavior in Bangkok.

[1]  James B Schreiber,et al.  Core reporting practices in structural equation modeling. , 2008, Research in social & administrative pharmacy : RSAP.

[2]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1992, Artificial Intelligence.

[3]  Andy P. Field,et al.  Discovering Statistics Using SPSS , 2000 .

[4]  Barbara M. Byrne,et al.  Structural equation modeling with AMOS , 2010 .

[5]  Ulf Kroehne,et al.  (Why) Should We Use SEM? Pros and Cons of Structural Equation Modeling , 2003 .

[6]  Matt Stevenson,et al.  Alcohol Use Disorders Identification Test , 2010 .

[7]  R. P. McDonald,et al.  Choosing a multivariate model: Noncentrality and goodness of fit. , 1990 .

[8]  Jin-Hyuk Chung,et al.  Analysis of traffic accident size for Korean highway using structural equation models. , 2008, Accident; analysis and prevention.

[9]  Barbara M. Byrne,et al.  Structural equation modeling with AMOS - basic concepts, applications, and programming, 2nd Edition , 2010, Multivariate applications series.

[10]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1996, Springer Berlin Heidelberg.

[11]  Pui‐wa Lei,et al.  Introduction to Structural Equation Modeling: Issues and Practical Considerations , 2007 .

[12]  B. Byrne.  Structural Equation Modeling With AMOS, EQS, and LISREL: Comparative Approaches to Testing for the Factorial Validity of a Measuring Instrument , 2001 .

[13]  M. Browne,et al.  Alternative Ways of Assessing Model Fit , 1992 .

[14]  R. MacCallum,et al.  Power analysis and determination of sample size for covariance structure modeling. , 1996 .

[15]  Thomas F. Golob,et al.  Structural Equation Modeling For Travel Behavior Research , 2001 .

[16]  P. Bentler,et al.  Comparative fit indexes in structural models. , 1990, Psychological bulletin.