Abstract: The structure equation model (SEM) is a multivariate technique for studying relationships among a set of substantively meaningful variables. The question as to which model best fits the data well, reflects the underlying theory, known as model fit, and is by no means much agreed upon. The present study aims at studying the effects of weights used in the original data to improve model fit indices within the framework of SEM. The weighted values ranged from 0.1 to 0.95; and varying sample sizes (50, 100, 200, 500, and 1000) were used in the present study. Moreover, the study also examines the performance of the weighted values across the symmetric, positive, and negative skewed distributions. To achieve this goal, a Monte Carlo simulation study was carried out using Python, with 500 iterations performed for each weight and sample size.
The results show that the p-value gets better in the weighted data in comparison with the values in the case of the original data, with the weights W = [0.2, 0.8] for the left-skewed distributions, W = [0.3, 0.9] for the symmetrical distributions and W = [0.75, 0.85] for the right-skewed distributions. Also, the results show that the larger the sample size, the greater the number of cases that achieve a better p-value with the weighted data. The results also show that the values of goodness of fit index (GFI) and root mean square error of approximation (RMSEA) get better in the weighted data in comparison with the values of the case of the original data, with the weights W = [0.1, 0.2] ∪ W = [0.8, 0.9] for the left-skewed distributions, W = [0.3, 0.15] ∪ W = [0.9, 0.95] for the symmetrical distributions and W = [0.7, 0.85] for the right-skewed distributions. Also, the results show that the larger the sample size, the lower the number of cases that achieve better GFI and RMSEA with the weighted data.
|
Keywords and phrases: structural weights, SEM, measurement weights, fit indices, model fit.
Received: June 17, 2023; Revised: July 30, 2023; Accepted: August 5, 2023; Published: August 26, 2023
How to cite this article: Mohammed Naji Al-Ghamdi, Ezz H. Abdelfattah and Ahmed Ezz, The impact of weighting data on some fit indices of structural equation models: a simulation study, Advances and Applications in Statistics 89(1) (2023), 73-106. http://dx.doi.org/10.17654/0972361723052
This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License
References: [1] M. N. Al-Ghamdi, E. H. Abdelfattah and A. Ezz, The effect of weighting data on the goodness of fit indices of the six sigma structural equation modeling, Journal of Mathematics and Statistics Studies 2 (2021), 36-49. https://doi.org/10.32996/jmss.2021.2.2.5. [2] M. N. Al-Ghamdi, E. H. Abdelfattah, F. Alam and A. Ezz, A proposed model for improving job performance in the technical and vocational training corporation in Saudi Arabia: based on combining six sigma methods and structural equation modeling, Academic Journal of Applied Mathematical Sciences 7(2) (2021), 129-136. https://doi.org/10.32861/ajams.72.129.136. [3] J. C. Anderson and D. W. Gerbing, The effect of sampling error on convergence, improper solutions, and goodness-of-fit indices for maximum likelihood confirmatory factor analysis, Psychometrika 49 (1984), 155-173. https://doi.org/10.1007/BF02294170. [4] P. M. Bentler, Comparative fit indexes in structural models, Psychological Bulletin 107 (1990), 238-246. https://doi.org/10.1037/0033-2909.107.2.238. [5] P. Bentler and D. Bonett, Significance tests and goodness of fit in the analysis of covariance structures, Psychological Bulletin 88(3) (1980), 588-606. https://doi.org/10.1037//0033-2909.88.3.588. [6] K. Bollen and J. S. Long, Editors, Testing structural equation models, Sage, Newbury Park, CA, 1993. [7] K. A. Bollen, Overall fit in covariance structure models: two types of sample size effects, Psychological Bulletin 107(2) (1990), 256-259. https://doi.org/10.1037/0033-2909.107.2.256. [8] F. Chen, K. A. Bollen, P. Paxton, P. Curran and J. Kirby, Improper solutions in structural equation models: causes, consequences, and strategies, Sociological Methods and Research 29 (2001), 468-508. https://doi.org/10.1177/0049124101029004003. [9] J. E. Gentle, Monte Carlo methods, S. Kotz and N. L. Johnson, eds., The Encyclopedia of Statistical Sciences, Wiley, New York, Vol. 5, 1985, pp. 612-617. [10] T. F. Golob, Structural equation modeling for travel behavior research, Transportation Research Part B: Methodological 37 (2003), 1-25. https://doi.org/10.1016/S0191-2615(01)00046-7. [11] K. G. Joreskog and D. Sorbom, LISREL8: User’s Reference Guide, Scientific Software, Mooresville, 1996. [12] M. Lei and R. G. Lomax, The effect of varying degrees of non-normality in structural equation modeling, Structural Equation Modeling: A Multidisciplinary Journal 12(1) (2005), 1-27. https://doi.org/10.1207/s15328007sem1201_1. [13] H. Liu, I. K. Jin and Z. Zhang, Structural equation modeling of social networks: specification, estimation, and application, Multivariate Behavioral Research 53(5) (2018), 714-730. https://doi.org/10.1080/00273171.2018.1479629. [14] R. E. Schumacker and R. G. ve Lomax, A Beginner’s Guide to Structural Equation Modeling, 2nd ed., Mahwah, NJ Lawrence Erlbaum Associates Inc., 2004. https://doi.org/10.4324/9781410610904. [15] J. H. Steiger, Structural model evaluation and modification: an interval estimation approach, Multivariate Behavioral Research 25 (1990), 173-180. https://doi.org/10.1207/s15327906mbr2502_4.
|