The Examination of Model Fit Indexes with Different Estimation Methods under Different Sample Sizes in Confirmatory Factor Analysis

Abstract

In adjustment studies of scales and in terms of cross validity at scale development, confirmatory factor analysis is conducted. Confirmatory factor analysis, multivariate statistics, is estimated via various parameter estimation methods and utilizes several fit indexes for evaluating the model fit. In this study, model fit indexes utilized in confirmatory factor analysis are examined with different parameter estimation methods under different sample sizes. For the purpose of this study, answers of 60, 100, 250, 500 and 1000 students who attended PISA 2012 program were pulled from the answers to two dimensional “thoughts on the importance of mathematics” dimension. Estimations were based on methods of maximum likelihood (ML), unweighted least squares (ULS) and generalized least squares (GLS). As a result of the study, it was found that model fit indexes were affected by the conditions, however some fit indexes were affected less than others and vice versa. In order to analyze these, some suggestions were made.

References

1. Boomsma, A. (1982). The robustness of LISREL against small sample sizes in factor analysis models. (Eds. K.
2. G. Jöreskog & H. Wold). Systems under indirect observations: Causality, structure, prediction,Amsterdam: North-Holland.
3. Byrne, B. M. (2010). Structural equation modeling with AMOS basic concepts, applications, and
4. programming. New York: Taylor & Francis Group.
5. Crocker, L. M. & Algina, L. (1986). Introduction to classical and modern test theory. New York: Holt,
6. Rinehart and Winston.
7. DeVellis, R. F. (2003). Scale development theory and applications, applied social research methods series.
8. California: SAGE Publications.

9. Fan, X. (1996). Structural equation modeling and canonical correlation analysis: What do they have in
10. common?. Structural Equation Modeling: A Multidisciplinary Journal, 4, 64-78.
11. Fan, X., Thompson B. & Wang, L. (1999). Effects of sample size, estimation methods, and model specification
12. on structural equation modeling fit indexes. Structural Equation Modeling, 6(1), 56-83.
13. Fan, X. and Sivo, S. A. (2007). Sensitivity of fit indices to model misspecification and model types.
14. Multivariate Behavioral Research, 42(3), 509-529.
15. Kaplan (2001), Structural equation modeling. International Encyclopedia of the Social & Behavioral Sciences,
16. -15222.
17. Kenny, D. A., & McCoach, D. B. (2003). Effect of the number of variables on measures of fit in structural
18. equation modeling. Structural Equation Modeling, 10, 333–351.
19. Kenny, D. A., Kaniskan, B. & McCoach, D. B. (2014). The Performance of RMSEA in Models With Small
20. Degrees of Freedom. Sociological Methods & Research, 1-22.
21. Kline, R. B. (2011). Principals and practice of structural equation modeling. New York. The Guilford Press.
22. Molwus, J. J., Erdogan, B. & Ogunlana, S. O. (2013). Sample Size and Model Fit Indices for Structural
23. Equation Modelling (SEM): The Case of Construction Management Research. ICCREM, 338-347.
24. Savalie, V. (2012). The relationship between root mean square error of approximation and model
25. misspecification in confirmatory factor analysis models. Educational and Psychological Measurement,
26. (6), 910-932.
27. Sayın, A. & Gelbal, S. (2016). Yapısal eşitlik modellemesinde parametrelerin klasik test kuramı ve madde
28. tepki kuramına göre sınırlandırılmasının uyum indekslerine etkisi. International Journal of Education
29. Science and Technology, 2(2), 57-71.
30. Schermelleh-Engel, K., Moosbrugger, H. & Müller, H. (2003). Evaluating the fit of structural equation models:
31. Tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research
32. Online, 8(2), 23-74.
33. Schumacker, R.E. & Lomax, R.G. (2004). A beginner’s guide to structural equation modeling. Lawrence
34. Erlbaum Associates, New Jersey.
35. Sharma, S., Mukherjee, S., Kumar, A. & Dillon, W.R. (2005). A simulation study to investigate the use of
36. cutoff values for assessing model fit in covariance structure models. Journal of Business Research, 58,
37. –943.
38. Tabachnick, B. G. & Fidell, L. S. (2007). Using multivariate statistics. Boston: Allyn and Bacon.