The Effect of Estimation Methods on SEM Fit Indices

Educational and Psychological Measurement - Tập 80 Số 3 - Trang 421-445 - 2020
Dexin Shi1, Alberto Maydeu‐Olivares2,1
1University of South Carolina, Columbia, SC. USA
2University of Barcelona, Barcelona, Spain

Tóm tắt

We examined the effect of estimation methods, maximum likelihood (ML), unweighted least squares (ULS), and diagonally weighted least squares (DWLS), on three population SEM (structural equation modeling) fit indices: the root mean square error of approximation (RMSEA), the comparative fit index (CFI), and the standardized root mean square residual (SRMR). We considered different types and levels of misspecification in factor analysis models: misspecified dimensionality, omitting cross-loadings, and ignoring residual correlations. Estimation methods had substantial impacts on the RMSEA and CFI so that different cutoff values need to be employed for different estimators. In contrast, SRMR is robust to the method used to estimate the model parameters. The same criterion can be applied at the population level when using the SRMR to evaluate model fit, regardless of the choice of estimation method.

Từ khóa


Tài liệu tham khảo

Asparouhov T., 2010, Simple second order chi-square correction scaled chi-square statistics

10.1080/10705511.2014.859510

10.1207/s15328007sem1302_2

10.1037/0033-2909.107.2.238

Bentler P. M. (1995). EQS 5 [Computer program]. Encino, CA: Multivariate Software.

10.1080/01621459.1979.10481600

10.1080/00273171.2014.933697

10.1080/00273171.2012.715252

Browne M. W., 1993, Testing structural equation models, 136

10.1080/10705511.2014.915373

10.1177/0049124108314720

10.1037/1082-989X.9.4.466

10.1080/10705510903203573

10.1080/10705511.2019.1637741

10.1080/10705519909540118

10.1007/BF02289343

Jöreskog K. G., 1988, LISREL 7. A guide to the program and applications, 2

10.1177/0049124114543236

10.1080/10705511.2018.1562351

Li L., 2006, Robust statistical tests for evaluating the hypothesis of close fit of misspecified mean and covariance structural models

10.1207/S15327906MBR3801_5

10.1037/1082-989X.1.2.130

10.1007/s11336-005-0773-4

10.1007/s11336-016-9552-7

10.1080/10705511.2016.1269606

10.1080/10705511.2011.581993

10.1080/10705511.2017.1389611

10.1080/00223891.2017.1281286

10.1080/10705511.2012.634724

Muthén B., 1993, Testing structural equation models, 205

Muthén B. O., du Toit S. H. C., Spisic D. (1997, November 18). Robust inference using weighted least squares and quadratic estimating equations in latent variable modeling with categorical and continuous outcomes. Retrieved from https://www.statmodel.com/download/Article_075.pdf

R Core Team, 2019, R: A language and environment for statistical computing

10.1037/a0029315

10.18637/jss.v048.i02

10.1080/10705510903203433

10.1007/BF00152011

Satorra A., 1994, Latent variable analysis. Applications for developmental research, 399

10.1177/0013164412452564

Savalei V., 2017, Paper presented at the Annual Meeting of the Society of Multivariate Experimental Psychology

10.1080/00273171.2018.1455142

10.1111/j.2044-8317.2012.02049.x

10.1080/10705511.2018.1449653

10.1177/0013164418783530

10.1080/00273171.2014.989012

10.1080/10705511.2017.1369088

10.1080/00273171.2018.1476221

10.1080/10705511.2019.1611434

Steiger J. H., 1989, EzPATH: A supplementary module for SYSTAT and SYGRAPH

10.1207/s15327906mbr2502_4

Steiger J., 1980, Paper presented at the annual Spring Meeting of the Psychometric Society

Xia Y. (2016). Investigating the chi-square-based model-fit indexes for WLSMV and ULSMV estimators (Unpublished doctoral dissertation). Florida State University, Tallahassee, FL.

10.1080/00273171.2018.1480346

10.3758/s13428-018-1055-2

10.1080/10705511.2010.489003

Thông tin
Thông tin xuất bản

Educational and Psychological Measurement

Tập 80 Số 3

421-445

Thông tin tác giả