Evaluating estimation methods for ordinal data in structural equation modeling

By Pui‐wa Lei. 2009

184 citations recorded by the source corpus.

Subfields: Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/ad6b0ce108f8b6cd122f8c1015f457c7dc90ad98

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

Citations in the corpus

Examining Chi-Square Test Statistics Under Conditions of Large Model Size and Ordinal Data (2018, background): “…under WLS (Bandalos, 2014; DiStefano & Morgan, 2014; Flora & Curran, 2004); however, if sample sizes were small and data moderately or extremely asymmetric, then problems, such as convergence or inaccurate parameter estimates, could be observed (Bandalos, 2014; DiStefano & Morgan, 2014; Lei, 2009).”
Comparison of Different Estimation Methods for Categorical and Ordinal Data in Confirmatory Factor Analysis (2015, result): “The literature has many studies that compared different estimation methods which have effects on results obtained (DiStefano, 2002; Forero et al, 2009; Hu, Bentler & Kano, 1992; Lei, 2009; Muthen & Kaplan, 1985; Rigdon & Ferguson, 1991; Yang-Wallentin et al., 2010).”
The performance of MLR, USLMV, and WLSMV estimation in structural regression models with ordinal variables (2014, methodology): “Some simulation studies examined the joint performance of both factor loading and inter-factor correlation estimates (see, e.g., Lei, 2009; Yang-Wallentin, Jöreskog, & Luo, 2010).”
A Comparison of Diagonal Weighted Least Squares Robust Estimation Techniques for Ordinal Data (2014, methodology): “Researchers have two options when using the DWLS estimator with LISREL: (a) inputting only asymptotic variances for estimation, or (b) inputting the full weight matrix and, as with WLSM and WLSMV, just the diagonal is inverted.”
Sample Size Requirements of the Robust Weighted Least Squares Estimator (2014, background): “Prior simulation studies have shown that, unlike full WLS, the robust WLS estimator possesses strong convergence properties even with small sample sizes and under non-normality in observed (Forero-Garcia et al., 2009; Lei, 2009; Muth n et al., 1997) and latent (Flora & Curran, 2004) response variabl”
Coefficient Alpha: A Reliability Coefficient for the 21st Century? (2011, methodology): “Flora and Curran (2004) and Lei (2009) demonstrated with computer-generated data that WLSMV yields relatively accurate results when the model is correctly specified, although slightly biased test statistics and standard errors are obtained when large models are estimated with small samples.”
An evaluation of WLSMV and Bayesian methods for confirmatory factor analysis with categorical indicators (2014, unknown): “With simulated data, WLSMV has shown to perform very well when models are correctly specified and sample size is not small (Flora and Curran, 2004) and to produce better results than ML (Lei, 2009).”
Current Methodological Considerations in Exploratory and Confirmatory Factor Analysis (2011, methodology): “Three viable alternatives for estimating FA models for ordinal data are robust continuous ML estimation, robust least squares (LS) estimation, and robust weighted least squares (WLS) estimation (Beauducel & Herzberg, 2006; Bentler & Yuan, 1999; Flora & Curran, 2004; Forero, Maydeu-Olivares, & Gallar”
Why Ordinal Variables Can (Almost) Always Be Treated as Continuous Variables: Clarifying Assumptions of Robust Continuous and Ordinal Factor Analysis Estimation Methods (2020, background): “There is a diversity of methodological literature addressing this issue (Dolan, 1994; DiStefano, 2002; Lei, 2009; Flora et al., 2012; Rhemtulla et al., 2012; Sass et al., 2014; Barendse et al., 2015; Asún et al., 2016; Li, 2016; Jia andWu, 2019).”
Probing for the Multiplicative Term in Modern Expectancy-Value Theory: A Latent Interaction Modeling Study. (2012, unknown): “…of the indicator variables and for clustering of the data (see below) and that has been shown to be robust when ordered categorical variables are treated as continuous (e.g., Beauducel & Herzberg, 2006; Dolan, 1994; Lei, 2009; Lubke & Muthén, 2004; Rhemtulla, BrosseauLiard, & Savalei, 2010).”