The effect of estimation method and sample size in multilevel structural equation modeling

By Joop J. Hox, Cora J. M. Maas, Matthieu J. S. Brinkhuis. 2010

334 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/f56933eaa0ce15d198a8a06ea8150fcea8689606

How researchers see this paper

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

Citations in the corpus

Multilevel Mediation With Small Samples: A Cautionary Note on the Multilevel Structural Equation Modeling Framework (2017, background): “Specific to ML-SEM, Hox and Maas (2001) and Hox, Maas, and Brinkhuis (2010) found that the number of clusters needs to be larger than 50 with more ideal data (low intraclass correlation and equal Level 1 sample sizes in each cluster) to produce reasonable estimates in the Level 2 portion of the…”
The Nexus of Ethical Leadership, Job Performance, and Turnover Intention: The Mediating Role of Job Satisfaction (2018, methodology): “Next, although the sample of this study was enough to notice important results in the mediation analysis, the future study needs to use large sample size which permits to analyse this model with a more robust analytical way like multilevel SEM [81].”
How Ethical Leadership Shapes Employees’ Job Performance: The Mediating Roles of Goal Congruence and Psychological Capital (2014, methodology): “Although our sample size was sufficient to detect significant effects in our mediation analyses, future research will require a larger sample size that would allow testing our model with a more sophisticated analysis technique, such as Multilevel Structural Equation Modeling (Hox et al. 2010).”
A Bayesian Approach to Multilevel Structural Equation Modeling With Continuous and Dichotomous Outcomes (2015, methodology): “Hox et al. (2010) looked at the properties of a number of frequentist estimators in the context of a multilevel confirmatory factor analysis (CFA) model, finding that robust maximum likelihood and the DWLS estimators implemented in Mplus produced accurate factor loading estimates with as few as 50…”