Although frequentist estimators can effectively fit ordinal confirmatory factor analysis (CFA) models, their assumptions are difficult to establish and estimation problems may prohibit their use at times. Consequently, researchers may want to also look to Bayesian analysis to fit their ordinal models. Bayesian methods offer researchers an effective means of estimating, testing, and interpreting ordinal CFA models. Unfortunately, there are few applied resources on the subject. The purpose of this article is to provide researchers with an introduction to the essential concepts, practice recommendations, and process of fitting ordinal CFA models using Bayesian analysis. Mplus 7.4 and data from the Pittsburg Common Cold Study 3 are used to example how researchers can set up their Bayesian models, conduct diagnostic checks, and interpret the results. This article also highlights the benefits and challenges of Bayesian ordinal CFA modeling. Accessed 1,146 times on https://pareonline.net from May 17, 2019 to December 31, 2019. For downloads from January 1, 2020 forward, please click on the PlumX Metrics link to the right.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Taylor, John M.
"Overview and Illustration of Bayesian Confirmatory Factor Analysis with Ordinal Indicators,"
Practical Assessment, Research, and Evaluation: Vol. 24
, Article 4.
Available at: https://scholarworks.umass.edu/pare/vol24/iss1/4