Using the Omnibus Maximum Likelihood F-Test in Confirmatory Factor Analysis Using R

Abstract

The utility of the maximum likelihood F-test was demonstrated as an alternative to the omnibus Chi-square test when evaluating model fit in confirmatory factor analysis with small samples, as it has been well documented that the likelihood ratio test (T_ML) with small samples is not Chi-square distributed.

Keywords:

• R function
• confirmatory factor analysis
• maximum likelihood F-test
• small sample size

Disclosure Statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Georgios D. Sideridis

Georgios D. Sideridis holds a PhD in child psychology from the University of Kansas. His research interests lie in Psychometrics, Item Response Theory, and Bayesian Inference. He is a survey methodologist at Boston Children's Hospital, Harvard Medical School.

Fathima Jaffari

Fathima Jaffari holds a PhD with a concentration in Quantitative Methods. Her research interests are in Psychometrics, Item Response Theory, Bayesian Inference, Structural Equation Modeling, Experimental Designs, and Monte Carlo Methods. She is a Senior Measurement specialist at the department of professional exams at National Center for Assessment-Qiyas and the Education & Training Evaluation Commission (ETEC).