argument 'group =' and model fit

[Christopher Bratt]

I would like to run a model on many groups, but not impose or test measurement invariance across these groups.

Am I correct in believing that I should use an R looping function because I cannot get model fit for each group when using the following lavaan argument?

 

group =

Christopher

[kinga.bie...@gmail.com]

Hi, 

I have the exact same question. 

Best, 

Kinga

[Terrence Jorgensen]

Am I correct in believing that I should use an R loop

If you want the test of fit in each group, then that is the simplest way, yes.  But summary() displays the chi-squared contribution from each group, and if there are no equality constraints, then the df are divided evenly between those groups, so you could always use pchisq() to calculate your p values manually from a multigroup model.

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam

http://www.uva.nl/profile/t.d.jorgensen

[Christopher Bratt]

Thanks for the clarification.