Model fit information-estimators&rota... PreviousNext
Mplus Discussion > Exploratory Factor Analysis >
Message/Author
 Gaye Ildeniz posted on Monday, September 10, 2018 - 10:10 am
I have been trying EFA with different estimators and rotations.
However, for example when I use MLR and promax or ULS and promax, I don’t get the same model fit information I would get with MLR & oblimin combination.
For example, there are no CFI/TLI or AIC&BIC values. Do you think it’s
because that mplus does not provide them with certain estimators/rotations or I might be doing something wrong?

Thank you
 Bengt O. Muthen posted on Monday, September 10, 2018 - 2:52 pm
The older rotation of promax has not been as far developed as the newer ones.
 Gaye Ildeniz posted on Tuesday, September 11, 2018 - 3:51 am
Thank you very much for your reply.

By 'older rotation of promax' do you mean the older versions of Mplus? Because I've used version 8.1. What are the newer rotation of promax? Could you please clarify this?

Thanks!
 Bengt O. Muthen posted on Tuesday, September 11, 2018 - 2:38 pm
No, I mean that Promax is an older rotation procedure - newer ones are better, such as our default Geomin. The Promax track of Mplus has therefore not been developed as fully as the newer tracks.
 Gaye Ildeniz posted on Wednesday, September 12, 2018 - 4:15 am
I see now, thanks a million!
 Gaye Ildeniz posted on Friday, September 21, 2018 - 4:46 am
Hi!
For an EFA with non-normally distributed data measured with a 5-point Likert-type scale (strongly agree-strongly disagree), would you recommend MLR & Geomin or WLSMV & Geomin? and why? Any papers explaining the reasoning behind?
Many thanks!
 Bengt O. Muthen posted on Friday, September 21, 2018 - 5:48 pm
If the variables have strong floor or ceiling effects, I would declare them as categorical; otherwise not. See also the FAQ on our website:

Estimator choices with categorical outcomes
 QianLi Xue posted on Friday, November 22, 2019 - 8:53 am
Why are CFI and TLI no longer part of the EFA output?
 Tihomir Asparouhov posted on Friday, November 22, 2019 - 10:37 am
In what circumstances? Can you send your example to support@statmodel.com
They should be in the output.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: