Robust Procdure?????? PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Amanda M. White posted on Monday, July 30, 2007 - 11:18 pm
Hi all,
I'm new to Mplus and have a question regarding using the use of a robust procedure. Here's the pertinent info about the study:

(A) I have continuous variables

(B)I have a about seven or so items in which the skewness ranges from
2.258 to 3.527 and the kurtosis ranges from 2.58 to 11.089

Is it best to use robust procedures on the original data, transform the data, or to do both?

Also, are there any citations I can use to support my decision?

Thank you so much for your help!!

Amanda
 Linda K. Muthen posted on Tuesday, July 31, 2007 - 12:01 am
I think there is a whole literature on transforming versus not transforming. I don't know of any one article. I would not transform unless I had a reason to do so. Otherwise, I would use a robust estimator.
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: