Multivariate normality testing PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Melinda Kavanaugh posted on Monday, March 04, 2013 - 6:48 am
Hello,
I am trying to test for normality in my simple path model.
I used tech13;
TYPE is Mixture;

but that did not work because I do not have categorical data.

I have all continuous variables. Is there a tech that addresses this?


Thank you,
Melinda
 Linda K. Muthen posted on Monday, March 04, 2013 - 8:21 am
TECH13 is for continuous not categorical variables. Use CLASSES = c (1); Also, testing for normality is not needed if you use the MLR estimator.
 Melinda Kavanaugh posted on Monday, March 04, 2013 - 1:53 pm
Thank you Linda. That was very helpful.
 Melinda Kavanaugh posted on Monday, March 04, 2013 - 2:29 pm
One more follow-up question. Do I report the scaling correction factors when using MLR? Or do I simply report the fit stats? I suppose I don't know what to do with the scaling factors.
Thank you again.
 Linda K. Muthen posted on Monday, March 04, 2013 - 3:38 pm
The scaling correction factor is used only for difference testing. Otherwise, just report the chi-square value.
 Melinda Kavanaugh posted on Tuesday, March 05, 2013 - 6:02 am
Thank you.
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: