Non-linear relationship PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Isabelle Dinslaken posted on Monday, October 19, 2015 - 1:18 am
Dear Dr. Muthen,
I’m analysing my data with structural equation modelling with Mplus for the first time and I have some general questions:
Is it allowed to expect a mediation between two parallel observed mediators?
Is it right to make a root transformation if I have a quadratic relationship between my mediator and my dependent variable (both observed variables)? And is it always necessary to make a transformation with my data if one of the preconditions isn’t fulfilled?
How do I know about the model fit if my dependent variable is categorical? Because than I won’t get a chi-square. Is it allowed to look at the model fit without specifying that the dependent variable is categorical to look at the chi-square and only specify that the dependent variable ist categorical in a further step of analysis to look at the model results?
Thanks a lot for your help.
I’m looking forward to hearing from you.
Yours sincerely,
Isabelle
 Linda K. Muthen posted on Monday, October 19, 2015 - 10:04 am
Yes, parallel mediators are allowed.

Yes, you can do a trasnforamtion.

Absolute fit statistics are not available with categorical outcomes and maximum likelihood estimation. They are for weighted least squares. Use the default WLSMV estimator if you want fit statistics.
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: