Uncorrelated exogenous variables in 2... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Thomas Carpenter posted on Thursday, July 26, 2012 - 6:23 pm
Hi there,

I have an additional SEM question. I have a have a 2x2 factorial experiment with a DV and a couple of mediators.

Typically exogenous variables are correlated. In this case, I have 3 IVs: A, B, and the AxB interaction. However, I KNOW they are uncorrelated because I made them orthogonal when I built the experiment. Do I still need to correlate them in the model? I can get additional degrees of freedom if I specify that they are UNcorrelated. However, I have been given mixed feedback on this, with some people saying it is fine to fix their correlations at zero, while others saying this violates assumptions. If anyone has any feedback on this, it would be greatly appreciated.

Thanks
 Linda K. Muthen posted on Friday, July 27, 2012 - 4:25 pm
In regression, the model is estimated conditioned on the observed exogenous variables. Their means, variances, and covariances are not model parameters. You should not mention these parameters in the MODEL command.
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: