|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Covariance btw Factors with Categoric... |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
I am studying comorbidity between latent internalizing and externalizing disorders. My data is based on both categorical and continuous indicators. In order to establish comorbidity, I was thinking to test the difference between my structural model when the latent factors (externalizing and internalizing) are allowed to covary versus a model where they cannot covary. How can I carry out this analysis? For the covariance model I was thinking to use a MIMIC model with the default WLSMV estimator. However it is unclear whether or not to use thresholds. Further, how do I specify a model where the factors are not allowed to covary (i.e. using the MPlus syntax). Thank you kindly for your assistance. |
|
|
To specify that factors do not covary, say f1 WITH f2@0; |
|
Back to top |
|
|