Meaning of variance in binary outcome PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Corinna posted on Tuesday, June 23, 2015 - 7:28 am
Dear Linda & Bengt,
I have a binary outcome in GMM. I know that with a categorical variable I do not have intercepts but thresholds. However, I do not understand the meaning of variances with a binary outcome. To what does the variance relate? To the probabilities? What is the meaning of a variance with a binary outcome?

I would be very thankful for a short answer.

With best regards,
Corinna
 Linda K. Muthen posted on Tuesday, June 23, 2015 - 8:44 am
Variances of categorical outcomes are not model parameters.
 Corinna posted on Tuesday, June 23, 2015 - 9:19 am
Dear Linda,
I am not sure if I expressed myself correctly. I am talking about residual variances of the growth parameters (intercept and slope) if you have a binary outcome.
If they are not model parameters, does that mean that I do not have to consider them at all in my model analyses? I am trying to find the meaning of these variances in a binary outcome... but if I do not have to consider them then perhaps I can just fix the variances and calculate and LCGA?
Thank you very much
 Linda K. Muthen posted on Tuesday, June 23, 2015 - 10:54 am
The growth factors are continuous latent variables not categorical. Their variances/residual variances are interpreted in the regular way.
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: