Survey longitudinal PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Scott Raynaud posted on Thursday, September 27, 2012 - 7:52 am
I’m looking at some longitudinal data with binary outcomes. These are all observed variables (various health problems such as cancer, heart disease, etc.), so no latent variables will be included. Covariates will be included (sex, age race). I would like to compare M/F adjusting for age and race. Is a growth mixture model appropriate? I only have two time points so I can’t fit quadratic effects but I can still perform the analysis in Mplus, correct? I’d also be interested in the probabilities of switching categories between time 1 and 2. I can conduct a latent transition analysis as well, right?
 Linda K. Muthen posted on Friday, September 28, 2012 - 9:56 am
I would not do growth modeling or growth mixture modeling with two time points and binary indicators. You can consider LTA.
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: