LGM with binary outcomes PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 gibbon lab posted on Thursday, October 13, 2011 - 7:12 am
Hi Professor Muthen,

I am interested in modeling the trend of a binary outcome with four time points. The outcome is smoking (yes=1/no=0). The data shows that the proportion of smoking increases along time points. Most subjects did not quit smoking during the study once they started. The patterns of outcomes for most smokers are 1 1 1 1 or 0 1 1 1 or 0 0 1 1 or 0 0 0 1. I am kind of worried about colinearity problems. Is latent growth model a good approach in this case? Thanks.
 Linda K. Muthen posted on Thursday, October 13, 2011 - 1:30 pm
This type of data could be problematic for a growth model. You might consider latent transition analysis for this type of data. See Slides 15-29 of the Topic 6 course handout.
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: