Two part gmm setup? PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Bruce A. Cooper posted on Monday, October 17, 2011 - 11:39 am
I have an 11-point pain scale, where pts who have no pain are coded 0, and those with pain are coded 1-10. This item is collected across 7 assessments. I'd like to see if there are latent growth classes for this pain variable, and it seems likely that a two part model for the pain item would be best for capturing that information given the large number of zeros at each assessment. Do you have any setups for a basic two part GMM or LCGA? Thanks! - bac
 Bengt O. Muthen posted on Monday, October 17, 2011 - 9:47 pm
This paper, although cross-sectional, may be helpful - see the figures for the setup:

http://www.statmodel.com/download/2partfmm.pdf
 Bruce A. Cooper posted on Tuesday, October 18, 2011 - 10:32 am
Thanks very much, Bengt - this will be very helpful.

The figures illustrate the logic for the analysis and the parameter estimates, but not the Mplus Model syntax that produced them. I hope you won't mind looking at an example of my syntax if need be!

Best, bac
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: