GGM or LGM PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Cathy Nylin posted on Monday, May 13, 2013 - 5:08 pm
I have read the chapters and papers on this site relating to growth mixture modeling and it seems appropriate to use for my data and questions. I am having a hard time understanding the example syntax in the example section and the user's guide, specifically which variables to use for u, y, x, w as shown in the examples. I have 2 questions:

1) I want to look at the relationship over time between school involvement (continuous variable measured at 3 time points) and alcohol use (semicontinuous variable measured at three time points, mostly 0s with rest skewed). I don't have any other covariates to include in this analysis. Should this be done with growth mixture modeling or as latent growth curves but only include nonzero data for alcohol use?

2) IF GMM is appropriate, where does the school involvement variable get included? It doesn't seem to fit as a w or x because there are multiple time points. Is there a chapter on this site or in a textbook that describes GMM step by step including how to do the syntax?
 Bengt O. Muthen posted on Monday, May 13, 2013 - 8:52 pm
I'm not sure, but it sounds as if alcohol use is your dependent variable for which you are interested in growth. And school involvement is your 'time-varying covariate". If that is correct and if there is a large percentage of zero alcohol use, there are several techniques to apply, all described in our Topics 3-4 and 6 handouts and videos. Many GMM papers are listed under Papers.
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: