Decision on Growth Mixture Model PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Anonymous posted on Sunday, March 06, 2005 - 8:43 am
Hello, I'm new to growth modeling. I have poisson data that I would like to examine for subgroups. The analysis that I want to perform is one that is presented by Nagin (1999). As I have read through the Mplus User's guide 3.11, which is awesome by the way, I am confused as to which GMM model I should use, and which plots command (1,2,or 3) will give a line graph with the subgroups? The key for me is to determine the subgroups. I think I understand how to put the covariates in to explain the subgroups at a conceptual level, but I need some guidance as to which model in the User's guide I should follow for the initial grouping? Thank you.
 Linda K. Muthen posted on Sunday, March 06, 2005 - 8:59 am
It would be Example 8.11 with the addition of covariates. The Nagin model is a LCGA not a GMM. Example 6.10 illustrates the addition of time-invariant and time-varying covariates. Ask for PLOT3. Then you will get as much as is available.
 Anonymous posted on Sunday, March 06, 2005 - 9:09 am
Thank you so much for the help and the quick response! The Mplus product is awesome.
 Annie Desrosiers posted on Wednesday, January 28, 2009 - 6:52 am

I'm doing a basic Latent Growth Model and I want to know if there any rule about how many covariates that we can put in the conditional model?

Thank You
 Linda K. Muthen posted on Wednesday, January 28, 2009 - 9:09 am
There is no rule that I know of regarding the number of covariates. The group of covariates should be thoughtfully selected.
 Carolin posted on Monday, July 25, 2011 - 2:16 am

is there any critical value concerning the smallest class size in growth mixture modeling? If so, do you know any references?

Thank you!
 Linda K. Muthen posted on Monday, July 25, 2011 - 2:31 pm
Not that I know of.
 Michael Lorber posted on Friday, May 02, 2014 - 5:05 pm
Dear Linda and Bengt,

Is it valid to estimate a growth mixture model for two separate DVs, such as two slopes and two intercepts? For example:

ix sx | x1@0 x2@1 x3@2;
iy sy | y1@0 y2@1 y3@2;

If it is valid, have you seen it in the literature?

 Linda K. Muthen posted on Saturday, May 03, 2014 - 11:42 am
One can estimate a growth mixture model for more than one process. I don't know of any reference. You could search under parallel or sequential processes.
 Michael Lorber posted on Saturday, May 03, 2014 - 8:13 pm
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message