Equivalent models in sequential LGCM PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Joo-Young Lee posted on Tuesday, April 27, 2010 - 10:18 pm
Hi,

I'm interested in a relationship of 2 growth processes and wish to examine which process precedes the other.

In order to do this, I analyzed 2 sequential latent growth curve models: regressing growth factors of A on growth factors of B and vice versa. Also, to examine the gender effect, I performed multiple group analysis in each model.

But one of our colleagues said that they are statistically equivalent and trying both models is not so appropriate.

What do 'equivalent models' mean?
And is it really so inappropriate to analyze that way?

If so, what other alternatives would you suggest in order to figure out which growth process comes first and to examine the gender effect additionally?


Thank you!
 Linda K. Muthen posted on Wednesday, April 28, 2010 - 8:21 am
Statistically equivalent models are models that obtain the same chi-square. Time-order is determined by when the events occurred. The choice of model should be based on theory.
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: