Modeling 3 waves of data as quadratic... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 daniel adkins posted on Thursday, February 14, 2008 - 10:11 pm
is there any trick for modeling 3 waves of data as a quadratic LGM, with means and variances estimated for all 3 growth factors? i've rearranged the data from 3 measurement occasions to an age metric with 14 ages. as i understand it, the model should technically be unidentified with 3 growth factor variances specified, but i wonder if i can use fiml or maybe a multiple group approach to get around this (or perhaps constrain the growth factor covariance=0?). any advice?
 Linda K. Muthen posted on Friday, February 15, 2008 - 6:21 am
If you have rearranged the data, you no longer have three time points. You now have 14 time points. Therefore a quadratic model is identified.
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: