Freed loading growth PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Yaacov Petscher posted on Tuesday, November 01, 2011 - 12:37 pm
Greetings - For a six time point model where the data demonstrate non-linearity, I've specified a freed loading growth model where time 1 is centered at 0 and time 6 is centered at 5. My understanding is that the slope will reflect the mean change from time 1 to time 6 while the freed estimate will provide an indication of general acceleration or deceleration across the other time points.

My raw means for time 1 and 6 are 9.9 and 13.9, respectively, and the predicted mean for the growth curve at time 1 is 8.9 with a slope of 1.15. Is accurate to state that the mean change from t1 to t6 is 1.15 but to get the fitted mean it is t1+(1.15*5)?
 Bengt O. Muthen posted on Tuesday, November 01, 2011 - 10:38 pm
The mean change from t1 to t6 is 1.15. The estimated mean at time t (t=1-6) is

[i]+ lambda_t*[s]

where [i] is the intercept growth factor mean, lambda_t is the estimated time score for time t (lambda_1=0), and [s] is the slope growth factor.

If you want the SE of the expression above, you define it as a New parameter in Model Constraint.
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