Oiman Kwok posted on Monday, October 31, 2011 - 1:58 pm
I am running a simple latent growth model (a simple linear growth model with only intercept and slope factors and no covariates) with 6 binary observed variables (0,1) using MLR with logit link, how can I interpret the means of the latent factors? Are they the logit coefficients?
Also, if I use the default estimation method (WLSMV), how will the means of the latent growth factors be interpreted?
Our default for binary growth is that the thresholds tau are held equal across time, the intercept growth factor mean is fixed at zero, and the slope growth factor mean is free. The model is given on slide 193 of the Topic 3 handout. With ML the slope mean is in logit scale. With WLSMV the mean is in probit scale.
Oiman Kwok posted on Monday, November 07, 2011 - 10:15 am
Dear Dr. Muthen,
Thank you very much for your reply and the very helpful information!