Growth model output PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Tiina Ojanen posted on Thursday, April 20, 2006 - 9:27 am
Dear Bengt & Linda,

I wanted to ask about the estimates for obtained variance in the level and slope factors in multiple indicator growth model: Should I report the standardized or unstandardized coefficients (i.e., bs or Betas?). At the moment, I'm using just the z-scores to indicate that there is significant individual variation in the level and slope factors. Betas in the output are 1.00 for both components, and the unstandardized coeffiecients practically zero (although the z-scores indicate clearly significant variation). Is it just a coincidence that the standardized coefficients both equal 1.00? Are there any differences in interpreting these coefficients (i.e., the variance of the level and slope) between the ordinary and multiple indicator growth models?

Finally, when there is negative residual variance in one of the three latent time scores in a multiple indicator growth model and this time score is fixed to be zero (T1@0), does this command set the variance, or the residual variance, of that variable to zero?

Thank you!
 Linda K. Muthen posted on Thursday, April 20, 2006 - 11:39 am
We most often report unstandardized coefficients. Note that the z-scores refer to the significance of the unstandardized coefficients. I would have to see your output to say anything about the two standardized coefficients of one. The growth factor parameters have the same interpretation whether the growth is in observed variables or latent variables (multiple indicators). In a growth model if t1 is the outcome at time 1, t1@0 refers to the residual variance. You would need to send your input, data, output, and license number to if you want comments on the coefficients of one.
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