The sign of the trend factor PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Simona Caravita posted on Friday, November 17, 2006 - 7:34 am
Dear Bengt and Linda,

Does the sign (positive/negative mean) of the trend factor in LGM influence on the interpretation of predictor variables' effects on this factor? Let say a variable A has a negative effect on a trend factor that has a negative mean, does this mean that A increases or decreses the decrease of this variable?

Futhermore, if there is negative correlation among the level and slope factors, would this point to floor or ceiling effects in that variable? If so, what could be done, and would the use of multiple indicator vs. the basic LGM matter?

Thank you!
 Bengt O. Muthen posted on Friday, November 17, 2006 - 11:13 am
I'll answer your questions in the order you stated them:

- no

- decreases (makes the slope even more negative)

- it might

- alternatives to try are multiple categorical indicators, 2-part modeling, and censored-normal modeling (all described in the Mplus UG)
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: