Mean of intercept when dividing by a ... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Diana posted on Saturday, January 13, 2018 - 6:36 pm
When running a growth curve model, I got the error that variance is greater than the maximum allowed of 1000000. I divided my outcome variables (rates of STDs) by a constant of 2 and that fixed the problem.
My questions are:
1. When interpreting the mean of the intercept do I need to multiply it by 2?
2. When interpreting the mean of the slope, do I need to multiply by 2?
Thank you!
 Bengt O. Muthen posted on Sunday, January 14, 2018 - 10:36 am
Instead of bothering with multiplication of estimates, why don't you instead re-define the variable? For instance, if a variable is recorded in centimeters and you have a variance of 1000000 (6 zeros), you can divide the variable by 100 - that is, define the variable in meters instead - which would cut down the variance by 100*100, which gives you a variance of 100 instead of the original 1000000. Dividing by 2 has no substantive meaning.
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