Standardization using TYPE=THREELEVEL PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Anonymous posted on Tuesday, January 02, 2018 - 7:49 am
Dear Profs. Muthén, I use a threelevel model and I want to standardize my L2- and L3-predictor variables.

(1) I'm not sure (and the manual doesn't indicate) whether Mplus takes the multilevel model into account when standardizing the variables, i.e., does Mplus standardize each variable based on it's unique realization on the specific level (e.g., using the aggregated mean and sd for a L2 variable)? Or does it use some kind of a global mean and sd for the standardization?

(2) When I use FIML: How are the means and sds for the standardization retrieved? Are the standardized values of the variables based on the FIML procedure or on the raw variables?

Thank you very much in advance!
 Bengt O. Muthen posted on Tuesday, January 02, 2018 - 2:18 pm
Standardization used in the Define command is based on the overall/global univariate statistics given in the output, not on multivariate FIML-estimated quantities.
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