Default centering for variables PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Giselle Guan posted on Wednesday, December 11, 2019 - 7:45 am
Dear Dr. Muthen,

After referring to the users' book and previous discussion I'm still confused about how MPlus deals with centering of variables not specified using "within" and "between" option.

The IV in my model has a ICC(1) larger than 0.2, so I think it's reasonable to model it at both the within and between level.

Accordingly, under the "DEFINE" section, I don't center it neither using "Center...(groupmean)" or "Center...(grandmean)"; in the "Usevariables" section, I don't specify the IV neither as a "within" nor "between" variable.

Under this situation, will Mplus automatically apply groupmean centering to the IV? or no centering is applied?

Many thanks for your patience and guidance!
 Bengt O. Muthen posted on Wednesday, December 11, 2019 - 3:29 pm
UG ex 9.1, part 2 on pages 274-275 show that latent variable decomposition is done automatically when an X variable is not declared as Within or Between. In the random slope case, only Bayes does this lv dcomp.
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