| Message/Author |
|
|
| Jinxin ZHU posted on Monday, October 13, 2014 - 9:18 pm
|
|
|
Dear Prof. Muthen,
I am now running a conditional latent growth model (with gender as covariate for i and s). Would you please help for the following 3 questions:
I want to report the means of the i and s, and ask Mplus to report technical 4 output but only estimated unstandardized mean for i and s were reported.
1.Do Mplus report the standardized means for i and s?
What I know is that we usually report undstandardized result, but the reviewer for my paper ask for the standardized coefficient for the result of LGM. I do not know how to answer.
2. Are standardized means for i and s meaningful?
3. Is there any paper about the interpretation of result of LGM about the standardized and unstandardized outout?
Thank you very much. |
|
|
|
|
1. Not directly - you would have to standardize them yourself using Model Constraint.
2. I don't see why one would want them standardized; that's typically for regression slopes.
3. Can't think of one off-hand.
You may want to ask these questions on SEMNET as well. |
|
|
| Jinxin ZHU posted on Thursday, October 16, 2014 - 10:16 pm
|
|
|
| Thank you very much! |
|
|
|
|
I estimated a LGCM with time-invariant and time-varying predictors. The reviewer wants me to also report standardized estimates. What standardized output would i use for the time-invariant predictors (dummy variables)? What output would i use for the time-varying predictors (continuous and mean centered)?
Model:
c1read on ct1learn cr1age ;
c2read on ct2learn ct2otlh cr2age ct2otath ;
c4read on ct4learn ct4otlh cr4age ct4otath ;
i s | c1read@0 c2read@1 c4read@3 ;
i on black hispanic asian other ses female;
s on black hispanic asian other ses female;
[ct2otlh ct4otlh ct1learn ct2learn ct4learn
ct2otath ct4otath black hispanic asian other ses female cr1age cr2age cr4age];
output: sampstat stand; |
|
|
|
|
Q1. STDY
Q2. STDYX |
|
|
|
|
| Thank you! |
|
|
| Back to top |
