Parameterization and standardization PreviousNext
Mplus Discussion > Categorical Data Modeling >
Message/Author
 Chen, He posted on Saturday, October 16, 2010 - 10:25 pm
Dear Linda,

I have 3 questions, thanks a lot!

Background: I use MPLUS 5.0, also set BOOTSTRP=1000.

1. Parmeterization: I run a path model with binary variables both influenced by and influence other binary variables. Should I must use "theta"?

2. Standardiezed coefficient: all the varibles in my model are binary. Is it appropriate for me to use "STD estimation" (from the output, only STD and STDYX available)?

3. CI: I also want to report confidence intervals for standardized coefficients. However, there is only standardized coefficients, but no CI or SE in the output. IS this normal? How can I calculate CI or SE?

Thank you so much!
 Linda K. Muthen posted on Monday, October 18, 2010 - 9:42 am
1. Yes.

2. You should use StdY which you can obtain by dividing StdYX by the standard deviation of x.

3. For a model like yours we do not give standard errors of the standardized coefficients so confidence intervals are not available.
 Chen, He posted on Tuesday, October 19, 2010 - 7:01 am
Dear Linda,

Thanks a lot!
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: