Indirect effects and imputed datasets PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Meghan Scrimgeour posted on Tuesday, June 09, 2015 - 2:22 pm
Is there any way to test indirect effects using imputed datasets? When using the Model Indirect command, I receive the error message that this is not allowed with imputed data. I am currently using MPlus version 7.2.

Thank you for your guidance!
 Bengt O. Muthen posted on Tuesday, June 09, 2015 - 2:39 pm
You can express the indirect effect in Model Constraint. For instance

Model Constraint:
new(indirect);
indirect = a*b;


where a and b are labels given for these parameters in the Model command.
 Meghan Scrimgeour posted on Monday, June 15, 2015 - 12:55 pm
Hi Dr. Muthen. Thank you kindly for your quick reply! The model constraint you suggested solved the issue I was experiencing.

As a follow-up question, is there a way to obtain bias-corrected bootstrap 95% confidence intervals with imputed datasets? When I include CINTERVAL(BCBOOTSTRAP)in the Output command, I receive the following error message: WARNING in OUTPUT command CINTERVAL option is not available with multiple imputation. Request for CINTERVAL is ignored.
 Bengt O. Muthen posted on Monday, June 15, 2015 - 2:04 pm
No, bias-corrected bootstrap 95% confidence intervals with imputed datasets are not available.
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: