|
|
Testing parallel mediator differences |
|
Message/Author |
|
Rich Mohn posted on Tuesday, March 14, 2017 - 12:18 pm
|
|
|
In PROCESS (model 4) you can test if the indirect effects from parallel mediators are different from each other - it subtracts one indirect from the other and creates a bootstrap CI. I've tried several ways to replicate this in Mplus and have been unsuccessful . . . any suggestions would be appreciated - thanks. The model I'm working with has multiple outcomes, which PROCESS can't handle, but I'd like to be able to test indirect effect differences. |
|
|
Use the Model Constraint command to form any difference you like, expressed in terms of parameter labels that you give in the Model command. |
|
Back to top |
|
|