Hello,
I am running a model using GSEM that has a binary independent variable (discrimination: yes or no), binary dependent variable (self-rated health: good or poor) and two mediators, one is continuous (stress level) and the other is binary (isolation: yes or no). My question is this: how do I calculate the indirect effects given I have a continuous mediator and binary dependent variable (the coefficient from continuous mediator -> binary dependent variable would be in log odds)? I have scoured this forum and seen similar questions asked without a response. Does nlcom work to calculate indirect effects with a continuous mediator and binary dependent variable? And is there a way to then calculate the total indirect effect based on one continuous mediator, one binary mediator and a binary dependent variable? I've read binary_mediation has been used to do this, but it seems like it is out of date (I am using STATA 17.0).
I am running a model using GSEM that has a binary independent variable (discrimination: yes or no), binary dependent variable (self-rated health: good or poor) and two mediators, one is continuous (stress level) and the other is binary (isolation: yes or no). My question is this: how do I calculate the indirect effects given I have a continuous mediator and binary dependent variable (the coefficient from continuous mediator -> binary dependent variable would be in log odds)? I have scoured this forum and seen similar questions asked without a response. Does nlcom work to calculate indirect effects with a continuous mediator and binary dependent variable? And is there a way to then calculate the total indirect effect based on one continuous mediator, one binary mediator and a binary dependent variable? I've read binary_mediation has been used to do this, but it seems like it is out of date (I am using STATA 17.0).
