[Mediation-information] mediation by levels of pre-treatment covariate
dtingley at gov.harvard.edu
Fri Sep 23 17:45:55 CEST 2011
We haven't implemented this option yet but have talked about it. An
inefficient way to do this is to run separate mediation analyses on subsets
of the data based on a dichotomized version of the pre-treatment covariate.
We hope to have some options here in the future.
On Fri, Sep 23, 2011 at 11:39 AM, Jonathan Renshon <jrenshon at fas.harvard.edu
> I know in the mediate software, you can allow the mediation to vary by
> treatment condition (interaction between treatment and mediator), but can
> you allow mediation to vary by levels (or presence) of a pre-treatment
> covariate (moderated mediation, I guess?)?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Mediation-information