[Mediation-information] medeff

[dustin tingley]

Stella-
I don't know Chen's paper. Our definitions come from an earlier statistics
literature. I suggest consulting one of our papers, available here
http://imai.princeton.edu/projects/mechanisms.html , where we define the
ADE (e.g., the Psychological Methods paper), or the Stata Journal paper
Raymond and I wrote for the medeff command.

There are no general rules regarding effect sizes. That is question and
discipline specific, just like with normal regression methods. Sorry I
can't be of more guidance on that.

best,
Dustin

Dustin Tingley
Government Department
Harvard University
http://scholar.harvard.edu/dtingley
Director, Program on Experience Based Learning in the Social Sciences
http://projects.iq.harvard.edu/peblss/
Contribute to the Research Reservoir:
http://projects.iq.harvard.edu/undergradscholars/pages/research-resources



On Fri, Apr 27, 2012 at 10:50 AM, Babalola, Stella O.
<sbabalol at jhuccp.org>wrote:

>
> Dear Dr. Hicks:
> I am using the medeff command in Stata to calculate causal effects. I
> assume that the measurement of ADE produced by the command is equivalent to
> Chen's d. Am I correct? Also could you provide me with an idea about how to
> classify the magnitude of the effects obtained. For example, what is the
> cut-off for small, moderate and large effects. Thank you. ~~Stella
>
> Stella Babalola, PhD
> Associate Professor
> Johns Hopkins University - Center for Communication Programs
> 111 Market Place, Baltimore, MD 21202
> Tel. 410-659-6300
> email: sbabalol at jhsph.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/mediation-information/attachments/20120427/c4047958/attachment.html>