Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: medeff results interpretation


From   "Brian Carolan" <brian@briancarolan.org>
To   statalist@hsphsun2.harvard.edu
Subject   st: medeff results interpretation
Date   Mon, 25 Mar 2013 12:19:48 -0700

Hello, 

I am using the user-written commands -medeff- in Stata/IC 11 for Windows
to test for causal mediation effects.

My treatment is binary (T), mediator is continuous (M), the outcome is
binary (Y), and x is a set of pre-treatment confounders. Here is an
example of my command: mediate (regress M T x) (probit Y T M x) ,
treat(T) mediate(M) sims(1000)

My question has to do with the interpretation of the ACME (average
mediation) and ADE (average direct effect). The value of the ACME is .01
(95% CI -.001, .028) and the ADE is .17 ((95% CI .073, .268).

Does this mean the average effect of the treatment variable on the
outcome that operates through the mediator is 1 percent? Also, does this
mean that that treatment increases the probability of the outcome by 17
percentage points?

Thank you for any help.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index