Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: medeff results interpretation
From
"Brian Carolan" <[email protected]>
To
[email protected]
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/