Re: st: Would _sgmediation_ work for dichotomous dependent variable?

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Sun, 1/8/10, Bo MacInnis wrote:
> Interested in using the wonderful _sgmediation_ command
> that Phil Ender provided, I wonder if the use of
> _sgmediation_ is restricted to linear regression with
> continuous dependent variable. I have a case where the
> dependent variable is categorical or even dichotomous, and
> I've been using probit for the main (total effect) for the
> dichotomous dependent variable. I'd like to use
> _sgmediation_ for the mediational analysis.

That is not possible. The estimation of mediation effects is
much more easier in linear regression. This kind of 
calculations do not generalize to non-linear models like
-probit-. A discussion of this issue can be found in: 

M.L. Buis (2010) "Direct and indirect effects in a logit 
model", The Stata Journal, 10(1), pp. 11-29.

This paper also contains one possible sollution.

Hope this helps,
Maarten (author of the above article)

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

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