Re: st: Confounding using chest

[Maarten Buis <maartenlbuis@gmail.com>]

On Sun, Feb 26, 2012 at 5:05 AM, Jennifer Calder wrote:
> I have been trying to use the chest command to check confounding in a binary logistic regression.   I am building the logistic regression using the purposeful selection.   I am not quite sure how chest is checking confounding or how the betas in the output table is calculated as well as how this program handles dummies, as I get as single coefficient  and percentage change in beta for dummy variables.

-chest- is user written, so per the Statalist FAQ you must say where
you got it from. This requirement is not to annoy you, but to help you
ask answerable questions. In case of user written software, there are
often multiple versions floating around in cyber space, and it
obviously helps when we are all talking about the same version.

To answer you question, did you read:

Zhiqiang Wang (2007) Two postestimation commands for assessing
confounding effects in epidemiological studies. The Stata Journal,
7(2): 183-196.
<http://www.stata-journal.com/article.html?article=st0124>

In general I am not too convinced by commands like these. The real
issue is to distinguish between confounding and intervening variables
and these kind of commands tend to distract from that.

Hope this helps,
Maarten

--------------------------
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/