If you are entering your 0/1 dummy variable directly into the logistic regression, then Stata does not "know" it is a dummy and treats it as it would any linear term... in other words the "reference category" will be 0 (as beta*0 = 0) and the odds ratio quoted will be for the category 1 (as beta*1 = beta).
If, on the other hand, you are (unnecessarily) using -xi- to declare your variable as a dummy then the default action will still be to omit the zero category (the first numerically or alphabetically) but this can be changed with -char varname[omit] #- or -char varname[omit] "string"- depending on the storage type... see the help for -xi-.
Hope this clarifies things for you.
David
[email protected]
-----Original Message-----
From: Carter Ivan Rees [mailto:[email protected]]
Sent: 07 October 2004 21:18
To: [email protected]
Subject: st: Reference Category for Dummy Variables
I am running a logistic regression and would like to know which category of a 0/1 dummy variable Stata uses as the reference category (first or last). The results I am getting right now are counter intuitive. I have checked my coding scheme and it is OK. Thanks in advance.
*******************************************
Carter Rees, M.A.
Associate Research Scientist
Wyoming Survey�&�Analysis Center
710 Garfield St.
Suite 320
Laramie, WY 82072
307-399-0496
*******************************************
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
