st: Odds Ratio and logistic regression

[BISSERY Alvine <Alvine.BISSERY@hop.egp.ap-hop-paris.fr>]

Hello all, 

I'm doing a logistic regression , with only one variable 

logistic Y X

The variable X is a dummy variable coded 0,1,2,3 , so I use the function:

xi:logistic Y i.X

This gives me 4 OR , and 4 CI

I used the command "tabodds Y X" to compare the resultas, and I've been very
surprised that the p value and CI are really different with these 2 methods
(p=0.024 or p=0.064)

Why the results are so different ? I know that the test used are differents
(Wald and Chi-2), but which test do I have to choose ?

Thank you for your help

Alvine Bissery
> 
*
*   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/