Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Alternative to Fisher's exact test for weighted observations


From   "Garth Rauscher" <garthr@uic.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Alternative to Fisher's exact test for weighted observations
Date   Wed, 25 Mar 2009 13:23:54 -0500

Dear listserve members,

I have data from a survey of 40 facilities and I am interested in examining
whether a higher minority caseload is associated with less facility reliance
on specialists. In other words, are minority patients disproportionately
cared for at facilities that rely more on generalists than specialists for
the procedure in question. I have three variables of interest: 

(1) Proportion minority patients (Minority: 1=>75%, 0=Less); 

(2) Reliance on specialists (Specialists: 0=Never, 1=Sometimes, 2=Always)

(3) annual procedure volume (volume). The total annual volume across the 40
facilities is about 150,000. 

I would like to obtain an exact Chi-square or other similar test that would
provide valid p-value for the association between Minority and Specialists,
with facility (not procedure) as the unit of analysis. I tried:

tabulate specialists minority [aweight=volume], chi2 exact

But cannot perform Fisher's exact test with weights. So I am looking for an
alternative method and hoping that there may be some ideas out there.

Thanks very much in advance.


Garth Rauscher
Division of Epid/Bios (M/C 923)
UIC School of Public Health
1603 West Taylor Street
Chicago, IL 60612
ph: (312)413-4317
fx:  (312)996-0064
em: garthr@uic.edu
 



*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index