This is what I had suspected and had been informed. To Bill Gould, Nick Cox or other STATA leaders...is there any plan or thought in the future of adding this functionality to STATA...there are loads of us who are forced to use SUDAAN to do such analyses, and for those of us with a long love affair with STATA (and less familiarity with SAS, on which SUDAAN is based) this seems hard, especially when STATA does have a set of -svy- commands. In my opinion, if this functionality were included, a large number of occasional SUDAAN users might migrate to STATA, since the SUDAAN site license fee is $700 per year for an individual site license (less for multiple licenses).
Because of this limitation, we are thinking of changing our project to analyze Medical Expenditure Panel Survey data instead, which can be analyzed with STATA...but it's hard, because we really want to work with NAMCS.
So here's a big addition to the STATA 9 wish list, guys!
Joe LeMaster
University of Missouri-Columbia
-----Original Message-----
From: Sayer, Bryan [mailto:[email protected]]
Sent: Monday, February 03, 2003 3:01 PM
To: '[email protected]'
Cc: LeMaster, Joseph W.
Subject: RE: Using STATA to analyze complex national datasets (NAMCS,
etc)
I'm pretty sure someone else has tried this, as I think I looked up the
documentation for them. Basically, the answer is no. Stata does only one
level of sample design. So unless you can reduce a more complex sample
design down to one level, it is not possible in Stata. One issue in
simplifying the sample design is that you can get increased variability in
the variance. So it isn't as simple as just using the highest level.
Perhaps if enough people lean on NCHS, they might come up with something.
Of course, not all NCHS datasets are like this. NHIS is useable in Stata
for instance.
Bryan Sayer
Statistician, SSS Inc.
[email protected]
-----Original Message-----
From: LeMaster, Joseph W. [mailto:[email protected]]
Sent: Monday, February 03, 2003 3:35 PM
To: [email protected]
Subject: st: Using STATA to analyze complex national datasets (NAMCS,
etc)
Colleagues,
A group of us here at the University of Missouri would like to use STATA to
analyze data from the National Ambulatory Medical Care Survey (NAMCS). The
National Center for Health Statistics provides variable and value label
information using SAS code, and SAS code to import the ASCII formatted data
into SAS. It is assumed that the user will use SUDAAN (free-standing or
called from within SAS) to analyze the data.
I was under the impression that STATA could be used to analyze this data,
but I'm not sure how. -svyset- allows for the specification of multiple
strata and primary sampling unit identifiers, but I cannot see immediately
how well it deals with multiple levels of nesting in the pasu's and strata,
which are (I gather) present in the NAMCS etc data.
We'd very much appreciate corresponding (or even collaborating!) with anyone
who has used STATA to analyze this data. A few of us here are regular STATA
users, but this is completely unfamiliar territory.
*********************
from the desk of:
Joseph W. LeMaster, MD MPH
Assistant Professor
Department of Family and Community Medicine
M239 Health Sciences Center, DC032.00
University of Missouri, Columbia
Columbia, MO 65212
Phone: 573-882-4991
FAX:573-882-9096
*
* 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/
