st: Re: Stratified Analyses

["Joseph Coveney" <stajc2@gmail.com>]

Charles Christian wrote:

Hello!  I have been using SAS for many years through my employer.  I
am now retired and looking at Stata as a software package for my own
use.

I have looked at what material I have been able to locate and am
unable to discern is Stata handles two things that SAS did:  1.)
Aggregated t-tests (controlling for various variables); and, 2.)
Stratified CMH and Wilcoxon analyses (again, controlling for various
variables).

Can someone advise if Stata handles those two types of analyses?

--------------------------------------------------------------------------------

I'm not quite sure what your first one is (it sounds like ANCOVA to me), but you
can take a look at the various commands that are briefly described at
http://www.stata.com/support/faqs/statistics/meta-analysis/ and see whether one
or more fits the bill, for example, -metareg-.

For your second, there is a user-written command, -emh-, that does what SAS's
PROC FREQ with the CMH option does. There is another command, -vanelteren-, that
specifically does stratified Wilcoxon tests, but -emh- can also do that.  It's
easiest to see more about these two commands by using Stata's -findit- command,
but if you don't already have Stata, then you can get an idea from
http://ideas.repec.org/c/boc/bocode/s457190.html and
http://ideas.repec.org/c/boc/bocode/s439101.html .

Joseph Coveney



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/