Re: st: comparison of means

[Roger Newson <r.newson@imperial.ac.uk>]

Yes, this can be done, using the -parmby- and -metaparm- modules of the 
package -parmest-, which you can download from SSC. This method can do 
an unequal-variance t-test on weighted or unweighted data.

For instance, if you have a Stata dataset with a variable -group-, equal 
to 1 for observations in Group 1 and 2 for observations in Group 2, and 
a sampling-probability weight variable -spweight-, and an outcome 
variable -y-, then you can use the code

parmby "regress y [pweight=spweight]", by(group) norestore list(,)
metaparm [iweight=(group==1)-(group==2)], list(,)

The first command will overwrite the dataset in memory with a new 
dataset, with 1 observation per group and data on the 2 weighted group 
means and their confidence limits, and list this dataset. The second 
command will create and list a temporary new dataset, with 1 
observation, containing the difference between the 2 weighted means, 
with its confidence limits and its P-value, calculated using 
Satterthwaite's unequal-variance t-test formula.

For more about -parmest-, you might like to refer to Newson (2008), 
which gives an example of unequal-variance t-tests using unweighted data.

I hope this helps.

Best wishes

Roger


Reference

Newson, RB. parmest and extensions. Presented at the 14th UK Stata User 
Meeting, 8-9 September, 2008. Download presentation from
http://www.imperial.ac.uk/nhli/r.newson/usergp.htm#uk2008


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: r.newson@imperial.ac.uk
Web page: http://www.imperial.ac.uk/nhli/r.newson/
Departmental Web page:
http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/popgenetics/reph/

Opinions expressed are those of the author, not of the institution.

On 12/04/2011 20:09, Abelev, Melissa wrote:
> Hi -
>
> I'm trying to do some simple descriptive statistics.  I want to do a
> two-group comparison of means.  The trick is that the data are weighted,
> and stata's "classical tests of hypotheses" doesn't allow me to do it.
> I found a journal article on a relatively complex calculation to do it
> by hand.  I'm hoping stata has a way to do this.  Is there?
>
> - Melissa
>
>
> *
> *   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/
*
*   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/