--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
--- On Tue, 20/10/09, Kate Rohrbaugh wrote:
> I have a dataset of about 300 observations and I am going
> to separate them into 4 groups -- not totally equal in size
> but based on certain characteristics. I want the four
> groups to all have similar distributions along a certain
> metric (i.e., cost) -- so the mean, median, SD, etc., are
> similar, so I want to sample observations from each
> group. I'll choose one of the four groups as my
> baseline (probably the one with smallest range or the fewest
> number of observations - but I'm open to any suggestions
> about that as well).
>
> What's the best approach for doing this? I suspect
> this question has been addressed numerous times in various
> places, so being pointed in the right direction is
> sufficient. Note: we still have Stata 9.2, so nothing
> too fancy.
Sounds like propensity score matching using a single variable.
This is implemented in Stata by Edwin Leuven and Barbara
Sianesi as the -psmatch2- package, which will run under
Stata 9.2. To install that package type in Stata:
-ssc install psmatch2-.
Hope this helps,
Maarten
*
* 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/
