[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: Balancing survey data in Survey package
This was discussed only recently. Richard
[Dick] Campbell asked a question and I replied.
(You might find it easier to search for
his name than for mine in the Statalist archives.)
If you type in Stata,
. search iterative proportional fitting
you will see pointers to programs by
. ssc desc mstdize
identifies a program discussed in the
thread initiated by Richard. In that
thread was mentioned the possibility of
using -glm- and offsets to do this.
I have no idea whether your problem includes
-svy- type complications, but I suspect that
none of the programs takes account of
survey design subtleties.
Walter R. Paczkowski, Ph.D.
> I'm a new user looking for help with the Survey package. I
> want to balance (also called raking) an Internet survey to
> make sample proportions (say for key demographics) agree with
> Census data. This is a very common practice that originated
> with Deming under the title iterative proportional fitting.
> I don't see anything that will do this, although I really did
> expect this very common procedure to be there. Any suggestions?
* For searches and help try: