Dear All:
I have a somewhat related question in order to do my weighted regression
I have to "reorganize" my strata because I have too many with just one
PSU. Pain staking but no problem.
But I do have one question. My data consists of several distinct
subsamples, such as one for foreigners, one for natives, one for a
"refreshment" sample, etc. One such subsample let's call it X refers to
a distinct population (immigrants) has a stratum in which the values for
the strata are all the same and the PSU (which is NOT a geographic one)
are all the same as well.
What shall I do with those observations? I may just have to drop this
whole subsample but I would rather not. Do you have any suggestions? I
would appreciate any advice.
Example:
1 obs = 1 individual
Obs Strata PSU
2091. | 12082002 0 |
2092. | 12082002 0 |
2093. | 12082002 0 |
2094. | 12082002 0 |
2095. | 12082002 0 |
|----------------------|
2096. | 12082002 0 |
2097. | 12082002 0 |
2098. | 12082002 0 |
2099. | 12082002 0 |
2100. | 12082002 0 |
|----------------------|
Richard, thank you for that link to that great discussion.
Also be advised that Winship's papers can also be easily accessed right
here:
http://www.wjh.harvard.edu/soc/faculty/winship/weights2003.pdf
Best wishes,
Sabine N. Merz
[email protected]
---cut--cut--
* Back in December there was a discussion as to whether you should
weight
at all; see Dick Campbell's message at
http://www.stata.com/statalist/archive/2003-11/msg00653.html
and the followups. Dick cited a very good 1994 article by Winship and
Radbill saying that the answer was usually no. Left unclear to me is
whether that is still true with Stata 2004 and its pweights and svy
commands. My reading of the Users Guide is that you should use pweights
rather than not weight at all. If there is a more contemporary
discussion
of these issues I would like to see it.
*
* 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/
