This seems to have come up in April 2006, but I cannot find a resolution in
the archives. The author indicated that inteff does not work with svy data.
At the same time, Stata Tech Support kindly confirmed the syntax but I
repeatedly get a conformability error r(503) for subsamples of the
population (inteff works for the full sample):
svy: logit HSdip Fpov90 gender3 povgen IV1...IVn if white==1
*or:
svy, subpop(white==1): logit HSdip Fpov90 gender3 povgen IV1...IVn *and
then, following either of the previous logit models: inteff HSdip Fpov90
gender3 povgen IV1...IVn if white==1
Various efforts (ignoring survey design, per author's guidance; changing
subsample; forcing Stata version 8.2, per an archived suggestion) have not
removed the problem. I consistently can only run inteff after a logit model
of the full sample.
Any ideas would be much appreciated.
Elizabeth Mumford
[email protected]
*
* 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/
