This seems to have come up in April 2006, but I cannot find a resolution in the archives. 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; 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/
