Is it acceptable for some subjects in my dataset to
have 2 different sampling weights? For example, say we
know that we undersampled the 0-1 yo's because, when
compared to national data for infant mortality rates,
our sample mortality rates are much lower. Thus, we
can "bring up" our sample mortality estimates to these
published levels through the use of correction
factors. However, we would not likewise correct the
weight the non-dead children because we don't know if
or by how much we undercounted these infants.
Furthermore, if we also correct for undersampling of
all infants (deceased and non deceased), wouldn't we
effectively be diluting
our mortality rates back the levels of the
undersampled population rates?
When correcting for the deceased infants, do we
include the correction factor in the sampling weight
or adjust for it after running the survey command?
Thank you.
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
*
* 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/
