Hi all,
Presently, we are analysing a survey dataset (health) with the following
sampling design:
1. 3 Strata,
2. Simple random sampling within each strata,
3. pweight is created for each Strata based on the respective age, gender and
probability of selection in the household. Weighting was used to
correct for the disproportionality of the sample with respect to the populations
of interest. The data were weighted using the Statistics Bureau data
so that the health estimates calculated would be representative of the
adult populations of those three states.
>From the STATA manual :
"Svy commands can produce proper estimates for subpopulations using the subpop
( ) option."
In particular, my interest is in the application of svyprop and subpop.
Then, my question is how can I convince my colleague to use STATA svy package
for analysis subpopulation estimates based on the above survey, for example the
health characteristics of the population of a town (could be anywhere in the
strata).
My colleague objections are
1. What assumptions does svy use for analysing subpopulation data?
2. What does STATA means by "proper estimates" for subpopulation data?
Hence, is there a convincing way to sway my colleague to using STATA subpop
command?
Regards
Hock
*
* 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/
