Analyzing the design variable of a survey

#1
Hi, this is a stupid question. I have a survey that is stratified on size of county in square miles and designation as urban or rural. For brevity, let's say there are 2 sizes.

So, in Stata I would set it up:

gen s_id = .
replace s_id =1 if loc=="Urban" & size==1
replace s_id =2 if loc=="Urban" & size==2
replace s_id =3 if loc=="Rural" & size==1
replace s_id =4 if loc=="Rural" & size==2

svyset id [pw=wt], strata(s_id)

If I want to get population estimates of "loc" do I need to remove it from s_id?

Or can I use the svyset above and run (assuming conversion of loc to a numeric variable):

svy: tab loc

?