Sorry to sneak in on this thread, but I can't figure out how to get
my question posted...
I'm trying to figure out how to something like a 3-way cross-
tabulation. I have a categorical behavior variable, and categorical
demographic variables. I want a chi-square on the demographics WHEN
the behavior variable is "1", so something like...
svytab ctsage sexor if CurrSmkr==1, subpop(femsub) ci
When I put in that CurrSmkr==1 (of a 1/0 variable) (I know that this
kind of specification is not advisable in svytab), which happens to
be about 64% of my sample, the demographic percentages are
calculated as proportions that add up to 64%. I either want Stata
to see this as the whole sample (100%) in this case, OR I want -
essentially - a 3-way comparison... sex X age when CurrSmkr=1 and
sex X age when CurrSmkr=0. Is that possible? Do I need a "by"
group, and if so, what is the syntax for that? I'm finding my
documentation less than helpful in this regard.
Thanks in advance; you all are terrific resources!
*
* 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/
