Burton sampling question

FD78@vaxb.acs.unt.edu
Sat, 11 Dec 1993 11:47:12 ECT

I believe that the sample you mentioned (involving white females, four
fourteen or older) would be valid for white females, fourteen or older,
who had an occupation. It would not be valid for white females, fourteen
or older, in general. In the first case (white females, fourteen or
older, who had an occupation), your student will have selected these
females in a systematic, unbiased manner. In the second case (white
females, fourteen or older, in general), the sample would not be
unbiased since white females of the right age would have been
arbitrarily excluded from the sample ("arbitrarily," that is, if your
student were studying white females of that age group as a whole).
If your concern centered around the use of a systematic sample
rather than a random sample, you can take comfort in the discussion of
systematic samples in Konrad H. Jarausch and Kenneth A. Hardy,
QUANTITATIVE METHODS FOR HISTORIANS (UNC Press, 1991), 68-70.
One caveat: this is the opinion of someone who has some experience
with sampling and quantitative methods but has received no formal
training in statistics. Someone with more formal training than I may
disagree. (But I don't think so.)

Richard Lowe
U. of North Texas