Lee, I think you do need to use svy commands for unclustered,
probability-weighted data. All the estimates I can think of involve the
weights as denominators; the weights themselves are random variables, and
division by random variables is not defined. So in effect, estimates from
finite-population probability samples are ratio-estimates if only because of
the random distribution of weights. Then Taylor Series or bootstrap methods
conservatively approximate the variance.
Arnold H. Levinson, PhD
Associate Scientist
Center for Research Methodology and Biometrics
AMC Cancer Research Center
Assistant Professor
Department of Preventive Medicine and Biometrics
University of Colorado School of Medicine
303.777.8801
[email protected]
fax 303.239.3394
*
* 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/
