Dear Statalisters
I am estimating a logit model with clustering. The unit of
analysis
is the record of an individual within an establishment. I am
using
the cluster() option to account for the fact that although
the sample
of establishments may be randomly obtained, individuals only
appear
if they are in one of the chose establishments.
However, my question is whether the cluster() option is the
appropriate method when the records of individuals are not a
random
sample taken from the random sample of establishments, but
rather a
survey. That is, they represent an 100% sample.
many thanks
Philip Stevens
*
* 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/
