Dear Yuval,
Because I lurk on the Statalist, my replies are late (and often lame, but that's another story).
As an afterthought to this discussion thread, I'd make a semantic distinction: your reweighted internet students still do not represent a *random* sample. Rather, they are *representative* of the population in the dimensions on which you post-stratified. They still might not look like the population in other dimensions -- either observed or unobserved.
regards,
Dan Waldo
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
