Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10 5/3/83; site cmu-cs-cad.ARPA Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!mhuxv!mhuxh!mhuxi!mhuxm!mhuxj!houxm!whuxlm!harpo!decvax!genrad!panda!talcott!harvard!seismo!rochester!cmu-cs-pt!cmu-cs-cad!mjc From: mjc@cmu-cs-cad.ARPA (Monica Cellio) Newsgroups: net.abortion Subject: Re: Statistics Message-ID: <283@cmu-cs-cad.ARPA> Date: Wed, 6-Feb-85 23:00:30 EST Article-I.D.: cmu-cs-c.283 Posted: Wed Feb 6 23:00:30 1985 Date-Received: Sat, 9-Feb-85 07:23:12 EST Organization: Carnegie-Mellon University, CS/RI Lines: 18 [ No quotes here; we all know the exchange by now... ] I am generally skeptical of polls because it is very hard to get a representative sample of the public (where "public" can sometimes be defined by the issue, e.g. only computer users for a programming language preference poll, etc). Any poll that does not have a representative sample is pretty much by definition invalid. If I see reference to a poll that does not mention that an effort was made to get a representative sample (how this was done always makes interesting reading, but is not essential), I usually don't give it a second look. The few exceptions are pollsters who reputedly always make this effort (pros such as Gallup). No offense meant, but you aren't one of those pollsters whose results I will consider semi-valid on the basis of your name alone. -Dragon -- UUCP: ...ucbvax!dual!lll-crg!dragon ARPA: monica.cellio@cmu-cs-cad or dragon@lll-crg