Reply to thread

Message: <blockquote data-quote="Beginning of the End" data-source="post: 4610238" data-attributes="member: 55271">One of the most basic things that go wrong with any poll is when you allow the sample to become non-representative. You're looking at a sample of gamers who (a) choose to participate in online communities; (b) choose to participate in this online community; and (c) choose to participate in this poll.All of those factors significantly warp the poll. I'm willing to accept that it's a fairly accurate assessment of the mood of the board (although even there the self-selection factor will have a significant impact), but for any wider demographic it's essentially irrelevant.Uh, no.First, the people who subscribe to Dungeon and Dragon are a self-selecting subset. Second, the method you're suggesting warps the result towards those who don't move. Finally, the method you're suggesting probably still warps the results to active players (who are more likely to respond to a mail poll like this).If you wanted better data, you'd perform the type of market research that Ryan Dancey had performed and then released publicly.</blockquote>