...I agree with Arcady. Your starting population, from which you're culling a sample, is limited to people who use the various forums and further to people who read your posts for the survey, and then again to people who chose to respond to it. Compare the actual total population of Poser users with the very specifically filtered population sample you're working with. I'm not talking at all about sample size, but rather sample quality. It certainly can't be said to be accurate to 7 or 8 percent as a representation of the entire Poser population, but only of that very filtered population. Some results could and probably do have a very large error with respect to the total Poser user base. Still a neat poll though. Probably tells us which way the wind is blowing, but not how fast ;)