• Hazzia@discuss.tchncs.de
      link
      fedilink
      arrow-up
      1
      ·
      edit-2
      9 months ago

      I get the reason people have been saying that on these poll posts, but I just want to point out that calculations for statistical significance exist specifically to avoid having to sample 90+ percent of a population to gain an adequite confidence level. Mind you a properly formatted survey will be designed to avoid over-representing certain subpopulations, which polls generally have about a 0% chance of doing, and for a poll that’s supposed to represent the whole nation the sample size is definitely way too damn small, so it’s still useless. But everyone implying that statistical probability can’t be calculated accurately without 100% sample data bugs me.