This seems like a worse own goal than “defund the police” as a slogan, best of intentions but the execution was so bad.

I’d put heavy money on Republicans running chunks of this interview hard during the run up to the midterms.

I have a sinking fear this will be one of those tragicomic moments of earnest, well intentioned progressives inadvertently helping trump, like when many sat out the 2024 election in a misguided attempt to somehow help Palestinians.

  • FiniteBanjo@feddit.online
    link
    fedilink
    English
    arrow-up
    1
    ·
    9 days ago

    I explained my reasoning pretty well in the comment. If you assume the sample accurately maps to the whole population then the numbers don’t add up, meaning some single digit percentage is lying

    • MyBrainHurts@piefed.caOP
      link
      fedilink
      English
      arrow-up
      1
      arrow-down
      2
      ·
      9 days ago

      As for the exit poll, amounts below 10% are generally considered response bias.

      But that’s not at all a thing in statistics. I’m not sure what you’re misunderstanding but you can’t just make up arbitrary cutoffs, that’s not at all how it works.I get that you’re unhappy you’re wrong but pretending stats are however you feel is just silly.

        • MyBrainHurts@piefed.caOP
          link
          fedilink
          English
          arrow-up
          1
          arrow-down
          2
          ·
          8 days ago

          Lol feel free to share a reasonable link but having done a heavy stats degree, I’ve never encountered this. I could call the former chair of statistics Canada and ask but again, pretty sure this isn’t a thing.

          It’s just so weird, when confronted with actual proof you’re going to make up statistical practices?

          The wildly charitable interpretation, maybe you are getting confused by sample size or something? It wouldn’t apply here but…

            • MyBrainHurts@piefed.caOP
              link
              fedilink
              English
              arrow-up
              1
              arrow-down
              2
              ·
              8 days ago

              Response bias exists, but there’s no “oh, under 10%, we chalk that up to response bias.”

              Again, you could prove yourself correct with a sinple link. According to you, it’s just a quick search away. If you’re telling the truth, well boy, isn’t it kind of weird to keep responding instead of taking the handful of seconds to prove yourself right?

                • MyBrainHurts@piefed.caOP
                  link
                  fedilink
                  English
                  arrow-up
                  1
                  arrow-down
                  2
                  ·
                  8 days ago

                  Again, you could prove yourself right. But right now, anyone with any smidgen of stats understanding knows you’re not telling the truth.

                  According to you, you could easily prove yourself right by sharing the results of a quick search…

                  • FiniteBanjo@feddit.online
                    link
                    fedilink
                    English
                    arrow-up
                    1
                    arrow-down
                    1
                    ·
                    8 days ago

                    You’re clearly upset I think you just need to walk away from this and be done, I’ll go ahead and give you last word: