• Rapidcreek@lemmy.world
    link
    fedilink
    arrow-up
    40
    arrow-down
    1
    ·
    2 months ago

    Pre-convention, ABC - Wash Post - Ipsos poll had Harris leading Trump 49% - 45% in a two person race.

    This one has Harris 52% - 46%.

    The gender gap is part of the convention bounce

        • whenyellowstonehasitsday@fedia.io
          link
          fedilink
          arrow-up
          2
          ·
          2 months ago

          so if a net change between the two is within the margin of error, you can’t state with confidence that it’s an increase and not just random noise

          • Rapidcreek@lemmy.world
            link
            fedilink
            arrow-up
            1
            arrow-down
            2
            ·
            2 months ago

            What? I can say with reasonably certainty that if other polls match their results, which they do, then the MOE is meaningless. I can also say that since the same pollster and ABC took both polls, ABC is straight lying when they say there was no bounce. Their own data says otherwise.

            • whenyellowstonehasitsday@fedia.io
              link
              fedilink
              arrow-up
              3
              ·
              2 months ago

              I can say with reasonably certainty that if other polls match their results, which they do, then the MOE is meaningless

              no that isn’t how statistics works

              Their own data says otherwise.

              if you look at only their own data then no, there was a statistically negligible increase

              • Rapidcreek@lemmy.world
                link
                fedilink
                arrow-up
                1
                arrow-down
                2
                ·
                2 months ago

                FFS. You don’t know how polls work. There are always small increases when you reach 50% The more important issue in these polls is that Trump has a ceiling of ~46%

          • Buffalox@lemmy.world
            link
            fedilink
            arrow-up
            1
            arrow-down
            1
            ·
            2 months ago

            I suppose that confidence would need to have it’s own margin of error. But it seems silly to say we can’t say anything from such a significant difference.
            I’d say the likelihood is still on the side that Harris is doing better than Trump.