• SheeEttin@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    ·
    1 year ago

    Lots of people. I have. The linked AP article says they polled 1165 people, which seems small compared to the 332 million, but the math checks out for a 95% confidence level and a 3% margin of error.

      • SheeEttin@lemmy.world
        link
        fedilink
        English
        arrow-up
        6
        ·
        1 year ago

        If you’re interested in their methodology, they’re named at the bottom of the linked AP article, and they have a page that answers that exact question.

        • surewhynotlem@lemmy.world
          link
          fedilink
          arrow-up
          2
          ·
          1 year ago

          “The poll of 1,165 adults was conducted Aug. 10-14 using a sample drawn from NORC’s probability-based AmeriSpeak Panel, which is designed to be representative of the U.S. population. The margin of sampling error for all respondents is plus or minus 3.8 percentage points.”

          So it doesn’t say how. I have this idea that they call people. I also have this idea that no one in certain age groups and demographics answers their phone, so are underrepresented.

          • SheeEttin@lemmy.world
            link
            fedilink
            English
            arrow-up
            3
            ·
            1 year ago

            And if you Google their name, they have a page that gives a summary of how they conduct their polls.

    • Bendavisunlv6@lemmynsfw.com
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      The thing is that they can be 95% sure they are within 3% of what they’d get if they polled everyone. But that’s it. The way people answer polls doesn’t always translate to the way they vote or do anything else.