The poll also highlights an ongoing issue for the Democratic Party in that Harris is considered less popular among voters than President Joe Biden.

  • BertramDitore@lemmy.world
    link
    fedilink
    English
    arrow-up
    37
    ·
    5 months ago

    A DailyMail.com survey of 1,000 likely voters

    Well that seems like some untrustworthy bullshit. Give me a sample size of 100,000 and maybe I’ll listen. Fucking polls…

    • ricecake@sh.itjust.works
      link
      fedilink
      arrow-up
      28
      ·
      5 months ago

      A thousand voters is actually a perfectly good statistical sampling for the US population.

      What matters is that

      • the daily Mail isn’t overly credible
      • they don’t mention how the thousand voters were selected.

      A thousand people is enough to get you into single digit margin of error for a population the size of the US. If you find them by dialing random landlines, or mailing surveys to people with a voter history, your sample biases towards the demographic that still has landlines or that’s willing to take the time to fill out a paper survey.

    • njm1314@lemmy.world
      link
      fedilink
      arrow-up
      14
      ·
      5 months ago

      That’s just an absurdly large sample size. 1000 is adequate if a little on the low side.

    • Jesus@lemmy.world
      link
      fedilink
      arrow-up
      3
      ·
      5 months ago

      A 100,000 person response rate for a snap poll is basically impossible. Not enough people say yes to taking political polls. Most polling is done with 1000-1500, and if you want higher response rates, you have to look at aggregators like 538.