Why I’m skeptical of some puzzling polls

  • jordanlund@lemmy.worldM
    link
    fedilink
    arrow-up
    4
    arrow-down
    5
    ·
    9 months ago

    There absolutely is. You follow them continually to establish a trendline.

    Since I’ve been following these key states, I’ve watched them go 50/50 to virtually all Trump, to all Trump by a wider margin.

    Following polls over time lets you see the momentum.

    By the time the election comes around, there should be no surprises.

    • protist@mander.xyz
      link
      fedilink
      English
      arrow-up
      7
      ·
      9 months ago

      If your goal here is to not be surprised, I hate to tell you polls have margins of error that mean surprises are inevitable. Also, many polls are not infrequently found to be outside their SE when the actual results come in.

      • jordanlund@lemmy.worldM
        link
        fedilink
        arrow-up
        5
        arrow-down
        2
        ·
        9 months ago

        Again, you can reduce the margin of error by plotting the trend line.

        It’s the same science for watching climate change:

        If one candidate is trending up towards election day and another is trending down you can tell which way it’s going to swing.