Skip to main content

View Diary: Ohio math: Willard must win non early votes by ???? (71 comments)

Comment Preferences

  •  including an internal poll ? what house effect? (0+ / 0-)

    no track record. giving five bars to Gravis. no.

    •  I have to agree (2+ / 0-)
      Recommended by:
      Vote4Obamain2012, FishOutofWater

      Nate has always been open with his egalitarian inclusion of polls, the argument being that a right wing house effect in some polls cancels out a left wing house effect in others.

      But what if a whole bunch of polls with no history and a right wing house effect hit at once? How do you deal with that.

      When we were ahead, people at DailyKos said to fight like we were behind. When we were behind, people at DailyKos said to give up.

      by NoFortunateSon on Sat Oct 13, 2012 at 07:47:56 PM PDT

      [ Parent ]

      •  I've never read Nate mention preparing his model (1+ / 0-)
        Recommended by:
        Vote4Obamain2012

        for that, so I don't know if it can deal with it.

        It would be doubly troubling, because that flood would move the mean to the right, and left-leaning and centrist polls would be weighted less as a result...

        But maybe there's a defense against built in against that.

        The dogmas of the quiet past, are inadequate to the stormy present

        by Inkin on Sat Oct 13, 2012 at 08:01:57 PM PDT

        [ Parent ]

        •  it doesn't simply depend on the mean (1+ / 0-)
          Recommended by:
          Vote4Obamain2012
          The basis for the comparison is a weighted average of the polls, with heavy emphasis given to a firm’s pollster rating in calculating its say in the consensus. Polling firms that we believe to produce more reliable results – based on their past performance and their methodological standards – have a lot more say in the calculation.

          In addition, in a new wrinkle in the model this year, the consensus is estimated solely from polling firms that include cellphones in their sample....

          link

          So, that's some protection. Hard to tell how much.

          Election protection: there's an app for that!
          Better Know Your Voting System with the Verifier!

          by HudsonValleyMark on Sat Oct 13, 2012 at 08:18:34 PM PDT

          [ Parent ]

      •  help me out on one thing here (2+ / 0-)
        Recommended by:
        Vote4Obamain2012, FishOutofWater
        Nate has always been open with his egalitarian inclusion of polls, the argument being that a right wing house effect in some polls cancels out a left wing house effect in others.
        I thought the argument was more that as long as his model can estimate the house effects, the polls don't need to be balanced by others with equal-and-opposite house effects.

        But as you point out, that may not work when the pollster doesn't have much history.

        Election protection: there's an app for that!
        Better Know Your Voting System with the Verifier!

        by HudsonValleyMark on Sat Oct 13, 2012 at 08:01:58 PM PDT

        [ Parent ]

Subscribe or Donate to support Daily Kos.

  • Recommended (129)
  • Community (61)
  • Memorial Day (31)
  • Culture (24)
  • Environment (22)
  • Law (20)
  • Science (20)
  • Rescued (20)
  • Civil Rights (20)
  • Labor (18)
  • Education (17)
  • Media (17)
  • Elections (17)
  • Trans-Pacific Partnership (16)
  • Marriage Equality (16)
  • Economy (15)
  • Republicans (14)
  • Racism (14)
  • Ireland (14)
  • Josh Duggar (13)
  • Click here for the mobile view of the site