Skip to main content

View Diary: Excluding GOP polls, still looking good for Dems (84 comments)

Comment Preferences

  •  That is not true... (0+ / 0-)

    And suggests that you haven't really paid attention to Nate's analysis.  Nate does not take wAA, Ras and other GOP polls "at face value" as you say.  He notes how heavily the partisan polls tend to lean one way or another, and compensates according to that bias.  Partisan polls are still useful, if for no other reason than they provide supporting evidence on which direction the numbers are headed and by how much.  But otherwise taking into account their house effect and including them in the model with modifications to account for that seems to be a reasonable system.

    •  But how? (3+ / 0-)

      He doesn't specify. Basically what he should be doing is ignoring the polls that don't have a robust methodology. If a poll includes land lines and cells, then include it. So, include Mason Dixon and Pew (even though they are skewed, etc.)

      But, why include Gravis Marketing or Purple strategies or ARG?

      He's being punked and so is TPM with their stupid banner showing a state as up or down with whatever the latest poll is...

      It probably won't impact his model in the end but it might.

      I like Nate but he's not infalliable.

      •  The RW pollsters are trying to create reality (0+ / 0-)

        Not measure public opinion. In the end, due to all the bedwetting Dems who need a lullaby every time a new right-wing poll comes out, they may very well succeed.

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site