Skip to main content

View Diary: A Surprisingly Predictable and Nationalized House Election? A preliminary look at 2012 candidates. (21 comments)

Comment Preferences

  •  incumbency (0+ / 0-)

    Have you tried running the regression using either:

    1. one dummy for a freshman incumbent, and a separate dummy for a longer-term incumbent, or

    2. scaling the incumbency variable by the number of terms, say by using the square root so that a 1-term incumbent would be coded 1, a 2-term incumbent 1.414, etc.

    I would guess that longer-term incumbency would count for more than being a freshman incumbent.

    SSP poster. 43, new CA-6, -0.25/-3.90

    by sacman701 on Wed Nov 14, 2012 at 09:30:21 AM PST

    •  The r-squared is already 0.93 (0+ / 0-)

      so there's not really all that much better it might do--but I might try the freshman idea later.

      27, Dem, Dude seeing a dude, CT-04(originally), PA-02/NY-14 (formerly PA-02/NY-12).

      by Xenocrypt on Wed Nov 14, 2012 at 10:15:02 AM PST

      [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site