Thursday, December 21, 2006

The problem with the Cook PVI

This is for the map and election geeks out there. Many times an organization will latch on to the Cook Partisan Voting Index or use another that uses the presidential vote to measure partisan leanings. This overstates the Democratic disadvantage in rural areas. What is needed, and what I do, is take minor races that people are not getting much information on to determine the partisan leanings of precincts or political districts. As shown in this look at Ohio and Indiana, this paints a very different picture. You can then compare actual performance of a local candidate with base performance. A good candidate with strong organization can beat a correctly computed PVI by 10 points or more.

No comments: