Red State, Blue State

INFO VISUALIZATION: BoingBoing posts this map:

Reader Jeff Culver in Seattle says:



"I was thinking today about how the 'red v. blue' states graphic is really misleading considering the slim margins that the candidates won some of those states by, so I sat down and created the map that's attached. In the dozens of hours I've been watching the news I haven't seen one like it, but thought that you and the BoingBoing readers might find it interesting. I think it definitely portrays our fellow states far differently than the extreme way we've been seeing to date."

The winner take all electoral college has to go.

Comments

[Anonymous]

only 38 states require the electoral votes to follow the popular vote.

[Dylan]

This map is interesting because it really puts to light the fact that while Bush won, it really was a 51% to 48% win. It would be great to see these kinds of maps for past elections, then animated through time to see the shift and flow of color through key time periods.

I found http://www.blogger.com/r?http%3A%2F%2Fwww.princeton.edu%2F%7Ervdb%2FJAVA%2Felection2004%2FElection2004textured.gif">this red/purple/blue results map by county to be interesting too, as you can see population densities overlayed onto the map along with the votes by county.

The third map I found interesting was http://www.blogger.com/r?http%3A%2F%2Fwww.esri.com%2Findustries%2Felections%2Fgraphics%2Fresults2004_lg.jpg">this 3D map of the US with election results by county. It gives you a much better idea of the population centers of the US, as well as the impact those population centers have on the overall vote.

[Anonymous]

If you're interested in visualizing election data - I've recently provided a small toolset based on GeoVISTA Studio, an open-source geovisualization kit that I work on at Penn State.

http://www.personal.psu.edu/acr181/election.html

I'm also sharing the data - since I think that's really important. One thing that struck me upon checking out the numerous cartogram pages and other 'purple' maps floating around now was how none of their owners were sharing the data they created - making it impossible to check their numbers or re-use that work in some other way.

The toolkit I've put up is designed for epidemiology, so it excels at exploring multivariate data - hence I've included a bunch of socioeconomic and other covariates in with the election data I created.