Historical election results
June 16, 2011 § Leave a comment
This scatterplot from Nate Silver’s FiveThirtyEight blog at the New York Times (http://fivethirtyeight.blogs.nytimes.com/2011/06/02/on-the-maddeningly-inexact-relationship-between-unemployment-and-re-election/) suggests there is no correlation between unemployment and a president’s re-election chances. I haven’t seen many news charts that set out to show a lack of correlation, but in this case it’s compelling. I also haven’t seen many scatterplots where the horizonal axis sits in the middle of the chart rather than the bottom, but it worked here and you can follow the chart easily.
While this is a great chart, what I really want to point out is the source of the election results. (This is Seth’s Sources after all!) Rather than the Federal Election Commission or Politics in America or whatnot, they come from David Leip’s Atlas of U.S. Presidential Elections (http://uselectionatlas.org/) When I first stumbled onto this source many years ago, I thought it was too homegrown and unofficial to use as a primary source. My opinion changed quickly. That’s the crazy thing about the Internet: someone who begins a Web site as an enthusiast or hobbyist can end up building a more complete, accurate and useful set of data than anything offered by the government or major publishers. Time and again, media organizations turn to David Leip’s atlas for historical election stats.
The atlas provides stats back to 1789. In addition to standard vote counts, you can get the size of the voting age population, turnout as a percentage of voting age population or registered voters, number of invalid ballots and a number of other statistics.