Last week we looked at an analysis done in R by the good folks at Offensive Politics, looking at the political climate surrounding the recent Senate election in Massachusetts. There were some very insightful comments (thanks, Revolutions readers!) about the design of the charts, especially in the choice of color schemes used (the originals didn't use a neutral white midpoint to represent no change). Well, Offensive Politics have taken another look at the charts, and produced some new ones which are a great improvement. With the revisions, the chart comparing the increase in Republican vote percentage from 2008 to 2010 looks like this:
Offensive Politics notes:
From this map we can see that from 2008 to 2010 the Republican vote percentage actually increased in every single township in MA. This is kind of shocking, especially when you consider that Democrats enjoy a 15 point registration advantage statewide.
Check the link below for all the updated charts, and of course the R code to create them.
Offensive Politics: Re-mapping Massachusetts Special election results
I left comments on the previous blog posting about these map projections. As the Revolution Blog is intended to help spread the value of R, it is unfortunate that the map that is being shown is flawed, and may convince some viewers that R isn't up to the task of mapping.
The map data needs to have a projection applied. I have indicated the command for this in my previous posting for one commonly used projection. With this, it will have an appearance that doesn't look distorted. In addition, the shapefile used includes waterways in the state/town boundaries, which creates a problematic appearance to the area at the Southwest of Cape Cod. This is easily fixed by using the US Census shapefile, available on the US Census website.
I have posted an example of the appearance of a map of MA, with these issues corrected at: http://pbleic.posterous.com/10573703
Other projections would give different appearances to the map.
Posted by: Paul Bleicher | February 04, 2010 at 10:22
Offensive Politics notes:
From this map we can see that from 2008 to 2010 the Republican vote percentage actually increased in every single township in MA. This is kind of shocking, especially when you consider that Democrats enjoy a 15 point registration advantage statewide.
How does a (presumably preexisting) registration disadvantage translate into suprise in an increase in the vote? I don't see how they are related. The registration disadvantage would translate into an expected *level* of the vote (and one might be surprised at that), but how does that relate to surprise at the *change* in vote unless the regristration difference got *worse*?
Posted by: GB | February 04, 2010 at 15:36