Mapping election results
Published Friday, November 11, 2005 by CCAer | E-mail this post
Mapping election results can be more of a challenge than one might initially think. Electoral districts generally are generally the same size in terms of population but not necessarily by area. Michael Gastner, Cosma Shalizi, and Mark Newman of the University of Michigan highlight some of the challenges faced
with mapping the United States presidential election results of 2004 and come up with varying successful solutions.
The cartogram seems to be a favourite solution to showing a more accurate weighting of population and this is one that Strategic Telemetry
has taken when mapping the results from the recent Virginia state election (pdf)
(which they describe as a “cartographic map”; or click here for a jpg version
). The problem with the cartogram is that is can obliterate anythng recognizable. In this map, for instance, it’s only because we know that it is Virginia that we recognize it as such.
has provided a better sol
ution that may or may not work with other jurisdictions. They have taken the example of the results of the Iowa caucus of 2004
. Instead of stretching or squishing the state out of physical shape, they have maintained its form and added scaled symbols to represent the size and proportion of the vote. The result is much cleaner and more understandable than a cartogram.