Distorting the US Election Results via "Cartograms"
Hilary Spencer
Thursday, 20 November 2008 23:51 UTC
Mark Newman, a physics professor at the University of Michigan, released several maps of election results in the United States. The first is one that we’ve all seen:
However, as Newman points out: “Looking at this map it gives the impression that the Republicans won the election handily, since there is rather more red on the map than there is blue. In fact, however, the reverse is true – the Democrats won by a substantial margin”. (This seems to be because the map violates the rather intuitive mapping between spatial attributes and numerical values — we assume that if something is bigger, then it has a larger value). Newman suggests correcting for this by scaling the states according to their population, or more accurately their electoral college votes:
Looking at the votes county by county, scaling by population, and using a red-blue color scale to represent the percentage of voters voting for a particular party (e.g. bright red = most votes were for John McCain; bright blue = most votes were for Barack Obama), we arrive at:
Newman has extended this idea to many other datasets via his Worldmapper project. There you can seem maps of the world scaled by population size, or carbon emissions in 2000.
If you’re curious about the methodology used to draw the maps, you can check out their PNAS paper: Michael T. Gastner and M. E. J. Newman (2004) Diffusion-based method for producing density equalizing maps Proc. Natl. Acad. Sci. 101, 7499-7504. (available via ArXiv) or M. T. Gastner, C. R. Shalizi, and M. E. J. Newman, Maps and cartograms of the 2004 US presidential election results, Advances in Complex Systems 8, 117–123 (2005).
Since the maps already involve some distortion in their projection into 2D, I wonder what this idea would look like on a globe.
Updated 21 November 2008 00:13 UTC
-
Replies
-
I just noticed that Mark Newman is also the author of “The first mover advantage in scientific publication”, which uses network theory to show that citations are preferentially made to earlier documents, even after advantages from the longer time of availability for citation are factored out.
-
We had some discussion at Friend Feed about this – and so did Noah on his Nature Network “Nothing’s Shocking” blog. Links to various interesting election map representations, including to these and to Noah’s post, are here. Fascinating stuff!
-
Sorry, Noah’s post is not at that link. It is here.
-
I am going blind – the link to Noah’s blog is in that Friend Feed discussion (two replies above this). There are also some other good map representations you can get to from there. Sorry for writing so many flaky comments – I have got too used to FF, where you can edit after writing!
-
According to the estimates on worldmapper, the world population in 2050 will be 9 billion. Already food is too expensive, add to that the increasing shortage in fussel fuels and water, how will this earth support such a population?
-
