The map of population size, U.S. Election

Although the popular vote was not a landslide, the map redrawn with the size of each state scaled to represent its population makes it clear that we are a majority blue nation. An explanation of the method used is here.

Via Steve Yelvington on Twitter.

This entry was posted on Wednesday, November 5, 2008, at 8:49 am and is filed under data, elections, graphics, maps, news. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

7 responses to “The map of population size, U.S. Election”

Kevin writes:

No, this is not correct. All that was done was resize the states based on electoral votes (not quite the same as population), and then paint it with a winner-take all color (either blue or red), regardless of the breakdown within the state of votes for either candidate.

A more representative chart uses county-level data, and applies a color that reflects the percent voting for one party or another, which is toward the bottom of the post. here
November 5, 2008 at 4:56 pm
Mindy McAdams writes:

The creator of the map wrote that this is “a map in which the sizes of states are rescaled according to their population.” So I think you are mistaken.

He does have another map, lower on the same page, that illustrates the electoral votes: “Wyoming, for instance, has approximately doubled in size, precisely because of the bias in favor of small states.”

And below that, he provides six other variations, with each map showing a different representation of the electorate — including county-level election results.
November 6, 2008 at 7:33 am
Kevin writes:

You’re correct that the site shows both a populattion-resized and an electoral vote-resized map.

What’s still wrong is the conclusion that the population-sized map demonstrates we’re a “majority blue nation.” That conclusion may be correct, but the map doesn’t show that. The map applies the winner-take-all outcome to the entire population of each state, regardless of the margin of victory.

If you assign all votes in a state to the winner (as the map does) then you’d think Obama got 70% of the popular vote. In fact, he got just under 53% of it.

And we need look only to 2000 to be reminded that it’s possible to win the electoral vote while not winning the popular vote. Your logic would conclude that in 2000 we were a “majority red nation,” which we were not.
November 6, 2008 at 8:56 am
Mindy McAdams writes:

@Kevin - Ah, now I understand. You’re right: When we see the county level, for example, it’s red all over. But many of those counties have very small populations.

So the map that we need would show each state, perhaps, in a shade of purple derived from the percentage of red plus the percentage of blue (which could be done in an RGB graphic). Then we would see some red, some blue — but LOTS of purple, yes? Sort of like this one –

http://www-personal.umich.edu/~mejn/election/2008/countymappurpler512.png

But maybe merged with the one I showed in my blog post, to indicate proportion of population per state.
November 6, 2008 at 9:22 am
Kevin writes:

Exactly. One could do a continuous-tone of color on a state, rather than county, level. I like that the county view better shows the cities within states, so you see contrasts between, say, Seattle and Eastern Washington state, or you see how Austin sticks out in Texas. Lots of cool opportunities here.

Incidentally, the Austin American-Statesman did a great story four years ago looking at the fraction of US population living in ‘landslide’ counties, and what it means for so many of us to live in places with little diversity of opinion.
http://www.statesman.com/specialreports/content/specialreports/greatdivide/120504divide.html
(sorry for the ugly url; I haven’t figured out how to link in comments here)

Kevin
November 6, 2008 at 11:43 am
John Conklin writes:

I would like to issue with you regarding your statement that “we are a majority blue nation”. If you mean’t it just for this presidential election, I agree but beyond that it means nothing. If things don’t go well for President Obama’s first term there’s an excellent chance there will be a Republican president four years from now. I believe the nation is still split about 50/50 in terms of left and right of center. I think the history of politics supports my position. Thanks for listening.
November 10, 2008 at 6:59 pm
Mindy McAdams writes:

@John Conklin - I agree with you, Obama could be a one-term president like Carter or Bush Sr. It may be that those who voted for him expect too much, and maybe it can’t get done in four years. If the amount of progress is too small, or he goes back on campaign promises, then the nation will flip again in four years.
November 11, 2008 at 7:32 am