She followed that up with another post focusing on one particular factor: the size of the geographic unit. Choropleth maps that shows data by municipality, county, region, state or country will look quite different, even if they show the same data. Averages tend to cancel out extremes. She gives the following examples:
Most of these more common map types focus on a particular variable that is displayed. But what if you have multiple variables that you would like to present on a map at the same time?
Here is my attempt to collect examples of multivariate maps I’ve found and organize them into a loose categorization. Follow along, or dive into the references, to spur on your own investigations and inspirations!
Jim’s examples of maps that display more than one variable include 3D maps, multicolour choropleth maps, multiple small maps, and embedded charts and symbols. Useful and enlightening.
France held the first round of its presidential election this past Sunday. Unlike U.S. presidential elections, it’s by popular vote, with the top two vote-getters moving on to a second round in two weeks’ time.
The major candidates’ support was distributed unevenly around the country. Media organizations used several different methods to show this. The New York Times used a choropleth map, showing who among five candidates (including Lassalle, excluding Hamon, who finished fifth but does not appear to have won a commune: ouch) finished first on a commune-by-commune basis. Of course, when you have four candidates finishing within a few points of one another, when you win a district, you don’t necessarily win by much. The print edition of Le Figaro included choropleth maps detailing five candidates’ regional support as well.
Both the Times and Le Figaro use geographical maps, which can be misleading because of the number of votes concentrated in large cities, as Libération’s Julien Guillot points out. (This comes up in most countries’ elections, to be honest—certainly the ones where it’s the popular vote, rather than the constituency, that’s being looked at.) Slate uses a cartogram to compensate for that. (Both of these pages are in French.)
For those seeking local results rather than analysis, several French media organizations provide them through a very similar map interface: see, for example, the online results pages for France 24, Le Figaro and Le Monde. Each begins with a map of France: clicking on a département provides results for that département that includes a map showing each commune, which can also be clicked on. For some reason neither France 24 nor Le Monde show actual vote totals at the local level, which doesn’t seem sensible in an election by popular vote.
Finally, a couple of outliers. This page looks at the results from all presidential elections under the French Fifth Republic. And this page marks the 56 communes in which Marine Le Pen received not a single vote.
Mapping U.S. election results by county and state is a bit different than mapping results by electoral or congressional district, because counties and states don’t have (roughly) equal populations. Choropleth maps are often used to show the margin of victory, but to show the raw vote total, some election cartographers are going 3D.
The map may not look advanced today, but in 1883 it broke new ground by enabling Americans to visualize the spatial dynamics of political power. Readers responded enthusiastically. One reviewer pointed to the Republican counties in Arkansas—something left invisible on a map of the Electoral College returns—and wondered what other oddities of geography and history might be uncovered when election returns were more systematically measured. In other words, the map revealed spatial patterns and relationships that might otherwise remain hidden, or only known anecdotally. Perhaps its no coincidence that at the same time the two parties began to launch more coordinated, disciplined, nationwide campaigns, creating a system of two-party rule that we have lived with ever since.
(This map also inverts the modern colours for the two main U.S. political parties: here the Democrats are red and the Republicans are blue. Those colours were standardized only fairly recently.) [Geolounge]
For other ways of mapping election results, see this gallery of thematic maps, which includes things like 3D choropleth maps, dot density maps, and all kinds of variations on cartograms and choropleth maps. There’s more than one way to map an election. [Andy Woodruff]
The BBC’s election night map is bare-bones, showing which side won which local authority, but not by how much. Appropriate for the moment, and for finding your locality, but not necessarily very revealing.
But the EU referendum isn’t like a general election, where each electoral district has roughly the same population, and counts the the same in parliament. In this case it’s the raw vote numbers that count, and local districts can vary in size by as much as a couple of orders of magnitude. So the Guardian’s approach (at right), a hexagon grid that combines a choropleth map with a cartogram to show both the margin of victory and the size of the electorate, is probably most fit for purpose in this case.
I’m actively looking for other maps of the EU referendum results. Send me links, and I’ll update this post below.
Mapchart is a quick and dirty way to make choropleth and other coloured outline maps for online use: choose a map (world, continents, some countries), assign a colour to the state, province or country, build a legend, export to image. [Boing Boing]
The Washington Post’s Christopher Ingraham compares two choropleth maps of U.S. population growth: while they look rather different, they use the same data. “The difference between my map and Pew’s—again, they both use the exact same data set—underscores a bit of a dirty little secret in data journalism: Visualizing data is as much an art as a science. And seemingly tiny design decisions—where to set a color threshold, how many thresholds to set, etc.—can radically alter how numbers are displayed and perceived by readers.” [Andy Woodruff]
(Worth mentioning that this is exactly the sort of thing dealt with in Mark Monmonier’s How to Lie with Maps.)