Borrowing a technique from Hollywood, historians at the University of Richmond have created animated maps that chart voting patterns in U.S. presidential elections since 1840.
The maps show county-by-county data for every major election year in which data are available, and that information shifts over time. One map, for example, highlights counties where the victor won by only a small margin. It reveals how "battleground states" have changed over the years. The maps are displayed as video montages, with each election year shown sequentially. A slow-fade effect—that's the Hollywood-inspired part—is used between maps, which helps highlight the changes.
Another series of maps plots the numbers of votes cast for third-party candidates in each county. It's more than you might think, given the reputation of the United States as a system dominated by two parties.
Leaders of the project, called Voting America, have coined a term for their images: "cinematic maps." They are examples of an emerging trend in social-science research in which scholars turn complex data sets into pictures to help reveal patterns across time.
The drive to create the Voting America project came from the University of Richmond's new president, Edward L. Ayers, a pioneer of digital technology in history research. His Valley of the Shadow project, an online archive of documents from two counties during the American Civil War, one in the North and one in the South, is one of the best-known digital history projects. Millions of visitors have explored the Web archive since it went online, in 1994.
When Mr. Ayers took the helm at the university last year, he led the development of a new center, called the Digital Scholarship Lab, to help humanities scholars use technology in their research. Voting America is the lab's first project, done by its staff to inspire other researchers at the institution to come up with ideas for projects of their own.
Pundits, Take Note
Mr. Ayers said he hopes that political pundits and others watching this year's election will explore the maps to get a better sense of its context. The main take-away point, he said, is that American elections have rarely been clear cut or predictable. "Pundits did a pretty bad job predicting that Barack Obama would emerge in the recent primary," said Mr. Ayers. "But this kind of ferment and change and surprise has been a hallmark of American politics" throughout the country's history.
Some of the maps in the University of Richmond project include a small video frame in the top right corner that features a scholar delivering commentary. That idea came from Mr. Ayers's son, Nathaniel, who is a programmer for the Digital Scholarship Lab.
"It immerses the viewer a little bit more, we're hoping," he says. "Whenever I see a presentation and am listening, I like to have something else to look at to kind of help bring across the point."
The scholars did not have to hunt down the historical data on voting and convert it to digital form themselves—they were able to find electronic copies of the data in existing sources. That let them focus on formatting it in new visual forms. "The reality of living in the digital age is all this information is already available," says Andrew J. Torget, director of the Digital Scholarship Lab.
In addition to the cinematic maps, the scholars created an interactive map of the United States that lets users zoom in on any county and see any of the selected data. (Think Google Earth with historical information.)
Several historians and other humanities scholars are beginning similar projects that turn data into pictures, says Annelie Rugg, director of the Center for Digital Humanities at the University of California at Los Angeles.
"Formerly this was data that was looked at as dates and maps in a two-dimensional, fixed representation," she says. "Now you can look at data as they evolve so you can see the dynamism and the process of change."
But doing such a project requires a culture shift for historians, who need to either learn the technology or learn to collaborate with others who have.
