genocide preconditions: U.Va. researcher creates racial data map online

TimesDispatch

A demographic map showing the racial distribution of the population of the United States has garnered national attention for its creator, a University of Virginia researcher.

Dustin Cable, a statistician in the Demographics and Workforce Group at UVa's Weldon Cooper Center for Public Service, used Census data and a computer template to create an interactive map of the United States with 308 million data points.

Each point represents a person counted by the Census Bureau, and the dots are color-coded by race. This allows users to look at the population density, distribution and racial segregation in one easy-to-use map, which allows users to scroll and zoom in and out.

"I wanted to create a visual of that 2010 Census data in a level of clarity not typically seen in Census maps," Cable said.

The map gained instant acclaim on the Internet. Writers at popular websites such as Gawker, Slate and Wired magazine highlighted the map, praising its ease of use and artful presentation. Writers also noticed a trend in diverse metropolitan areas - people of different races still tend to cluster in separate areas.

Qian Cai, director of the Demographics and Workforce Group, called the map "a masterpiece." It conveys a lot of information that would usually have to be plumbed from dozens of charts, tables and maps. Cable's map, she said, replaces "thousands of pages of reports."

It's also a great example of how researchers can make data accessible to the general public, Cai said. That's becoming increasingly important as the volume of data increases and it plays a bigger role in policymaking and research, she said.

"Everybody is talking about data-based decision making ... so more people are data users," Cai said. "To be able to present data in an easily understandable way is crucial."

Cable developed the map using a template created by Brandon Martin-Anderson, a graduate student at the Massachusetts Institute of Technology. Martin-Anderson's original map, developed with the Python programming language, was black and white and did not include the racial data Cable added. He made the coding available so people could build off the map, he said.

"The Census just gathers so much information ... you could ask any demographic question at all," Martin-Anderson said. "You could do a dot map coded by income, coded by household size, coded by anything you could think of."

Martin-Anderson said he liked Cable's adaptation, in part because smaller-scale racial distribution maps, showing racial distributions in Chicago and other metropolitan areas, inspired him to create the original national dot map. But those other racial distribution maps lacked the scale, the simple interface and beauty of Cable's map.