The Huffington Post has compiled and mapped news reports of gun-related homicides and accidental deaths in the U.S. since the school shooting in Newtown, Conn.
For 98 days, a team of Huffington Post researchers scanned news reports from around the country and added them to a spreadsheet to be sorted by date, city and state. The maps were part of Jason Cherkis’ series about gun violence in America.
Cherkis writes, “In the first week after the Newtown, Conn., massacre on Dec. 14, more than 100 people in the U.S. were killed by guns. In the first seven weeks, that number had risen to at least 1,285 gunshot killings and accidental deaths. A little more than three months after Newtown, there have been 2,244.”
The resulting interactive map, “Mapping the Dead, Gun Deaths Since Sandy Hook,” is a breathtaking display of where 2,244 people died in America. You can see the names of each one of the victims, and the map links every data point to a news story about the incident.
“To us, the most shocking thing is the magnitude of it,” Huffington Post Data Editor Andrei Scheinkman said by phone. “You can see how many reported deaths there were and how it affects every corner of the country.”
How they created the map
Scheinkman said his team considered many colors for the display but settled on red. The team did not add any sounds to the map, but did add animation “to give a sense that as time passes, these incidences keep happening.”
D3 allows users to move in on a section of a map just by clicking on it.
“A lot of news organizations are using this,” Scheinkman said. “It is a designed for exactly this kind of project, displaying data and letting users interact with the data.”
A look at the data
The Huffington team did not try to analyze the data by comparing it to previous years. The data is not weighed by population density or by weapon or ammo used, and it does not specify which shootings were out of self defense and which were not.
“Our focus was on trying to just convey the sheer number of incidents and emphasis the tragedy of each one,” Scheinkman said.
Given that the map only features deaths that the news media have reported on, there’s no question that the map under-reports the number of people who died in the 98 days the team collected the data.
Slate and the Twitter feed @GunDeaths is trying a different method. They are crowdsourcing gun deaths and have collected details of more than 3,000 gun deaths since the Newtown shooting.
That display is sortable by the victim’s age, gender, location and date. When the user clicks on one of the icons representing a death, a victim’s name, the city and date of the shooting appears. The site also asks the readers to report any inaccuracies and links to the source of the information.
What makes these types of projects work
The Huffington Post has found that while these interactive projects require significant time investments, they pay off.
Scheinkman says previous interactive data projects his team built increased “time spent on page” to more than five minutes, which is significantly more than any other feature on the site.
Accompanying stories help add context to the data projects. One of the hallmarks of the map is how Cherkis puts a face on the story of some of the individuals who make up the 2,244 victims.
Stories about a random shooting that killed a 9-year-old in a passing car, an innocent bystander who was shot, and a 21-year-old woman who was an unintended shooting victim reflect the heartache behind the statistics.
Related training: The Poynter Institute is helping journalists cover the debate over guns, gun control and gun violence by holding a McCormick Foundation-sponsored seminar in Chicago April 1-3. We will live stream coverage of our “Covering Guns” conference from Loyola University on Poynter.org.
Because of the overwhelming demand from applicants for our Chicago event, Poynter and the Scripps Howard Foundation are going to hold a second “Covering Guns” conference on the campus of the University of Maryland, July 10-12. You can choose how much you pay for the training.
Correction: This article originally stated that the map increased time spent on site to more than five minutes; it has increased time spent on the page to more than five minutes.