Around 11 am Pacific on January 20th, while newly-inaugurated President Trump finished a celebratory lunch in the Capitol Rotunda, Magdalena Cerdá noticed something different about the White House's website: All of its references to climate change had disappeared. Cerdá is an epidemiologist at UC Davis' Violence Prevention Research Program, which focuses on another politicized region of science---gun violence. So she knew what that meant.
"It was a real call to action," Cerdá says. With links to climate data vanishing, she worried the same thing could happen to gun violence data on websites belonging to the Centers for Disease Control and Prevention, the Bureau of Justice Statistics, or the Bureau of Alcohol, Tobacco, Firearms and Explosives. "I was on Amtrak between Berkeley and Sacramento," she says. "So I sent an email to Garen Wintemute saying we needed to start downloading our data immediately."
Wintemute, epidemiologist and director of the Violence Prevention Research Program, was prepared. After seeing that climate scientists were systematically downloading crucial information from federal databases, he had drawn up a spreadsheet of the gun-related datasets he uses every day: lists of gun licensees, retailers, and manufacturers; gun tracing data; firearm-related death and injury numbers sorted by categories like race, location, or age. "I basically walked around the building saying, 'Get it done now,'" Wintemute says. So on inauguration day, as Cerdá says, the Violence Prevention Research Program was less of a lab and more of a "little downloading bootcamp."
This wasn't just alarmism. "I’ve been through it before," says Wintemute. During the Clinton years and early in George W. Bush's presidency, he worked with a group of academics who partnered with the Bureau of Alcohol, Tobacco, Firearms and Explosives to study the inner working of criminal gun markets. "We had the reports ready for 2001 and 2002, but their publication was suppressed," says Wintemute. "We were ordered to destroy our copies of the documents."
Wintemute wasn't hankering to repeat that experience, so he instructed his team---barely 10 people, which isn't surprising in this notoriously under-funded field---to create a compendium of gun-related data as a failsafe. "Everyone stopped what they were doing and gathered around a big table in the center of the office," says Aaron Shev, the team's senior statistician. "We had Dr. Wintemute's list, and wrote things up on a whiteboard, assigning jobs. Then we sat there for a day, through lunch, in this frenzy of downloading."
It's not as if they were hacking, or even using some secret scientists-only login. "Everything we downloaded by definition is public information," says Wintemute. "We didn’t go behind any firewalls." So the fact that scientists are worried they'll lose access should probably give you pause.
In many cases, federal information is vital to research in these fields. "I was scared," says Veronica Pear, a data analyst at the Violence Prevention Research Center. She's using federal data for a paper on firearm mortality---tracking hotspots in California between 1999 and 2015---and her work is almost complete. Since the federal query system makes it easy to search the data but cumbersome to download, Pear had never bothered to save the information to her computer. To catch up, "I had to enter around 50 different queries," she says. "I felt frantic."
After about five hours spent in a flurry of activity, the team had downloaded everything they could think of. Now, it's stored on a secure server at UC Davis, ready if gun violence researchers ever do lose access to federal data on firearm licenses, sales, use in criminal activity, and deaths. It's not as large of an effort as the climate data scraping executed in recent weeks---the datasets numbered in the tens rather than thousands---but that doesn't mean the UC Davis team's work was insignificant. Thanks to funding issues and opposition from the National Rifle Association, very little gun-related data exists. Every scrap counts. "There aren’t a lot of different kinds of data, but they are foundational," says Wintemute. "Every research study on firearm violence begins with a statement on the size of the problem. That's what these data provide."
Should the information disappear from the web, the UC Davis team will have no qualms sharing it. "I think of it is as a public service for the scientific community," says Pear. "And for me, it feels more important now than ever to be vocally on the side of truth."
But for now, they're just watching and waiting. "Every day I go onto the sites, I hold my breath," says Pear. "'Is it going to be there?'" When science becomes politicized, the only thing for researchers to do is let the data speak for itself. The UC Davis team wants to make sure the data is still around to do the talking.
