After the presidential election in 2020, when Georgia launched runoff elections for the state’s two Senate seats, officials in Cobb County on the outskirts of Atlanta announced that they were going to shut down several early voting sites in diverse neighborhoods. But a recently launched nonprofit stepped in, using anonymized cell phone data to show the impact that the change would have on voters.
“We were able to use the data we had already collected to show how shutting down those specific polling locations would disproportionately impact voters of color,” says Daniel Wein, cofounder and head of partnerships at the nonprofit, called the Center for New Data. “And in two days, that information was published and was circulated in the community. Stacey Abrams tweeted about it. And Cobb County reversed several of the poll closures.”
The nonprofit, part of the current cohort at the impact tech accelerator Fast Forwardmakes use of the type of data typically used by advertisers—location data from smartphones—to understand how long voters have to wait in line; in 2020the nonprofit said it would receive voter wait time data from the November 3 election, collected by location-data mining companies X-Mode Social and Veraset, as soon as November 4. The methodology for using this data to analyze disparities at polling places came from a study, published in 2019, that analyzed anonymous location data from 10 million smartphones at 93,000 polling places to create a detailed picture of wait times. In Black neighborhoods voters waited 29% longer to vote than those in white neighborhoods, per that study. They were also 74% more likely to end up waiting more than half an hour.
The Center for New Data plans to share information with multiple nonpartisan voting rights organizations in the 2022 midterm elections. In many cases, the data can be used while the election is still happening. If the early vote period lasts for three weeks, for example, the team can identify which polling locations are underserved in time for potential changes to be made. “Our data produces the evidence that can be taken to a county or an other election administration body,” says Steven Davenport, cofounder and executive director of the Center for New Data. And we’ve seen that they do actually have time to open locations or change the way they’re administering the election in time for it to make a difference and actually expand voting opportunities.”
That information is critical, says Valencia Richardson, who focuses on local election compliance issues in the South at the nonprofit Campaign Legal Center and will use the data in the midterm elections. “Their data will serve to help us identify problem areas, so that we can advocate for better resource allocation,” she says.
In states where voting is already difficult—like Alabama, where there is no early voting and few people are allowed to vote by mail—the length of lines at polls can mean the difference between someone votes or not. That is particularly true for marginalized voters, many of whom find it difficult to take time off of work to vote when the wait time is two hours.
“This is a methodology for using mobility data to monitor and measure voting access that had been sitting out in the policy ecosystem for a couple of years,” says Wein. “But what we wanted to do was take this data and actually make it accessible to the voting rights organizations that are doing work on the ground, so that they can learn from and respond to voting access issues in their local communities.”
Voting rights organizations already gathered anecdotal data about long lines, but the Center for New Data offers something much more comprehensive. “There are many places in Georgia where we simply don’t have eyes, and the CND’s data can give us a bird-eye view of Georgia that can help us better identify counties and precincts that have historical problems,” says Juan Manuel Balcazar, campaigns and data director at Fair Fight, the Georgia-based nonprofit founded by Stacey Abrams to fight voter suppression.
In Iowa, stats from the Center for New Data about the impact of reduced voting hours on the Latino vote are being used as evidence in a current lawsuit, LULAC vs. the Iowa Secretary of State. The data can also be used to advocate for longer-term changes. “We are empowering on the ground voices,” says Davenport. “There’s people on the ground who are stuck with anecdotes, and we’re giving that contextual data so that they can be trusted and that governments can be accurately informed to make the right decisions to broaden access. And we’re also using big data to paint a big picture that can show systematic disparities.”
That’s not to say, of course, that advocates will have an easy job. In 2021, 19 states passed 34 laws that restricted access to voting in various ways, from new requirements for mail-in voting in Texas that led to thousands of ballots being rejected in primary elections this year—including those of many seniors and people with disabilities—to a ban in Georgia on bringing water or snacks to people waiting in line to vote as part of a bill known as SB 202, and a new law in Iowa that allows criminal prosecution for election officials who don’t purge voter rolls. The John Lewis Voting Rights Advancement Act, which would restore and strengthen voting protections, failed to pass the Senate last November and failed again in January. After Trump lost the last election, his allies have worked to replace election officials across the country. Regardless of what the data proves, some election officials may be unsympathetic to any attempts to make voting more accessible.
Still, the data can help. Voting rights advocacy “has always been challenging,” says Fair Fight’s Balcazar. “Unfortunately, now we are seeing bold and egregious attacks on democracy and the freedom to vote. This wave of anti-voter laws have created disproportionate burdens on voters of color, enabled voter and election worker intimidation, and facilitated election administration power grabs. However, we are fighting back, and our advocacy and activism has made [Georgia bill] SB 202 is less harmful than it was initially meant to be and [helped] mitigate some of its impact. As a data-driven organization, Fair Fight has always fought voter suppression with qualitative and quantitative proof. Our data collected, including that of CND, have been an important asset in our fight.”