Why we should be talking about the Police Department’s stop-and-frisk data

On April 26, three days before protestors accused Mayor Jim Kenney of backtracking on his campaign promise to end stop-and-frisk, the Philadelphia Police Department released more than two years of stop-and-frisk data.
Through maps and charts, the data shows who’s getting stopped — overwhelmingly young black men, plus where and how often. (Our sister site Generocity has a breakdown of the data here.) The PPD also released data around shooting victims, including officer-involved shootings. It’s part of the White House’s Police Data Initiative, which encouraged police departments to release non-crime data that could increase accountability and transparency. It will publish new data every day with a two-week lag time.
See the data

An attempt to change the narrative, with data.

The data release, which PPD research director Kevin Thomas called “courageous,” feels like a big deal for two reasons. It’s an example of the city releasing data around a controversial policy, one that caused civil rights attorneys to sue the city in 2010 and has continued to plague the city, as shown by the recent stop-and-frisk town hall with Mayor Kenney. (This dataset was in fact born of the 2010 lawsuit. The city agreed to collect the data as part of a settlement.)
Thomas said the PPD has done a lot of work to understand the data better, since it’s been auditing it since 2011. “We’ve gotten a lot more comfortable with the data,” he said.
But it’s also the first example we’ve seen of the city attempting to use data to tell a story, to reclaim a narrative. Instead of just publishing the raw data, the PPD used Esri’s Map Journal tool to provide more context around the data.
“It’s very easy to take law enforcement data out of context and tell a story that fits your own narrative,” said Thomas, who oversaw the release. “The challenge is how to release data like this and bring context to the conversation.”

It's very easy to take law enforcement data out of context.

The PPD’s solution was publishing stop-and-frisk data visualizations alongside maps of the city that show poverty and crime rates, plus a demographic breakdown. It also featured pointers on how to interpret the data, encouraging viewers to take into account socioeconomic and crime data.
“The question is often raised about whether a specific racial or ethnic group is over represented in pedestrian and vehicle stops,” the Map Journal presentation reads. “On the surface, a cursory examination of the racial distribution of the pedestrian and vehicle stops, may suggest this is the case. However, as noted previously, it is important to consider all of the circumstances when interpreting raw statistical data.”
But that’s a misleading way to characterize the data, said civil rights attorney David Rudovsky in an interview with Technical.ly Philly.
Rudovsky’s law firm Kairys, Rudovsky, Messing & Feinberg was one of the parties who sued the city in November 2010 over stop-and-frisk. Along with the American Civil Liberties Union of Pennsylvania, his firm continues to analyze the legality of the PPD’s stops.
In a recent analysis, they controlled for demographic, economic and crime factors — data that the PPD’s map shows — and found that a racial disparity still existed.
“We agree with [the city] that disparity in numbers alone doesn’t prove anything,” Rudovsky said. “But we controlled for those factors in the regression analysis and it doesn’t explain the disparity.”
There’s also one crucial missing data field, which would help tell a more complete story of this data and how race affects it: the “why” behind each of the stops.
Thomas said the data doesn’t include reasons for each of the stops because of a technical issue.

An incomplete picture.

That data is entered in a free text format, and it would have been too difficult to rid it of identifying details like names — data that the PPD is bound by federal law to keep confidential, he said.
(When it comes to releasing city data, confidentiality is a major concern. Think of how the city took pains to make sure it wasn’t revealing the identities of undercover police officers with its recent release of city employee salary data and how it’s working to scrub city expenditure data of details like credit card numbers and witness information.)
It’s a shame that’s missing from the release because it’s an important part of the stop-and-frisk story.
In the most recent audit of the city’s stop-and-frisk data, Rudovsky’s firm and the ACLU found that police officers were still stopping people without “reasonable suspicion.”
“On reasonable suspicion issues, the data continues to show very high numbers of illegal stops and frisks,” the March 2016 report reads. “For the First and Second Quarters, 2015, plaintiffs found that 33% of all stops and 42% of all frisks were without reasonable suspicion.”
What’s more, Black people were stopped without reasonable suspicion more often than White people, the report said, “demonstrating that police were using a higher threshold of ‘reasonable suspicion’ for stops of White suspects.”
A racial breakdown of stop-and-frisk data is incomplete if it doesn’t address this point, Rudovsky said.
Still, it’s impressive that the PPD’s Map Journal presentation doesn’t dance around the issue of race. Two of its pages specifically call out the issue, one entitled “Racial Breakdown of Car and Ped Stops” and a map that plots “Car and Ped Stops by Race.”
The PPD’s new effort is not perfect, but it’s a start.
We can get behind more context around open data. City GIS chief Mark Wheeler tweeted an important reminder about it last week.
https://twitter.com/Wheelmrk/status/728318826249650176
It’s also another way to make open data accessible to the public, which Chief Data Officer Tim Wisniewski has aspired to do. Wisniewski said he was thrilled that the PPD had taken this approach to publishing stop-and-frisk data.
“It’s a dream come true,” he said.
But like with any storyteller, there’s power in creating a narrative. Power that can be used in many different ways. How will the city use it?