Impact Brief: A Roadmap Breathes New Life into News That’s Local and Impactful

When Phoebe Barghouty set out to earn a master’s degree in journalism from Stanford, she had two key goals in mind: to cover law enforcement and the military and to do it using tactics that go far beyond classic gumshoe reporting.

She got what she wanted—and then some. Five years later, Barghouty is at the center of a groundbreaking Stanford initiative to support investigative journalism using advanced computational methods. Among other things, the Stanford Computational Journalism Lab collects public data often hidden from view, turning the disparate records into massive databases of government records, software for uncovering meaningful patterns in the information, and online tutorials showing anyone—including journalists and policymakers—how to use it.

The first of these online repositories, called the Stanford Open Policing Project, contains some 100 million records of traffic stops in states, counties, and cities around the country. Using a novel test for discrimination developed by one of the project’s leaders at Stanford, the data show that police have a lower threshold for searching Black and Hispanic drivers than white drivers.

“In a world of shrinking resources for reporters, particularly those in local newsrooms, important stories that have major social benefits too often go untold,” said James Hamilton, a professor of communication and cofounder of the Stanford Computational Journalism Lab.

To Hamilton, the challenges journalists now face constitute a market failure that academic institutions like Stanford are uniquely positioned to fix. “If we can lower the cost of finding those stories through better use of data and algorithms, stories can be told that hold institutions accountable,” said Hamilton, an economist by training whose new book, Democracy’s Detectives: The Economics of Investigative Journalism, makes the case that $1 invested in in-depth reporting can yield more than $100 in benefits to society.

LESSONS IN PERSEVERANCE

The policing project is the first in a series of steps Hamilton and his collaborators at the Stanford Computational Journalism Lab are taking. The effort is part of an interdisciplinary program that includes courses, academic research, and partnerships with media organizations on data-driven storytelling. Work is also underway to use local public records for building national databases that can yield academic research, journalistic stories, and policy change.

Other faculty leading the program, part of the Stanford Journalism and Democracy Initiative, are Cheryl Phillips, a visiting professor in the Department of Communication who was previously the data innovation editor at The Seattle Times; and Sharad Goel, an assistant professor in the Department of Management Science & Engineering, who designs computational tools for studying criminal justice.

100 million: The number of records of traffic stops in states, counties, and cities around the country held in the online repository, the Stanford Open Policing Project.

From the start, student journalists have been on the front lines of the policing project, as well as similar effort underway to gather data on people imprisoned for not paying their debts. In early 2015, Barghouty was one of Phillips’ students assigned to file public records requests for traffic stop data for all 50 states.

The process took two years. And it was eye-popping. Some states denied the requests outright; Louisiana explained that it has a formal policy against racial profiling and doesn’t see the need to collect traffic stop data. Others delivered, but in digital formats that were unreadable or lacked key information. CDs without labels would arrive in unmarked envelopes.

“I’ve had people get angry, hang up, or try to intimidate me,” said Barghouty, who now serves as a liaison between government agencies and the Stanford data scientists who clean up and analyze the records for the Computational Policy Lab, an interdisciplinary team of researchers, engineers, and journalists that Goel established. “Whenever we would reach roadblocks, Cheryl would look at us and say, ‘Now what are you going to do?’ She taught us that the answer can never be to just give up.”

TOOLS + TRAINING = IMPACT

Just as critical to the work is Goel’s team of engineers. Before joining the Stanford faculty in 2014, Goel had published seminal research on discrimination in stop-and-frisk practices in law enforcement in New Jersey and New York. Hamilton had read his research and suggested that Phillips, with whom he had just founded the Computational Journalism Lab, reach out to Goel. The two had lunch—and the Open Policing Project was born.

Goel pulled together a small team of engineers and also developed a more nuanced and statistically valid way of inferring if discrimination was at play when an officer pulled over and searched a motorist based on the race or ethnicity of the driver. The details of Goel’s new discrimination measure were detailed in a 2017 working paper about the policing project’s early findings.

Those results, based on state traffic stop records from 2011 to 2015, found that Black and Hispanic drivers were more likely than whites to be cited, searched, and arrested, but less likely to have drugs, guns, or other contraband on them.

Their work didn’t stop there. Goel’s team at the Computational Policy Lab is now working with agencies like the Nashville Police Department to suggest strategies for reducing bias in its policing.

Phillips also reached out to the Poynter Institute to help sponsor on-site trainings for journalists so they could make sense of the policing data.

More than 150 reporters from the around the country convened in Chicago and Phoenix, where they learned, among other things, how to navigate Excel and the “R” programming language underlying the statistical analyses and graphics.

“As a reporter, you can get data but you don’t necessarily know what to do with it,” said Mike Beaudet, an investigative reporter with WCVB-TV in Boston who attended two training sessions and turned to Phillips for additional guidance when producing his own story based on the data. “Cheryl and her team gave us a roadmap so we could understand what the data meant.”

Beaudet says that also knowing what the data revealed on a national scale helped inform his own story, which looked at both the disparate treatment of minority motorists in Massachusetts and concerns among some policymakers that a law requiring police departments to collect stop data had ended. “They also gave us a national perspective, which allowed us to tell the local story but also put it in context for how things look around the country,” said Beaudet, who also teaches journalism at Northeastern University. “That’s really valuable.”

This issue brief describes how teams of researchers and leaders in government, business, and nonprofits can work together to generate new ideas, insights, and solutions to make progress on social problems. Stanford Communication Professor Jay Hamilton was a member of Stanford Impact Labs’ design team and now serves on the Faculty Advisory Board. This brief was written prior to the launch of Stanford Impact Labs to show how new evidence and insights developed jointly by scholars and external practitioners can inform policies and programs to improve lives.

Stanford Impact Labs invests in highly motivated teams of researchers and practitioners from government, business, nonprofit organizations, and philanthropy. These teams—impact labs—work together on social problems they choose and where practical progress is possible. With financial capital and professional support from Stanford Impact Labs, they can rapidly develop, test, and scale new solutions to social problems that affect millions of people worldwide.

Learn more about the work Stanford Impact Labs is investing in at impact.stanford.edu.