Race Behind Bars

Race Behind Bars

Organisation: The New York Times (United States)

Publication Date: 04/07/2017

Size of team/newsroom:large


Much attention gets paid to racial disparity in policing and in the courts, the "front end" of the U.S. justice system. This investigation focussed on the "back end" -- disparities that continue through incarceration and release. After officials dismissed a series of highly publicized prison brutality incidents as isolated cases, the New York Times obtained a database of prison disciplinary incidents and investigated how guards punish prisoners on a systemic basis and found that minority inmates were more likely to get punished; were often punished at a higher rate for infractions involving no physical evidence (such as disobeying orders); were more likely to face time in solitary confinement as a consequence of their infractions; and were sentenced to longer average stints in isolation. On top of that, the racial disparity continues when inmates appear before the parole board. While the board rarely grants parole to violent offenders of any race, the Times found the board to be more willing to give low-level offenders, like burglars, a chance if they were white. The day the second piece was published, the governor of New York ordered an independent investigation of the system.

What makes this project innovative? What was its impact?

Many academic studies and newspaper projects have been devoted to bias in the front end of the justice system. Shootings involving police; highway stops; search-and-frisks; disparate sentencing. But except for a handful of individual cases, there have been few data-driven reports that look life behind bars or prison release. New York State had never released prison disciplinary data before we obtained it, and as we later learned, they did not track discipline by race internally. In fact, to conduct our analysis, we had to merge the discipline data with a separate data set of incarcerated individuals so we could put the disciplinary incidents in the context of race, age, gender and severity of the crime of conviction. Likewise, there have been relatively few studies of parole. Not only did our story unearth the pattern of disparity among low-level offenders, but it also delved into the politics and general administrative problems plaguing the parole board. The impact of the story was quick. Governor Andrew Cuomo, who had previously stated that problems in the system were isolated incidents, ordered an independent investigation of racial inequities in the 54 state prisons. And the stories were reprinted in local papers throughout the state. It should also be noted that while the data work provided the backbone for the investigation, it would not have been possible without years of beat reporting on the topic. New York has a tough FOIA law, so some of the data we obtained was only released to us because of the trust developed over years between reporters and sources.

Technologies used for this project:

The data work for the story involved merging an array of datasets about the prison system. We obtained data on disciplinary incidents via FOIA, and in a separate FOIA, obtained rosters of people currently or formerly incarcerated. Sources began sending us snapshots of the prison rosters at various time periods, which we used to fill in missing information about inmates in the disciplinary database. Because this was a collaborative project with three reporters and several editors, we built an interactive version of the database in Ruby on Rails, which allowed all team members to query the data in dozens of ways at any time. This became even more important when our document FOIA requests came in. To go deeper into the details of how cases played out, we requested hundreds of incident narrative reports and parole board transcripts. These documents were also posted on our Rails site, with an interface for team members to tag and classify the documents. Lastly, when reading the parole transcripts showed us that the board decisions often seemed arbitrary, we created a "game" for our readers that enable them to read snippets of cases and try to guess how the parole board decided.
Follow this project

Comments (0)

You have to be connected to contribute

You have to be connected to follow

Leave this project and no longer be informed about this project

By joining this project, you will be informed by email when an update or a new contribution is posted on the website.

Thank you for your active participation !

The GEN Community Team