Justice for Some

Justice for Some

Organisation: Pittsburgh Tribune-Review (United States)

Publication Date: 04/11/2016

Size of team/newsroom:large


In the wake of several major cases of police officers shooting blacks in recent years, the Tribune-Review decided to look at how often the 94 U.S. Attorneys offices nationwide prosecuted civil rights complaints against police officers. The perception from groups like the ACLU, the NAACP, Black Lives Matters as well as some in the general public was that it didn't seem like the feds prosecuted very many cases. The Trib analyzed more than 3 million records dating back to 1995 to show the declination rate by US Attorneys presented with civil rights complaints was more than 90 percent each year. Over the last two decades, US Attorneys declined to prosecute police officers for civil rights complaints an average 96% of the time. Our story addresses several key reasons why that is, the message it sends to the public, reaction, etc.

What makes this project innovative? What was its impact?

Our project is innovative because it has never been done before on such a scale. The Trib went back to 1995 to start because the federal anti-crime legislation went into effect that year. Only one study has ever looked at the prosecution rate and that was done in 1995 and reported in 1996. It was already several months old when publicly released and the only thing available to cite since. The Trib decided to look at a two-decade/20 year history to determine pattern over much lengthier period of time that includes the period between the anti-crime bill enactment and the last two years of racial divide over police shootings to get a much deeper, complete picture of whether there was any pattern by the feds and what that pattern was and why. The Trib analyzed more than 3 million records to spotlight the outcome of civil rights complaints filed against law enforcement officers. We were interested in whether federal prosecutors brought any charges in response to an alleged criminal incident -- such as a civil rights violation by police -- than the number of people they prosecuted. To prevent any uncertainties, our analysis also focuses ONLY on cases with CLEAR OUTCOMES: either prosecutors declined to prosecute anyone in the case, prosecutors filed charges and obtained convictions, or charges were filed and defendants were acquitted.

Technologies used for this project:

We utilized annual reports from the US Department of Justice, and decided our best source of information from them was the National Caseload Data that the department releases each month on its website. The information was downloaded as textfiles. EditPadPro7 was used for some editing. To get answers out of the data, we had to import it into a MySQL database. That involves downloading about 50 GB of data for each month for each year since 1995 and then using a combination of Python and MySQL scripts to process the text files and import them into the database. Following that, we used R programs scripts to query the database and analyze the results. Early on, we realized there were two main ways to analyze the data: by defendants or by cases. We chose case analysis for a couple of reasons. One is that most of the information that would make a defendant-based analysis interesting and informative, such as race, gender and age of defendants and victims, is redacted from the files made available The other is, as we said above, we were more interested in whether prosecutors brought any charges in response to an alleged criminal incident involving law enforcement officers. The information was then put into Excel spreadsheet for reporter to use as reference when making calls, etc. We also used this info to create an interactive map via Google for the newspaper's website at TribLIVE.com, allowing readers to click the U.S. Attorney's office district for their state or area within a state or any of the 94 districts nationwide and see the rate at which civil rights cases against police were declined by federal prosecutors there. (Map can be found at the story link site above)


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