Should Prison Sentences Be Based On Crimes That Haven’t Been Committed Yet?

Should Prison Sentences Be Based On Crimes That Haven’t Been Committed Yet?

Organisation: FiveThirtyEight (United States)

Publication Date: 04/10/2016

Size of team/newsroom:large


This collaboration between FiveThirtyEight and The Marshall Project examined the role of risk assessments in the U.S. criminal justice system. These survey- and interview-based tools aim to gauge how likely an offender is to commit another crime. They have often been used to determine parole and bail, but recently some states have expanded their use to a new frontier: criminal sentencing. Over the course of several months, a team of reporters, editors and graphic journalists from both publications examined the how risk assessments are used in several U.S. states, with the goal of giving readers an evenhanded look at their benefits and drawbacks. These rubrics can be morally uncomfortable – the pre-crime of “Minority Report” is often mentioned by way of comparison. The models can include factors like employment, education and marriage status that, taken together, can act as a strong proxy for race, potentially exacerbating disparities in criminal sentencing. They may also seem to punish offenders for crimes that haven’t been committed yet, perpetuating a cycle of incarceration. However, states that use risk assessments in sentencing are trying to reduce their prison populations by keeping low-risk offenders out of prison. The tools proponents also argue that informal risk assessment is already built into the criminal justice system, and that formal rubrics with a paper trail are preferable to the unrecorded biases of parole officers and judges. Most importantly, they argue that these systems work – while there isn’t enough data to be definitive, in certain areas risk assessment has proven to reduce recidivism. This was a difficult piece of journalism: There were few simple conclusions or right or wrong answers. We aimed to give readers with an informative and careful presentation of the issue through our reporting while engaging them through interactive graphics in the tough policy choices that are made every day in the criminal justice system.

What makes this project innovative? What was its impact?

This project involved months of research and extensive interviews, but also included heavy-duty data analysis to produce interactive graphics. Taken separately, a traditional written story on risk assessment may have been less engaging for readers while a largely interactive story may have been cold. Together, these components made for a more effective look at a complex topic. This project was also distinctive in its choice of subject matter. The emergence of risk assessment in (adult) criminal sentencing is relatively recent and has received little sustained or in-depth coverage. Finally, the collaboration between the two graphics departments was a technical achievement, with code co-written and edited through GitHub by members of both teams.

Technologies used for this project:

This project included two interactive components, one built in d3.js and the other in react.js. The “Risk Assessment Doesn't Eliminate Bias” interactive required extensive analysis of American Community Survey data, performed using R. As mentioned above, this project also made use of GitHub’s collaborative features, allowing both the Marshall Project and FiveThirtyEight graphics team to work off of the same codebase.
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