Project description:

This project is an effort to visualize the death toll of policing in the United States and to explore some of the dynamics that underlie this phenomenon. We draw on data from Fatal Encounters and put it in conversation with information about the governance of the places where these deaths occur.

Impact reached:

Police brutality is a rampant and significant issue for many communities across the United States.  Complex and sensitive in nature, this issue is even further obscured by the lack of government transparency surrounding issues of police brutality.  There currently are not any government led initiatives that provide clarity on who is being killed by police when and where.

In light of such government failures, academic efforts have resulted in national databases about police killings.  In an effort to put this novel open source data to good use, our article attempts to shed light on the issue of police brutality through a visual narrative of storytelling that allows the reader to explore the data for themself.

In publishing this work, we have received an honorable mention at the Pudding Awards for best visual stories.  Regardless of this success, our project has had the important impact of illustrating fact from fiction surrounding issues of police violence.

Techniques/technologies used:

The main tools used for this data journalism project were the Python programming language, the D3.js Javascript Framework, and GitHub.  The Python programming language was used to collect, clean, and wrangle data from multiple data sources (i.e. fatalencounters.org, 2016 county presidential election returns data, police employment and population data).  With the Python programming language, exploratory data analysis was also conducted.  The D3.js Javascript Framework was used to manipulate the data with the Javascript programming language in order to build interactive web visualizations.  In particular, the final beeswarm visualization of this article was produced using the D3.js force layout.  Finally, GitHub was used to not only manage this project, but to also host the associated article webpage for this project.

What was the hardest part of this project?

Police brutality is a sensitive, heart-wrenching subject that can be very difficult to report upon effectively.  For this reason, we spent a considerable amount of time planning and discussing the narrative, storyline, and accompanying visualizations of this piece.  While we wanted the article to be engaging, we also wanted to ensure that the piece was presented in good taste.  We strived to guarantee that while exploratory in nature, the seriousness and gravity associated with the piece was not lost.  Ensuring a balance between exploration and seriousness was difficult in completing this project.

What can others learn from this project?

First, this project is a good example of how to work with unique and unconventional data sources. The Fatal Encounters dataset documents a reality that is quite literally not documented anywhere else, and the data is presented as it is collected. It was a challenge to sort through all of this information and decide, by ourselves, which part we want to focus on, how we want to analyze it, and how we want to present it. Similarly, it was a challenge to put different datasets in dialogue with each other. Yet, while these decisions were challenging, they allowed us to write a meaningful and substantial story. 

Further, this data on its own is malleable – it can be interpreted in different ways, and used for different purposes. It was important for us to not simply “let the data speak for itself”, but to give a strong historical and political context to the numbers. Beyond providing essential information to understand the data, it also explains why and how we came to certain interpretations and conclusions. 

Lastly, this article provides valuable insight on data visualization. The visual presentation of data undoubtedly has enormous impact on how well it is understood or assimilated by readers, and we were extremely careful in our choice of visualizations; our goal was for the visuals to give clarity to the numbers, not to change their meaning. The article also exemplifies how the D3.js Javascript Framework can provide a profound contribution to the coherence and clarity of a data-driven story.

Project links:

www.leonardonicoletti.com/work/deathtoll

pudding.cool/process/pudding-cup-2020/

github.com/lnicoletti/police_violence