2015 Police Killings Statistical Report
This topic may be sensitive to some readers
For my Statistical Graphics & Visualization Course, we found data on 467 documented police killings in the United States from January to June 2015, using a dataset compiled by FiveThirtyEight from The Guardian’s reporting. Our goal was to understand when and where these incidents occurred and to explore how demographic, socioeconomic, and geographic factors relate to patterns in police use of lethal force. And we tried to answer our 4 research questions.
Overall, the project highlighted complex relationships between policing outcomes and broader structural factors, while underscoring the importance of careful normalization and statistical analysis when interpreting sensitive real-world data.
Key Figures:
1. Shows the locations & frequency of police killings.
Killings cluster geographically, and interpretations change when using per-capita rather than absolute counts. Incidents also occur more frequently on weekdays than weekends.
2. Shows the economic characteristics related to the police killings
Police killings disproportionately occur in regions with higher unemployment, and areas with higher white population shares tend to have lower poverty rates.
Research Questions:
What patterns exist in when & where police killings occur?
What are the economic characteristics of the areas where victims were killed?
Are there racial disparities in who is killed and how?
Do armed victims die different deaths than unarmed victims?
3. Shows whether there are racial disparities in police killings
Yes—age distributions differ by race, and a chi-squared test shows that race and cause of death are statistically associated rather than independent.
4. Shows how victims die from police killings
Armed victims did not die substantially differently than unarmed victims in terms of cause of death. in both groups, police gunshots are by far the most common cause of death.
