Police Shootings in the US — a simple data analysis.
Police brutality has been a hot topic in recent years, with protests springing up in different countries, and hashtags trending across various social media platforms. Numerous videos showing police officers exercising undue or excessive force against an individual have trended, and oftentimes caused outrage on social media. In a number of cases, these brutal scenes have ended in the death of the affected individuals. This is why I was keen to explore the police shootings dataset found on Kaggle, to see what could be learned from it.
The dataset is a record of shootings carried out by police officers in the United States, between 2015 and 2020. It contains a list of people that were shot by the police, with dates, and across various cities. It also has details of the victims’ age, gender, race, if they were armed or not, and so on.
I explored the last 3000 rows of the dataset, which consisted of fatal shootings that occured from 2017 to 2020. I used Python Data Science libraries for my analysis. The findings are as follows.
This is the link to the complete data analysis on Github.
General Statistics
Most of the victims were young people, between the age of 30–40 years, with men making up about 95% of them. Only approximately 5% of victims were female. In terms of absolute numbers, more White people were fatally shot (1,498 deaths), followed by Black people (798 deaths) and Hispanic people (565 deaths). However, normalizing these numbers by the demographic composition of the country showed that Black people suffered almost three times more fatal police shootings than any other group. In 2020, roughly 0.1 out of 100,000 White people were fatally shot by the police, compared to 0.25 out of 100,000 Black people.
About 96% of victims were shot, while 4 % were both shot and tasered. In some cases, the individuals were armed. Guns were the most held weapon followed by sharp objects. 200 people out of the total 3000 were unarmed, while 4 had explosives.
The number of shootings continued to remain high (over 800 per year) from 2017 to 2019, with little change, then dropped by half in 2020. This drop is unlikely to be linked to body cam usage, as the use of body cameras actually fell in 2020, after very minimal uptake in the first 3 years.There were no body cam recordings for a huge proportion of these fatal events.
Going forward...
It is good to finally see a fall in the number of police shootings from 2020. I wonder if protests, or passers-by recording incidents have influenced this, and of course if lockdowns have saved some people from more than Covid 19. I’d love to analyse the data at the end of 2021 to see if this downward trend is maintained.
Thank you for reading my very first report!! I hope you enjoyed it. Please share your thoughts to help me improve.
