Exploratory Data Analysis: Visualizing Police Killings in the U.S for the year 2015.

The Counted is a project by the Guardian working to count the number of people killed by police and other law enforcement agencies in the United States throughout 2015 and 2016, to monitor their demographics and to tell the stories of how they died. The data is available for download at kaggle.com.

A Brief Description:

The data contains a total of 1146 observations — victims of the U.S police killing for the year 2015 — and a total of 14 features — attributes of each victim.

The features recorded for each victim is described below…

1. id: A unique number given to each victim
2. Name: The name of the victim
3. Age: Age of the victim
4. Gender: The gender of the victim
5. Race/Ethnicity: Race/ Ethnicity of the victim
6. Armed: Was the victim armed? If yes, what type of Arm?
7. Month: The month the incident occurred
8. Day: The Day the incident occurred.
9. Year: The Year the incident occurred (2015)
10. Street Address: The street where the killing occurred
11. City: The city of the killing
12. State: The State where it happened
13. Longitude: The Longitude(coordinate) of the killing
14. Latitude: The Latitude(coordinate) of the killing
15. Classification: How the victim was killed, i.e gunshot/vehicle
16. Law Enforcement Agency: The office responsible for the attack

The Visualization:

Gender: It is not far-fetched to say that most of the victims of police killing in the U.S are male as it consists of over 95% of the total data recorded.

Age Distribution: The age was fairly normally distributed with a long tail and the age group of within 25–35 having the highest number of occurrences.

The small bar between 0–10 is Jeremy Mardis, a 6 year old boy that was not armed and was killed by gunshots in November 2015. He was ‘White’ and the law enforcement agency responsible for the killing was Ward 2 city marshals, Los Angeles.

Race/Ethnicity: More ‘White’ was killed than all other ethnicity/race combined. Over 50% of the people killed by the U.S police force in 2015 are White and the Arab-American recorded the lowest number of victims with less than 1% of the total data recorded.

Suspect Armed?: Over 50% of the total victims are armed either with a firearm or a knife. Over 30% of the total data recorded are with No firearm or with a Non-lethal firearm or its unknown if they were armed or not.

Nature of Death: Pretty much cause or nature of death was accounted for with almost 90% of the victims being killed with gunshots. Other forms were; Death in custody, Struck by vehicle, Taser and others( which is less than 1%).

Killings Per Months: The killing was at its peak in the month of July with over 125 victims and was at its lowest in the month of June with 80 victims.

Killings Per States: 10 States accounted for over 50% of the killings out of the 51 states recorded with California topping the list with 211 recorded incidence which represents 18.4% of the total data.

Race/Ethnicity Based on Arms: This shows the rate killing each ethnic groups based on arms.

Almost 50% of the ‘White’ killed were carrying firearms and the Arab-America has the lowest rate of victims with firearms. Also less than 30% of the Arab-American killed were with firearm.

Nature of Death Based on Race/Ethnicity: As I’ve said earlier, almost 90% of the victims were killed with gunshots. All the Arab-American cases recorded were with gunshots which makes the only group with a 100% nature of death. Native American has the highest number of percentage that died in custody and Asia/Pacific Islander has the highest percentage of victims that died by ‘Taser’. Note that these rates are relative to the total number of each group recorded.

And this come to the end of the exploratory data analysis for U.S Police killings in the year 2015.

The data and the code can be downloaded from github.com.

Leave your comments, questions and what you would like me to add to the post and don’t forget to recommend.

Cheers

Ede