Read on Twitter
Here are some charts showing various relationships between police killings, all and unarmed victims, race and gender.
All analysis is original and the raw data was obtained from http://mappingpoliceviolence.org , the most complete dataset available.
All analysis is original and the raw data was obtained from http://mappingpoliceviolence.org , the most complete dataset available.
The overwhelming number of people killed by police are male.
In terms of percentage, 95% of police killing victim’s are male. Tramsgender victims are 0.1%.
In terms of race, more white people are killed by police than other races. This is because white people still are a majority of the population in the USA, comprising 62% or so.
Here are the actual population numbers for 2017.
This chart attempts to correct for population differences by finding the number of police killings per 100,000 population. Using this metric, black Americans are almost 3x as likely to be killed by police.
But is population by race the correct normalizing factor? If the goal is to measure police racial bias, in general, police can only possibly kill those they interact with. One imperfect way to measure this is the total number of crimes by race.
This chart shows the total number of crimes committed by people of each race annually. It is noteworthy that more than 7% of Native Americans are arrested for crimes each year, more than 5% of Blacks, 2% of White people, and just 0.5% of Asians.
We hypothesize that the number of police killings should be proportional to the number of crimes, since in general, these encounters create a window for police to kill. It is noteworthy that the bars are now all similar, potentially indicating minimal racial bias.
Now let’s consider the emotionally charged data related to killing of unarmed people. http://Mappingpoliceviolence.org defines unarmed as not holding any weapons when killed, or holding a toy weapon.
If we normalize police killings by race population, it is clear that unarmed Black Americans are killed at a much higher rate than any other racial group. It is also important to note that the overall rates are tiny, only 1 of 1 million black people will die while unarmed.
If we normalize the number of unarmed police killings per crime rate by race, the rates are once again very similar. There is a slightly higher rate for blacks than whites. The very small rate for Native Americans may have to do with isolation on reservations with tribal police.
It’s instructive to examine police killings by race and gender.
Gender clearly plays a vastly more influential role in police killings than race. Why are males 20-30 times more likely to be killed? Is this evidence for gender bias in law enforcement? Or are there other reasons?
Gender clearly plays a vastly more influential role in police killings than race. Why are males 20-30 times more likely to be killed? Is this evidence for gender bias in law enforcement? Or are there other reasons?
Another statistic to probe evidence for racial bias in policing is to divide the number of unarmed victims by total victims for a given race, to measure if police are perhaps targeting unarmed people. There is a slightly higher rate for unarmed shooting of black people.
It is also instructive to analyze trends over time. Here, we see that the relative rates of killing unarmed black vs. white people has dropped, with 2019 being the first year when unarmed whites were killed at a higher rate than unarmed blacks.
The total number of police killings is incredibly consistent from year to year, varying in a range from 1050 to 1100, independent of the political party in power.
If we zoom in on the number of unarmed black people killed by police, we see a significant drop since 2016, with 2018-2019 having the lowest rates on record.
We can look for evidence of systemic racism in the larger society by comparing the rates that police officers are charged with a crime as a result of an officer involved killing. Based on 8 years of data, police are twice as likely to be charged after killing a black person.
It’s interesting to look for patterns based on the age of those who are killed by police. Here, we show the distribution of black victims of police killings. It is striking that there is a large bump from 20 to 25 years of age.
In contrast, older white men tend to be the victims of police shooting. The percentage of whites steadily rises all the way to 27 years old.
Among black men it is especially young men(<25 y.o.) who are killed by police. It seems urgent to understand the societal causes for this.
Among black men it is especially young men(<25 y.o.) who are killed by police. It seems urgent to understand the societal causes for this.
