The police and systemic violence against Blacks is a reality.
I have been hearing this word a lot (systemic) in phrases such as systemic racism and systemic violence.
It means system wide, or whole body (government, public).
Do the numbers support this as a reality, or is it a perception?
For police shootings, I have found the following:
https://www.statista.com/statistics/585152/people-shot-to-death-by-us-police-by-race/
Which shows in 2019, 370 White people shot to death by police, 235 Black people, and 158 Hispanic
Population percentage of the above is: 60.4% White, 13.4% Black, 18.3% Hispanic
By population, you would expect about 90 black people to be shot to death instead of 235 if we are doing equal opportunity police shooting.
If you dig deeper though, you will find that black people commit about 52% of the homicides in the USA, while making up only 13.4% of the population. If you just go with violent crime in general (rape, aggravated assault, manslaughter, murder, armed robbery) black people commit 38.5% of those crimes.
So just like a logger is far more likely to be killed by a falling tree than a person working in a office in the city, a black person is more likely to be in a situation where they get shot because they are committing more violent crimes per population than white people.
This doesn't mean that the reasons they are in that situation are not problems we should fix (poverty, education, steering away from gangs) but it might mean the police shootings are not really systemic racism, but rather a side effect of a wealth racism.
On this specific metric, analyzed in this specific way, no effect of systemic racism show up conclusively.
However, there are many others where it does. This article, for example points to some of them:
https://www.crf-usa.org/brown-v-board-50th-anniversary/the-color-of-justice.htmlThey, in fact, cite some RAND corporation studies.
(I mistakenly used to associate them with libertarian ideas for the word "rand". It seems they do real data driven research too, so I need to pay more attention to them in future. "unbiased"/"centrist" POV is of course more valuable because it is likely to be less influenced by political BS.).
Raw data is a bitch. You will have a lot of noise in the data due to many complex factors. Sometimes the noise will drown out the signal. e.g. a specific data point can act very differently in the middle of the curve vs. at the tail. How do I konw? Just contemplate the effect/utility of $1 for a median American household earning $50k vs Jeff Bezos. Similarly, when US has 10 deaths per 100k, it is much easier to change vs. Australia's 1.6 per 100k.
So you need to look at it from many different angles and only *then* construct a story - especially in a complex sociological issue. The more different angles you look at, the more likely you are to overcome the confounding factors that are a bane of any honest boffin. These confounding factors can just make the data appear random, or may even point you in the incorrect direction.
After confounding factors come the innate bias of the person. e.g. I used to think of myself as centrist, but of late have seen myself alarmingly close to most liberal positions. So I should require extra dose of evidence to reach any position that agrees with the liberal political point of view. If your "bias" is conservative, you should ask for the same on the other end of the spectrum.
Hence the importance of looking at as many (preferably mutually independent) such angles as possible.
I can tell you that I have seen several such studies showing a persistent presence of systemic violence, and it has a negligible probability that they are all pointing to the same direction by chance.
But again, I am no expert on this. I have no training in sociology. So it is possible that the 5 (or 10) such studies I have seen are indeed hand-picked by a biased media to show to the common public, and that there are 100 others that point in the opposite direction that I have no idea of. If so, please point me to them and I will happily adjust my stance.