How Poorly Informed We Are Re: Police Killings

[marshwiggle]

So we should expect black Americans to commit exactly 13% of the crimes and thus be apprehended by police at a rate of 13% of the total? On what basis?

Our baseline assumption should be that blacks are killed by the police at the same proportion that they appear in the population. If the proportion deviates significantly from that, then we should ask why.

So by that logic, if black children are killed in drive-by shootings at a higher rate than they appear in the population, should we assume that gangs are biased against black children?

[Sun_Worshiper]

The average American is totally clueless about data and statistics, so this doesn't surprise me. Same is true if you ask Americans how much foreign aid we give or how many immigrants their are around the world. My strong suspicion is that people on the left and on the right have erroneous beliefs about data that correspond to their priors.

But if the author is really concerned about getting the data right then he should note that 27% remains a disproportionately large number, since blacks make up about 13% of the population. He should also cite some of the other (and better) statistical analyses showing a less rosy picture, if he wants readers to really understand the issue:

• https://www.pnas.org/content/116/34/16793
• https://www.cambridge.org/core/journals/american-political-science-review/article/administrative-records-mask-racially-biased-policing/66BC0F9998543868BB20F241796B79B8

So we should expect black Americans to commit exactly 13% of the crimes and thus be apprehended by police at a rate of 13% of the total? On what basis?

Our baseline assumption should be that blacks are killed by the police at the same proportion that they appear in the population.

Again, why? I do not understand this. Different demographics have different lifestyles, ways of speaking, values that are emphasized, attitudes, family settings etc. Getting killed by a policeman is not like experiencing an earthquake or being born on the fourth of July. It is not exactly random. It involves interaction with authority, for one thing. I don't see any reason to assume whites, blacks, Asian Americans, Latino Americans, Indian Americans all have the same attitude towards compliance with law, or trust of police for example.

If the proportion deviates significantly from that, then we should ask why.

Well sure, let's do that. but to assume racism is the answer is only a guess.

Did I assume racism? No. Try reading my comment all the way through: "If the proportion deviates significantly from that, then we should ask why. The article you posted doesn't offer any insight, but there could be any number of reasons, such as biased policing or blacks living in high-crime areas. If you are actually curious to find out what is going on, then I'm sure you can find some peer reviewed studies that offer insight."

So we should expect black Americans to commit exactly 13% of the crimes and thus be apprehended by police at a rate of 13% of the total? On what basis?

Our baseline assumption should be that blacks are killed by the police at the same proportion that they appear in the population. If the proportion deviates significantly from that, then we should ask why.

So by that logic, if black children are killed in drive-by shootings at a higher rate than they appear in the population, should we assume that gangs are biased against black children?

Again, you have misread or misunderstood my post. Re-read my entire comment and you'll see that by the logic in my post we should not make any assumption about why we see a relationship between race and police brutality, and we should instead look for (or conduct) further research to answer the question.

[mahagonny]

No, I reiterate, I think I would be more surprised to find out that members of any demographic genre or age had identical rates of lethal encounters with police (either way). Although delving into our various questions further isn't a bad idea.

[Sun_Worshiper]

No, I reiterate, I think I would be more surprised to find out that members of any demographic genre or age had identical rates of lethal encounters with police (either way). Although delving into our various questions further isn't a bad idea.

You should not be surprised that police violence towards individuals of different groups is not identical, but you should be surprised/concerned if differences exist that are unlikely to be due to chance alone.* Some people will assume that the difference exists because of policing, some will assume that it is because blacks commit more crimes or tend to live in high-crime areas, but one would need additional evidence to substantiate those assumptions. This is why delving further into these questions is necessary.**

* A statistically significant difference is one that is probably not due to chance alone.
** I assume that many people have already delved into these questions, which is why I encouraged you to look into the existing research, if you actually want to know what is going on instead of falling back on your own prior assumptions.

[mahagonny]

No, I reiterate, I think I would be more surprised to find out that members of any demographic genre or age had identical rates of lethal encounters with police (either way). Although delving into our various questions further isn't a bad idea.

You should not be surprised that police violence towards individuals of different groups is not identical, but you should be surprised/concerned if differences exist that are unlikely to be due to chance alone.* Some people will assume that the difference exists because of policing, some will assume that it is because blacks commit more crimes or tend to live in high-crime areas, but one would need additional evidence to substantiate those assumptions. This is why delving further into these questions is necessary.**

* A statistically significant difference is one that is probably not due to chance alone.
** I assume that many people have already delved into these questions, which is why I encouraged you to look into the existing research, if you actually want to know what is going on instead of falling back on your own prior assumptions.

OK, I get it. I see what the homework assignment is, professor, thank you.

I should think you would be interested in why people on the left are so vastly  far off in their beliefs about the number of blacks killed in 2019 by police and the proportion of blacks to whites killed by police in that year, and only in one direction, because it might make your party look like a bunch of fools, but that's your affair. I remain disappointed that these misperceptions exist, but whereas I have shifting more to the right lately, it might represent an opportunity.

[Sun_Worshiper]

No, I reiterate, I think I would be more surprised to find out that members of any demographic genre or age had identical rates of lethal encounters with police (either way). Although delving into our various questions further isn't a bad idea.

You should not be surprised that police violence towards individuals of different groups is not identical, but you should be surprised/concerned if differences exist that are unlikely to be due to chance alone.* Some people will assume that the difference exists because of policing, some will assume that it is because blacks commit more crimes or tend to live in high-crime areas, but one would need additional evidence to substantiate those assumptions. This is why delving further into these questions is necessary.**

* A statistically significant difference is one that is probably not due to chance alone.
** I assume that many people have already delved into these questions, which is why I encouraged you to look into the existing research, if you actually want to know what is going on instead of falling back on your own prior assumptions.

OK, I get it. I see what the homework assignment is, professor, thank you.

Just trying help you understand some basic statistical concepts, since based on your posts you appear not to. No need to get upset.

I should think you would be interested in why people on the left are so vastly  far off in their beliefs about the number of blacks killed in 2019 by police and the proportion of blacks to whites killed by police in that year, and only in one direction, because it makes your party look like a bunch of fools, but that's your affair. I remain disappointed that these misperceptions persist, but whereas I have shifting more the the right lately, it might represent an opportunity.

I am interested in why people are often wrong about data/statistics. Look at my first post in this thread, in which I said the following: "The average American is totally clueless about data and statistics, so this doesn't surprise me. Same is true if you ask Americans how much foreign aid we give or how many immigrants their are around the world. My strong suspicion is that people on the left and on the right have erroneous beliefs about data that correspond to their priors."

[mahagonny]

I expect there are many things about which many of us are woefully unformed, but usually that ignorance is not the basis for social justice movements.

[apl68]

Stepping back from all the arguing that's been going on above, it is concerning to see that large stretches of the American public believe that unarmed Black men are being slaughtered by the hundreds, even thousands, by police officers when this simply isn't true.  It indicates that inflammatory rhetoric on the subject has created a great deal of serious misinformation.  This is never good when dealing with such an emotive issue.

There's plenty of evidence that we need reforms in police practices in many police departments around the country.  The need goes far beyond the extreme problem of one or two shootings of unarmed Black men per month.  The activists calling for reform are risking real damage to their cause by engaging in such careless rhetoric.  They're driving away people whom polls show would be willing to back moderate and sensible efforts at reform.

[marshwiggle]

There's plenty of evidence that we need reforms in police practices in many police departments around the country.  The need goes far beyond the extreme problem of one or two shootings of unarmed Black men per month.  The activists calling for reform are risking real damage to their cause by engaging in such careless rhetoric.  They're driving away people whom polls show would be willing to back moderate and sensible efforts at reform.

Regarding the issue of police shootings of people having mental health crises, the history is important to remember, because it involves both sides of the political spectrum. In the 80's and 90's there was a desire to get people out of mental institutions. From the left, this was to give people more autonomy, and from the right there was a desire to reduce costs. The problem was that in the hurry to get that done, sufficient community supports weren't put in place to make these people with real challenges able to live on their own. This resulted in homelessness, etc.

In many cases of police being called to deal with people having mental health crises, it is actually the families of the people themselves making the call. This makes a crucial point: if the people closest to them who understand them the best are calling the police, it means these people with the most insight feel that the use of force provided by the police is needed, at least as an option.

Police did not choose to be the automatic option for dealing with mental health crises; it was thrust on them. Blaming them for having to deal with situations so complex that even the family members who know the victims best aren't able to handle it on their own is unfair.

Calls to "defund the police" in order to put money into more appropriate mental health interventions runs the risk of repeating the error of the 80's and 90's;  if the infrastructure isn't in place before making the changes, there will be chaos and damage that will make things worse. In jurisdictions where other interventions are in place, it requires more funding, at least in the short term, to set up these programs and get them operating. The only savings can be realised after they begin to take effect.

[mahagonny]

Stepping back from all the arguing that's been going on above, it is concerning to see that large stretches of the American public believe that unarmed Black men are being slaughtered by the hundreds, even thousands, by police officers when this simply isn't true.  It indicates that inflammatory rhetoric on the subject has created a great deal of serious misinformation.  This is never good when dealing with such an emotive issue.

Uhm...yeah.

[Descartes]

The problem is that the conversation around this and other hot button issues is decidedly unacademic, and academics themselves are lumped in with politicians, business leaders, and any other group as a collective to threaten with "get on the right side, or we'll cancel you."  I've seen it with one of my favorite (I say that ironically) local activists who has no education beyond high school and who often parrots talking points and phrases used in higher ed discourse on the topic, but then will become vitriolically angry with any actual, real academic critical study that doesn't simply scream the narrative over and over.  This person excuses themselves for this by claiming her "lived experience" informs her more than studies or papers.

I can barely even read comments sections anymore because of all the horrible logic, poorly constructed arguments and bad thinking.  One of my favorite tropes is "Y is an example of white privilege because person Y was treated differently by a federal police agency in Seattle than person Z was treated by the NYPD.  Person Y was white and person Z was black; see the difference?"  Of course this ignores all sorts of other variables, such as a federal law enforcement agency vs. a local one, a different state with different laws, courts, and standards than another state or the federal system, variances in local policy from two cities 2,000 miles away from each other, and a myriad of other variables; but no, they are simply convinced that because Y got let out on bail, or didn't get shot, or whatever else, it is because Y is white when Z's outcome was different (sometimes also with different sets of facts to boot!)

[Kron3007]

The problem is that the conversation around this and other hot button issues is decidedly unacademic, and academics themselves are lumped in with politicians, business leaders, and any other group as a collective to threaten with "get on the right side, or we'll cancel you."  I've seen it with one of my favorite (I say that ironically) local activists who has no education beyond high school and who often parrots talking points and phrases used in higher ed discourse on the topic, but then will become vitriolically angry with any actual, real academic critical study that doesn't simply scream the narrative over and over.  This person excuses themselves for this by claiming her "lived experience" informs her more than studies or papers.

I can barely even read comments sections anymore because of all the horrible logic, poorly constructed arguments and bad thinking.  One of my favorite tropes is "Y is an example of white privilege because person Y was treated differently by a federal police agency in Seattle than person Z was treated by the NYPD.  Person Y was white and person Z was black; see the difference?"  Of course this ignores all sorts of other variables, such as a federal law enforcement agency vs. a local one, a different state with different laws, courts, and standards than another state or the federal system, variances in local policy from two cities 2,000 miles away from each other, and a myriad of other variables; but no, they are simply convinced that because Y got let out on bail, or didn't get shot, or whatever else, it is because Y is white when Z's outcome was different (sometimes also with different sets of facts to boot!)

Just as one person's lived experience is not overly convincing or relevant, neither is presenting this "study" in isolation while ignoring the many, many, studies and reports that show police treat minorities different than whites (it is not only about how many are killed).  Even if it were only about the absolute numbers of deaths, the original link neglects to compare the absolute number of black deaths to that of whites to provide any context.  13-25 dosn't sound like a big number but there is no reference point.  Ten pounds dosn't sound very heavy, but it is if we are talking about hamsters... 

Sure, this link shows that people do not have a good idea of the actual numbers, but that would be true of almost anything you ask them that is even slightly political.  You could do the exact same thing regarding problems with illegal immigration and you would likely find the same results, only in reverse.





   

[marshwiggle]

Just as one person's lived experience is not overly convincing or relevant, neither is presenting this "study" in isolation while ignoring the many, many, studies and reports that show police treat minorities different than whites (it is not only about how many are killed).  Even if it were only about the absolute numbers of deaths, the original link neglects to compare the absolute number of black deaths to that of whites to provide any context. 13-25 dosn't sound like a big number but there is no reference point.  Ten pounds dosn't sound very heavy, but it is if we are talking about hamsters... 

"Context" isn't unique. The number of blacks killed by white police could be put in "context" by:

• the number of blacks killed by other blacks who aren't cops
• the number of whites killed by black cops
• the number of arrests of blacks by white cops

and so on.
Picking any one other number for context creates a different story.

[Kron3007]

Just as one person's lived experience is not overly convincing or relevant, neither is presenting this "study" in isolation while ignoring the many, many, studies and reports that show police treat minorities different than whites (it is not only about how many are killed).  Even if it were only about the absolute numbers of deaths, the original link neglects to compare the absolute number of black deaths to that of whites to provide any context. 13-25 dosn't sound like a big number but there is no reference point.  Ten pounds dosn't sound very heavy, but it is if we are talking about hamsters... 

"Context" isn't unique. The number of blacks killed by white police could be put in "context" by:

• the number of blacks killed by other blacks who aren't cops
• the number of whites killed by black cops
• the number of arrests of blacks by white cops

and so on.
Picking any one other number for context creates a different story.

Thanks for agreeing with me that it is important to take a more holistic look at the issues and not focus on any one statistic in isolation.