Event-level prediction of urban crime reveals a enforcement bias in US cities
nature.com
nature.com
Which is weird, because during the pandemic a whole bunch of police were super vocal about not wanted to enforce 'unjust mask mandates' yet they'll go and terrorize minority communities with glee
They are definitely part of the problem, but you're right that it extends far beyond just them
https://www.rutgers.edu/news/bad-policing-bad-law-not-bad-ap...
That's the data, not my opinion.
The system can be racist even if its constituents aren't. It shouldn't be a new or unfamiliar idea to anyone here that complex systems can have unexpected outcomes, biases, and emergent properties.
Further reading (I'm not typing all of this again):
Let me outline the counter claim. Police spend more effort on actually solving crimes in rich neighborhood (as per article) but they still need arrests in poor neighborhoods so they arrest people more randomly and on flimsier pretexts. When people complain, they throw out the line you have above.
that's because when the police kill people, there are no repercussions for them, no matter how much obvious bias, incompetence, disregard for human life, and malice is apparent in their activity. police can literally murder people on a whim and 99% of the time not even be suspected of any crime much less prosecuted. The George Floyd murder required that it was videoed from start to finish for there to be any chance of an actual criminal trial proceeding. All of the police murders that we hear about now are all news because there is now video. Think of all the decades and decades when there was no video, how many people must have been murdered on a whim, without the event even being considered a crime. But even today, obviously immensely negligent acts, such as the murders of Tamir Rice and Eric Garner, still basically have zero repercussions for their killers.
This is nothing like the usual "murder rate", where the violence that occurs between people who are not cops is all considered to be crime, and is handled as such. There's no disagreement on what justice looks like. Murder is bad, but aribtrary murder by state actors with no accountability is worse.
> Imagine being a police officer risking your life every single day.
please review the accounts of Uvalde as children and teachers were murdered over an hour while heavily armed and armored police stood around and did nothing but prevent anyone else from being able to help. There are many ordinary jobs that are statistically much more dangerous than that of the police including things like roofing, truck driving, construction (source: OSHA https://www.invictuslawpc.com/most-dangerous-jobs-osha/)
Same with medical professionals. Medical errors kills over 250k people each year in the US. It generally isn't criminal if they make a judgement that results in someone's death. There's been a number of serial killers in medicine because we give medical professionals the benefit of the doubt due to the nature of the profession.
My point is that these professions aren't your average jobs. It's normal for doctors and police to deal with life and death, and to make decisions that could set off a sequence of events that quickly results in death. So either we have to say that nobody deserves to make these decisions, or we have to hold them to realistic standards.
Indeed, and it doesn't show up on the CDC's leading causes of death list, even though it should rank in third place.
> Analyzing medical death rate data over an eight-year period, Johns Hopkins patient safety experts have calculated that more than 250,000 deaths per year are due to medical error in the U.S. Their figure, published May 3 in The BMJ, surpasses the U.S. Centers for Disease Control and Prevention’s (CDC’s) third leading cause of death — respiratory disease, which kills close to 150,000 people per year.
> The Johns Hopkins team says the CDC’s way of collecting national health statistics fails to classify medical errors separately on the death certificate.
https://www.hopkinsmedicine.org/news/media/releases/study_su...
Personally, I think unchallenged hero worship of medical professionals has something to do with it. Similar to the way widespread [albeit less universal] hero worship also shields police.
Yes, absolutely. That's what needs to start happening with regard to police. You've understood perfectly.
0. https://www.prisonlegalnews.org/news/1995/dec/15/contract-ph...
Also we’re not talking about medical professionals here and mixing them into the conversation is muddying the water.
If that was the case, then why isn't there 250x the public outrage when a medical professional makes an error that kills someone, since they kill 250x more than police do each year[0]?
And I'm not muddying the waters, so I don't appreciate the accusation of bad faith. I think the professions are comparable in nature, but there's a huge discrepancy in how they're treated, despite a massive body count difference.
0. https://www.prisonpolicy.org/blog/2020/06/05/policekillings/
You're being accused of muddying the waters for comparing medical professionals fucking up difficult procedures to cops using illegal escalations of violence against unarmed civilians. I happen to agree.
And it's a problem that only exists in the US. Other industrial nations don't have anywhere near the police killings we do per capita. I'm sure the reply will be that those places are more homogeneous, which is very "it deports the minorities or it gets police brutality"
When cops actually face anything resembling a consequence for taking human lives we can talk about what public outrage is "disproportionate"
Racist strawmanning
They do not have qualified immunity.
It is treason.
(It is abuse of the powers assigned by the State.)
Now imagine being a logging worker, risking your life every day[1], and not having anyone ever say this about you.
Deadliest Catch, similarly, highlighted the risk faced in one of the other most dangerous jobs, offshore fishing.
Defender problem. Just like how pen testers are insufferably smug that throwing random input at your endpoint causes unexpected problems. Nevermind that building secure endpoints is harder than building random fuzzers.
Don't look up
I have never seen a valid social science result derived by this method. This is just fancy model lovers dabbling.
If the patterns they claim are real, they can be exposed with much simpler and direct methods. Let's talk after that's been done.
No social scientist worth their salt would just report some model-generated plots and claim it represented something about the real world. That's just hypothesis generation. Confirming the hypothesis requires direct data analysis.
Will it? Are you looking at reported crimes or investigated crimes? If the cops never show up do people stop calling? I’m not sure how a simple plot tells the whole story here.
0.90 AUC doesn't mean everything you can get this model to tell you is real, especially if you are trying to tell a causal story.
Now, show me a plot of your data with a trend line and I'll often be able to tell you if there's a real pattern in your data.
This isn't a new or surprising principle, and it also applies to much simpler models. Any scientist knows that, if you fit a line to data, you better plot the data with the line before you make any big claims about what that line tells you about the world, because the data underlying that fit could look lots of completely different ways with different implications [1]. These authors did the equivalent of fitting the line and telling us all about its formula and the big implications of the formula without ever plotting it with the data.
A cutting-edge model can provide interesting leads but they need to be confirmed by a known reliable method if a serious question hangs in the balance.
The authors developed an impressive model and got interesting results, and I appreciate the stated intent to shed light on enforcement bias patterns. However, I disagree with their claim that they have successfully done so in this paper. They have only provided an interesting prediction which needs to be verified with careful data analysis.
If increased crime in poor areas leads to decreased enforcement, we should be able to see this with well-chosen analysis, reporting and visualizations. We should be able to see examples of the claimed phenomenon happening in full context and have the opportunity to consider alternative causal explanations. If the claim still stands after this has been done, it deserves to be taken seriously. At present, the complexity of the model and the fact that Granger causality != Real causality prevent us from drawing any conclusions.
(The above does not refer to the claim about rich communities draining resources from poor communities, which seems to me like pure speculation. I can draw no logical connection between this claim and the results in the paper.)
There are complications, though. In the 90s, leadership figures in minority communities rightly called out under policing as a serious issue that hurt their communities and needed addressing. Nowadays enforcement actions, although increasing overall quality of life and decreasing violence and premature death, are highly politically sensitive, and police in those communities can expect lots of push back and legal actions without the support of political leadership.
It's simple enough to say "well police officers should enforce perfectly," and that would of course be ideal. But in a world where perfect solutions are the enemy of the good, what's the best course of action? There are always tradeoffs, and activists need to recognize them.
...and stopping people for random pat-downs, pulling people over for the slightest traffic infraction or "smell of pot" and then emptying everyone out of the car because someone acted "suspiciously" and searching the car.
That's what activists and leaders in minority communities are referring to when they talk about under and over policing. You can have a community that is both under and over policed.
The typical pattern is: little focus on enforcement around crimes that affect people's lives, and lots of enforcement on "harming society / law and order" type crimes.
Another example of over-policing: SWAT teams showing up with door-busters and sub machine guns and stun grenades to serve no-knock warrants on suspects with no history of violence, over non-violent drug crimes.
Is the answer basically "officers only use force against individuals who are actively committing violence against them at that same very moment and otherwise just exist and don't engage with any particular person more or less than any other"?
In other words, cops abandon poor neighborhoods when crime goes up there and mobilize to rich neighborhoods when crime goes up there. I’d be interested to know the other 6 metro areas, although I have a suspicion of which ones they are and their flavor of DA.
If you work a low-end shift job and miss a shift, you're likely to get fired almost immediately, or put on some sort of suspension, or otherwise punished via poor shift assignments and the like. At the very least, you lose income for the shift, and that is likely a substantial amount of money for you. The "court fee" is a substantial amount of money to you. You end up with a public defender who has little influence with the other officers of the court - the judge, the DA, etc. They have a docket a mile long for that day and needs to triage their time and resources, so they push you to plea out so they have time for the guy who is facing a felony or even capitol charge. If convicted your choices are jail or fines, and the fines might be days or weeks of pay you don't have.
If you work a professional job, you have "personal days" you can cash in, your employer provides you with things like free legal assistance via EAPs - or you probably already have established professional relationships with at least some sort of attorney, who you can afford, and who knows someone who knows the judges and DAs in the district you're going to be appearing in, and maybe the whole thing just goes away. The court fees are a rounding error in your weekly pay. And the laws are all "jail time or X dollar fine." So even if you're found guilty - the fine is a rounding error to you.
Typically you see under-enforcement of crimes that have direct victims and real impact on people's lives - property and violent crime (say, a stolen car, or stolen bike that leaves the victim unable to get to work/child care, or the victim injured and unable to work, if only temporarily) that directly impact a person's life...and over-enforcement of general "orderly society" crimes and regulations that don't have a direct victim or impact one people's lives.
For example, you'll see cops take 20 minutes to respond to calls of domestic violence, which might come down to being short staffed.
But isn't it strange that every patrol cop in the neighborhood never fails to make sure to check that the corner sausage cart vendor and ice cream van have every single permit in order? And how there's always time to stop someone for a non-functional tail light? And to stop random dudes for pat-downs?
While in the US they have to come up with a billion bullshit reasons to stop someone which mostly impacts poor people in bad cars.
it's not that surprising. I'm sure malice is part of the explanation, but it's also just easier. the ice cream van either has the permit or they don't. they're not going to run away. in 30 minutes, the officer gets to add another citation to the monthly tally. it's a nearly certain outcome that takes little effort.
the stolen car or bike is already gone by the time the officer shows up to the scene (if they bother to). it takes an indeterminate amount of time to track down the stolen item, and even more to gather enough evidence to actually charge someone. it sucks, but most people who are evaluated on metrics try to avoid spending a lot of time on tasks where they might come out empty-handed. it's a hard organizational problem to incentivize this sort of work.
Ingestion is a manual process, and only happens when someone comes to them with a large amount of papers.
Sci-hub uses an automated process. However that process has been halted for a while since the ongoing legal drama in India.