Ferguson

[civvver]

No I was just saying those stats are meaningless without context.

This article is much more incriminating. Sounds like there are a ton of issues there. Also shocked how the department is only 8% black when the community is 2/3rds.

http://www.cnn.com/2015/03/04/politics/ferguson-justice-report-shocking/index.html

 

[EgonSpengler]

This article is much more incriminating. Sounds like there are a ton of issues there. Also shocked how the department is only 8% black when the community is 2/3rds.

http://www.cnn.com/2015/03/04/politics/ferguson-justice-report-shocking/index.html

It looks like the article at least partially answers your earlier question:

What you need to do is see the percentage of suspects vs actual arrests or something, see if whites are being overlooked and let off the hook while blacks aren't.

African-American drivers were twice as likely as white drivers to be searched during traffic stops, but 26% less likely to be found in possession of contraband.

 

[Timsup2nothin]

African-American drivers were twice as likely as white drivers to be searched during traffic stops, but 26% less likely to be found in possession of contraband.

When the NYPD was confronted regarding very similar results from their 'stop and frisk' program the blue wall of justification immediately trumpeted:

"See, the program is working! We've trained the minorities to not carry contraband!"

Fortunately when the federal judge stopped laughing and caught his breath he landed on them with both feet.

 

[Bamspeedy]

No I was just saying those stats are meaningless without context.

This article is much more incriminating. Sounds like there are a ton of issues there. Also shocked how the department is only 8% black when the community is 2/3rds.

http://www.cnn.com/2015/03/04/politics/ferguson-justice-report-shocking/index.html

And doesn't that stat need context? What is the racial makeup of the applicants to the police department? (maybe there is indeed discrimination here as well). They don't pick random people off the street to fill the positions.

 

[JollyRoger]

They don't pick random people off the street to fill the positions.

Instead, the pick random people off the street to fill with bullets.

 

[Commodore]

When the NYPD was confronted regarding very similar results from their 'stop and frisk' program the blue wall of justification immediately trumpeted:

"See, the program is working! We've trained the minorities to not carry contraband!"

Fortunately when the federal judge stopped laughing and caught his breath he landed on them with both feet.

I actually wrote a paper on racial disparity in the US criminal justice system and I included a section about the NYPD "stop and frisk" program. That program was so hilariously racist that I was literally laughing while doing my research on it and reading the NYPD's defense of the program.

 

[onejayhawk]

Instead, the pick random people off the street to fill with bullets.

You've been complaining that it wasn't random. Make up your mind.

J

 

[JollyRoger]

You've been complaining that it wasn't random. Make up your mind.

J

I say, I say, that there was a joke, son. Of course I meant random non-white people.

 

[reddishrecue]

Ferguson has been going on way after osama bin laden and the world trade center.

 

[Ajidica]

I actually wrote a paper on racial disparity in the US criminal justice system and I included a section about the NYPD "stop and frisk" program. That program was so hilariously racist that I was literally laughing while doing my research on it and reading the NYPD's defense of the program.

The NYPD could be hilariously racist in more areas than that. When my uncle worked as a cop there, whenever he had to meet his quota for parking tickets, he and other officers would just go down to the black movie theaters, confident they could find expired meters to ticket or traffic violations on the car.

 

[civvver]

The police chief resigned and there was a celebratory protest during which two police officers were shot.

http://www.cnn.com/2015/03/12/us/ferguson-protests/index.html

Obviously the actions of one individual shouldn't implicate an entire community. But it's pretty screwed up when they are finally getting some legal recourse and this happens. It makes it quite hard to sympathize. Both the officers shot weren't even part of the Ferguson department.

 

[Farm Boy]

Originally Posted by CNN
African-American drivers were twice as likely as white drivers to be searched during traffic stops, but 26% less likely to be found in possession of contraband.

Now, I'm not amazing at math, but if we stipulate that bias is causing black drivers to be pulled over for less, meaning we're catching a less reasonably suspicious cross section of the black population by profiling them; if we stipulate that that less suspicious cross-section is being searched at double the rate of more suspicious white people, wouldn't that mean the likelihood of finding something should have a bigger discrepancy than 26%? Doesn't this actually show, within this sample size, the opposite of what is being implied(other than the raw disparity in how suspicion is treated)?

 

[Hygro]

Now, I'm not amazing at math, but if we stipulate that bias is causing black drivers to be pulled over for less, meaning we're catching a less reasonably suspicious cross section of the black population by profiling them; if we stipulate that that less suspicious cross-section is being searched at double the rate of more suspicious white people, wouldn't that mean the likelihood of finding something should have a bigger discrepancy than 26%? Doesn't this actually show, within this sample size, the opposite of what is being implied(other than the raw disparity in how suspicion is treated)?

Unless police pulling people over is not tightly correlated with the rate of contraband material.

 

[Farm Boy]

I would assume it's not. But also the decision to search in both instances, white and black people who have been pulled over would then also need to have extremely low correlation. I mean, that's kinda the point? That the PD are pulling over black drivers and searching them at such high rates that they are drag netting cross sections of the populace that are doing absolutely nothing illegal. Whereas, presumably, that isn't happening to nearly such an extent for white drivers. Should we be of the takeaway that all drivers being searched, white and black, are probably just being harassed for the sake of harassing them, just that blacks harassed somewhat more and because of that drive with dope on them less frequently?

 

[EgonSpengler]

Now, I'm not amazing at math, but if we stipulate that bias is causing black drivers to be pulled over for less, meaning we're catching a less reasonably suspicious cross section of the black population by profiling them; if we stipulate that that less suspicious cross-section is being searched at double the rate of more suspicious white people, wouldn't that mean the likelihood of finding something should have a bigger discrepancy than 26%? Doesn't this actually show, within this sample size, the opposite of what is being implied(other than the raw disparity in how suspicion is treated)?

You might be misreading the sentence (or I might be). It means that a Black person is more likely to be pulled over than a White person, but is less likely to be carrying contraband. I'm no good at math either, but as I understand it, this doesn't mean that the Black people pulled over had 26% less contraband, in aggregate, than the White people who were pulled over.

 

[hangman]

Now, I'm not amazing at math, but if we stipulate that bias is causing black drivers to be pulled over for less, meaning we're catching a less reasonably suspicious cross section of the black population by profiling them; if we stipulate that that less suspicious cross-section is being searched at double the rate of more suspicious white people, wouldn't that mean the likelihood of finding something should have a bigger discrepancy than 26%? Doesn't this actually show, within this sample size, the opposite of what is being implied(other than the raw disparity in how suspicion is treated)?

You mean like some kind of Bayesian caveat?

So if we have...

P(B) -> probability of being black in Ferguson
P(D) -> probability of any person having contraband
P(D | B) -> probability of contraband given that a person is black
P(B | D) -> probability of being black given that person has contraband

(and a similar scheme for white people with P(W))

So what we want to know is which of P(B | D) and P(W | D) is larger.

P(B | D) = P(D | B) * P(B) / P(D)
P(W | D) = P(D | W) * P(W) / P(D)

P(B) = .68
P(W) = .32
P(D | B) = .20
P(D | W) = .33

(Correct me if I'm wrong, I'm not totally sure about the numbers)

If we divide the first equation by the second, we can eliminate the need for knowing P(D), although I'm sure the exact number is around somewhere.

P(B | D) / P(W | D) = [P(D | B) * P(B)] / P(D | W) * P(W)]

So P(B | D) / P(W | D) = (.20 * .68) / (.33 * .32) = .136 / .105

This means that a person possessing contraband in Ferguson is 29% more likely to be black than white. Therefore, if black residents are being searched at a rate significantly higher than 29% more than white residents, the police force is most likely racially biased. We know that black residents are stopped 85% of the time, meaning white residents are stopped 15% of the time or less.

P(S | B) = .85
P(S | W) = .15

.85 / .15 = 5.67
.68 / .32 = 2.32

So for a randomly selected stop as it is, the suspect is 5.67 times as likely to be black, even though a randomly selected Ferguson resident is only 2.32 times as likely to be black. If we account for the likelihood of black residents having contraband, an appropriate ratio would be 2.99 black residents per white resident. Therefore, a black resident is searched 5.67/2.99*.32 (?) = 28% more often than they should be if the police weren't racially biased.

 

[Farm Boy]

I guess my confusion wasn't the conclusion that the police force was biased. I guess my confusion was the implication(was it actually implied or did I misread?) that the police force was biased and targeting black people for contraband enforcement while black people were also less likely in raw probability to be carrying contraband. Which I wouldn't find impossible or hugely unlikely, just that the numbers didn't seem to back the phrasing/implication. Does that make sense or am I way overthinking?

 

[Timsup2nothin]

The police chief resigned and there was a celebratory protest during which two police officers were shot.

http://www.cnn.com/2015/03/12/us/ferguson-protests/index.html

Obviously the actions of one individual shouldn't implicate an entire community. But it's pretty screwed up when they are finally getting some legal recourse and this happens. It makes it quite hard to sympathize. Both the officers shot weren't even part of the Ferguson department.

The lesson is:

When the police sacrifice all relatedness with the public in which they have to operate and make them "the enemy" the enemy they created will sometimes shoot back.

This lesson should be taken to heart by all the "good cops" who don't actively participate in the behaviors that create the conditions. If they don't take action and choose to be part of the 'blue wall of unity' they are just as likely to take the consequences as anyone else. Through a scope their uniform looks just like the rest.

Interesting that the police chief finally resigned. Fair to assume he will be getting a substantial pension and preferential treatment from law enforcement for the rest of his life. Progress.

 

[hangman]

I guess my confusion wasn't the conclusion that the police force was biased. I guess my confusion was the implication(was it actually implied or did I misread?) that the police force was biased and targeting black people for contraband enforcement while black people were also less likely in raw probability to be carrying contraband. Which I wouldn't find impossible or hugely unlikely, just that the numbers didn't seem to back the phrasing/implication. Does that make sense or am I way overthinking?

No, that's legitimate. I updated my last post with more info.

 

[Timsup2nothin]

I guess my confusion wasn't the conclusion that the police force was biased. I guess my confusion was the implication(was it actually implied or did I misread?) that the police force was biased and targeting black people for contraband enforcement while black people were also less likely in raw probability to be carrying contraband. Which I wouldn't find impossible or hugely unlikely, just that the numbers didn't seem to back the phrasing/implication. Does that make sense or am I way overthinking?

Actually, you are making sense. If they only stop and search whites who are really drawing attention as 'likely suspects' while stopping every minority housewife, businessman, or common worker, then yes it makes sense they have a higher contraband rate among the stopped whites without necessarily indicating that whites in general carry more contraband.

It does also indicate though that there are valid indicators that can be used to determine who should be checked, and those valid indicators are being overruled by a general 'screw with the minority folks' approach.

Bottom line, the cops are being sent out with a 'stop fifty people' mandate, stopping ten who merit the search and then filling out their assignment with minorities who just happen along while ignoring whites who just happen along. The 'stop fifty no matter how many really need it' mandate is the problem, and if that was applied without regard to race the white housewives, businessmen, and common workers being stopped would have been able to get the practice discontinued.