Dashcam images reveal where police are deployed

Yea my apologies, my statement was ambiguously two-parted

Stop and frisk was ruled unconstitutional due to general shitting on the 4th amendment.

Separately, the federal government still claims that it's disproportionately targeting minorities (and that it's still occurring generally).

My vague recollection of the decade-ago ruling was that it didn't have any 14th amendment or w/e statements

No. The big question is whether we want to live in a society that requires the police to randomly stop and frisk people in order to stop crime. There is crime, I don't think anyone can dispute that.

Honestly, I personally have no problem with criminal-looking individuals being stopped and frisked in high-crime areas. And I say this as someone that's been profiled as a white person in a black country, and held at gunpoint by assault-rifle wielding police.

> Why are some areas of towns so racially segregated?

In NYC's own words:

https://a816-dohbesp.nyc.gov/IndicatorPublic/beta/data-stori...

The problem with that methodology is that it doesn't only hide race. Criminals tend to be repeated offenders. It only proves that the police couldn't recognize them.

They got paid to enforce the law.

It's important to note that common sense itself is a bias. It's a bias towards what we already think. Don't mistake that as trying to say that for "we should avoid priors" - that's not possible. What would it take for you to soften (not move) your priors on the common sense that "police aren't co-creators in crime through the act of observation"?

Are you arguing that more police does help against crime, but does not raise the amount of crime discovered by the police?

How is it a poor argument? If a crime is committed but there's no relevant reporter to observe it, it won't get counted as "crime", correct?

I think it also depends on your definition of crime or the type of crime that is being committed as outside of the scope of truly violent crimes or crime where there is a clear victim there are a lot of crimes that really should not be a crime in the first place (like possession of certain plants) and there may be many instances of truly victimless crimes that would go "uncaught" and thus untracked if an authority figure was not observing them.

Common sense is a poor argument.

"Crime" has multiple elements: the law, the act, the observation, and the prosecution.

We disproportionately outlaw things that poor people do. ("The law, in its majestic equality, forbids rich and poor alike from sleeping under bridges.") We disproportionately observe poor people, as shown by this article. We disproportionately convict poor people, who can't afford bail or good legal representation.

Yet people always want to jump to that other element. The disproportionate crime rates can't be due to all those other things, it must be due to them committing more crimes.

The only thing we can say for sure is that poor people commit different crimes. They use different drugs and in different settings; they are more likely to commit theft than embezzlement; they are more likely to cause property damage than violate building codes. The rest is made murky by an unfair legal system and a societal tendency to treat crimes of poverty as individual in nature and crimes of wealth as systemic failures.

Not in every case.

https://www.livescience.com/59329-drug-alcohol-addiction-wea...

> By age 22, lifetime rates of addiction to drugs or alcohol were 11 to 16 percent among women from affluent families, which is similar to the national norms; the rates were 19 to 27 percent among men from affluent families, twice the national norms, according to the study.

> But a more troubling trend emerged by age 26: Lifetime rates of addiction to drugs or alcohol were 19 to 24 percent among women from wealthier upbringings and 23 to 40 percent among men from those families. These rates were three times higher than the national average for women, and two times higher in men.

This does't match up with drug arrest rates.

Murder is just one crime. Take for example Wage Theft. Which accounts for as much as $50 billion a year ^1 (more than robberies).

And yet, you won’t see cops arresting a manager at a store for stealing from their workers.

[1](https://inthesetimes.com/article/wage-theft-union-labor-bide...)

Interestingly, you are correct, but you omitted the rest of the paragraph where they articulate why, emphasizing that across booth residential and commercial neighborhoods there is a striking difference in policing.

> We note that we report racial and socioeconomic disparities without attempting to control for other covariates for two reasons. First, if New York City residents of different races face different levels of police deployments, that disparate impact is itself important; it can also bias algorithms trained on downstream policing data irrespective of the true causal mechanism. Second, controlling for other factors in policing data can be difficult to interpret, introducing concerns about omitted variable bias and model misspecification which make it difficult to identify which factor is truly the “cause” of higher police deployments. For the sake of transparency and simplicity, therefore, we report results by stratifying each variable separately, noting that these disparities are themselves important but that multiple causal mechanisms may underlie them.

Misled implies a degree of intent, the wording is ambiguous but I don't assume ill intent by the author, especially when they later spend a paragraph explaining why they did not slice by covariates.

Let’s also quote why

“We report racial and socioeconomic disparities without attempting to control for other covariates. First, if New York City residents of different races face different levels of police deployments, that disparate impact is itself important; it can also bias algorithms trained on downstream policing data irrespective of the true causal mechanism. Second, controlling for other factors in policing data can be difficult to interpret, introducing concerns about omitted variable bias and model misspecification which make it difficult to identify which factor is truly the “cause” of higher police deployments. For the sake of transparency and simplicity, therefore, we report results by stratifying each variable separately, noting that these disparities are themselves important but that multiple causal mechanisms may underlie them.”

Hah, just linked to the Minneapolis counterpart study above. Yes, this is an elegant natural data experiment, thanks for sharing this version that touches a larger dataset.

I'd be surprised if it was crowd sourced. Even in fairly rural areas the prices of all stations are regularly updated. Crowd sourcing things like traffic data works as even a few people driving down a road can passively report data. It seems unlikely that manually entering gas data that changes daily would be sufficient to achieve such good coverage.

> Before describing the details of the framework, the high-level intuition is that we are reweighting the Nexar data sample, which is sampled from a non-representative set of locations, so that it matches the locations where different demographic groups actually live. For example, to calculate the police deployment levels that Asian residents of New York City experience, we reweight the original data sample to upsample neighborhoods with larger Asian populations.

> Overall, our estimation procedure compensates for two types of potential bias. Equation 2 compensates for a data bias, reweighting the Nexar dataset (which is sampled from a set of locations which does not necessarily match the population distribution; Figure 1) to match the population distribution of demographic subgroups. This is conceptually similar to inverse propensity weighting procedures [4] which are used to compensate for non-representative data in other settings. Equation 3 compensates for imperfect model performance, and allows us to check that model performance is unbiased (i.e., calibrated) across demographic subgroups.

Section 4.1 goes into the mathematical functions they use to address the data set.

Section 3.2 describes the data set and how it is geographically distributed.

> Data was provided to us by Nexar in two phases. Phase 1 consists of 3,987,835 images sampled prior to September 1 2020, and is extremely geographically and temporally skewed. Geographically, it is concentrated within the boroughs of Manhattan and Brooklyn, and does not contain data from the boroughs of Staten Island, Queens, and the Bronx at all; temporally, it overrepresents data from Thursday nights. Phase 2, which constitutes the majority of the dataset, consists of 20,816,019 images sampled after October 4 2020, and is much more geographically and temporally representative: it is sampled at all times of the day, on all days of the week, and also covers the entire geographic area of New York City.

> Because Phase 2 is much more representative than Phase 1, we conduct our primary analysis of disparities using only data from Phase 2. We additionally conduct numerous validations and bias corrections, described in §4.1, to compensate for non-representative sampling in the dataset. Geographic and temporal coverage during the Phase 2 period is very good. Specifically, 100% of hours during the Phase 2 period are covered; 99.6% of Census Block Groups (CBGs)3 have at least one image, with a mean of 168.2 images per CBG; 88% of roads contained within the borders of New York City are covered by at least one image, using data from OSMNX [6]. Figure 1 summarizes geographic data availability; Figure S1 summarizes temporal data availability.

Do you disagree with the statement?

So, it's research into a topic you don't think should be researched?

Do you have any problems with the methodology or conclusions? Like, science can still be science even if it's pursued in the name of social justice. If they are cooking the books, then that's one thing, but if they are making a hypothesis, gathering data, and evaluating that hypothesis fairly using real data, then it's just science.

I didn't say there was nothing wrong with crime in SF. I just shared a link showing (just as your does) that its crime rate is significantly lower than a lot of cities.

For some reason a certain segment of the national media and populace is always talking about the San Francisco crime rate but seemingly never covers stories about crime in Phoenix, Charlotte, Denver, Dallas, or Indianapolis.