Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Earth

Exposure To Air Pollution is Linked To an Increase in Violent Crime (economist.com) 85

Breathing dirty air is linked to aggressive behavior, according to a new paper by Jesse Burkhardt and his colleagues at Colorado State University and the University of Minnesota. Using crime data from the Federal Bureau of Investigation and air-pollution data from the Environmental Protection Agency, the authors analyse the link between air pollution and violent crime in 397 American counties between 2006 and 2013. From a report: They find that a 10% increase in same-day exposure to PM2.5 (particulate matter less than 2.5 microns in diameter) is associated with a 0.14% increase in violent crimes, such as assault. An equivalent increase in exposure to ozone, an air pollutant, is associated with a 0.3% jump in such crimes. Pollution levels can easily rise by much more than that. Last November, owing to wildfires, PM2.5 levels in San Francisco rose seven times higher than average. Correlation is not causation of course (there may, for example, be a third variable affecting both pollution and crime) and the authors are cautious not to speculate about the precise mechanism by which contaminated air might lead to more rapes or robberies. This is not the first time researchers have identified a relationship between pollution and crime. In the 1970s America banned lead-based paint and began phasing out leaded petrol; two decades later, crime fell. Many researchers now argue that the two developments were linked. In a paper published in 2007, Jessica Wolpaw Reyes, an economist at Amherst College, estimated that the drop in lead exposure experienced by American children in the 1970s and 1980s may explain over half of the decline in violent crime in the 1990s.
This discussion has been archived. No new comments can be posted.

Exposure To Air Pollution is Linked To an Increase in Violent Crime

Comments Filter:
  • by 0xdeaddead ( 797696 ) on Friday October 11, 2019 @11:15AM (#59296328) Homepage Journal

    Oh wait.

    • Re: (Score:3, Insightful)

      by drinkypoo ( 153816 )

      Ask the million Uighurs in concentration camps if China is violent.

      Or ask a Tibetan, if you can find one.

      • Re: (Score:2, Interesting)

        So the Chinese government is locking up the Uighers because of air pollution? I've heard some loony environmental theories before but this one takes the cake.
        • You still haven't heard that one. I was speaking to whether China was violent, not about whether air pollution was the cause. It can't be helping, though.

          • You still haven't heard that one. I was speaking to whether China was violent, not about whether air pollution was the cause. It can't be helping, though.

            So wait, correct me if i'm wrong, you are saying that an ethnic minority in China is locked up because they are violent?

            I thought they were locked up because they were Uighurs.

            • So wait, correct me if i'm wrong, you are saying that an ethnic minority in China is locked up because they are violent?

              Because China is violent. I didn't say the Uighurs were especially violent.

              I thought they were locked up because they were Uighurs.

              Their ideology is abhorrent to TPTB in China, and they are racially disadvantaged so the government can easily demonize them, gaining the support of the population. Anyone different is a target.

    • Re: (Score:2, Troll)

      by cayenne8 ( 626475 )
      Actually I was about to ask if that means that Chicago and Baltimore are a couple of our two most polluted cities in the US?
      • Depending on how any underlying mechanism behind this correlation works, it might not matter. If we assumed a simple linear relationship (which is very unlikely, but simple) then doubling the pollution would only see a ~1% increase in crime. So even if pollution is responsible, it's probably not even the dominant factor let alone one of the top among many. That isn't to say we shouldn't study it more, because certain types of pollution may have far more significant affects and reducing lead exposure is beli
    • Ask anyone who isn't a Party member.

    • by hey! ( 33014 ) on Friday October 11, 2019 @11:45AM (#59296492) Homepage Journal

      You can't compare across cultures and governments. Chinese citizens live in a surveillance state. In a way, Americans do too, but it's the private sector doing it and it doesn't care what you do, only what you'll buy.

      The most robust retrospective comparison you can make is between the same places on low-pollution days vs. on high-pollution days, which IIRC is what the study did: correlate crimes to same-day pollution, further controlling for confounding weather variables like temperature. This will eliminate the confounding variables of comparing different kinds of places.

    • Therefore China is the most violent place on earth

      It's difficult to discern considering that China keeps information about law enforcement entirely secret. Remember, in China people get disappeared and they have an active organ harvesting program. It's entirely possible that they do have a big problem with violent crime.

    • The real question is what sort of moron trusts any of the statistics coming out of China at all. Crime statistics in countries without a vested central interest in downplaying metrics are barely trustworthy as it is.

  • Since the paper was published in an Elsevier-owned journal, I can't read it. That's an awfully small percentage, can anyone with institutional access read the paper and share the P-values with us?

    • by nomadic ( 141991 )

      The second link leads to a newspaper article that has graphs from the paper which show the 95% confidence interval for the data.

  • by Kiaser Zohsay ( 20134 ) on Friday October 11, 2019 @11:19AM (#59296358)

    But correlation is also not coincidence. It is entirely possible that both of these exposures come from living in a large city.

    • by txmason ( 882110 )
      That was exactly my thought.

      https://tylervigen.com/page?pa... [tylervigen.com]
    • Yup. Of course they can test this: put a bunch of people in a room and pipe some diesel exhaust in there. Let's see how long it takes before they are at each other's throats.
    • by pr0t0 ( 216378 )

      One would hope that would have occured to the researchers and that, at a minimum, cities of similar size and density were compared when factoring violence vs. air pollution.

      • by Shotgun ( 30919 )

        But, comparing a Los Angeles, Detroit, or San Francisco to a well run city will also lead to such correlations. The problem isn't correlations, though. The problem is the medias penchant for reporting stories about single factor regression studies.

        As you indicated, but did not state explicitly, the validity of the study rests entirely on what factors they controlled for.

    • Poverty is probably the linking factor, even within the same city, rich people will live in better filtered houses, drive newer cars (and be in neighbourhoods with newer cars, less pollution, etc.

    • But correlation is also not coincidence.

      Really? So spurious correlations is not a thing?

      https://twentytwowords.com/fun... [twentytwowords.com]

    • Of course. Always makes me facepalm when I read this sort of bullshit.

      You know that piezzoelectric discharges are linked to lung cancer? people exposed to over 20 piezzoelectric discarges a day less than 15 centimeters from their face, have about 90% higher chance of dying from lung cancer than people who average 1 such discharge per month.

      Interestingly, this trend only happened over past 2-3 decades, before that it was ferrocerium exposure. But as pliezzo lighters got more popular, this became the dominant

    • Maybe it's exposure to Democrat shitholes that causes violence. They also have the most dense air pollution.
  • > . Last November, owing to wildfires, PM2.5 levels in San Francisco rose seven times higher than average. Correlation is not causation of course (there may, for example, be a third variable affecting both pollution and crime) and the authors are cautious not to speculate about the precise mechanism by which contaminated air might lead to more rapes or robberies. Well talk about a non story lol.. man. I'm sure there gonna be ton of stupid media outlets that will take this and completely misrepresent it
  • How about "densely populated area with low social equality" as the common cause for both?

    Do I get a Nobel Prize now?

    • Write a paper and you may just get one. They have awarded many to ideas with far less merit. Make sure there's plenty of 2d and 3d charts or an easy-to-read abstract. (The committee members are often not too bright.)
      Of course, it is the least Nobel of the prizes.

    • by nomadic ( 141991 )

      You don't think they accounted for confounding factors? Why not?

      • by doom ( 14564 )

        No, it always requires Some Guy on the Internet to explain to researchers that they should think about economic factors. Clearly they always forget about them, or else we wouldn't need a few hundred people to bring this point up in the comments, over and over again.

    • by skids ( 119237 )

      How would population density increase and social equity decrease in line with "same-day exposure" to pollution (which is what they say they measured)? Those variables generally don't vary much on a short timescale.

  • They find that a 10% increase in same-day exposure to PM2.5 (particulate matter less than 2.5 microns in diameter) is associated with a 0.14% increase in violent crimes, such as assault.

    That means instead of 10,000 "violent crimes" a county will see 10,014 "violent crimes" committed...

    Even if the researchers can factor out every other social, economic, racial, and historical factor than can influence behavior, that is an AMAZINGLY SMALL increase - well below any reasonable margin of error I'd expect to see in any professional study.

    • by nomadic ( 141991 )

      There's a difference between effect size and margin of error.

    • by jwhyche ( 6192 )

      That is so far in the margin of error that I have to seriously doubt someones math skills. No respectful scientist would ever both to publish a paper like this that kind of margin. I literally means that a very small number of samples has to be wrong to throw the whole study off.

  • So, what are the chances that the "tough on crime" groups will add pollution control to their platforms?

    • Probably not much considering the effect size is small. There's also a question of diminishing returns as you reduce pollution.

      But let's look for intelligent ways to reduce crime (or pollution) for its own sake, instead of trying to attach unrelated goals as a primary driver. This never works out well in practice and only pisses off people who catch on to what you're doing. In same cases it makes them resentful towards the initial cause, which is part of why I suspect that many conservatives are opposed
      • by AmiMoJo ( 196126 )

        This study is useful because it creates an additional public health reason to fix pollution. It gives people an interest in cleaning up the environment because it protects them from crime (although the effect will be long delayed).

        It also helps debunk some really nasty "group X is genetically more violent" by providing yet more evidence that it's environmental.

        • by Shotgun ( 30919 )

          As the ggp pointed out, the correlation rates fall into the noise. If you use it to push for removing whatever you consider pollution or to change a strongly correlated stereotype, it will cause people to tune you out and add you to the "making shit up" category.

        • It also helps debunk some really nasty "group X is genetically more violent" by providing yet more evidence that it's environmental.

          The effect sizes here are so small it's not terribly useful for anything like that. Fortunately studies controlling for other factors have already shown that poverty and single-parent households explain almost the entirety of any gap in black-white crime differences. Or more succinctly, poor people without fathers tend to commit crime disproportionately regardless of race. It just happens that this occurs more often in the African American population in urban areas than other populations.

  • Correlation is not causation... Yep that's the one.
  • Sooo... (Score:4, Informative)

    by the_skywise ( 189793 ) on Friday October 11, 2019 @11:32AM (#59296422)
    If I stand directly by my tail pipe and breathe in the fumes - I'll turn evil?
  • Maybe reducing the violent crime is what lowered the lead exposure.

  • In The Cloud Minders, there was a population that lived in a city in the sky that devoted their lives to art, philosophy, music and leisure; a city of intellectuals, and a population of miners on the ground that spent their brutish lives mining "zenite", and were known for their violence and a perception of their stupidity that were discriminated against by the Stratos city dwellers, but it all stemmed from an invisible gas emitted by the zenite.

    https://memory-alpha.fandom.co... [fandom.com]

  • "Correlation is not causation of course (there may, for example, be a third variable affecting both pollution and crime) and the authors are cautious not to speculate about the precise mechanism by which contaminated air might lead to more rapes or robberies. "

    Hmmm, dirty polluted air and there is more likely to be a dense population of working class poor, living on a lower incomes with probably attendent high violent crime rates caused by a lack of opportunities for social mobility. Can I have a government

  • Residential zones in close proximity to industrial zones are undesirable.

    Cost of living in these areas in therefore lower.

    Poor people have a greater incentive to be commit violent crime. (give me your money/stuff)

    As someone mentioned above correlation != causation.

  • by DaveV1.0 ( 203135 ) on Friday October 11, 2019 @11:59AM (#59296538) Journal
    Pollution is more common in urban areas because people are packed in close together.
    Violence is more common in urban areas because people are packed in close together
    • This doesn't fit with what they found. They found an increase in violent crime even given short term increases in air pollution in the same locations. Also, they controlled for a whole host of variables including population density. If there's any criticism here it should be in the other direction: when one controls for many variables, there's a risk that subtle issues in how one controls for them will create statistical artifacts.
    • How "closely packed" people are in urban areas doesn't seem like it would vary up and down by >10% on a daily basis.

  • by www.sorehands.com ( 142825 ) on Friday October 11, 2019 @12:13PM (#59296566) Homepage

    I have found that 100% of violent criminals have suffered from repeated exposure to dihydrogen monoxide. Therefore, it must be banned.

    Aliso Viejo almost banned it, but the dihydrogen monoxide cartel blocked it.

  • Wow such percentage increases
  • The poor live a life of strife and woes. They live in the cheapest dwellings available and those neighborhoods have low rents because they are stuck near the worst sources of pollution. After all, you can't find a trash dump or sewer plant in a rich neighborhood. You also don't tend to find rich housing next to railroads, industrial facilities, or busy highways. So just how can the cause be pinned down? Is it due to simply living in poverty or is it due to the high levels of pollution in their neighborh
  • Poor neighborhoods, which have more violent crime compared with more affluent neighborhoods, are also more polluted.

    Separately, what is the relation between particulates, ozone, and temperature. It's long proven the relationship between higher temperatures and higher violent crime. Perhaps particulates and/or ozone is just a proxy for temperature?

    I don't find 0.14% and 0.30% compelling numbers.

  • ... that lead pollution maps from leaded fuel matched crime rates almost exactly in the USA, and crime rated dropped with the introduction of unleaded fuel.

    I wonder if this is also lead, or unrelated.

  • Here's a Reddit dedicated to spurious correlations. Apparently you can pick any one of them and expect it to be your next Slashdot post.

    https://www.reddit.com/r/rando... [reddit.com]

  • Pollution is more common in urban areas because people are packed in close together. Violence is more common in urban areas because people are packed in close together NordVPN Cracked [itsaboutpc.com]
  • I find this mistake is often made even by experienced engineers. The fact that something correlates does not prove causality. People with problematic backgrounds often reside in polluted area's, simply because living there is cheaper/achievable. This proves exactly nothing.

  • Most crime is by petty criminals. Petty criminals live in the crappiest part of town. The crappiest part of town has the most air pollution. Correlation is not causation. Any further questions?

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Working...