http://www.usatoday.com/story/news/nation/2014/01/01/chicago-homicide-rate-drop/4281539/

Chicago has had one of its lowest killings in decades.

Does anyone know of any sociological studies looking into the "why" of this?

I remember reading a while back on a white supremacist website that the civil rights era was the cause of the increase. It seems like rates are now back to pre-civil rights era levels, so I guess we can't blame the blacks anymore, and we have more Hispanics than we can shake sticks at.

Have we reached a level of standard of living where even poor people aren't desperate enough to kill? Are we sufficiently placated by Facebook and vines? Is Christianity finally working? It atheism finally working?

Or have the criminals got us trained well to roll over if we're being robbed that they don't need to murder?

I wish there were an easy way to get a broader view of what's happening.

1/1/2014 7:27:12 PM

gun nuts will need a new example to use

1/1/2014 7:29:01 PM

Looks like big drops came during Clinton, and small drops during Obama, for a political spin.

1/1/2014 9:00:03 PM

I'm assuming those percentages are per 100k people?

The murder rate in NY has also dropped quit a bit. I wonder if it has more to do with the Mayor and Governor than the president?

[Edited on January 1, 2014 at 10:29 PM. Reason : adfs]

1/1/2014 10:27:45 PM

Pb

1/1/2014 10:33:31 PM

People aren't killing each other for crack anymore like they were in the 70s and 80s

1/1/2014 10:37:55 PM

while you guys are armchair analyzing, don't forget lag

rates are going to lag behind whatever caused them, so you need to look at whatever cause some amount of time ago.

(and Pb is a great example of this, the correlation matches only after you adjust for lag)

1/1/2014 10:39:02 PM

Isn't that what everybody does in TSB, armchair analyze? Except for maybe you, because you are just so much smarter than the rest of us.

[Edited on January 1, 2014 at 10:43 PM. Reason : dsaf]

1/1/2014 10:43:25 PM

What is Pb?

And I am not too sure about the lag thing, depending on what the cause is. Clinton famously pushed more law enforcement funding, which would filter through more quickly.

If the cause was better education though, that could have a multi-year lag.

And I doubt it's governor-lead, since the effect seems to be nationwide.

The first spike was in the 20s and the second spike in the 60s, both peaked at similar levels, could be interesting correlations there.

1/1/2014 10:46:42 PM

http://www.motherjones.com/environment/2013/01/lead-crime-link-gasoline

1/1/2014 10:53:25 PM

Interesting read.

1/1/2014 11:13:14 PM

Since the trend is worldwide, its probably not Clinton

(of course its multivariate, but Clinton is probably not the significant factor)

1/1/2014 11:14:18 PM

The lead thing is a very interesting hypothesis.

I would love to see some correlation coefficients of atmospheric lead and other variables to crime rates. Taking that article at face value though, I would have to give the credit to lead (although the drop off towards the end needs splainin').

1/1/2014 11:21:45 PM

There is a ton more scholarly work on the worldwide lead link, that article is just a quick intro

1/1/2014 11:23:55 PM

i think the lead thing is very compelling (and I thought that correlation was pretty well known?). There could be other factors converging,too (the ol' Freakonomics example regarding abortion, etc).

1/2/2014 12:32:35 AM

It's possible it's well known, but of the correlation is as strong as beig suggested by that article, it's worth being a major policy initiative.

1/2/2014 12:49:16 AM

i think it's a combination of many factors, but lead was the biggest driver for a while

1/2/2014 9:25:10 AM

I tweeted that article and it's gotten 5 retweets already (the most ever for anything I've tweeted...). Looks like it's not THAT well known.

1/2/2014 9:35:55 AM

I know a lot of people have read it, but Freakonomics has a very compelling argument to explain the drop in crime. It offends a lot of people, but it makes sense to me.

1/2/2014 1:01:17 PM

it would be pretty straight forward to calculate the correlation, assuming decent data, of how much abortion or lead affects crime.

1/2/2014 2:56:42 PM

i wouldn't ever call multivariate statistics "pretty straight forward", its certainly not to most people.

Here are some studies:

The Relationship between Lead and Crime*
http://hsb.sagepub.com/content/45/2/214.short

Understanding international crime trends: The legacy of preschool lead exposure
http://www.sciencedirect.com/science/article/pii/S0013935107000503

Association of Prenatal and Childhood Blood Lead Concentrations with Criminal Arrests in Early Adulthood
http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.0050101

Bone lead levels in adjudicated delinquents: A case control study
http://www.sciencedirect.com/science/article/pii/S0892036202002696

The Relationship Between Lead Exposure and Homicide
http://archpedi.jamanetwork.com/article.aspx?articleid=190628

The urban rise and fall of air lead (Pb) and the latent surge and retreat of societal violence
http://www.sciencedirect.com/science/article/pii/S0160412012000566


[Edited on January 2, 2014 at 3:05 PM. Reason : .]

1/2/2014 3:01:59 PM

(i work around a lot of analytics nerds)

1/2/2014 3:10:17 PM

The suggestion I heard was that it was the legalization of abortion which drive down the criminality rate.

1/3/2014 11:59:18 AM

Fertility rates were dropping anyway. Even if abortion had never been legalized, the vast majority of the trend that ^ references would still exist.

Between 1960 to 1975, fertility rates fell from about 3.6 to 1.8. It's hard to understate just how insane that shift was. That's 1 and 1/2 fewer kids per family.

That trend was also a slow decline. Roe v. Wade was a 1973 decision. Even if there were a sudden legal change that resulted in more abortions, fertility rates themselves don't reflect it. The human condition is more complicated than what the law says. Japan is a better example of the effect of abortion. They legalized it earlier, and they saw a faster decline in fertility rates right around that time.

Abortion isn't alone among family planning tools either. Birth control was introduced earlier. The options available to the average person for family planning slowly and steadily improved over time. This is more realistic and more consistent with the fertility rates trend.

Decline in crime was 1980 or 1990 to 2010 roughly. It could still be continuing now. Property crime exhibits similar trends, but the window is a little different for all of them. There probably is a generational link between the two things, but it's fuzzy. It's also not super duper convincing since crime rates only declined by a factor of 2 or so. With the changes in development patterns and work patterns, we probably spend half the time outside than what we used to. So a factor of 2 isn't as convincing as it might initially sound. The shift exists, but there is an abundance of causal candidates that you can try to connect it with.

That said, I do want to read this book:

http://en.wikipedia.org/wiki/The_Better_Angels_of_Our_Nature



Violence has always been declining, and I think we might be hasty to connect the recent trend to something specific. I frankly don't believe the numbers from the 1950s. They were probably under-reported. Society didn't operate the same back then. I bet you'd find even lower "crime rates" if you looked at the year 1800. Lots of people were substance farmers. Those folks didn't handle conflicts like we do today. The rise in reported crime is probably more of a reflection of our changing community structure. Government has replaced many roles that used to be provided by family. Go read about the modern Amish rape cases if you want to know what relying on family for these functions is realistically like. We've been slowly crawling out of that agricultural way of life since the Sumerians.

But the incidence rates of violence has been declining faster in modern times. The problem with explaining that is that so many things follow the "hockey stick" curve of modern society. All metrics moved at once, but the most straightforward explanation for this is population. When the number of hands increases, the number of goods produced does the same. But this leaves something unexplained.

Population density is probably more important, and it has a more direct reflection on way of life. Increasing density leads to both the reporting in the first place, as well as the ultimate decline as it literally pacifies people. That's the deeper explanation.

You should be suspicious of statistics that just happen to terminate on one end of the graph, and that's what we see here. Every google search you do will return the same general window. That's because it's the only data that's worth a crap.

1/3/2014 3:19:50 PM