News for hackers. Stuff that matters.
Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by ves on Tuesday May 03, 2016 @09:07PM (#11625475)
How is this even remotely good science? So far as I can tell, they didn't admit the possibility that this can be either complete noise (if you have, say 100 diseases, and you're looking for time-series correlations, or performing tests with too low an alpha or power) or caused by a confounder.

The task of making causal claims is a difficult one, and not one to be made lightly. The literature is replete with examples of this going terribly wrong, even with perceived physical justifications. See

[0]: tylervigen.com

[1]: http://nerdsonwallstreet.typepad.com/my_weblog/files/datamin...

[2]: http://www.huffingtonpost.com/2013/04/16/reinhart-rogoff-aus...

[3]: http://www.economist.com/node/5246700

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by tinalumfoil on Tuesday May 03, 2016 @09:55PM (#11625627)
From the paper:

> Our second identification strategy, introduced by Troesken (2006) and Clay et al. (2014), exploits the fact that more lead particulate will leach into water with low pH levels. If lead exposure has a causal effect on homicide, more acidic water should be correlated with higher homicide rates only in cities that used lead water pipes.

So there's more causal evidence than appears in the Mother Jones article. The two references for their "identification strategy" don't seem particularly well-cited though. [1][2]

[1] http://www.mitpressjournals.org/doi/abs/10.1162/REST_a_00396

[2] https://mitpress.mit.edu/books/great-lead-water-pipe-disaste...

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by josu on Tuesday May 03, 2016 @09:26PM (#11625530)
The article is bad (MotherJones), the paper may be better [1]. I've only skimmed it, but it looks at the correlation between distance from a lead factory and homicide rate, which probably increases the signal/noise ratio.

[1] http://scholar.harvard.edu/files/jfeigenbaum/files/feigenbau...

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by ves on Tuesday May 03, 2016 @09:44PM (#11625592)
Do you not admit the confounder that proximity to a lead factory should also correlate with socioeconomic status?

I don't have time to check out the article this week, but will soon.

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by Steko on Wednesday May 04, 2016 @01:00AM (#11626297)
3 Which cities used lead pipes?

An important concern in a cross-sectional study such as ours is that the cities that installed lead pipes in the nineteenth century might have differed from the cities that did not in unobserved ways that were correlated with their homicide rates. Based on available data on the characteristics of the cities in our sample, we find little evidence that the cities that used lead pipes differed in observable ways from the cities that did not. Although larger cities, denser cities, and cities with comparatively low rates of home ownership were more likely to have used lead pipes, so were better-educated cities. High-homicide southern cities, meanwhile, were less likely to have used lead pipes. When we predict whether cities used lead pipes using all covariates, the only difference that remains statistically significant is the city population.

Using the Integrated Public Use Microdata Series (IPUMS) sample of the 1900 census, we calculate several city-level covariates that could be correlated with both lead pipe use and crime.17 These include each city’s population, population density, home ownership rate, and literacy rate, the share of each city’s population composed of African Americans, foreign-born residents, and single men aged 18 to 40, and the share of each city’s employed population working in manufacturing. We also create covariates for cities’ latitude and longitude...

> I don't have time to check out the article this week, but will soon.

Maybe reading first before accusing them of bad science is a good idea?

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by peejaybee on Tuesday May 03, 2016 @10:16PM (#11625701)
"Third, a city-level analysis such as ours does not suffer from a common problem with individual-level studies, namely that individuals exposed to lead typically come from poor neighborhoods with both low-quality housing and high crime rates. Common confounders of the lead-crime relationship in individual-level studies pose less of a problem in our city-level study, where in most cases the entire city population was exposed to lead through water."

Paper's here: http://scholar.harvard.edu/files/jfeigenbaum/files/feigenbau...

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by maxerickson on Tuesday May 03, 2016 @10:09PM (#11625680)
They use rail distance to lead refineries as an instrument variable. They briefly discuss direct poisoning from a refinery.

The real kicker is that lead pipes combined with acidic water have a higher effect than lead pipes combined with basic water.

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by sdenton4 on Tuesday May 03, 2016 @09:33PM (#11625560)
The real targets here are the crime rates, and the diseases known to be side effects of lead poisoning. And the question is, can we use high level aggregates to observe lead poisoning caused by industrial design decisions?

We've seen massive drops on the level of violent crime in the US (and beyond) over the last twenty or thirty years. And the attribution of that drop has huge ramifications. Should we double down on three-strikes laws, or can the drop be attributed to environmental factors like ambient lead levels due to leader gasoline?

What I see here is increased evidence for the lead poisoning hypothesis. We know that lead poisoning causes violent behavior; the evidence is indicating that elevated crime rates may be due to undiagnosed poisoning.

Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by Retric on Tuesday May 03, 2016 @10:21PM (#11625717)
A lot of crimes in prisons goes unreported, so total crime stats are harder to calculate. EX: Murder seems to clearly be down, male on male rape may be up.
Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by lostlogin on Wednesday May 04, 2016 @03:19AM (#11626756)
A lot of non prison crime goes unreported too - the classic unreported crime is rape. Here it suggests 68% is unreported. https://rainn.org/get-information/statistics/reporting-rates
Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by MacsHeadroom on Wednesday May 04, 2016 @08:31AM (#11627828)
Female on male rape is up too, outside of prisons of course. Federal statistics in the past 4 years place unreported female on male rape about as high as all male on female rape. They still call it "made to penetrate" and classify it as 'mere' sexual assault though.
Re:Lead Water Pipes in 1900 Caused Higher Crime Rates in 1920? by cuchoi on Tuesday May 03, 2016 @09:52PM (#11625619)
Did you read the paper? (I am asking sincerely)