Lead and Violent Crime

I may be made of as much as 50 percent cheese.

(Ooh... new page. I could quote what I'm responding to, but may have as much as a 100 percent chance of not doing so.)

SpacePPoliceman wrote:

Which article, what data led to that conclusion, and what data do you have to counter that conclusion?

This one, beats me, and none because I'm not the one arguing for or against it. This was just the bit that seemed to stick in KG's craw as unsupported by current trends in crime rates by age.

Well, then, if that turns out to be the case, I look forward to his answers.

Stengah wrote:
SpacePPoliceman wrote:

Which article, what data led to that conclusion, and what data do you have to counter that conclusion?

This one, beats me, and none because I'm not the one arguing for or against it. This was just the bit that seemed to stick in KG's craw as unsupported by current trends in crime rates by age.

It's also unsupported by his reference, which says that elevated blood levels show a pattern as a significant factor in up to that amount of certain types of violent crime, but that same paper also plainly leaves room for other significant factors as well. I'd call that bad reporting more than anything else, and I think that's been said before both in this thread and in other articles linked on the subject.

Jonman, I stand corrected.

Stengah wrote:
Hypatian wrote:

Er. Did anyone ever claim that? I'm pretty sure we'd all agree that claims of the form "Y can only ever be caused by X" can almost always be rejected without further examination.

One of the articles claimed "Gasoline lead may explain as much as 90 percent of the rise and fall of violent crime over the past half century."

I thought I already debunked that claim as an instance of a journalist not quite getting things right.

Just to be perfectly clear, none of the studies linked from the Mother Jones article ever said or concluded that leaded gasoline explained or may have explained 90 percent of the violent crime.

OG_slinger wrote:
KingGorilla wrote:

OG this is what the article says, and this is what I am taking issue with:

Gasoline lead may explain as much as 90 percent of the rise and fall of violent crime over the past half century.

It says to me that gasoline lead is the predominant culprit.

I am not arguing the adverse effects of lead on development. I am not disagreeing that there is no safe amount of lead. I am stating that the claim that atmospheric lead is to violent crime what atmospheric CO2 is to global temperature.

This might be a case of when a journalist quite report the science right.

The actual study (page 32) said that the R2, the coefficient of determination, between lead (23-year lag) and all violent crime was .9. The reporter might have misinterpreted that as saying lead was responsible for 90% of violent crime when, instead, it just meant that the regression line fit the observed data well (really well).

Reyes' study was the only one to make claims on an actual percentage decrease in violent crime. She said the data implied that a 56% decrease in violent crime between 1992 and 2002 could be attributed to the lowering of lead levels.

OG_slinger wrote:
Stengah wrote:
Hypatian wrote:

Er. Did anyone ever claim that? I'm pretty sure we'd all agree that claims of the form "Y can only ever be caused by X" can almost always be rejected without further examination.

One of the articles claimed "Gasoline lead may explain as much as 90 percent of the rise and fall of violent crime over the past half century."

I thought I already debunked that claim as an instance of a journalist not quite getting things right.

Just to be perfectly clear, none of the studies linked from the Mother Jones article ever said or concluded that leaded gasoline explained or may have explained 90 percent of the violent crime.

Indeed you did, well, you suggested it was probably a case of bad journalism, so maybe debunked is too strong a word, but still, you agreed with KG that the 90% claim isn't supported by anything in the actual studies. My response was to Malor's post that KG was fighting the whole thing, and not just specific parts. Which is, you know, why I quoted that when I made my response.

Here is Steven Novella's take, a clinical neurologist, most famous for the Skeptic Guide podcast.

http://theness.com/neurologicablog/i...

He doesn't really dispute the Mother Jones article, especially after Kevin Drum reduced the 90% with a 50% figure of the rise and fall of crime correlation.

goman wrote:

Here is Steven Novella's take, a clinical neurologist, most famous for the Skeptic Guide podcast.

http://theness.com/neurologicablog/i...

He doesn't really dispute the Mother Jones article, especially after Kevin Drum reduced the 90% with a 50% figure of the rise and fall of crime correlation.

50% correlation with crime rates "above normal" is much easier to accept than 90% causation. Novella agrees with the basis of what the Mother Jones article says (there's a connection between lead and crime rates) but the conclusion Novella makes is a fair bit different than the one the MJ author makes:

If we accept the 20% figure (crime that is lead related), which seems plausible, then this indicates a significant role for lead, but lead is certainly not the only important factor.

Which is saying that even if lead is a factor in 20% of all crime, it's A factor, not the the factor. I think the biggest issue with the MJ article is that it's worded in a way that indicates that lead is sole factor for that % of crime, or the rise in crime rates above "it's natural level." Even in the correction to the original, the author says

It's true that one researcher has suggested that lead can explain 90 percent of the rise and fall of crime, but that's very much the high end of the estimates in the field. I'm a lot more comfortable with an estimate of around 50 percent, something I should have made clearer in the text of my piece. In other words, lead probably explains a very big chunk of the rise and fall of postwar crime in America, but it doesn't trump everything else. Drugs, poverty, urban gang warfare, education, policing tactics, and other things also play a role.

The way I read this, he's saying lead is responsible for ~50% of the crime rate above "natural" and all those other factors are responsible for the other ~50%. If he were to say that lead is one of multiple factors in 50% of the crime rate above natural, then he'd find much less opposition.

Stengah wrote:
goman wrote:

Here is Steven Novella's take, a clinical neurologist, most famous for the Skeptic Guide podcast.

http://theness.com/neurologicablog/i...

He doesn't really dispute the Mother Jones article, especially after Kevin Drum reduced the 90% with a 50% figure of the rise and fall of crime correlation.

50% correlation with crime rates "above normal" is much easier to accept than 90% causation. Novella agrees with the basis of what the Mother Jones article says (there's a connection between lead and crime rates) but the conclusion Novella makes is a fair bit different than the one the MJ author makes:

If we accept the 20% figure (crime that is lead related), which seems plausible, then this indicates a significant role for lead, but lead is certainly not the only important factor.

Which is saying that even if lead is a factor in 20% of all crime, it's A factor, not the the factor. I think the biggest issue with the MJ article is that it's worded in a way that indicates that lead is sole factor for that % of crime, or the rise in crime rates above "it's natural level." Even in the correction to the original, the author says

It's true that one researcher has suggested that lead can explain 90 percent of the rise and fall of crime, but that's very much the high end of the estimates in the field. I'm a lot more comfortable with an estimate of around 50 percent, something I should have made clearer in the text of my piece. In other words, lead probably explains a very big chunk of the rise and fall of postwar crime in America, but it doesn't trump everything else. Drugs, poverty, urban gang warfare, education, policing tactics, and other things also play a role.

The way I read this, he's saying lead is responsible for ~50% of the crime rate above "natural" and all those other factors are responsible for the other ~50%. If he were to say that lead is one of multiple factors in 50% of the crime rate above natural, then he'd find much less opposition.

Now it just seems like Drum's making up sh*t. He had absolutely no research backing his claim that 90% of the violent crime of the last century could be explained by lead so now he's only claiming 50%.

While there is research backing a similar number--the Reyes study found that the reduction of lead was responsible for a 56% reduction in violent crime--it only covered the period between 1992 and 2002, not the entire century.

The study Novella brings in simply cloudies the picture. For one thing, it's behind a pay wall, so there's no way to read it. But the study (apparently) talks about all crime, which is not what both studies cited in the MJ article focused on. They focused on lead and violent crime. So we have a rather large case of trying to compare apples to oranges.

Either way, Drum's not doing himself any favors by not admitting he f*cked up the 90% thing. I don't give a care what percentage he's comfortable with because his personal opinion doesn't matter. There's actual research out there with actual percentage of crime reduction attributable to the reduction of lead and he should simply point to that instead of pontificating about everything else that might have contributed.

It also doesn't help that his wishy-washy correction largely invalidates the first section of his article, which essentially said that the usual suspects of drugs, poverty, abortions, police tactics, etc. were all bullsh*t and didn't adequately explain the drop in violent crime.