September 25, 2012

"More Americans now commit suicide than are killed in car crashes as miserable economy takes its toll.”

That's the headline asserting causation based on correlation and... there actually isn't even correlation.

Quite apart from that, if car crashes decrease and suicides increase, maybe all that is really happening is that the car crash method of suicide is becoming less popular. That is, previously, some suicides were hidden within the car crash numbers. Suicide numbers will go up if the method used for a suicide makes it crushingly clear it wasn't an accident. I'll speculate that guns cause a high proportion of suicides in part for this reason.

29 comments:

Paul Zrimsek said...

Well, the last thing I remember, Doc, (strike)she started to swerve(/strike)I took those 30 Nembutals.

Amexpat said...

Well the correlation can be that people wanting to commit suicide can't afford to do it by car because of the bad economy.

TMink said...

Actually, guns never kill themselves. They have a suicide rate of 0.

As do SUVs.

Trey

Widmerpool said...

Of a piece with this. The narrative must be supported!

Fun with Stats!

Tarzan said...

Here's the money quote from the ensuing discussion on Volokh:

"As to the link between suicide and unemployment, keep in mind is works in two directions. Those that have passed away are no longer be counted as looking for work, although they may still vote in Chicago."

Dust Bunny Queen said...

More people commit suicide when the economy sucks, the lose their homes, jobs, their posessions and they lose hope of ever being able to climb out of the financial hell that they are in.

It has nothing to do with cars. I blame Obama and the Democrat controlled Congress.

traditionalguy said...

IMO hopelessness is a disease that kills. False hopes in delusions that man is the measure of all things is the entry point for the disease.

The antidote to the humanism disease is still Christianity.

edutcher said...

Back when the networks would breathlessly report the number of highway deaths over various holidays, I remember some comic doing a riff on that saying, "Highway accidents are down 40% this Labor Day. You're not tryin-n-ng".

Paul Zrimsek said...

Well, the last thing I remember, Doc, (strike)she started to swerve(/strike)I took those 30 Nembutals.

W-o-on't come back from Dead Man's Curve!

Ambrose said...

Burying the lede. The real news is that technology has made cars much safer.

edutcher said...

PS DBQ is right about economic conditions and suicide. Back during the Depression, Billie Holliday recorded a song so sad it became known as, "The Suicide Song", because it was more than some people could stand.

Colonel Angus said...

Interestingly enough, the guy who has presided over this horrendous economy seems to be a shoe in for re-election if the polls are even remotely accurate.

When he said the American people are easily confused, he may be more right than he knows. They evidently have a death wish as well.

Colonel Angus said...

He being Bill Clinton.

Nonapod said...

I can't seem to find it on Volokh or Political Animal, but regarding the increase from 2000 to 2009 of 10.44 to 11.77, what do those numbers mean? I assume they don't mean 11.77 deaths out of every 100 are due to suicides?

dbp said...

Technology has made cars safer, true. But also, in a down economy people drive a lot less: Less disposable income means fewer pleasure trips, fewer jobs mean fewer people commuting to work.

Other than that, the chart over at Volokh does show a lot of correlation beteen unemployment and suicide.

Texan99 said...

It's obviously the kale.

Christopher in MA said...

PS DBQ is right about economic conditions and suicide. Back during the Depression, Billie Holliday recorded a song so sad it became known as, "The Suicide Song", because it was more than some people could stand.

"Gloomy Sunday." It's a Hungarian song. The composer killed himself.

karrde said...

@Nonapod,

those numbers are likely the rates-per-100K-population.

I think that the surprising news is the decrease in deaths by automobile accidents that enable this comparison.

If I read the numbers correctly (from the CDC's WISQARS tool), I get a decline from ~42000 deaths per year in auto accidents to ~36000 per year.

The ~42000 deaths/year rate was fairly flat from 2000 to 2007, then began to trend downwards in 2008.

Current data is only available up to 2010.

I blame the advance of safety technology in the autmotive world. (Remember, the average car on the road has safety technology that was cutting-edge a decade ago.)

Larry J said...

While I don't have a source, I remember hearing highway death tolls of about 50,000 a year back in the 1960s and even before. The death toll is much lower today despite about twice as many cars on the road and more miles are being driven. You have to remember that cars weren't required to have seat belts until sometime in the 1960s. Today's cars have as standard equipment not only seat belts but airbags, crumple zones, anti-lock brakes and many other safety features. Some of those ideas were being tested in the 1960s but it took decades of development to make them commonplace and affordable.

Smilin' Jack said...

Suicide numbers will go up if the method used for a suicide makes it crushingly clear it wasn't an accident. I'll speculate that guns cause a high proportion of suicides in part for this reason.

Can someone explain how that statement makes any sense at all?

Kirby Olson said...

Christopher Walken's role in which he looks over at Woody Allen and talks about how all it would take is to veer into the other lane... Is that in Annie Hall?

Peter said...

It's hard to stop counting the errors and omissions in the linked article.

For example, the "suicides are terribly undercounted" theme of the first paragraph seems to be intended to imply that unemployment-related suicide rates might be even worse, yet it leaves open the possibility that the correlation is due to changes in the method chosen in attempted suicides.

The second paragraph says, "there does seem to be a correlation between the suicide rate and the unemployment rate," yet it's too much trouble to actually calculate the correlation? After all, there are two variables, so were talking about one number. Even though one can see that the local peaks in the unemployment rate don't seem to be accompanied by matching local peaks in the suicide rate (as one would expect if there were a high correlation between the two).

But it's largest weakness is that it doesn't even consider confounding factors-- of which there are an all but unlimited number.

For example, problem gamblers have a high suicide rate, and legal gambling has become much more available in most of the USA since 1981 yet they don't look for a correlation between the two.

kentuckyliz said...
This comment has been removed by the author.
Sigivald said...

Dust Bunny: Look at the charts here (at Volokh) and you'll see that the suicide rate has precious little correlation to the economy.

One might think that a bad economy "should" cause people to kill themselves, but it doesn't appear to do so to any significant extent.

Or at very least, some other factor(s) are greatly overwhelming it in the data.

kentuckyliz said...

^ I tried some strikethrough coding methods and they didn't work.

The only HTML allowed is a for links, bold, and italics. Spam prevention donchaknow.

Not easy to find in Blogger's help FAQ but I found it.

Paul said...

Ann,

This suicide thing is kind of like mass murderers. Yea they use guns NOW, but 50 years ago they just set fire to whole buildings while the exits were blocked..

Now yes there IS more suicides today than car accidents. A) the population is larger, B) gas prices are HIGHER, C) yes the economy is worse and that does affect suicide rates (as well as murder/suicide rates) but I bet some car deaths ARE suicides.

So HOPE for sone CHANGE in 2012 (and it looks like the Republicans will keep the house and MAYBE take the senate, regardless of what happens to 'the one'.

Robert Cook said...

"More people commit suicide when the economy sucks, the lose their homes, jobs, their posessions and they lose hope of ever being able to climb out of the financial hell that they are in.

"It has nothing to do with cars. I blame Obama and the Democrat controlled Congress."


You're leaving out the Republicans in Congress, George Bush, Bill Clinton, George Bush, and Ronald Reagan.

In short, our current dire economy is the outcome of decades of intentional actions meant to benefit the wealthy elites, of policy decisions agreed to by both Republicans and Democrats.

There are many reasons to deplore Obama, (claiming a right to murder, acting on that claimed right, protecting torturers, prosecuting whistleblowers, etc.), but Obama is to blame for our current economy only to the extent that he has picked up the tail and is helping drag the carcass along with those who came before him to preserve the appearance of life still present in the dead dog.

MadisonMan said...

Corvairs for everyone!

Dust Bunny Queen said...

Corvairs for everyone!

Pintos!.

Althouse: "Suicide numbers will go up if the method used for a suicide makes it crushingly clear it wasn't an accident. I'll speculate that guns cause a high proportion of suicides in part for this reason."

Can someone explain how that statement makes any sense at all?

You can accidentally take an overdose of pills. Slip and accidentally fall off a cliff or down stairs or maybe off of a bridge.

It is hard to accidentally put a gun barrel in your mouth and pull the trigger. It is not easy to accidentally put a gun directly to your temple and pull the trigger.

Not all gun deaths are the result of deliberate suicide. Accidents can happen, it is just easier to determine that it wasn't an accident in the gun related cases than it is in an overdose of drugs and alcohol.

MadisonMan said...

Pintos

What's that movie that has a car just barely tapping a pinto's read and the whole thing explodes. Makes me cackle everytime.

(Goes to look on youtube)

Ah, yes. Top Secret. That backwards talking scene makes me laugh everytime too.