I've said this for years, but when I say it, I'm the crazy guy who wants people to die in car accidents.

Anyway, that study, among others, is discussed in "THe Armchair Economist". ANd if you want to reduce car fatalities, then instead of seatbets, why not mandate that steering wheels be fitted with a giant spear aimed that driver's heart. I guarantee that you'll see accidents (and fatalities) down to near zero in no time.

I get the point about the silliness of legislating the "fine print of our lives".

However, I don't think I accept the premise that once I put on my seatbelt, I feel less vunerable and *therefore* are more likely to drive recklessly.

Couldn't you just as easily argue that the act of putting on a seatbelt reminds you of the inherent dangers of driving and therefore causes people to drive *more* safley.

(the link to the pdf didn't seem to work, so I don't have any real "facts" to base my comment on)

Also, I'm not sure that anyone is claiming seatbelts help save the lives of those *outside* of the cars in a automobile accidents, however, I know *I* would feel more comfortable being in a car accident with mine on, rather than off.

I *am* interested in reading the study you linked to because as I'm sure you know AUA, people can come up with statistics to prove anything, 14% of all people know that.

sorry, another thought...

it's also like saying...

"highway speed limits kill more people than they save because while a person is driving under the speed limit, they percieve they are safe and therefore become less careful about their driving"

Well you also have to determine if you're talking about the driver wearing seatbelts or passengers in the vehicle, particularly minors. Choosing to not wear your own seatbelt is a different dilemma than choosing to not ensure your kid has one on in the car.

The link to the pdf is fixed.

Sorry about that .pdf link. This should work for you:
http://www.geog.ucl.ac.uk/~jadams/PDFs/failure%20of%20seatbelt%20legislation.pdf

After reading it and re-reading it, nothing in it appears to be doubletalk or junk science. But only way to know is to look for yourself.

The fact is, mandatory seatbelt use hasn't lowered the incidence of road fatality. The theory for this paradox is Risk Compensation, but the fact is that it exists.

And of course, all of that just goes to support the notion that laws, even when they seem innocuous, can often be more dangerous to abide by than ignore.

But you're assuming a correlation between seatbelt use and highway accidents. It's not a matter of whether seatbelt use has reduced fatalities altogether, but reduced fatalities as a percentage of highway accidents. If accidents in general have gone up considerably over the last decade let's say, is that because of seatbelt use or just the fact there are more people driving now than 10 years ago? And if they have gone up, has the fatality rate gone up at the same level? There may just be a considerably larger amount of accidents.

there's no way to know whether it's the seatbelts or just something else that has happened historically. there might be something fundamentally different about the types of countries that have laws vs. don't. there's really just no way to know. i mean we've all heard the argument that legalized abortion caused the reduction in the crime rate, but how many people believe it? there are certainly many counter-theories.

Awesome.... that my prediction came true.

AUA - please come back to the blogosphere. You are the only one who can match the dark wit of RCR.

Also, doesn't forcing people to wear seatbelts only keep those bad drivers alive just a little bit longer so that they end up killing more people?

When you're discussing evidence on a well researched social issue, instead of cherry-picking single studies to back your pet beliefs, why not do what social scientists do: start looking at meta analyses of available research to find emerging consensus points around what's been discovered.

One on this issue:
http://ip.bmjjournals.com/cgi/content/full/6/2/82

Punchline: There's adequate evidence, across issues of safety legislation, to show that the dynamic of risk compensation is very real. Make things safer, and people will behave in more risky ways. The question is: will they do so to a degree that offsets entirely any safety advantages?

For seat belt laws, risk compensation does lower the effectiveness of seat belt regulations, and it does shift the risk burden somewhat, BUT not enough to overshadow their overall public health advantages.

Yes, people overestimate the safety advantages of seatbelt requirements. But the safety advantage is still there, and it's still considerable.

p.s. Speaking as someone who teaches graduate courses in research methods, let me suggest that you don't use rhetoric like "irrefutable evidence" when describing research around complex social patterns and behavior--it's nonsensical.

Ian - because this is teh internets and we dont have 5 years for AUA to give us his conclusive findings on why the seat belt law is dumb.

That healthy mass of tangential information does not refute Adams' claims. Did you read it, or was this a homework assignment in analyzing analyses? The lecture does not address the veracity of Adams' conclusions. This is particularly surprising, as you cherry-picked (to borrow a term) a lecture delivered by a member of an organization called "Highway Safety North." Despite this, it merely said that Adams had critics . . . as do we all.

It drawing erroneous correlations between bicycle helmets and safety belts (which are addressed as equivalent, in that analysis, from the perspective of Risk Compensation.) But then, most bicycles don't weigh two tons and travel at speeds in excess of 70 miles per hour.

You seem pretty adamant about seat belt safety laws. I'm glad they're there, so you'll know to put one on.

Excuse the grammatical error at the start of the second paragraph. I was striving to clean out all my vitriol (you'll notice I was only about half finished doing so) when my low battery warning went on. Didn't want to lose it all; hit Create Post.

And thanks, Phil. It's hard to write a strong and compelling dissertation when #1: one doesn't really care, and #2: it's done on a coffee break.

So AUA, you're alternating between "I don't care that much" and self-reported problems of too much vitriol in your posts? All that anger with all that apathy? Interesting. But, a tangent, so moving on...

I chose that lecture--a presented paper actually--because it was widely accessible and was a well cited review s of the literature and debates on the topic. As with any meta analysis, the point is not that it's original research countering any one paper, but that it weighs the available research against one another and provides cites for you to explore and evaluate the judgments.

If you require a roadmap, to dig deeper into the arguments around Adams' and related arguments, a good place to start would be the numerous cites listed under the section "Risk Compensation in 2000".

Again, the recurring theme in the literature is that risk compensation to required seat belt use occurs, but not enough to offset the safety gains.

Debating the topic isn't of great importance to me, as the facts are pretty clear. Patronizing, self-aggrandizing arm-wavers like yourself, on the other hand, tend to get my hackles up.

You clearly have strong opinions about what the data indicates. If your view was researched with the same effort you poured into misunderstanding even *this* simple exchange, then I think your remarks and references deserve even less consideration than I've already given.

AUA, *shrug* you can justify your lack of response however you like.

This isn't an obscure issue; most folks doing social research (myself included) follow the lit on risk compensation because it plays out in all sorts of human endeavors. Ok, let me put it this way: I'd be very surprised if my colleagues in any social psychology, political science, sociology, public health, or economics department weren't familiar with it.

And so you know, the area of seat belts is one of the places where the evidence is strongest that people don't entirely compensate for reduced risk: i.e. of all the issues to pick to argue against the "nanny state" with the rationale that citizens will just compensate and no benefit will be achieved, seat belts remains one of the weakest and refuted-by-research arguments.

Moreover, you'll notice that in all sorts of areas where this comes up: public health campaigns, etc., the consensus is still to do work around increasing public safety. Why? Because, even in areas where risk compensation is a more powerful counter-acting force than in the seat belt example, risk compensation only rarely actually overshadows the aggregate benefits.

And if you were curious, probably the "best"--saddest, actually--example of risk compensation over-compensating for improved safety is around AIDS treatment and prevention in developing countries, which is why we've needed more public education campaigns around the risks of AIDS to counter-act the risk compensating behavior.

Yes, risk compensation is a reason to knock people down who get a bit overly enthusiastic about safety regulations; 'Look, you're injecting this into a social system that will compensate, and it won't have the degree of advantages you're expecting,' is something that needs to get said more often. But it's not (very often) a refutation against such measures. And it's certainly not--looking at the breadth of research on the topic--a refutation against seat belt requirements.

Anyhow, you said you're done with this. I've said my peace. And while this has been nasty, I suspect folks unfamiliar with the arguments have had new stuff to mull over on the issue. Cheers.

I like it when the sun is really really hot and you get your shorts wearing ass into the car and accidentally sit on the metal buckle of a seat belt that's been exposed to direct sunlight and burn the back of your leg. That's the danger I'm talking about.