?

Log in

No account? Create an account
Lakeshore
An author of no particular popularity

Jay Lake
Date: 2007-04-30 05:48
Subject: [culture] Risk and probability: the school bus problem
Security: Public
Tags:culture, guns, personal, politics
Lots of great stuff in comments on the risk post from this weekened. kadath identified the issue of risk transfer as the free rider problem. I'm not sure I agree with her, or if it is the free rider problem, it's a special case of it. Nonetheless, that link is a fascinating read.

There's a problem related to risk transfer which lies in how we conduct risk assessment. Untrained humans are notoriously bad at inuiting the probability of events. If we had a decent grasp of probability, Las Vegas would be a grubby mining town in the middle of the desert. (This, incidentally, is another argument against Intelligent Design/Creationism, if there were any point in arguing logically with those viewpoints -- like the knee, the spine and the reproductive system, our wired-in grasp of mathematics and probablity leaves a lot to be desired.)

I had a friend years ago in Texas who at the time was the father of two grade-school aged kids. G insisted on driving his kids to school every day, because as he said to me, "Do you have any idea what kind of people drive school busses?" He was convinced his children were far safer in his Toyota than in a school bus. This despite the fact that school busses have an incredibly low death rate per passenger mile. Insofar as I can tell, most school bus deaths happen when the bus falls off of or into something (a bridge, a quarry, a lake).

(School bus design is also a fantastic example of the risk transfer discussed in the earlier post. In the case of school busses, the risk transfer is highly deliberate as the result of a social decision making process that values protection of the children on board more highly than the protection of any third party involved in a school bus accident.)

I tried to point out to him that a random vehicle running a stoplight had a far better chance of injuring or killing his kids than pretty much anything that could happen to a school bus in Austin traffic, regardless of how incompetent the driver. G said, "Yes, but I'd be able to avoid the accident."

People evaluate risk in terms of their perceived ability to control the situation. G's fears for his kids when his hands weren't on the steering wheel trumped any logic or statistics. This is also why so many people are afraid of flying. Airplanes are far safer than automobiles, but as a passenger, we're utterly passive. We can't even back seat drive an airliner.

Again, I find the school bus problem, and risk/probability assessment in general, fascinating for its own sake, but this kind of thinking also bears on everything from gun control to healthcare spending to tax policy. It's profoundly illogical and profoundly human to think this way -- we all do it. The fact that statistics are so malleable and so poorly understood by the vast majority of people only reinforces the tendency, because we can cherrypick data to suit our prejudices, whatever they may be.

So what would a society be like where everyone had a firm grasp of risk and probability? Should the mathematics and sociology of the school bus problem be taught early and often in the schools? I dunno, but there's plenty of food for thought here.
Post A Comment | 35 Comments | | Flag | Link






User: dirkcjelli
Date: 2007-04-30 13:15 (UTC)
Subject: (no subject)
When it comes to the education one requires to be a functioning member of society in a liberal representative democracy, we have more fish to fry than that...
Reply | Thread | Link



Brian Dolton
User: tchernabyelo
Date: 2007-04-30 13:17 (UTC)
Subject: (no subject)
"This, incidentally, is another argument against Intelligent Design/Creationism, if there were any point in arguing logically with those viewpoints -- like the knee, the spine and the reproductive system, our wired-in grasp of mathematics and probablity leaves a lot to be desired."

I think you must have missed the latest theory - the "Not Very Intelligent Design" approach...
Reply | Thread | Link



(no subject) - (Anonymous)
User: ellameena
Date: 2007-04-30 15:06 (UTC)
Subject: (no subject)
Actually, one other wrinkle to the vaccination situation is that vaccinated children are not 100% safe. So when vaccination rates decrease, a certain number of children whose parents took the risk are still going to get sick.
Reply | Parent | Thread | Link



(no subject) - (Anonymous) Expand
(no subject) - (Anonymous) Expand
(no subject) - (Anonymous) Expand
(no subject) - (Anonymous) Expand
User: ellameena
Date: 2007-04-30 15:05 (UTC)
Subject: (no subject)
You should read a book called _Why People Believe Weird Things_. The phenomenon you're describing is called superstitious preconditioning, which is basically believing in a cause-and-effect relationship which defies the laws of statistics. It causes people to believe they are "on a roll" and also that it always rains when they wash their car, etc. Intellectuals, atheists, and others are not immune to this effect. It's probably a side effect of the ability our brains have to learn from experience. I rather suspect, actually, that the smartr you are, the more vulnerable you are to this effect, because your brain is very good at making connection. You need training in order to sort the information out and correctly identify the connections you make as "theories."

Animals do it, too. My male cat caught a mouse once, and he took it into the bath tub to play with it. Later, our female cat entered the room and found a mouse in the bathtub! Years later, she still believes that mice come from the bathtub, and she hops in there and peers into the drain to check on a regular basis.

What is surprising, then, when you think about it is not our lack of comprehension of statistics, but the fact that we ever comprehend them at all.

By the way, as an example of so-called "smart people" believing something that is statistically untrue--how many people do you know who are afraid of "filling up" all of our landfill space? If you do the math, we have enough space on earth to throw garbage away until the sun goes nova, but most educated people believe that we are "running out of space" for land fills. (There are other reasons to reduce consumption, but lack of landfill space is not one of them.)
Reply | Thread | Link



Jay Lake
User: jaylake
Date: 2007-04-30 16:03 (UTC)
Subject: (no subject)
Yeah. I catch myself doing this sort of thing all the time, for example, some of the quirks in how I personally handle postal mail. It makes me laugh at myself, but it also makes me question myself. This is some *deep* wiring in our brains.

Thanks for the book reference!

(BTW, as to landfill, that's a very similar problem as the "stand on Zanzibar" effect, about running out of living space. The issue quickly elaborates with nuance, such as "living space near available water and arable land" or "landfill with appropriate soil and geology at accessible distances from urban areas." But that's all engineering -- completely agree with you about the effect.)
Reply | Parent | Thread | Link | Expand



In a heaven of people only some want to fly
User: chipmunk_planet
Date: 2007-04-30 15:18 (UTC)
Subject: (no subject)
I drive my kids to school 1) because otherwise they'd have to go out in freezing weather a half hour to an hour earlier, 2) I don't like them standing out on a street corner unsupervised for a half hour, 3) I know where they are, and 4) the drive to and from school is personal time with my children. I don't feel like that's so nutty.
Reply | Thread | Link



Jay Lake
User: jaylake
Date: 2007-04-30 15:53 (UTC)
Subject: (no subject)
That's not the least bit nutty. Those are excellent reasons for driving your kids to school.

G's stated reason was road safety, which is nutty.

For the record, Mother of the Child drives the_child to and from school everyday, for pretty much your same reasons, and she drives as Scion xB, which would be toast if it got hit by so much as a minivan. Neither she nor I claim this is a safety decision, however.
Reply | Parent | Thread | Link



bodandra
User: bodandra
Date: 2007-04-30 15:44 (UTC)
Subject: Busing vs. parent driving
Drivers of school buses are trained 1. in how to drive safely; 2. handle children; and 3. remain upbeat for the kids.

I rode the bus to school from 7th grade until I graduated.
That was a good thing, because I was so shy that I'd never talk until one of the other students engaged me in conversation.
I still keep in touch with a few of them now ... smiles.
My mom told me when (our) bus driver died ... of old age, of course.

Reply | Thread | Link



User: (Anonymous)
Date: 2007-04-30 15:45 (UTC)
Subject: (no subject)
The classic example of this is railways. Statistically you are at least an order of magnitude safer on a train than in a car. Yet if you ask people why they do not ride trains, one of the most common excuses is that trains are "too dangerous". This is the case even in the UK where train ridership is much more common.

There are two main reasons for these perceptions. The first, as you note, is that people will always think that they, personally, will be able to avoid car accidents because they are better drivers. In taking public transit they are trusting the driving to someone else, which is something they are not prepared to do. The other is that car accidents are not news. Thousands of people die every year on the roads. Mostly you never hear about these accidents. But if one person is killed in a train crash then it is in the papers every day for the next week.
Reply | Thread | Link



Kevin Standlee: Wig Wag
User: kevin_standlee
Date: 2007-04-30 18:56 (UTC)
Subject: (no subject)
Keyword:Wig Wag
What she said. I know I've talked to people who are convinced that trains are terribly dangerous, but that's because train crashes are news precisely because they are unusual.

There was a high-profile derailment with loss of life in the UK a couple months ago. It had been a couple of years since the previous passenger fatality in a train crash. In the intervening time, thousands of people died in automobile accidents on Britain's roadways, but that's not news, so people think roads are safe and trains and planes are dangerous.
Reply | Parent | Thread | Link



russ: quo vadis
User: goulo
Date: 2007-04-30 15:53 (UTC)
Subject: (no subject)
Keyword:quo vadis
Your friend G not only puts his kids in more direct danger from traffic accidents; by taking them in a separate vehicle, he causes more pollution, which causes more health problems for everyone, not just his kids who are supposedly "benefitting" from riding in the car.
Reply | Thread | Link



dragonkal
User: dragonkal
Date: 2007-04-30 15:56 (UTC)
Subject: (no subject)
I don't yet have anything of value to contribute to this discussion, but wanted to thank you for the excellent subject. I'd never thought of risk transfer before, and now I can't stop pondering it. What a fascinating and often unconsidered aspect of our societal structure!
Reply | Thread | Link



User: dsgood
Date: 2007-04-30 16:35 (UTC)
Subject: (no subject)
People who know the odds in theory don't seem to act much differently. Futurologists don't seem to be thinner than people outside their field. So far as I know, they don't vote differently than their backgrounds etc. would predict. And as far as I know, they aren't spectacularly good at investing.
Reply | Thread | Link



bemused_leftist
User: bemused_leftist
Date: 2007-05-01 17:39 (UTC)
Subject: (no subject)
Thinner? Otherwise very good points.

But still ... how good is the data the futurologists have? That's an area of speculation at best, and stacked data can be profitable to the stackers. An example less like reading tea grounds would be helpful.
Reply | Parent | Thread | Link



lordofallfools
User: lordofallfools
Date: 2007-04-30 16:41 (UTC)
Subject: (no subject)
I don't think that teaching statistics and probability would do much to help. Any skillful surveyor can skew statistics to read what they want them to read. There should, in fact, be an inherent (but benign) mistrust in anyone who quotes statistics. :-)

But teaching (or helping students acquire) and exercising critical thinking and logic would be helpful. This is, I maintain, why literary studies in middle and high school are so important: students learn to evaluate and critique information that can have multiple 'correct answers' and to logically support their conclusions.

Literary critique and analysis are essential, IMHO, to building a healthy, thoughtful society.
Reply | Thread | Link



Jay Lake
User: jaylake
Date: 2007-04-30 16:50 (UTC)
Subject: (no subject)
Man, that is one of the best statements I've ever heard about studying literature.

:: doffs cap ::
Reply | Parent | Thread | Link



User: ladylike4
Date: 2007-04-30 16:50 (UTC)
Subject: (no subject)
Wow! I have to say I question the parents who take there kids to school in the district I drive for. The parents park like maniacs and basically never use the crosswalks (where there are people posted to stop traffic). They are just idiots and I think to myself they are teaching there kids these crazy behaviors and we wonder what is going on and where kids get there quirks.

Of course, not all the parents are this way. I do say a majority though. We drivers talk about someone is going to get killed very frequently.

Jennifer (www.ladylike4.com)
Reply | Thread | Link



User: deangc
Date: 2007-04-30 19:34 (UTC)
Subject: (no subject)
Jay sez: The fact that statistics are so malleable and so poorly understood by the vast majority of people only reinforces the tendency, because we can cherrypick data to suit our prejudices, whatever they may be.

I would argue that statistics are malleable ONLY because they are so poorly understood. If more people understood, people wouldn't be able to get away with the widespread abuse that they do.

Properly used, statistics are as ironclad as any other branch of mathematics. But even some of the really smart people that I know, including a Ph D. in chemistry, have trouble with statistics, and it's no wonder that what is a good and staggeringly useful branch of mathematics gets a bum rap.
Reply | Thread | Link



User: deangc
Date: 2007-04-30 19:37 (UTC)
Subject: (no subject)
So what would a society be like where everyone had a firm grasp of risk and probability?

The lotteries would all go bust, for starters.
Reply | Thread | Link



browse
my journal
links
January 2014
2012 appearances