[Read Part 1: Risk | Part 2: Risky Business | Part 3: Risk & Reality | Part 4: Risk & Politics | Risk, Reality, & Bullsh-t ]

My dad was a gambler. He was an unusual gambler, because he won (sometimes quite a lot); in part this was because he was a brilliant mathematician – he’d been a cryptographer in Army Intelligence in Hawaii and India during World War II – and because he understood that the places to win lots of money were the places where the betting public…he had a less polite word for them…mispriced the real odds. A horse might really be a 25 – 1 shot, but if the parimutual odds on it were 500 – 1 and the conditions were right, it could definitely be a worthwhile bet. A series of those bets would certainly result in a profitable season.

What that gave me was an appreciation for the smell of cigar smoke, an eye for judging things in terms of risks and odds, and the sense that public perception of odds and costs might not really reflect the true risks involved.

Here, in this blog, we’re talking about the possible emotional mistreatment of children by teachers who use them to act out their political agendas…at the expense of the children’s well-being.

I wrote about this recently, and asked if it was a ‘problem’ or an ‘incident’; there’s an important difference here that I should elaborate on.

Bad things happen. My local paper today has an article about a young woman here in L.A. who was abducted by a slimeball who promised her a shot at celebrity and instead murdered her and dumped her body in the Hollywood Hills. This is a horrible event…but is it a ‘problem’ in our community? A ‘problem’ to me is something that materially affects the community as a whole. In reality they are just two points on a spectrum; there is no magic threshold where ‘incidents’ become a ‘problem’. But we treat them differently; we externalize the costs of problems while we leave individuals to deal with the consequences of incidents.

Another way of looking at it would be to ask how safe is your car? In reality, it could be made much safer…we could, for example, use racing-type fuel cells essentially rubber bladders inside of reinforced gas tanks), which would effectively eliminate fires as a consequence of accidents. But in reality, these cells are very expensive, and relatively few people are injured or die in automobile fires. So how much are we all prepared to spend – to bet, in essence – in order to get what kind of payoff – how many lives saved?

Sometimes the bet (cost) is financial – adding race-car quality fuel cells to cars would cost hundreds or even thousands of dollars to the cost of a car. Sometimes it is behavioral – getting people to drive slower. Each one has a payoff (increased safety) and a cost (financial or behavioral).

In an ideal world, we’d carefully look at the payoff and the cost of our choices, and do what good managers do – pick the high-payoff/low-cost actions to do first, and work our way down from there.

We seldom do.
We don’t for two basic reasons; because we misperceive the true odds – we think we are more likely to win (or lose) than we really are, and we don’t look at the total cost of the choices we make, because we try hard to externalize the costs – to get other people to pay on our behalf.

We misperceive the true odds for a variety of reasons. One of the ones that I tend to focus on is that we misperceive things because the dramatic nature of the event, and how it interacts with preexisting ways in which we view the world, and both reinforces what we already believe and serves as a factual argument to support our beliefs.

So let’s take a case and explore it a bit.

Over at Armed Liberal, I raised the point that gun-control opponents typically leverage children’s injuries and deaths from firearms as a justification for stricter regulation of guns.

My response was that this was, to be generous, disingenuous as swimming pools are far more hazardous to children than firearms are, and we don’t see an aggressive effort to ban or restrict the ownership of swimming pools.

So I went to the CDC data. Take a look at the tables below; in the ‘accidental’ category, drownings represent 10 times the deaths caused by firearms (868 vs 85); when we include homicide and suicide, drownings still represent more than twice the deaths (881 vs 410). After adding homicides and suicides, firearms deaths (410) represent less than 20% of the deaths caused by car crashes (2,220).

Here are some numbers, from the CDC data for the year 2000. These are summaries that I did from their tables of the causes death for children 1-4, 5-9, and 10-14. Here are the Top 10 overall causes of death:

All Deaths	12,392	100.00%
Unintentional Injury	4805	38.78%
All Others	2,459	19.84%
Malignant Neoplasms	1,434	11.57%
Congenital Anomalies	894	7.21%
Homicide	727	5.87%
Heart Disease	452	3.65%
Suicide	307	2.48%
Influenza & Pneumonia	190	1.53%
Chronic Low. Respiratory Disease	190	1.53%
Septicemia	162	1.31%

Drilling down into the ‘Unintentional Injury’ category gives us this table:

MV Traffic	2210	45.99%
Drowning	868	18.06%
Fire/burn	564	11.74%
Suffocation	268	5.58%
“Pedestrian, Other”	166	3.45%
Other Land Transport	144	3.00%
Firearm	85	1.77%
Poisoning	77	1.60%
Fall	73	1.52%
Natural/ Environment	67	1.39%
Struck by or Against	64	1.33%
Other Transport	52	1.08%
Unspecified	50	1.04%
Other Spec., classifiable	38	0.79%
Pedal cyclist, Other	32	0.67%
Other Spec., NEC	24	0.50%
Machinery	18	0.37%
Cut/pierce	5	0.10%
TOTAL	4805

This obviously doesn’t include all the children who die from firearms; the homicide and suicide classifications obviously include deaths that would change these numbers. Taking the homicide and suicide causes of death, and mapping the clear overlaps (firearms, drowning, suffocation, etc.), we get a final table that looks like this:

MV Traffic	2220	40.50%
Drowning	881	16.07%
Fire/burn	601	10.96%
Suffocation	494	9.01%
Firearm	410	7.48%
Pedestrian, Other	166	3.03%
Other Land Transport	144	2.63%
Poisoning	98	1.79%
Fall	73	1.33%
Natural/ Environment	67	1.22%
Struck by or Against	64	1.17%
Other Transport	52	0.95%
Unspecified	50	0.91%
Cut/pierce	50	0.91%
Other Spec., classifiable”	38	0.69%
Pedal cyclist, Other	32	0.58%
Other Spec., NEC	24	0.44%
Machinery	18	0.33%
TOTAL	5482

Look, each of these deaths is itself an unspeakable tragedy. As a father, it is unimaginably painful for me to look at these tables and recognize that each number could represent one of my sons.

But if our goal is to try and reduce the overall number of these tragedies, and we have a finite amount of effort, political capital, and money doesn’t it make sense to spend the effort, political capital, and money where it will have a greater effect?

In other words, how do we make good bets?

How much has Canada spent on their failing gun registry? How many lives did it save, and how many would have been saved had those dollars been spent elsewhere…for example on giving car seats away, on drivers training for young mothers, on subsidized loans for swimming pool fences, or on water-safety classes in kindergartens and elementary schools?

And a part of it becomes one of setting priorities – of making good bets.

We do this in part by trying to agree, as a society when something is an ‘incident’ – even if tragic – and when it becomes a ‘problem’ – an issue where there are enough related incidents that by changing something we get a positive return on our effort.

We need to be conscious of getting a return, because we don’t live in a world in infinite effort or resources; we have to choose the places to spend attention and effort.

That’s the danger, as I see it, of misperception of risk. We spend what limited resources and attention we have on low-impact, high-cost solutions where we could have been solving high-impact low-cost problems.

And part of what we do is to place unrealistically low perceived risks on things that we control, or where we disproportionately benefit, and outrageously high perceived risks on things where we have little control or gain little direct benefit.

Having a swimming pool is probably significantly more dangerous that having a gun; more than twice as many children die from drowning as from firearms, and fewer people have pools than guns. But first, those who have pools benefit from them, and (often tragically) misperceive the risks involved. And secondly, the dramatic image of a child shot and killed or shot and wounded is so great – even though the real agony of a parent who loses a child found quietly drowned in a back-yard pool is certainly no less.

And for those who have a strong set of beliefs about the world – that guns are bad – those dramatic images drive your perception of the world, regardless of the true level of problem or the cost of dealing with it.

If you already believe that guns are bad, then every incident feeds your perception and is used to strengthen your argument that they are a ‘problem’. The young woman was murdered by a male sexual predator; if you believe that all males are sexual predators, and that all relations between men and women are defined by this – here’s evidence.

And the same thing is true if you believe that liberals are bad; every incident of bad behavior by a liberal is amplified into a groundswell of abusive teachers.

Why is this a problem?

Because on one hand, we live in a world where incredible anxiety about objectively low risks changes our public policies and gets us to spend limited attention, energy and wealth on solutions.

And on the other, we ignore, or lack the will to spend the resources on objectively more serious risks that we undervalue and so we remain exposed to them.

Conversely, recent studies have been done among gay and bisexual men, American heterosexuals and women in Africa which show that they dramatically underestimate their risk of STD’s and AIDS – meaning that the behavioral changes that would make it less likely that they would contract or spread HIV aren’t on their plate, because they don’t see any reason to make those changes.

“At the XIV International AIDS Conference in Barcelona, Dr. Ron Valdiserri, deputy director of the U.S. Centers for Disease Control and Prevention (CDC) programs, released alarming data on the occurrence of unrecognized HIV infection and misperception of risk among young African American “men who have sex with men” (MSM) – the center’s technical designation for gay and bisexual men.

This latest CDC data revealed that nine out of ten HIV positive African Americans who participated in a study of young gay men were unaware of their HIV status.

Dr. Valdiserri also released data that indicates that rates of HIV infection have stabilized in the U.S. in recent years. Meanwhile, the widespread use of antiretroviral therapy in the U.S. has dramatically reduced the number of AIDS cases and deaths in adults and children since 1996. Since 1998, the reported AIDS cases have remained at approximately 40,000 AIDS cases per year in the U.S.

And it’s not limited to gays:

New Survey Reveals Americans Underestimate Their Risk for Contracting Genital Herpes
April 26, 2000
Contact Tracey Adams
Director, Media Relations
(919) 361- 4811
(919) 361- 4815 (fax)

Research Triangle Park, NC — A new survey by the American Social Health Association (ASHA) shows that many adults in a high-risk category scored themselves as having no risk of contracting genital herpes, one of the nation’s most common sexually transmitted diseases (STDs).

In a Web-based survey of 1,414 individuals conducted for National STD Awareness Month, ASHA found that 46 percent of men and 39 percent of women believe they are not at risk for genital herpes despite the fact that they currently have multiple sex partners. This significant misperception of risk contrasts sharply with high knowledge scores on several questions, including risk factors.

“This is what we call ‘the herpes disconnect,” says Linda L. Alexander, PhD, FAAN, president of ASHA. “Three out of four people will tell you the ways that genital herpes is transmitted and will tell you that having multiple partners is a risk factor. But when you ask them about their own risk, all too many throw out the facts and minimize the risk.”

Or to the US:

In eight countries in which HIV/ AIDS has reached epidemic levels, with 5% or more of the population infected, a significant proportion of sexually active girls (aged 15-19) think they face no risk of contracting the disease, according to results of recent surveys.

In three countries – Haiti, Zambia and Zimbabwe – the misperception of risk was shared by 63%, 52% and 50% of girls, respectively, and by more than 40% of girls in three other countries surveyed.

So we drive SUV’s we choose for perceived safety (where the numbers are relatively clear there is no safety advantage) while talking on our cellphones (where the numbers are becoming clear that there are substantial risks from their use while driving and the evidence on other health effects is murky) to the home we bought far from powerlines (where there is no evidence of health effects) to microwave our dinners (which can generate greater EMF than living next door to powerlines).

It’s insane. We make bets each time we make a decision based on risk. We bet that the bank will be solvent, the car safe, the medicine will work without ill effect. But we’re choosing to make the sucker bets while leaving all the winning bets unmade. No wonder we’re going broke.

And beyond this, I’ll suggest that there is a deeper philosophical problem.

The flip side of risk is hope. Risk is the perception of negative consequence; hope the desire for the positive outcome.

We are driven by both, by a desire to avoid the bad and to grasp the good.

The world today suffers from a crisis of hope.

I wonder how deep it is, or if it is primarily in the educated media-spinning classes which makes it appear pervasive.

This is Part I
This is Part II
Part III is here
Parts IV > VI aren’t written yet.