"Wait, I thought people had a 'right to die'..."

"... so why is preventing suicide a reason for gun control? Oh, who am I kidding — at the NYT, anything is a good reason for gun control."

Since most of the gun deaths are suicides, you'd have to throw away your best statistics if you didn't count suicides. But then they never stop to wonder how many of those suicides were people who felt an uncontrollable murderous urge. What if Adam Lanza had only shot himself? We'd never know what horrible plan he'd left unfulfilled.

"There’s a new stigma in town: guns + mental illness = violence."

Says Scott Bryant-Comstock, president of the Children’s Mental Health Network, "And it should break the hearts of advocates nationwide."

Quoted in the Capital Times (a Wisconsin newspaper) under the headline "Some say efforts to boost mental health treatment to combat gun violence are misguided." Also quoted, Jeri Bonavia, director of the Wisconsin Anti-Violence Effort:

"We know that there are some problems with getting mental health records into the background check system, and I think that needs to be addressed," she says. "But it can’t be that we turn our attention just to mental health issues related to gun violence because people suffering from mental illness make up a very small percentage of the perpetrators of gun violence."....

A poll this week by Quinnipiac University shows that more than 90 percent of American voters support background checks for all gun buyers, which would close the so-called gun show loophole. And that's where Bonavia says Wisconsin should be focusing its effort. Her group is currently in the midst of a petition drive to urge Walker to propose background checks. 

So first Bonavia implies that we ought to make policy based on the percentages. But then she says, make a pervasive law that applies to everyone, without mentioning the very small percentage of perpetrators of gun violence within the truly vast category of Americans who buy guns. And by the way, the category "gun violence" lumps things together. Gun control has become a hot issue because of a few massacres. If you make a category out of the set of incidents that has inflamed present-day opinion, people suffering from mental illness seem to be 100% of the perpetrators! You only get your very small percentage if you throw in other types of incidents, such as gangsters wiping each other out. Wake me up when 90% of Americans want to do something about that. And explain to me how background checks have any curative power over that problem.

The appeal to statistics and reason falls flat when you shape it to suit the policy you already want.

"Only 4 to 5 percent of violent crimes are committed by people with mental illness," Dilip Jeste, the president of the [American Psychiatric Association], says in a statement. "About one quarter of all Americans have a mental disorder in any given year, and only a very small percentage of them will ever commit violent crimes."

See what I mean? Questions for Dr. Jeste: 1. What percentage of school shootings are committed by persons with mental illness? 2. If we cut the category "violent crimes" down to massacre-type shootings where the motive isn't robbery and the victim isn't someone with whom the shooter has a personal dispute, what percentage of those crimes are committed by persons with mental illness? 3. If we break the category "mental disorder" into subparts, so that depression and schizophrenia aren't lumped together, is there any category within which you cannot say that only very small percentage will ever commit violent crimes? 4. In your effort to shield the mentally ill from unnecessary stigma, are you giving cover to a set of persons who could and should be identified as dangerous? 5. What are the interests of the psychiatric profession that could affect whether you are giving truly honest answers to all of these questions, including this one?

"In some cases, [Nate] Silver tends to attribute successful reasoning to the use of Bayesian methods..."

"... without any evidence that those particular analyses were actually performed in Bayesian fashion."

For instance, he writes about Bob Voulgaris, a basketball gambler,

Bob’s money is on Bayes too. He does not literally apply Bayes’ theorem every time he makes a prediction. But his practice of testing statistical data in the context of hypotheses and beliefs derived from his basketball knowledge is very Bayesian, as is his comfort with accepting probabilistic answers to his questions.

But, judging from the description in the previous thirty pages, Voulgaris follows instinct, not fancy Bayesian math. Here, Silver seems to be using “Bayesian” not to mean the use of Bayes’s theorem but, rather, the general strategy of combining many different kinds of information.

Why are gun-death statistics inflated with gun-suicide numbers?

"I thought we had a right to die...."

Obviously, one reason is: to get bigger numbers. But I think the people that lump gun deaths together believe (or want others to believe) that guns are really dangerous. When it comes to suicide, there are 2 ways to think about the deadly effectiveness of guns: 1. For those who really want to kill themselves, guns are a sensible choice, or 2. The scary deadliness of a gun tempts weak/impulsive persons to go ahead and do something that wouldn't happen otherwise.

You can easily see that those 2 ways to think represent the mindsets that lead to libertarian or authoritarian answers to all sorts of questions. #1 would allow the individual to make his own decisions and to take care of himself, and #2 thinks the individual — call her Julia — needs to be helped and protected (even from herself).

Sorry to go all gender-y, but I'm interested in talking about suicide and attitudes about guns in the context of gender difference, because 4x as many men as women commit suicide and 56% of male suicides use firearms compared to only 30% of female suicides. Those statistics are skewed by the fact that guns are an effective method. It might be that the gender disproportion is because men choose the method that leaves fewer survivors of attempts at suicide. I note that 40% of female suicides use "poisoning" (presumably, that includes drug overdosing). What's the proportion of females attempting suicide by poisoning to females succeeding in killing themselves with poison?

If you have a fantasy of rescuing those who are in the process of committing suicide, you might think taking guns away will give you a better shot.

ADDED: It occurred to me, after the Sandy Hook murders, that blaming guns is a secular substitute for blaming the devil. People find it too challenging to figure out why a human being would do this terrible thing and they latch on to the idea that the gun made it happen. Suicide presents a similar challenge, and one way to fathom it is to say: It was the gun. Isn't it like saying the devil made him do it? The gun/the devil is a great go-to answer, freeing you from wracking your brain about the workings of the human mind.

"We can’t go out and lock up all the socially awkward young men in the world..."

"But we have to try to prevent the unpredicted."

Says a psychiatrist quoted in a WaPo article titled "Predicting violence is a work in progress."

The photos at the link — a line-up of Jared Lee Loughner, James Holmes and Adam Lanza — create the impression that you can tell by looking at them, especially at their eyes. But if you look long enough, you'll see the 3 men are quite different. Only Loughner is smiling. One may sense that madness radiates out of his face, but if you can exclude what you know — that's a photo taken after his shooting spree — he may seem like a fairly ordinary guy. Holmes's face, especially stuck between those other 2, looks open — even empathetic — and sad. Lanza looks abnormal, like an alien. The oddness is enhanced by knowing that this isn't a mug shot like the other 2. Is that his driver's license pic? It's hard to believe — in this age of digital photography — that a picture that came out that bad wouldn't be trashed. If I were diagnosing Lanza from that photograph, I'd say his problem was anorexia. What that boy needs is cheeseburger*... and a better haircut.

An analysis of 20 studies published three years ago found that schizophrenia increased the risk of acting violently fourfold in men and even more in women. The risk of schizophrenics committing homicide was 0.3 percent — more than 10 times greater than the average citizen.

What is the risk of serious violence — not just homicide — for schizophrenics who are also young and male? What is the risk for young, male schizophrenics of the paranoid subtype? If we're going to reason from statistics, we need to be able to look at the numbers in different ways. I suspect that the 0.3 figure — which screams you can't just institutionalize schizophrenics — is massively diluted by including large numbers of females, over 30s, and the nonparanoid subtypes (disorganized, catatonic, undifferentiated, and residual).

Back to the WaPo article:

John Monahan, a University of Virginia psychologist... and many others came up with a constellation of “risk factors” and “protective factors” for violent behavior.... [that] the presence of a mental disorder as only a small contributor to risk, outweighed by other factors such as age, previous violent acts, alcohol use, impulsivity, gang membership and lack of family support.

Gang membership? Now, you've got a list of factors that's off the subject of mental illness and more about a young male's social context.

“From our research, we could quickly distinguish between a patient whose chance of being violent was 1-in-10 from one whose was 1-in-2,” [Monahan] said.

Some statistics skepticism: he's saying "violent," not homicide or even serious violence.

[A British Medical Journal analysis found that of] the people predicted to “violently offend,” 41 percent did. Of those predicted to be nonviolent, 91 percent were. In practical terms, that meant that if authorities used the tools for the purposes of public health, they’d have to detain two people to prevent one from becoming violent.

That is patently defective reasoning. Where did they draw the line in scoring individuals using their set of factors? Show me how the factors were scored and who ended up being violent. Why have you simply divided people into 2 groups? Look at different subsets within the predicted-to-be-violent group with the highest scores. For example, if you break out 10% of them, the predicted-to-be-violent ones with the highest scores, what percentage of them went on to commit acts of serious violence? If that approaches 100%, then the "practical terms" about the fairness of detention would look entirely different.
________________________________

*Adam Lanza was a vegan — "He didn't want to hurt animals."

"The liberal news and culture website Slate published an online map Friday that it says will show a 'daily tally of gun-deaths in America'..."

"... basing that tally on incidents collected by an anonymous Twitter user. The organization, however, won’t say who is behind the tweets or how reliable its data is as a statistical compilation."

Gun deaths? So... just mix in the accidents and the suicides. Now, personally, as you may know from previous discussions on this blog, I think suicide is murder. Morally, it should be recognized as the deliberate killing of a human being. But many people withhold moral judgment and express only sadness and sympathy. I am sympathetic when the self-killer is mentally ill, but I'm also sympathetic when those who murder others are mentally ill.

But what's with grouping self-murder and murder of others for the purposes of Slate's gun-death map? It's a great way to boost the numbers. Google statistics, and I think you'll see that in the United States, there are twice as many suicides as murders, and that proportion is also true within the category of gun deaths. Shooting oneself is the most common method of suicide.

I realize some people think that there would be less suicide if people didn't have guns at hand. But I wonder how many of the mentally disturbed/emotionally overwrought men who shoot themselves to death are self-executors — that is, persons who feel a compulsion toward violence and stop themselves. There are murderers — like Adam Lanza — who murder and then kill themselves. There is no sympathy for these suicides, yet many people think: I wish he'd killed himself first.

If he had, it probably wouldn't have made the news, but if you'd read his story, you would have reacted the way you usually do to suicide: How sad... if only he'd perceived that suicide is a permanent solution to a temporary problem... whatever it is you usually think when you read that an isolated, bullied 20-year-old man has shot himself in the head. Would you ever think to speculate that he knew that he had an irresistible, evil impulse to gun down children, and, thank God, he had one last shred of humanity and turned the gun on himself?

Gun deaths. You can collect them on a map as faceless pins, all looking alike. It's all numbers. No individuality. What are you counting? I don't know what these pins represent. Use a different color for suicides. And which suicides are uncontrolled mental illness? Who is checking out because of uncontrolled physical pain? Identify the justified killings, done in self-defense. I want to know which ones are criminals gunning down other criminals. I'd like to think about policy with some complexity and without irrational fear mongering.

Ironically, Slate seems to want to inspire urgency about the loss of human life, but it is reducing the lives we're supposed to care so much about to a bunch of dots who matter only because they're a massive crowd. Where is the individuality? There is particularity to each death, despite the common element of the gun.

"Concealed carry in a gun free zone really makes it safe."

"What are the odds that two people have a gun in a gun free zone? It's the old bombs on an airplane trick. Always pack a bomb in your luggage."

The number of women not in the labor force hits an all-time high as 324,000 women leave in the last 2 months.

According to the Bureau of Labor Statistics.

The labor force includes people who are unemployed but actively looking for work.

This year (in both January and April), only 57.6 percent of the women in the civilian noninstitutional population were in the labor force. That is the lowest rate of labor force participation by American women since April 1993, according to historical data maintained by BLS.

The rate of female participation in the civilian workforce peaked twelve years ago--in April 2000--when hit 60.3 percent...

For both males and females combined, the rate of participation in the labor force dropped to 63.6 percent in April—the lowest rate since December 1981.

Recently, however, women have been leaving the labor force in larger numbers than men.

It would be interesting to know why this has happened. Is it a cultural trend, with families restructuring to single-earner form? Are more women going the "Julia" route, finding succor in the arms of the government? Or is this mostly the government's recategorization of the unemployed, done to keep us from seeing the real unemployment percentage?

"Need to prove something you already believe?"

"All you need are two graphs and a leading question."

(Thanks to my colleague John Ohnesorge for this very funny link.)

At the Statistics Tree Café...

... you can celebrate World Statistics Day.

The Narcissism Epidemic?

It's everywhere. It seems.

That got me thinking of "An Outbreak of Autism, or a Statistical Fluke?"

The puzzling dearth of women in computer science and the annoying lack of statistical competence in the NYT.

Here's an article about how few women there are in the field of computer science, written by Randall Stross (which I noticed because of its rank on the NYT most-emailed list). It begins this way:

ELLEN SPERTUS, a graduate student at M.I.T., wondered why the computer camp she had attended as a girl had a boy-girl ratio of six to one. And why were only 20 percent of computer science undergraduates at M.I.T. female? She published a 124-page paper, “Why Are There So Few Female Computer Scientists?”, that catalogued different cultural biases that discouraged girls and women from pursuing a career in the field. The year was 1991.

Computer science has changed considerably since then. Now, there are even fewer women entering the field. Why this is so remains a matter of dispute.

What’s particularly puzzling is that the explanations for under-representation of women that were assembled back in 1991 applied to all technical fields. Yet women have achieved broad parity with men in almost every other technical pursuit. When all science and engineering fields are considered, the percentage of bachelor’s degree recipients who are women has improved to 51 percent in 2004-5 from 39 percent in 1984-85, according to National Science Foundation surveys.

When one looks at computer science in particular, however, the proportion of women has been falling....

Now, wait a minute. You can't compare the average of all the fields to the number in one particular field, then assert that the one field stands out from all the others -- or even "almost" all the others. The numbers that make up that average could be all over the place, with many lows balanced by highs. They could be drastically skewed by the inclusion of some science field that is unusually attractive to women -- or unattractive to men. I wish the NYT would link to the NSF surveys so I could see for myself what is inside that 51%. Also, unstated, is the fact that more women than men receive bachelor's degrees these days. What percentage of female college graduates major in science and engineering, and what percentage of male college graduates major in science and engineering?

Anyway, the failure of women to enter computer science is especially interesting if it is true that it's the only field -- or "almost" the only field -- that women have shunned as they pour into the rest of science and engineering, but I'm not convinced it's true. If it is, though, maybe it's a bit puzzling. One professor, we read, theorizes that in the past "young women earlier had felt comfortable pursing the major because the male subculture of action gaming had yet to appear." So there's this idea that the key to getting more women to enter the field is to entice young girls to play computer games. Indeed, there was a "girls game movement," but it's already failed.

There are other theories too: women who like computers prefer to do website design, women are more sensitive about being regarded as nerds, etc. These theories already contain the belief that women's interests differ from men's. That being the case, why not just say that fewer women are interested in computer science? Presumably, the answer is that because the percentage of women in computer science has been falling over the years, it probably doesn't reflect an innate gender difference. If it's something out there in the culture, then, supposedly, it's something that can -- and should -- be manipulated.

I think there are at least 3 shaky assumptions in the previous 2 sentences but I won't lengthen this post by belaboring them.

I'll lengthen this post by pointing to the news that Barack Obama might appoint Larry Summers as Secretary of the Treasury, and some women are displeased:

A controversial comment at a Cambridge conference may cost former Treasury Secretary Lawrence Summers a second stint in the Cabinet.....

In 2005, [Nancy Hopkins, a biology professor at the Massachusetts Institute of Technology] walked out of an academic conference after Summers, the keynote speaker and the president of Harvard University at the time, said that innate differences between men and women might be one reason fewer women succeed in science and math careers.....

And now women’s groups have expressed so much outrage over Summers’ possible appointment that, according to top Democratic sources, his name may even have been stricken from the short list....

Just after Obama won, National Organization for Women president Kim Gandy told the Huffington Post she had “mixed feelings” about Summers, saying he doesn’t “get” the economic implications of gender-based wage disparities.

The New Agenda, a nonpartisan women’s rights group, issued a press release, saying Summers’ “record of derogatory comments aimed at women ensures that his selection would be divisive and thus distract from efforts to fix the economy.”

How much does the work of the Secretary of the Treasury have to do with getting the implications of gender-based wage disparities? And does NOW really get the implications of gender-based wage disparities or does it simply invoke them to get attention and try to appear relevant and powerful? Is there some innate gender difference that makes women want to stand between the new President and the man who might be the best person for what is a phenomenally important job?

IN THE COMMENTS: Joan writes:

I read the article yesterday when I saw the headline -- couldn't resist. I graduated from MIT and worked for 15 years as a software developer. I did not major in computer science. My informed opinions on why fewer women are choosing computer science:

1. You can easily work in computer science fields without a computer science degree. The joke at MIT back in the late '80s was it didn't matter what you majored in, we'd all end up writing software anyway. It was true for about 80% of the people in my living group, at least for portions of their professional careers.

2. As far as choosing CS as a career, the field is dominated by people (both men and women) with stunted emotional and social maturity. High-school level drama in the workplace is wearing and unpleasant.

3. The work is challenging and can be really fun. It can also be a real grind, and the cyclical nature of new product releases means you have to work overtime for extended periods every year. The pay is good, and that is one form of compensation. But the work itself is ephemeral, and this is the key to why I don't work in software anymore: If I kill myself to get this release out, the software will be used for 6 months, maybe a year, until the next release. It never ends, and there's no perceptible benefit. Aren't you tired of the new versions of your favorite software continuously appearing, laden with feature-bloat and a host of new problems?

I'm teaching now because I get a sense of fulfillment, and because it works with my own children's schedules. The money is horrifically bad compared to what I was making as a project lead at Oracle, but money is not my only concern.

Regarding Larry Summers, he was right when he talked about the innate differences between men and women -- women self-select into professions they enjoy, just as men do, and those who deny this are insufferable. I stopped donating money to MIT after the Nancy Hopkins incident. They should have repudiated her, and instead she was lauded. I'm disgusted by how PC my alma mater has become over the years, and question the quality of the education kids are getting there, if they have idiots like Nancy Hopkins on staff.