Friday, January 24, 2003
Posted by Glen Whitman at 4:31 PM
And Who Says Accidental Gun Deaths Are All Bad?I heard on the radio this afternoon that a man in Virginia (I think) accidentally shot and killed himself while trying to beat his dog to death with the butt of his shotgun.
UPDATE: And here's the link, courtesy of Eugene.
Thursday, January 23, 2003
Posted by Glen Whitman at 5:54 PM
Squeals on Wheels?Today's (23 October) Dear Abby column is more interesting than usual. Read the letter for the full details; here's the adumbrated version. The letter writer has a friend with muscular dystrophy. He is mostly paralyzed and could die any time. Yet, like most men, he still has sexual needs. The letter writer arranged for his friend to have a date with a female "escort" -- yes, that kind of escort -- and the encounter apparently went quite well. But the man's parents (who care for him) were appalled, and they have banned the writer from seeing his friend.
Now, this is interesting to me for a couple of reasons. First, like all libertarians, I think prostitution should be legalized. But I don't think prostitution is a "good thing." In general, I think it's a poor personal choice. I support legalization because (a) I think prohibition creates even greater harms, and (b) I am unwilling to substitute my personal moral judgment for that of another competent adult. Here, however, we have one of the rare cases in which I think prostitution is not just tolerable, but actually desirable. Abby gets it exactly right in her response:
It is wrong for a person in control to project his or her own moral values on another adult who is dependent. For parents to confine an adult child, to prevent that person from having relationships, and to discourage that person from living life as fully as possible is to me both cruel and ignorant.Second, there is a twist in the story that I left out above: when the writer contacted the escort service and explained the situation, the manager offered the escort's services for *free*. Does that mean this wasn't prostitution after all, because the price charged was zero? From a legal standpoint, I suspect this encounter was totally legit, unless the escort herself was still paid and the money came out of the business's general revenues. However, my understanding is that escort services usually handle only the "date" portion of an escort's business; the prostitution, if any, is left as a transaction strictly between the escort and her customer (so that the escort service is shielded legally). I think this case highlights the arbitrariness of a law that bans certain acts of sex simply because money is involved. What if, hypothetically, the escort service had not offered the service for free, but had instead offered a discount? Would it matter how big the discount was? In this case, it's hard for me to find a relevant moral distinction between sex for free and sex for a nominal payment. But perhaps I'm being misled by the fact that I think prostitution is desirable in this situation anyway. I'd like to hear someone who is opposed to prostitution make the case for a meaningful distinction.
I would only hope that someone in your community who understands this could intercede and explain to Kent's parents that there is room for nontraditional relationships in cases like this one.
Posted by Glen Whitman at 1:56 PM
Glengarry Glen RossEugene is excited about this list of name frequencies collected by the Census. I think it’s pretty cool, too. But in looking at it closely, I noticed something slightly misleading about it: alternate spellings of the same name are listed separately. Naturally, I noticed this because I looked up my own name in the list. “Glen” is the 188th most common male name in the U.S., at 0.094% of the male population. But “Glenn” is the 117th most common male name, at 0.167% of the male population. Adding these together, they constitute 0.261% of the male population, making Glen/Glenn the 68th most common male name (between Bruce and Brandon). Of course, that’s assuming that all the other names are distinct. But if we combine Shawn (92nd) and Sean (94th), we find that Shawn/Sean comes in 42nd.
And then there’s the problem of nicknames or related names. William comes in 5th, Billy comes in 73rd, and Bill 159th. Combining them, they jump past Michael (4th). But that’s not accounting for all the Mikes out there.
Since this is Census data, I assume that seeming nicknames might be the actual names on birth certificates. So arguably, these are not people with the same name after all. But on the other hand, many of the people categorized under the standard names may go by the nicknames. Given the subjectivity of deciding what constitutes the “same” name, I can’t fault the Census for taking the objective spelling-only approach.
Wednesday, January 22, 2003
Posted by Glen Whitman at 11:35 AM
A Request for Bad StatisticsFor an MBA course I’m teaching this semester, I will be giving a lecture on the use and abuse of numbers and statistics, and also a lecture on the use and abuse of tables and figures. I plan to structure both lectures in “greatest misses” format – using examples of people employing numbers and figures in bogus or misleading ways. If you’ve read Darrell Huff’s book _How to Lie With Statistics_, you know the kind of thing I’m talking about; unfortunately, Huff’s examples are rather dated, and I’m looking for more recent ones. I have lots of old examples in my head, but I really prefer documented cases – “I saw an article one time that said…” isn’t good enough.
So, I’m hoping to solicit my readers’ assistance. If you know of any good examples of numbers/stats/tables/figures being used badly, please send me an email. I need full references or webpage locations if at all possible. I’m looking for examples in various categories, including:
• Confusion of mean and median (or mean and mode, or mode and median). For example, it’s often observed that the average (i.e., mean) time it takes a woman to conceive a child is several months, but most women (i.e., the mode) conceive much more quickly – within a couple of months. The average is higher because of a small number of women who take a really long time to conceive.
• Confusion of absolute numbers with changes in numbers, or confusion of changes in numbers with percent changes in numbers. For example, during the 2000 election season I recall hearing that Houston’s air quality had actually gotten worse than L.A.’s – but it turned out that while Houston’s had gotten worse over time (compared to Houston before) and L.A.’s had gotten better (compared to L.A. before), L.A.’s air quality was still worse than Houston’s.
• Providing figures that are meaningless without a figure for comparison. For example, saying that some very high percentage (say, X%) of all traffic fatalities involve people not wearing seatbelts, without telling us what percentage of *all* drivers wear seatbelts. (If that figure were also X%, we might conclude that seatbelts have no effect!)
• Creating a pie chart that leaves out the “other” category. For example, showing a pie chart with market shares for only the top four firms in an industry, even though together they constitute only a fraction of the market.
• Creating a line graph that does not show a trend over time, but instead connects unlike things. (“What’s this big jump in the line show? Oh, it shows the jump in GDP you get when you go from Mexico to Japan.”)
This is not a comprehensive list, of course. I’d appreciate any ideas or contributions you have.
Labels: damn lies and statistics
Monday, January 20, 2003
Posted by Glen Whitman at 5:06 PM
My Chickens Come Home to RoostSo when I moved to a new apartment back in July, I was excited by the prospect of a new telephone number that might actually spell something. The Pacific Bell customer service representative gave me three choices of phone number, and some back-of-the-envelope scribblings revealed that one of them was 761-AUTO. I thought that was reasonably cool, so I picked it.
I'd had the number for about two days before I left town for a week-long seminar. I came home to an answering machine with about 30 messages -- all people wanting to advertise their cars in Auto Trader Magazine, which apparently had the number before I did. Some of the callers were pretty steamed about my failure to get back to them. "What the FUCK, man," said one of them, "why don't you people pick up your FUCKing phone???" As message after message rolled out of my machine, Olivia laughed her ass off at my expense -- since I had, of course, brought this on myself. Shortly thereafter, I called Pac Bell and got my phone number changed. End of story… or so I thought.
Last week I received a form letter from Pacific Bell -- now SBC Pacific Bell, and soon to be just SBC -- informing me that, due to my poor payment history, certain restrictions were being placed on my account: I must pay all bills within 15 (not 30) days, and I can't run up more than $200 in charges without special permission. How strange, I thought. I knew I'd occasionally paid a bill a day or two late, but hardly often enough to justify a deadbeat classification. So I went looking through my old bills and online banking records. It took a while to figure out, but I finally realized that Pacific Bell hadn't received one of my online payments, even though my bank said it had cleared. I'd continued to pay the additional monthly charges, but the amount due on my September bill continued to hang around for three months, seemingly unpaid, and presumably lousing up my credit. And what, you ask, happened to that errant payment? Turns out that after getting the 761-AUTO number changed, I'd never altered the account number in my online banking settings. Somehow the phone company still managed to register my subsequent payments, but that first payment under my new (post-AUTO) number got lost in the vortex.
I tried to sort all this out on the phone, but I still have to get documentation from my bank to prove the payment was actually made. Once I straighten things out with the phone company, I suppose I'll have to contact the credit agencies, too. And to top it all off, my current phone number doesn't spell a damned thing -- too many 1's and 0's, I'm afraid.
Sunday, January 19, 2003
Posted by Glen Whitman at 5:50 PM
Mutatis MutandisMark Kleiman makes a plea to libertarians and libertarian-leaning folk to flee the Republicans and join forces with the Democrats: "And when Glenn Reynolds … figures out that, while most Democrats take anti-libertarian positions with some discomfort, and largely out of perceived political necessity, the current Republican crew is deeply and abidingly anti-libertarian in the personal sphere and not really opposed to corporatism in the economic sphere, I'll be happy to welcome him as well."
Aye, but here's the rub: the current Democratic crew is deeply and abidingly anti-libertarian in the economic sphere and not really opposed to paternalism in the personal sphere. Much as I like Mark, he himself provides a nice example of the latter point - consider, for example, his opposition to the notion that "the blamelessness of the tempter follows from the responsibility of the tempted." Summary version: I persuade you to eat one more jelly doughnut, and you do so, to the detriment of your waistline. According to Mark's theory, I may be partially to blame for your eventual heart attack, especially if I'm a big bad corporation. Mark is not a politician, of course, but we all know how this kind of position gets translated into policy when the Democrats have their way.
Not that the Republicans are any great shakes. Fact is, neither party is abidingly libertarian, and neither can be relied upon to defend liberty even in the areas where we expect them to. Republicans are supposed to be for free enterprise, and then Bush supports the steel tariff. Democrats are supposed to be for freedom of expression, and then Bill Clinton signs the Communications Decency Act. Mr. Hobson, what horses do you have for rent today?