These are just random, lack-of-sleep thoughts; I have no particular point, in case you were wondering.

Vance of Begging to Differ takes issue with FactCheck.org’s claim that the Bush administration’s claim that the average tax cut is $1,586 is “misleading,” because using the mean instead of the median† is improper. Vance writes:

I can think of a valid justification for either measure. If you’re trying to understand the overall economic effects of the tax cuts, for example, an average is entirely applicable.

In the case where data is “normally distributed”—following the “bell curve” known to statisticians—the mean and the median are essentially the same.* When they differ, the data is said to be skewed, and measures of central tendency and dispersion that assume a normal distribution (like the mean) are generally misleading, as they don’t properly describe the distribution. The income distribution, for example, is skewed right.‡

To cast things in non-mathematical terms, when people think about averages they are thinking in terms of things that are most typical, rather than in terms of distributions. And, in general, the median better reflects this perception of average than the mean. While there may be technical value to the mean for specialists and those who want to engage in further analysis, I think the median does a better job of reflecting the “most typical” observation in most data patterns.

Robert Prather is continuing his semi-quixotic effort to get Virginia Postrel back as editor of Reason; I wholeheartedly support any and all efforts in this regard.

My daily routine now goes something like this:

1. Visit the Chronicle jobs web site.
2. Visit HigherEdJobs.com.
3. Fight with APSA’s eJobs system to get it to show me the last few days’ postings—bearing in mind that the “show jobs posted in the last two days” function doesn’t actually work because their website’s database isn’t synced with the actual jobs database, so jobs actually “posted” today may have been entered into the system several days ago. Also bear in mind that the “Full Professor” job at Rockford College (that I didn’t apply for, hence why I’m mentioning it by name) is actually a junior-level position, that jobs that don’t list your field on the main list may actually be looking in your field when you get to reading the actual text of the posting, that the same job at one college is in the database twice. Oh, the “print” version of the page actually takes up more paper than the non-print version. (Did I mention that eJobs sucks?)
4. Write cover letter(s) as needed.
5. Make pretty mailing label(s).
6. Stuff cover letter, vita, and other requested materials in big manila envelope(s).
7. Weigh and put stamps on envelope(s).
8. If before 1 pm, stick envelope(s) in mail box. If before 4:30 pm, get in car and go to post office to stick envelopes in mail box.

Writing cover letters at this point is a simultaneously easy and hard process. It’s easy in the sense that after you’ve written 50 of the damn things, one you’ve already written is pretty close to the one you “need” for the particular job. It’s hard in the sense that you have to remember which of those 50 letters is the right one to massage for the particular job in question.

It’s also hard in the sense that you have to show some enthusiasm for the job on paper—which for me entails doing some basic research about the college and putting in some thought as to I’d fit in there, something that makes me a bit more emotionally invested in a process that more closely resembles a meat market than anything a reasonable person would want to be emotionally attached to. It’s hard not to go from “I’d enjoy the opportunity to teach at X because I can contribute in ways A, B, and C” to actually feeling like you’d enjoy going to college X—and thus running the risk of being disappointed if you don’t get to go to college X for whatever reason, even if it’s not the “dream job” you expect to be doing when tenure time rolls around.

I’m feeling terribly conflicted this morning. The Baseball Crank has a lengthy post on the same-sex marriage issue, in which he makes—and highlights—the following prediction:

Gay marriage will become the law of the land without any state legislature ever having voted it into law, without a majority of either house of Congress ever having voted in favor of gay marriage, without any statewide popular referendum ever having voted in favor of gay marriage, and without any state or federal constitutional provision ever having explicitly authorized it.

I think the first part of the Crank’s premise is incorrect—Massachusetts’ legislature will probably vote it into law later this year, albeit under court duress—but otherwise, there’s not a word there I’d disagree with. And efforts to analogize this “struggle for equality” with those of racial minorities don’t track—those groups were deliberately and systematically excluded from political participation through ordinary legislative channels, against the plain text of the 14th and 15th amendments to the Constitution and numerous federal statutes, and thus their recourse to the judiciary was justified. Tyranny and oppression is being confronted with policemen on horseback, unjustly imprisoned, blasted with fire hoses, and lynching; being deprived of legal recognition of the fact you’ve set up housekeeping with someone of the same gender doesn’t quite fit into that category.

Is the only justification needed for anyone to get a victory in the courts something along the lines of “we couldn’t get the legislature to vote for it“? I find this a profoundly disturbing question. And I say that as one of the tiny minority of people in my state who would support legal recognition of same-sex marriages. What am I supposed to tell my students? “Well, the Supreme Court doesn’t trust your representatives to do what’s right, so they’ve decided to decide on everyone else’s behalf what your laws should be.” I don’t remember seeing that in Federalist 10.

On the other hand, I think Kate Malcolm is probably right that the judgment of history may well see those who oppose same-sex marriage today in much the same light as we (well, most of we at any rate) today see the segregationists of the 1950s and 1960s. So in the end I end up just feeling conflicted about the whole thing.

I see someone else is trying to make a Python port of the Textile 2 syntax. My approach so far has been a straight port of the Perl code by Brad Choate, with some minor tweaks (using urlparse and mimetypes, for example), rather than trying to hack Mark’s existing module.

Now that I’ve been hacking away for a week (and am about 700 lines away from being done—I just started on format_table, which looks downright nasty), I’m becoming convinced that the smarter plan would have been to write a lexer from scratch for the Textile markup, rather than trying to use the regex-happy approach Choate used (which works far, far better in Perl than in Python).

I think Brock below is being a bit obtuse in claiming the President’s position on a federal marriage amendment is an endorsement of “enshrining anti-gay bigotry into the United States Constitution”—particularly since that self-same alleged anti-gay bigotry is essentially the law of the land as of February 25, 2004. And I also think it’s absurd to criticize the president for not living up to one’s own fantasies about him, as Andrew Sullivan has done.

Funnily enough, my thoughts on the matter, from a policy perspective, generally coincide with those of Steven Taylor—although I personally do not share Taylor’s “moral objections” to homosexuality. As a supporter of same-sex marriage, I firmly believe the process that has been used to this point by its more overzealous proponents—particularly the extralegal behavior of officers of the City and County of San Francisco—is likely to energize enough additional support for FMA for it to pass, particularly if, as I expect will happen, Congress calls for ratification by special state conventions.* But my emotional reaction to the president’s support for FMA is closer to Tim Sandefur’s—which was perhaps even stronger than Brock’s.

Elsewhere: Dan Drezner is hosting a discussion of the politics of the proposal.

In lieu of accomplishing anything worthwhile today (except more job applications), I replaced the server that Signifying Nothing runs on this morning. The previous server was a discontinued Compaq corporate desktop box (Pentium II 300 MHz) I picked up an eon ago for $400 from Buy.com with 160 MB of RAM and a noisy* 20 GB hard drive. The current server is a tower-case Pentium III, 450 MHz with 256 MB of RAM and a 60 GB hard drive that is basically the reconstituted version of a (then top-of-the-line) PC I paid good money for from Quantex about five years ago—the only newish components are two 10/100 Ethernet cards† and a spare CD burner I had lying around the house. I did a fresh Debian sarge install using the Beta-2 installer from CD-R (which went smoothly after I gave up trying to do a network install instead), then updated to unstable and a 2.6.3 kernel (the last part being the most painful step, since I had trouble with the initrd support until I finally gave up on that).

I was hoping the new box would be less noisy, but to no avail (fan noise has replaced drive whining)... I’ll probably have to clear some space and move the box under the desk to get some noise relief. Actually, what’s more likely is that I’ll get my replacement laptop in a few days then go back to rarely using the computer that is right underneath the SN box, so the noise won’t bother me any more.

The Secretary of Education (thank Dick Nixon for that great idea) talked to some governors today. His speech, er, didn’t go very well:

Education Secretary Rod Paige said Monday that the National Education Association, one of the nation’s largest labor unions, was like “a terrorist organization” because of the way it was resisting many provisions of a school improvement law pushed through Congress by President Bush in 2001.

To his credit, at least Paige isn’t bothering to claim he was “quoted out of context,” or similar such nonsense:

His initial remark was described by four governors and confirmed by the Education Department. “The secretary was responding to a question,” said Susan Aspey, a spokeswoman for Mr. Paige. “He said he considered the N.E.A. to be a terrorist organization.”

But he did offer a sorta-kinda retraction later:

After his remark had begun circulating, Mr. Paige issued a statement saying he had gone too far in describing the union as a terrorist organization. “It was an inappropriate choice of words to describe the obstructionist scare tactics that the N.E.A.’s Washington lobbyists have employed against No Child Left Behind’s historic education reforms,” he said.

“As one who grew up on the receiving end of insensitive remarks,” said Mr. Paige, who is black and was born in a segregated Mississippi, “I should have chosen my words better.”

Now, in general I’ll be first in line to criticize the NEA, who are among the worst kind of rent-seeking interest group to pollute the waters of the Potomoc basin. Their lobbyists routinely try to delude the public into believing that their members’ best interests (which is what the NEA lobbys for) somehow coincide with the best interests of American children (which, more often than not, is what the NEA lobbys against).

However, the NEA is in no way, shape, or form a “terrorist group.” Al Qaeda is a terrorist group. Hamas is a terrorist group. Palestinian Islamic Jihad is a terrorist group. Shining Path is a terrorist group. Terrorists generally blow stuff up, kill and maim people, and the like. The NEA, by contrast, is a group of middle-class workers who peacefully lobby for their preferred public policies through non-violent activity. They don’t even qualify for the mantle of “shakedown artist,” unlike leftist fellow-travellers like Ralph Nader and Jesse Jackson. I think it’s time for Mr. Paige to ease himself into a nice, early retirement—not just because he’s the Education Secretary, but because this sort of rhetoric is on the level of the “Bush=Hitler” analogy and should be beyond the pale.

Elsewhere: John Cole agrees, Jonathan Wilde at Catallarchy.net thinks it’s about time someone told the truth, and David Bernstein misses the point entirely, as my co-blogger would probably expect.

James Joyner finds Howard Kurtz in today’s Washington Post acknowledging many of the same sins of the pundit class that SN did almost two weeks ago.

James Joyner has linked a column by Bob “Endangering National Security Since 2003” Novak on the wrangling between Capitol Hill and the White House over the six-year transportation reauthorization bill, coined SAFETEA. As usual, the debate is mostly about how much money to spend and where to find the cash; many House members from both parties want an increase in the federal fuel excise taxes to fund a larger spending program of $375 billion over six years, while the White House wants to limit spending on highways and mass transit to $256 billion.

I’m a bit late to the party on this one, but in case you haven’t heard—Ralph Nader will run in 2004 as an independent presidential candidate. What does it mean? Juan Non-Volokh and Glenn Reynolds think it might invigorate efforts to improve ballot access for third parties; some Democrats are apoplectic; Robert Garcia Tagorda thinks it may help Democrats; and Steven Bainbridge, Steven Taylor, and James Joyner used the occasion to dump on third-party candidates in general.

Steven Taylor has posted the most recent iteration of the infamous Toast-O-Meter.

Lily Malcolm visited her local Radio Shack and CompUSA stores today, and came back with this nugget of knowledge:

If the CompUSA people really knew much about computers, they wouldn’t be working at CompUSA.

In all seriousness, there are a number of solutions to the problem of getting data from one’s old computer to one’s new computer. If both computers have an Ethernet jack, the preferred option is to either (a) connect both computers to an Ethernet hub or switch using regular Ethernet cables or (b) get a “crossover Ethernet cable.” You can also get something flashy that will walk you through the procedure, like LapLink, but I’m pretty sure the basic software is built into recent versions of Windows… you may need to set up “Home Networking” to do it.

Me? I usually just pull the hard drive out of the old PC, slap it in the new one, and copy the files that way. It’s normally faster, but far more intimidating for the novice.

Greg Goelzhauser at Crescat Sententia considers the use and abuse of behavioral economics by legal scholars, jumping off from this interview with Berkeley economist Matthew Rabin. Greg writes:

That humans fall prey to a variety of heuristics is nothing new or extraordinary. What is important for law and economics is if some of these heuristics lead the relevant actors to systematically err in their decisionmaking. Unfortunately, many of those writing in the behavioral law and economics field care little about whether the actors they are concerned with actually rely on the heuristics attributed to them or, if they do so happen to rely, whether reliance actually leads to systematic error. The reason? These are often difficult empirical questions.

I’m not sure that this is an accurate characterization of what a heuristic is; it’s not simply a matter of “falling prey” to them, as many are reasonable shortcuts. You assume that the store with the cheapest price on a product is advertising it, rather than calling every store in town. Instead of digging through the platforms to find the most stridently anti-war presidential candidate, you assume the candidate making the most noise about the war is that candidate.

Now, as Tversky and Kahneman have pointed out, some heuristics do lead to systematic, non-random error. And some of those errors are big enough that the cost associated with the error is larger than the deadweight loss of not using the heuristic (calling every store in the world or spending hours reading the minutae of Lyndon LaRouche’s campaign platform).

Greg’s larger point—that social scientists and legal scholars often assume away the “difficult empirical questions” associated with determining whether systematic error exists—is well-taken, but I think characterizing heuristics as something we “fall prey” to assumes away the more important question of whether the systematic error involved in using heuristics outweighs the costs we avoid by using them.

It seems the “enterprise web filter software” that Brock tested in September isn’t the only popular censorware product that blocks a large number of weblogs; Eric of Classical Values took SonicWALL’s web filter for a spin and, shall we say, was unimpressed. (Link via Tim Sandefur.)

The Baseball Crank has all the nasty stuff the Republicans never got to say about Howard Dean neatly collected in a single post. It might come in handy, just in case Dean ever tries to get elected to the city council in your town.

Brock Sides points out a public policy reason for states to subsidize universities (one that I thought about, but didn’t mention, in my prior post):

State support of higher education may be a rational investment by the state in its own tax base.

There is some merit in this argument, if the marginal increase in tax receipts due to residents’ higher education exceeds the amount of subsidy required—bearing in mind that, ceteris peribus, many of those residents would have gotten a collegiate education anyway. And it’s certainly an empirically-testable proposition, although one that’s difficult to examine in a single blog post.*

Anyway, as a graduate of two state universities and as someone with about a 70% chance of working for a state-supported institution next year (under the completely unreasonable assumption that I have an equal chance of being offered every single job I’ve applied for), self-interest—if nothing else—compels an end to this discussion.

John Cole is right to be disgusted by ex-Colorado coach Gary Barnett’s remarks about former CU player Katie Hnida, who was allegedly raped by one of her former teammates. Barnett said, apparently in response to an inquiry from a reporter as to the reason for Ms. Hnida’s departure from the team (and transfer to New Mexico) in 2000:

It was obvious Katie was not very good. She was awful. ... Katie was not only a girl, she was terrible. OK? There’s no other way to say it.

Not only does the comment show a lack of seriousness by Barnett, it also makes me question his abilities as a coach and leader. Good coaches don’t speak ill about their players in public; that’s close to the cardinal rule of coaching. What a dipshit, and good riddance.

More on this story in Friday’s New York Times.

Matt Stinson thinks Julian Sanchez’s argument by analogy on the term “unborn child” fails. Julian argues:

If you don’t share their view about the moral status of the fetus, that’s like calling a pile of bricks an “unbuilt house” or, for that matter, a blank screen an “unwritten blog post.” Let’s not give them this one.

On the other hand, Matt says:

I’m pro-life, though not stridently so, but would a pile of bricks, without human action, begin to form a house over a period of nine months, unless you smashed those bricks down with a sledgehammer, and would a blog post begin to appear on that blank screen unless you pressed the delete key repeatedly?

Luckily enough, however, Smokey the Bear may still call discarded lit cigarettes “potential forest fires.” Or something. Semantics was never my strong suit…

Will Baude is the latest to jump into the public universities argument (roughly between Jack Balkin and the Volokh conspirators); Will writes:

I don’t particularly have a problem with government involvement in the private education market—either through direct subsidy (which is probably unnecessary) or through regulating the likely capital market failure. In other words, government-guaranteed student loans are great; a “graduate tax” could accomplish the same thing.

I used to generally agree with Will on this point; however, I’ve come to think that government subsidies—like guaranteed student loans, Pell grants, and student loan interest credits—make public and private universities insensitive to price as a rationing mechanism. This leads to much of the same problem we see in the health care market: most consumers don’t discriminate on the basis of price, because they have no personal stake in the price of service. In the case of higher education, the problem is more subtle, as at least there are direct costs to the consumer—they just aren’t felt until after college, due to in-school deferments of loan interest and principal payments. Regardless, this allows universities to increase tuition and fees at rates well in excess of inflation.

The disconnect between price and demand also allows universities to use price as a “prestige” factor; although virtually nobody actually pays $40,000 a year to go to Harvard, the price premium makes it appear as if you’re getting a better education than you would paying $15,000 to go to Americana State University. (You probably do get a better education at Harvard, but I suspect the premium is not worth $100,000.)

There are good reasons to criticize public subsidies of state universities—particularly in a poor state like Mississippi—but public subsidy of colleges and universities in general bears considerable scrutiny as well.

Update: Will Baude responds:

One thing to think about—

The reason, in general, that American[a] State U has a tuition of 15,000 to Harvard's 40,000 has a lot to do with the subsidies that American State provides to its U. To be sure, some private colleges are cheaper than others, but lots of kids I knew did indeed take price (and their financial aid packages) into account when choosing between them. And the diversity of price in private universities is pretty small—I don't know whether that's due to a universal-ness of costs (I doubt it) or more likely because demand is fairly price-inelastic. While it's true that subsidies (and to a lesser degree, loans) encourage that elasticity, it's not actually clear that's bad. On the one hand, some kids go to Harvard who really should have gone to Miss. But on the other hand, some kids go to Miss who otherwise wouldn't have gone at all.

Just a thought.

My experience as an undergrad (granted, 5+ years ago) was that there was more price differentiation among private universities; I know the tuition at Rose-Hulman was significantly lower than that of Georgetown, and the price differential was more than could be justified on the basis of cost of living differences between D.C. and Indiana (not to mention that Rose-Hulman is a superior academic institution to Georgetown). There may be less differentiation among elite-tier private institutions like Chicago, Stanford, and the Ivys, however (some of that used to be due to now-illegal agreements among the Ivys to limit financial aid awards to exceptional students).

Now it is true that price does matter to some people, even with government subsidies (both to universities and students). On the other hand, I find it difficult to justify subsidizing a flagship public university like Ole Miss on the backs of working class people; that being said, Ole Miss may be something of an aberration in this regard, although I suspect a number of other colleges, like the University of Alabama, Auburn, and LSU, are similar “blue blood” state universities (to say nothing of elite-level state universities like UC-Berkeley and Michigan, which are far more selective).

Alex Tabarrok endorses an econometrics text that makes two rather bold simplifying assumptions:

Stock and Watson use a “robust” estimator of standard errors right from the beginning. This means that they can dump an entire chapter on hetereoskedasticity and methods of “correcting” for hetereoskedasticity (these rarely worked in any case.)

They do not waste time discussing the difference between the t-distribution and the normal-distribution. Instead, they assume reasonably large datasets from the get-go and base their theorems on large-sample theory.

I can sort-of-see the value of always using heteroskedasticity-consistent standard errors (although I think it’s better to model the heteroskedasticity if you can), but dispensing with the t distribution seems to be a bridge too far. Large sample theory is nice, but (a) common econometrics software (e.g. Stata, LIMDEP, and R) uses the t distribution even into sample sizes in the 100s, so you need to discuss it anyway, and (b) there are plenty of theorems that can only be tested with small samples due to data limitations. Now, these may be less problematic in the large-n world that economists inhabit, but I’d have real trouble justifying such a text for a graduate seminar in political science methods (undergrads rarely get beyond bivariate regression).

Since John Kerry’s alleged “zipper problem” has been debunked, Andrew Sullivan notes that Sid Blumenthal (not to be confused with Atrios) thinks John Kerry should sue the Sun for libel. Funnily enough, Jeffrey Archer had much the same idea under similar circumstances, but it didn’t quite work out the way he planned…

Update: Conrad has thoughts in a similar vein. And thanks to Glenn Reynolds for the link!

CalPundit and Steve Verdon are among those noting a report from the Union of Concerned Scientists over the Bush administration’s use and alleged abuse of science. Steve writes:

Personally I think the notion of impartiality is misleading. All scientists have their own views on the issues and particularly the area they are researching. ... Of course, the fact that scientists and researchers themselves have their own views and biases does not let the Bush Administration off the hook when it comes to possibly distorting science. However, it cannot be ignored that the Union of Concerned Scientists can also be said to have an agenda and that this agenda may be playing a role as well in this report when that agenda diverges from the agenda of the Bush Administration.

There’s not much to disagree with in either post, but something to bear in mind is that science is always politicized when it is used to make political decisions; there’s no way around it. For example, if a hypothetical study shows that tightening emissions standards will save 3000 lives a year, but cost consumers $100 billion per life saved, politics is going to decide which figure gets emphasized.

Update: More at the Dead Parrot Society.

Michael Jennings has uncovered a bit of a visual oxymoron over at TransportBlog.