Tuesday, July 31, 2007

Learn math to learn science

USA Today reports:
Philip M. Sadler of the Harvard-Smithsonian Center for Astrophysics and Robert H. Tai of the University of Virginia surveyed 8,474 students taking introductory science courses at 63 U.S. colleges and universities. Their findings are reported in Friday's edition of the journal Science. ...

Using a scale of 0-to-100 points, Sadler and Tai found that every year of high school math a student took added 1.86 points to their grade in college chemistry. Taking chemistry in high school added 1.72 points to the college grade, but taking biology or physics in high school had no significant impact on the college chemistry grade.

Likewise, students taking college biology got a 1.84 point boost for each year of high school math. Taking high school biology got them an extra 1.35 points, but high school chemistry and physics had no significant effect.

And for physics, each year of high school math added 1.28 points, high school physics gave a 1.32 point boost, while high school biology and chemistry had no impact.
In other words, if you want to be a scientist, then study as much math as you can. You can learn science in college.

Update: A reader points out that maybe taking high school math and getting good college science grades are just both symptoms of a hardworking overachiever.

Freakonomics and Freedomnomics

I enjoyed the Freakonomics book, and now just just got its rival, Freedomnomics. Freakonomics is based on the work of hot-shot economist Steve Levitt, who is mainly famous for the theory that legalized abortion reduces crime. Freedomnomics is written by John Lott, who is best known for his study showing that guns deter crime.

Both Levitt and Lott have had their embarrassments. Levitt had to admit that his abortion study had a critical coding error that made the abortion/crime effect seem much larger. Lott once lost some data with a hard disc crash.

Levitt trashed Lott in his book and elsewhere, and now Levitt has had to retract some of what he said. Levitt put down a Lott article as not being peer-reviewed, when in fact it was and Levitt himself was one of the reviewers!

Update: Sailer has more info.

Update: Here is more on the abortion-crime dispute:
Further, in 2005, The Economist magazine reported that two economists with the Federal Reserve Bank of Boston determined that the Levitt-Donohue study was filled with statistical errors. The magazine opined, "To be politically incorrect is one thing; to be simply incorrect, quite another."

Levitt was traveling this week and deferred comment to Donohue.

In a written statement, Donohue did not comment on either study. He instead raised questions concerning Lott's research methods and said, "I am a social scientist, however, so Lott's behavior has in my mind, put him outside the bounds of scientific discourse."
I think that it is a little strange that Levitt uses a spokesman to make a personal attack on Lott, instead of just defending his own research.

Saturday, July 28, 2007

Rise of hedge funds

I thought that big company CEOs were overpaid, but read this:
A study by Steven N. Kaplan and Joshua Rauh finds that it's not corporate honchos who are filling up the ranks of the filthy rich. It's hedge fund managers. Or, as Kaplan and Rauh put it, "the top 25 hedge fund managers combined appear to have earned more than all 500 S&P 500 CEOs combined." The hedge fund guys are profiting not because there's been a shift in social norms favoring the mega-rich. It's just that a few superstars are now handling so much capital.
This New Yorker article says that the hedge funds are making their managers rich, but the funds aren't necessarily doing that well for investors.

Update: Here is an example of a hedge fund doing badly. The Harvard endowment invested $500 million in Sowood's Alpha hedge fund, and the fund lost most of its value just in the last month.

Thursday, July 26, 2007

Explaining Ohio domestic violence

The Ohio supreme court has dismissed a domestic violence indictment because of Ohio's Defense of Marriage Amendment. I expect the gay lobby to soon be arguing that resistance to same-sex marriage is allowing men to beat their girlfriends in Ohio.

Assault and battery are still crimes in Ohio, just as they are elsewhere. What is really peculiar about the law in Ohio and some other states is that the domestic violence lobby has convinced them to punish spousal battery more severely than stranger battery. I think that the idea is that betrayal by a spouse is somehow worse than being attacked by a stranger.

I don't really agree with the premises behind these domestic violence laws, but if you accept them, then the Ohio court ruling is not surprising. The court is merely saying that if the greater penalty is based on the partners being married, then the law requires that they really be married before that greater penalty is applied.

This court ruling is not an unintended consequence. It is merely distinguishing those who are married from those who aren't, and that is apparently what the people of Ohio wanted to do.

Tuesday, July 24, 2007

NASA doesn't think Moon landing was a hoax

Here is an interview with the head of NASA:
Question: I live about 200 miles down southeast of here, at the home of slapstick living and most of the folks down there are convinced that the Moon landing were a hoax, and wrestling’s real. [Laughter] Do you have any ideas on how to speak to these people?

Griffin: I don’t. [Laughter] [Applause] I mean, what can we say? I mean, all the guys who were still alive and a couple of the ones who are not have all become personal friends of mine, and in the intervening decades. I don't think they're lying to me. I just don’t. I don’t know what to tell people who say the Moon landings were a hoax.
Weak answer. He doesn't think that the astronauts are lying?! And how is he making friend with dead astronauts anyway?

Bad mom sues rink for neglecting her delinquent son

Here is a silly lawsuit:
Van Dusen is the mother of the notorious 10-year-old involved in the Amherst Pepsi Center dispute. Charles Schmidl and his 5-year-old son were at the center for a mid-April skate when, he said, the 10-year-old taunted him, shot hockey pucks at him and his son and dared him to do something about it.

Schmidl, a 38-year-old autoworker, said he searched for 40 minutes in vain for staff, security or the boy's parents before taking matters into his own hands. He dragged the obnoxious kid off the ice, at which point Charlene Van Dusen appeared and had him arrested.

Amherst cops say witness accounts largely confirm Schmidl's story. They also say the 10-year-old has a history of trouble at the rink and that his family is "familiar" to Amherst police, and not for the right reasons. ...

The plot thickened Friday, with the news that Van Dusen -- mother of the wayward kid -- intends to sue the Town of Amherst and its Recreation Department. She claims that the lack of supervision at the rink resulted in "physical injuries and emotional trauma" to her son.

Essentially, Van Dusen is suing the town for not doing her job.
Too bad no one is allowed to drag the mom out of town.

Monday, July 23, 2007

Companies cannot give H-1B numbers

A letter in today's Si Valley paper says:
H-1B visas driven by desire, not need

It is often stated that this nation is in dire need of highly skilled foreign workers. How would we know that? A recent article in the Mercury News (Page 1A, July 15) interviewed a number of people regarding H-1B visas. Among them were representatives for Cisco and Hewlett-Packard. Interestingly, none of them were able to state the number of visas needed. State and federal representatives were just as ill-informed.

Filing multiple visa applications for the same position or applying for an over-estimated number of visas suggests that the companies first get as many H-1B visa employees as they can and then fill the rest with the locals. Perhaps this is not the way the program should be used.

Zoltan Lukman
Santa Clara
He's right. The companies just want cheap labor. If the foreigners were really filling a labor shortage, then the jobs would go unfilled when the H-1B quotas are filled. But no, the companies are always able to hire local engineers when they cannot get the H-1B visas.

Sunday, July 22, 2007

Why do they hate us?

Pakistani-American Mohsin Hamid tries to explain:
The residue of U.S. foreign policy coats much of the world. It is the other part of the answer to the question, "Why do they hate us?" Simply because America has -- often for what seemed good reasons at the time -- intervened to shape the destinies of other countries and then, as a nation, walked away.
In other words, they hate us because we are not more imperialist in imposing our values on Third World countries. They realize that they'd be better off if America ruled the world.

Friday, July 20, 2007

Checkers is mostly solved

The NY Times reports:
Jonathan Schaeffer, a professor of computer science at the University of Alberta, set out on his checkers-playing quest in 1989, aiming to write software that could challenge the world checkers champion. He and his colleagues finished their computations 18 years later, in April.

"From my point of view, thank god it's over," Dr. Schaeffer said.

Even with the advances in computers over the past two decades, it is still impossible, in practical terms, to compute moves for all 500 billion billion board positions. Instead, the researchers took the usual starting position and then looked only at the positions that would occur during the normal course of play.

"It's a computational proof," Dr. Schaeffer said. "It's certainly not a formal mathematical proof."
Then it is not a proof at all. A proper computational proof would have examined all the possibilities, and proved that those really were all the possibilities. It appears that Schaeffer just computed end-game scenarios for some list of likely positions, and didn't prove anything at all about the other positions.

That is not a proof.

Here is another description of what they did:
The Alberta researchers exhaustively checked the final stages of play for any arrangement of 10 pieces or less, identifying which of the 3.9 x 1013 positions won, lost or drew.

Next they identified 19 representative opening sequences and searched through subsequent moves for the easiest-to-find connections to final positions—win, lose or draw. To save time, they ignored pointless back-and-forth moves or those that did not turn a draw into a win.
This doesn't sound like a proof to me.

Wednesday, July 18, 2007

German language is dying

Stefan Klein complains that the German language is on the verge of atrophy among scientists:
By now, English is the sole language used in lectures in 250 out of 1,976 advanced educational fields in Germany ("master's degree" programs). Should this development continue, it would mean certain death for German as a language of research.
German scientists speaking in Germany at a German conference to a mostly German audience will usually speak in English. He also complains about the difficulty in translating concepts like the "long-distance ghostly effect".

The trend is inevitable. English is the world's standard language.

Tuesday, July 17, 2007

America No Longer Tallest Country in the World

AP reports:
America used to be the tallest country in the world. From the days of the founding fathers right on through the industrial revolution and two world wars, Americans literally towered over other nations. ... But just as it has in so many other arenas, America's predominance in height has faded. ...

And those height differences translate into real benefits. A number of studies have shown that disease and malnutrition early in life -- the same things that limit a person's height -- increase a person's chances of developing heart disease and other life-shortening conditions later on. Though tall people are more likely to get cancer, they suffer less mortality overall than short people.

International statistics bear it out. Life expectancy in the Netherlands is 79.11 years; in Sweden it's 80.63. America's life expectancy of 78.00 years puts it in somewhat shorter company, just above Cyprus and a few notches below Bosnia-Herzegovina. ...

"American children might consume more meals prepared outside of the home, more fast food rich in fat, high in energy density and low in essential micronutrients," wrote Komlos and co-author Benjamin E. Lauderdale of Princeton University. "Furthermore, the European welfare states provide a more comprehensive social safety net including universal health care coverage."

In the United States, by comparison, an estimated 9 million children have no health insurance.
See also this 2004 New Yorker article, and this 2006 Komlos-Lauderdale paper, and this 2007 paper.

This is being used as an argument for socialized medicine.

The obvious explanation is that the USA has had massive Mexican and Oriental immigration. The study authors don't control for this directly, but have an argument that it is not the explanation. Besides, they argue, except for African pygmies, any population can grow as tall as any other.

I am not convinced. Some European countries have longer life expectancies also, but it is not necessarily because of better diets or better health care. The Dutch used to be shorter that other Euros, and now they are taller, but it doesn't seem to have anything to do with health care.

Monday, July 16, 2007

Poll shows liberals want to censor political web sites

A Rasmussen Poll reports:
Americans are evenly divided as to whether or not the government should "require all radio and television stations to offer equal amounts of conservative and liberal political commentary." The latest Rasmussen Reports national telephone survey found that 41% favor that proposal and 41% are opposed. The concept, known as the "Fairness Doctrine" in legislative circles, has been gaining ground on Capitol Hill ever since public opposition forced the Senate to back down on the immigration issue.
Immigration? The immigration bill failed when the Democrats who control Congress failed to cut off debate on the subject. They tried to get a vote with no debate, and their plan was ruined by a public debate on radio and elsewhere.
An interesting dynamic of the public debate is that liberals are more supportive of the "Fairness Doctrine" than conservatives. ... A separate survey found that the general public tends to believe that major broadcast networks have a liberal bias. ... Even those who are politically liberal are more likely to see the New York Times as biased in favor of liberals than conservatives.
So this liberal push for "fairness" appears to be driven by an attempt to censor opposing viewpoints. Here is the real kicker:
A large segment of the public would like to extend the concept of the Fairness Doctrine to the Internet as well. Thirty-four percent (34%) believe the government should "require web sites that offer political commentary to present opposing viewpoints." Fifty percent (50%) are opposed.
Wow. These folks really want the govt to censor the political content of web pages?! How would that even work?

Saturday, July 07, 2007

Origin of the zero

SciAm magazine reports:
The first recorded zero appeared in Mesopotamia around 3 B.C. The Mayans invented it independently circa 4 A.D. It was later devised in India in the mid-fifth century, spread to Cambodia near the end of the seventh century, and into China and the Islamic countries at the end of the eighth. Zero reached western Europe in the 12th century.
There was no year zero. The year 1 BC was followed by the year 1 AD. That is because the zero had not been invented yet. So when was it invented? It turns out that, as far as we know, it could have been invented in what would have been the year zero!

Friday, July 06, 2007

Decline of civilization

Pat Buchanan on McLaughlin Group today says:
The rise of women to power in a civilization is very often the mark of its decline.
Eleanor Clift is flabbergasted.

Ten pop psychology theories

Sailor recommends a fun pop psychology article with theories for:
Men like blond bombshells (and women want to look like them)

Most suicide bombers are Muslim

Having sons reduces the likelihood of divorce

Men sexually harass women because they are not sexist
Some of this sounds bogus to me, but the theories are amusing anyway.

Thursday, July 05, 2007

Spinning US Census data

WSJ reports:
Census demographer Martin O’Connell explained that the survey results are weighted by age, race and gender. If a 26-year-old black man is married to a 35-year-old white woman, and if 26-year-olds and blacks are underrepresented in the survey sample, then his interview may be weighted more heavily than hers. "There's no final step to make sure the number of husbands and wives is exactly equal," Mr. O’Connell told me. "If we did that, we would have to reweight the data again, and the data would endlessly spin around.” Nonetheless, he says of the American Community Survey numbers, "They are the best numbers you will ever get, any place."
They are already spinning the data.

There was some controversy over whether the 2000 Census should weight the data for Congressional apportionment. As you can see, spinning the data has some adverse consequences that are subject to political manipulation.

Tuesday, July 03, 2007

Females bear grudges

The John Horgan George Johnson video blog has good commentary on current scientific news. Among other things:

Females bear grudges more than males. My pet theory attributed this to boys playing team sports, but Horgan says that the same thing has been observed in monkeys, chimps, and bonobos.

The new studies are puzzling, and the old ones on things like the Flynn Effect still have no good explanation.

Evolution by natural selection is a tautology. There won't be any more paradigm shifts because evolution can absorb any truths that might be discovered.

Mindreading terrorists

Security technology writer Bruce Schneier writes:
the theory says that people infer the motives of an actor based on the consequences of the action. So people assume that the motives of a terrorist are wanton death and destruction, and not the stated aims of the terrorist group ... This certainly explains a great deal about the U.S.'s reaction to the 9/11 attacks. Many people -- along with our politicians and press -- believe that al Qaeda terrorism is different, and they're just out to kill us all.
The question is whether we should judge al Qaeda terrorists by what they say or what they do. The cited paper says that terrorists are ineffective because governments ignore the stated wishes of the terrorists. (Here is a sample of what al Qaeda says.)

I don't believe in mindreading, and I don't believe in assuming that the stated goals of a bunch of terrorists are their real goals. Why believe that a bunch of terrorists are telling the truth? If they are willing to murder innocent civilians to advance their agenda, then they are surely also capable of lying.

If ignoring the stated goals of the terrorists is causing them to be ineffective, then so much the better. Schneier seems to be arguing that this means that we are wrong about al Qaeda, but as I see it, it implies that we are right.

Sunday, July 01, 2007

Bomb ended the war

AP Tokyo reports:
The Japanese defence minister, Fumio Kyuma, came under fire for saying the US atomic bombings of Japan during the second world war were inevitable. In a speech on Saturday, Mr Kyuma said: "I understand that the bombing ended the war, and I think that it couldn't be helped."
He stated the obvious. Not only did the atomic bomb end the war, it saved millions of lives, kept Japan from Soviet domination, and helped keep the peace for 50 years.