Impact Factors

Frances Woolley makes a great point in Impact Factors and how they vary by field and by sub-field. This issue comes up in medical research all of the time. If I try and be really methodological, a statistics or epidemiology journal has a much smaller imapct factor than a medical journal. On the other hand, if I focus on the drugs and go to a pharmacy journal then impact factors are even lower.

So it is a constant dilemma about how to focus one's research. I would like to think that tenure and promotion committees are sophisticated about such issues but one does worry . . .

I suspect that the same issues are replicated with grant funding.

Airline Security

This is a nice point from Megan McArdle on the current value of airline security procedures:

Somehow, this seems like a questionable reaction to two attacks that failed. Especially since they failed for the same reason that any similar attack is likely to fail: the amount of explosives you can smuggle in your underwear or shoes is necessarily small, meaning that you need to be in the cabin to detonate them if you want to be sure that you'll bring the plane down. And it's really hard to set your underwear, or your shoes, on fire without your fellow passengers noticing. In Asia, I've never been required to have my shoes scanned--not even to get on a US bound flight. And yet, we have not been confronted with a rash of exploding planes out of Taipei or Saigon.

I am forced to fly a lot for work and I have the misfortune of living far away from family. Yet I would gladly never fly again if I could arrange it. It's really become a miserable process. At some point one really would like to avoid the whole mess.

History question of the day

This excellent column by Steven Pearlstein got me thinking, is hard-currency populism a new phenomena? I realize a lot separates Bryan and Palin/Beck/etc. but this would seem to be 180.

Maybe this is making outcomes worse?

From an interesting interview with a (formerly) obese cardiologist:

In 2003, Gary Foster, now director of the Center for Obesity Research and Education at Temple University, conducted a survey of primary-care doctors about obesity. More than half viewed obese patients as ugly and noncompliant. A third saw them as weak-willed and lazy.

and

He understands that doctors can be frustrated by their patients' failure to lose weight, but he sees it as no worse than many other difficult-to-treat diseases. "My fellow colleagues are understanding of cancer even when it recurs and recurs, and they'll say, 'Well, that's the disease,' " he said.

I think that this might have some serious ramifications for the treatment of obesity by the medical profession. Patients seem as noncompliant are less likely to receive treatment which can make the health risk associated with obesity more difficult to treat.

It's true that there is some factor that is driving the current obesity epidemic (and the even mroe cocnerning diabetes epidemic) that is, my definition, modifiable. But perhaps we should focus on what that factor is? After all, our grandparents were much less obese (on average) and that suggests that the focus shouldn't necessary be at the level of the patient but rather at how we have changed our culture.

Seniors in space

From 1996:

Now here are two seemingly unrelated facts.

Fact One: 30% of Medicare expenditures are incurred by people in the last years of their lives.

Fact Two: NASA spends billions per year on astronaut safety.

Maybe you see where I am going....

Why not shoot the elderly into space? Stay with me. Because I'm not just thinking about the budget here. I'm talking about science. Just think how many more manned space operations NASA could undertake if they didn't have to worry about getting the astronauts back.

Now, I'm not saying we don't try to get them back. We just don't make such a big deal about it. That way we don't have to use the shuttle every time, which is very expensive. Put an old Mercury capsule on top of a Saturn rocket, fire it up, and see what happens. And if the "Houston, we've got a problem" call comes, Mission Control can simply reply, "Best of luck. We're rooting for you."

-- Al Franken, in "Rush Limbaugh is a Big Fat Idiot"




From the AP:

Two scientists are suggesting that colonization of the red planet could happen faster and more economically if astronauts behaved like the first settlers to come to North America — not expecting to go home....

"You would send a little bit older folks, around 60 or something like that," Schulze-Makuch said, bringing to mind the aging heroes who saved the day in the movie "Space Cowboys."

You can read the actual proposal here but be warned, it does not improve under close scrutiny.

Things you can do if you're a charter school

Make demands of parents:

Parent involvement has been a surprisingly central feature of many of the proposals for new and restructured charter schools in California. State legislation mandates that parents be involved in the governance of charter schools, but broader kinds of parental involvement seem to be contemplated by those who are designing these new schools. In many of these schools, parent involvement is much more than simply a requirement to volunteer assistance or to help with their children's homework. Parents are seen as central adults in a more inclusive school community--participants who share time and expertise with the school's students as a whole. For example, a majority of the first chartered schools were planning to have parents and other community members as instructors in the school building, and several expected to sponsor training in tutoring methods and parenting techniques for use at home. In fact, a survey of thirty-four of the first forty-four schools chartered by the state found that in more than 50 percent parents are required to sign contracts and to participate in certain activities (Dianda & Corwin, 1994). One recently approved charter school, for example, intends to require parents "to volunteer a minimum of three hours per month at the school." Another stated in its charter: "Parents, by signing their child's registration form, commit themselves to at least 2 hours of school service per month.... Any student accepted on an above mentioned agreement will meet a prescribed written contract and will understand, if the contract is broken, said agreement will be revoked and the student will be disenrolled."

In many ways this is a good thing and I'm not saying that charter schools should stop, but it does provide another example of the near impossibility of making meaningful comparisons between schools that have the option of easily 'disenrolling' students and those that don't.

Is it too late tonight to write a post on the card game Eleusis?

Yes.

Something you have to post immediately if you see it at two a.m.

I'm generally not that impressed with this sort of thing but obviously I have to post this.

From Psychology Today:

There is thus no indication in any of the ethnographic evidence that any sustained nocturnal activities occur in traditional societies, other than occasional conversations and singing, in these tribes. It is therefore reasonable to infer that our ancestors must also have limited their daily activities to daylight, and sustained nocturnal activities are largely evolutionarily novel. The Hypothesis would therefore predict that more intelligent individuals are more likely to be nocturnal than less intelligent individuals.

Not much time for blogging...

... or even reading other people's blogs. I've been busy and based on the conversations I've had recently, Joseph is inundated.

Regular blogging will resume soon. Till then you can kill some time with the only popular board game I know of* that has imperfect information but no random elements.


* And no, I wouldn't call this a popular board game.

More wheelbarrows

I suggested in a previous post that the surreal images and anecdotes of hyper-inflation (particularly in early Twenties Germany) have a more powerful hold on the imagination than do the deceptively benign images of deflation.

Today I learned (via Media Matters by way of Krugman) that hyper-inflation warnings are showing up on Glenn Beck's show:

As I told you at the beginning I'm not an expert. I am not a guy to listen to for financial advice. I'm an American with an opinion. Period. I don't know that the Weimar republic isn't going to happen here. I don't know if it is going to happen here. But I will tell you this: Those same damn experts told me two years ago that the Fed wouldn't do what they did yesterday. those same damn experts told me four years ago the housing market was fine. When will these experts lose a little bit of credibility? When will we start listening to our own guts, and to common sense?

While on another Fox show the host is calling for a return to gold not only to back the dollar but as an alternative currency.

Observational Research

Andrew Gelman quotes a discussion the design of clinical trials by John Langford. It's based on work by Amy Harmon on people who fail to get the drug in the active arm of a randomized controlled trial.

The case that she begins with:

But when Mr. Ryan, 22, was admitted to the trial in May, he was assigned by a computer lottery to what is known as the control arm. Instead of the pills, he was to get infusions of the chemotherapy drug that has been the notoriously ineffective recourse in treating melanoma for 30 years.

The question is:

With reasonable record keeping of existing outcomes for the standard treatments, there is no need to explicitly assign people to a control group with the standard treatment, as that approach is effectively explored with great certainty.

I see this as an argument for Observational (instead of experimental) research. The reason that experimental research is valuable is that you can rule out confounding and estimating quantities like the counter-factual outcomes is a lot easier (as counterfactuals have the nice properties of being missing completely at random).

If you only test the novel drug in the trial then you are assuming that there are not other changes (in population composition, in other forms of care) that do not explain some or all of the variability. You can use statistical adjustment to reduce differences but then things hinge on model specification and whether there are unmeasured differences.

I worry that we are some distance from trusting data where two analysts can get two different answers making different sets of assumptions (which years are the control years? who is included? which covariates are included?).

That being said, many key breakthroughs are entirely observational in nature and it would be good to see more use of this study design. But it is also clear that there are a lot of blind alleys that occur when non-randomized data is used (statins and cancer, anyone?).

[In a larger sense, I think that this is the danger of outside experts. Sometimes they point out that the Emperor has no clothes. However, they may not realize that is because it is currently bath-time . . . ]

The short, sad life of Toxie the toxic asset

Another exceptional example of financial journalism from This American Life/Planet Money. It's a remarkably entertaining hour but it still manages to be as or more informative as anything you're likely to read on the subject.

As always, the download is free but the show could really use the money if you care to donate a couple of bucks.

Bad polls, good polls, better polls

You might have expected Nate Silver to sleep through the rest of the week after the elections, but the Red Bull must flow freely at 538. Anyone who is serious about polls should read these two recent posts:

Rasmussen Polls Were Biased and Inaccurate; Quinnipiac, SurveyUSA Performed Strongly

and

When ‘House Effects’ Become ‘Bias’

Even more interesting is Mark Blumenthal's article on the remarkable accuracy of both camps internal polling in Nevada.

Not All Polls Were Wrong In Nevada

The Wheelbarrow Problem -- is the imagery of deflation influencing the debate?

Many economists (most notably Paul Krugman) have been arguing that deflation is a more immediate and far greater danger than inflation and recent data seems to back them up, but they don't seem to be getting much traction in the press. I wonder if part of this failure comes from the fact that the imagery of deflation seems so benign ("That'll be a nickel for the soda pop, sonny.") while the imagery of hyper-inflation is frightening and often surreal.

Here's how George Goodman put it in Paper Money:

In retrospect, you can trace the steps to hyperinflation, but some of the reasons remain cloudy. Germany abandoned the gold backing of its currency in 1914. The war was expected to be short, so it was financed by government borrowing, not by savings and taxation. In Germany prices doubled between 1914 and 1919.

After four disastrous years Germany had lost the war. Under the Treaty of Versailles it was forced to make a reparations payment in gold-backed Marks, and it was due to lose part of the production of the Ruhr and of the province of Upper Silesia. The Weimar Republic was politically fragile.

But the bourgeois habits were very strong. Ordinary citizens worked at their jobs, sent their children to school and worried about their grades, maneuvered for promotions and rejoiced when they got them, and generally expected things to get better. But the prices that had doubled from 1914 to 1919 doubled again during just five months in 1922. Milk went from 7 Marks per liter to 16; beer from 5.6 to 18. There were complaints about the high cost of living. Professors and civil servants complained of getting squeezed. Factory workers pressed for wage increases. An underground economy developed, aided by a desire to beat the tax collector.

On June 24, 1922, right-wing fanatics assassinated Walter Rathenau, the moderate, able foreign minister. Rathenau was a charismatic figure, and the idea that a popular, wealthy, and glamorous government minister could be shot in a law-abiding society shattered the faith of the Germans, who wanted to believe that things were going to be all right. Rathenau's state funeral was a national trauma. The nervous citizens of the Ruhr were already getting their money out of the currency and into real goods -- diamonds, works of art, safe real estate. Now ordinary Germans began to get out of Marks and into real goods.

Pianos, wrote the British historian Adam Fergusson, were bought even by unmusical families. Sellers held back because the Mark was worth less every day. As prices went up, the amounts of currency demanded were greater, and the German Central Bank responded to the demands. Yet the ruling authorities did not see anything wrong. A leading financial newspaper said that the amounts of money in circulation were not excessively high. Dr. Rudolf Havenstein, the president of the Reichsbank (equivalent to the Federal Reserve) told an economics professor that he needed a new suit but wasn't going to buy one until prices came down.

Why did the German government not act to halt the inflation? It was a shaky, fragile government, especially after the assassination. The vengeful French sent their army into the Ruhr to enforce their demands for reparations, and the Germans were powerless to resist. More than inflation, the Germans feared unemployment. In 1919 Communists had tried to take over, and severe unemployment might give the Communists another chance. The great German industrial combines -- Krupp, Thyssen, Farben, Stinnes -- condoned the inflation and survived it well. A cheaper Mark, they reasoned, would make German goods cheap and easy to export, and they needed the export earnings to buy raw materials abroad. Inflation kept everyone working.

So the printing presses ran, and once they began to run, they were hard to stop. The price increases began to be dizzying. Menus in cafes could not be revised quickly enough. A student at Freiburg University ordered a cup of coffee at a cafe. The price on the menu was 5,000 Marks. He had two cups. When the bill came, it was for 14,000 Marks. "If you want to save money," he was told, "and you want two cups of coffee, you should order them both at the same time."

The presses of the Reichsbank could not keep up though they ran through the night. Individual cities and states began to issue their own money. Dr. Havenstein, the president of the Reichsbank, did not get his new suit. A factory worker described payday, which was every day at 11:00 a.m.: "At 11:00 in the morning a siren sounded, and everybody gathered in the factory forecourt, where a five-ton lorry was drawn up loaded brimful with paper money. The chief cashier and his assistants climbed up on top. They read out names and just threw out bundles of notes. As soon as you had caught one you made a dash for the nearest shop and bought just anything that was going." Teachers, paid at 10:00 a.m., brought their money to the playground, where relatives took the bundles and hurried off with them. Banks closed at 11:00 a.m.; the harried clerks went on strike.

The flight from currency that had begun with the buying of diamonds, gold, country houses, and antiques now extended to minor and almost useless items -- bric-a-brac, soap, hairpins. The law-abiding country crumbled into petty thievery. Copper pipes and brass armatures weren't safe. Gasoline was siphoned from cars. People bought things they didn't need and used them to barter -- a pair of shoes for a shirt, some crockery for coffee. Berlin had a "witches' Sabbath" atmosphere. Prostitutes of both sexes roamed the streets. Cocaine was the fashionable drug. In the cabarets the newly rich and their foreign friends could dance and spend money. Other reports noted that not all the young people had a bad time. Their parents had taught them to work and save, and that was clearly wrong, so they could spend money, enjoy themselves, and flout the old.

The publisher Leopold Ullstein wrote: "People just didn't understand what was happening. All the economic theory they had been taught didn't provide for the phenomenon. There was a feeling of utter dependence on anonymous powers -- almost as a primitive people believed in magic -- that somebody must be in the know, and that this small group of 'somebodies' must be a conspiracy."

When the 1,000-billion Mark note came out, few bothered to collect the change when they spent it. By November 1923, with one dollar equal to one trillion Marks, the breakdown was complete. The currency had lost meaning.

What happened immediately afterward is as fascinating as the Great Inflation itself. The tornado of the Mark inflation was succeeded by the "miracle of the Rentenmark." A new president took over the Reichsbank, Horace Greeley Hjalmar Schacht, who came by his first two names because of his father's admiration for an editor of the New York Tribune. The Rentenmark was not Schacht's idea, but he executed it, and as the Reichsbank president, he got the credit for it. For decades afterward he was able to maintain a reputation for financial wizardry. He became the architect of the financial prosperity brought by the Nazi party.

Obviously, though the currency was worthless, Germany was still a rich country -- with mines, farms, factories, forests. The backing for the Rentenmark was mortgages on the land and bonds on the factories, but that backing was a fiction; the factories and land couldn't be turned into cash or used abroad. Nine zeros were struck from the currency; that is, one Rentenmark was equal to one billion old Marks. The Germans wanted desperately to believe in the Rentenmark, and so they did. "I remember," said one Frau Barten of East Prussia, "the feeling of having just one Rentenmark to spend. I bought a small tin bread bin. Just to buy something that had a price tag for one Mark was so exciting."

All money is a matter of belief. Credit derives from Latin, credere, "to believe." Belief was there, the factories functioned, the farmers delivered their produce. The Central Bank kept the belief alive when it would not let even the government borrow further.

But although the country functioned again, the savings were never restored, nor were the values of hard work and decency that had accompanied the savings. There was a different temper in the country, a temper that Hitler would later exploit with diabolical talent. Thomas Mann wrote: "The market woman who without batting an eyelash demanded 100 million for an egg lost the capacity for surprise. And nothing that has happened since has been insane or cruel enough to surprise her."

update: restored the missing words in the title.

Not quite grasping the placebo concept

Marcus A. Winters writing in the Manhattan Institute's City Journal says:

The charters’ random-selection policy has allowed Hoxby’s team to evaluate them with a randomized field trial (RFT), the kind of experiment often described by social scientists as the “gold standard” in research design. RFTs resemble the clinical trials performed by pharmaceutical companies, which test a new drug by assembling a pool of subjects, randomly assigning subjects to receive either the drug or a placebo, and then comparing the subjects’ condition. Similarly, Hoxby took a pool of subjects (students applying to New York City charter schools); took advantage of the random nature of the lotteries, which assigned the subjects either to charters or to traditional public schools; and then compared their academic achievement. Because access to a charter school is the only meaningful difference between the two groups—both applied to charter schools, after all—comparing the groups’ later achievement really will tell us what effect charters have on student performance.

I've seen researchers and commentators ignore potential issues with open-label studies (see here and here for background), but Mr. Winters is the first not only to mention placebos but to compare them explicitly to public schools. At the risk of beating a man while he's down, the use of placebos is based on the assumption that knowing that you are being treated can and often does produce a "meaningful difference between the two groups."

Caroline Hoxby never mentioned placebos in this paper and I'm certain she would never suggest that being randomly assigned a public school is equivalent to receiving a placebo in a double blind test.

Unfortunately Dr. Hoxby is not the one writing the articles.