Thursday, December 20, 2007

Direct Instruction

Liam Julian makes a useful comparison here between Direct Instruction -- a curriculum that has proven advantages but that, as I noted a couple of years ago, is nonetheless criticized by the likes of Jonathan Kozol for being too "scripted" -- and the practice of medicine:
It is foolish to believe that big-hearted 22-year-olds will know, intrinsically, the best way to teach reading to a class of second graders, just as it is foolish to think that newly minted doctors can on their own derive the best way for treating a particular pathology. Surely, though, veteran teachers can teach without rigid, Direct Instruction curricula? Not necessarily. A comparison to the medical profession suggests that even the most grizzled teachers (and their students) may benefit greatly from scripted procedures.

In the December 10th New Yorker, Atul Gawande writes about the intensive care units of hospitals, which specialize in saving people whose bodies have undergone seemingly untreatable damage. But it isn't easy. A study by Israeli scientists found, Gawande writes, "that the average [I.C.U.] patient required 178 individual actions per day, ranging from administering a drug to suctioning the lungs, and every one of them posed risks." And although the nurses and doctors in the study erred in just one percent of those 178 actions, in such critical situations, even the smallest mistakes can have disastrous consequences.

In 2001, Peter Provonost, a critical-care specialist at Johns Hopkins, decided to try making a basic procedural checklist for I.C.U. doctors. His first list was focused on tackling one problem, line infections, which are common in the I.C.U. and can be deadly. Provonost hypothesized that he could curtail such infections by concentrating on simple practices--washing hands, cleaning the patient's skin, etc.--that might easily be overlooked in hectic I.C.U. environments. He convinced the hospital administration to implement the checklist, and "to authorize nurses to stop doctors if they saw them skipping a step."

After a year with the checklist, "the ten-day line-infection rate went from eleven per cent to zero." In one hospital, "the checklist had prevented some 43 infections and eight deaths, and saved two million dollars in costs."

Direct Instruction attempts to do something similar in schools--to make a checklist of sorts for reading instruction, so the basics don't go overlooked. * * *


Monday, December 17, 2007

Eva's Note

The other night, my wife was trying to practice a few Christmas carols with the kids. My daughter Eva, who turned 6 in November, was being very whiny and was trying to bang on the piano and disrupt the singing. So my wife told her to leave while she was acting like that. Eva disappeared upstairs. A few minutes later, she came back with the following note written on a dryer sheet. Spelling's not so great, but I thought it was a cute note (esp. the end lamenting her "sad hart"):

Saturday, December 15, 2007

Notes on Blogging

Julian Sanchez has some good thoughts here:
This probably sounds a little odd coming from me, but a lot of the habits blogging implants really are pretty destructive. I've obviously decided it's worth it to keep doing it, on net, but I try to remind myself of all the unhealthy tendencies blogging encourages. Most obviously, it is just absolute poison for a writer to get too accustomed to reading and writing in chunks that average 300-500 words. As you get hooked on the instant gratification of firing off "pieces" that each take a half hour, your inclination and facility at crafting sustained arguments really does get degraded. This is compounded by the bloggy focus on timeliness: It always feels as though the most vital thing you could possibly be writing about is whatever all the other bloggers are discussing right this second.
I posted this in the comments:
Heresy! If an intelligent human being asks himself, "What is the most important task to which I could apply my mind at this time," the answer should always be: Whatever other bloggers are frenetically discussing. What separates us from the apes, after all, is that we write out lengthy opinions on Scott Thomas Beauchamp, Huckabee's latest quip, Clinton's press release about Obama's kindergarten paper, etc., etc. This is all going to be of long-lasting significance to the American project, if not the universe.

Tuesday, December 11, 2007

Thought-Provoking Posts

I meant to recommend these earlier:

1. Joe Carter: "Our Tortured Silence: The Shameful Response of Christians to Waterboarding," and the followup, "Anthropos and Enemy: Further Thoughts on Waterboarding."

2. David Friedman's reflections on his experiences with unschooling: here and here. A good but unorthodox observation about "the standard model of K-12 schooling, public and private:"
1. That model implicitly assumes that, out of the enormous body of human knowledge, there is some subset that everyone should study and that is large enough to fill most of thirteen years of schooling. That assumption is clearly false. Being able to read and do arithmetic is important for almost everyone. Beyond that, it is hard to think of any particular subject which there is a good reason for everyone to study, easy to think of many subjects outside the standard curriculum which there are good reasons for some people to study.

Monday, December 10, 2007

Vouchers in Florida -- More Evidence of Improvement

A team of respected education researchers -- Cecilia Elena Rouse, Jane Hannaway, Dan Goldhaber, and David Figlio -- has just released a study of how public schools improved due to Florida's accountability program (which, among other things, allowed students trapped in failing public schools to obtain vouchers).* The study is called, "Feeling the Florida Heat? How Low-Performing Schools Respond to Voucher and Accountability Pressure." It's based on what the authors deem a "remarkable" five-year survey that attempted to elicit responses from every public school in Florida, "coupled with detailed administrative data on student performance."

The main finding:
We analyze the impact of the accountability system on Florida’s students and schools using a three-part analysis. First, we estimate the effect of the accountability system and the threat of becoming voucher eligible on student test score performance, both in the short-run and in the longer term. Second, we study the effects of the reform on school policies and practices. Finally, we attempt to determine if the policies appear to affect student achievement or explain the change in student performance. We find that student achievement significantly increased in elementary schools that received an “F” grade by between 6 to 14 percent of a standard deviation in math and between 6 to 10 percent of a standard deviation in reading in the first year. Three years later the impacts persist.

Importantly, we also detect specific school policy changes implemented by the schools that explain part of these increases. Specifically, when faced with increased accountability pressure, schools appear to focus on low-performing students, lengthen the amount of time devoted to instruction, adopt different ways of organizing the day and learning environment of the students and teachers, increase resources available to teachers, and decrease principal control. These, combined with other policies, explain more than 15 percent of the test scores gains of students in reading and over 38 percent of the test scores gains of students in math, depending on the model specification. As such we find evidence that schools respond to accountability pressure in educationally meaningful ways.
This seems to be an important confirmation of earlier results by Jay Greene and Marcus Winters, Chakrabarti (three papers: here, here, and here), Sandstrom and Bergstrom, Belfield, Auguste and Valenzuela, West and Peterson, and Caroline Hoxby.

Of course, if vouchers put pressure on failing public schools to get their acts together, then that could complicate any attempt to figure out whether the voucher students had improved. They may indeed have improved, but if the public schools improved to the same degree, it could misleadingly look as if there were no effect from vouchers at all.

*The voucher aspect of the Florida program was struck down by the Florida Supreme Court on the absolutely Orwellian ground that it violated the Florida constitution's guarantee that the state should provide a "uniform, efficient, safe, secure, and high quality system of free public schools." I say Orwellian because vouchers would never have been triggered in the first place unless a given public school had hopelessly failed to meet those qualifications.


Monopolies in Network Industries

About four years ago, I wondered whether monopolists in network industries face the same incentives as monopolists elsewhere. George Priest of Yale Law School has just released a useful and short analysis of similar topics:
Rethinking Antitrust Law in an Age of Network Industries

Yale Law School

Yale Law & Economics Research Paper No. 352

Economists have recognized the increasing role of network industries in our modern economy and have substantially advanced the understanding of network economics. This paper discusses how the special economic features of networks and, in particular, practices that networks adopt to enhance network benefits, requires a reconceptualization of modern antitrust analysis. The proposition is demonstrated by the example of several recent antitrust prosecutions of network practices where the economics of networks were largely ignored. The paper also discusses many cases in the antitrust canon that are more adequately analyzed when the network character of the practice is taken into account. The paper propose a reorganization of antitrust analysis to distinguish the fundamental economic analysis of network practices from the analysis of horizontal and vertical industrial practices.

Monday, December 03, 2007

Review of Carr's The Big Switch

Nicholas Carr has written an excellent book, The Big Switch: Rewiring the World, from Edison to Google,* which will be released on Jan. 7, 2008. The grand theme is how the world will change due to the ever-more-pervasive presence of computers. Under that grand theme, the book is then divided into two main parts.

In the first half of the book, Carr contends that the information technology departments of large businesses are about to become obsolete, as are large software companies like Microsoft (one chapter title: "Goodbye, Mr. Gates" -- I wonder how many readers will be aware of the allusion . . .). This is because computers are about to undergo the same transformation that happened with electricity a hundred years ago, an interesting story that Carr recounts. Whereas lots of individual firms used to produce their own electricity, they stopped doing so as soon as it became easier for electric companies to transmit power over longer distances. After all, if you're trying to make clothing or cars, why bother being an electric company unto yourself as well? The same thing is already happening as to computers, both hardware and software. It's becoming more feasible for relatively centralized companies to offer computer services to multiple other companies, which will eventually see that it's not worth the trouble to maintain their own set of hard drives, software, etc. In Carr's words, "Much of the traditional hardware business would simply disappear." (p. 79).

One warning that Carr makes, however, is that while the transformation of electricity into a centralized utility encouraged the production of many other things that depend on electricity (hence making possible the rise of the American middle class), the transformation of computers into a utility may not have the same effect. In fact, it may be more likely to exacerbate the "winner-take-all" trend of growing inequality. YouTube and Flickr are good examples discussed here: The creators of those services became enormously wealthy, but most of the actual material was supplied for free by hobbyists (who, needless to say, did not become wealthy). As Carr says, "the World Wide Computer provides an incredibly efficient mechanism for harvesting the economic value of the labor provided by the very many and concentrating it in the hands of the very few." (p. 143).

* * *

The second half of Carr's book is about the way that computers are used, and how this will transform society. One good example: Because of computers, intellectual content is more and more offered on an unbundled basis. (An earlier blog post by Carr touched on this topic.)

For example, you don't have to buy cable; instead, you can download individual TV shows (that's what I do). You don't have to subscribe to an entire newspaper to get the news; you can read individual articles that show up on blogs or Google (and advertisers are more and more able to target their advertising only to those articles that are popular). You don't have to buy full albums; you can buy individual songs on ITunes.

The crucial point is that when unbundling becomes more common, cross-subsidization must go down. The sports pages (popular) are not as likely to be cross-subsidizing the international reporter stationed in Zimbabwe (not popular). As a result, mainstream outlets are going to have a harder time paying for anything that doesn't have broad popularity and that doesn't specifically draw in advertisers on its own -- including international news or niche interests. As Carr says, "When bundled into a print edition, hard journalism can add considerably to the overall value of a newspaper. . . . Online, however, most hard journalism becomes difficult to justify economically. Getting a freelance writer to dash off a review of high-definition television sets -- or, better yet, getting readers to contribute their own reviews for free -- would produce much more attractive returns." (p. 155).

In other words, Carr is making the opposite of Chris Anderson's "Long Tail" argument. Or rather, he's showing where Anderson's argument doesn't apply. Anderson has a point when he says that companies like Amazon or Netflix can make a lot of money from the "long tail" of products, in which the sheer number of not-individually-popular books or movies can aggregate to large sales. But it's hard to see how that argument applies where the "long tail" had depended for its existence on cross-subsidization by the more popular products, and where the cross-subsidization is coming to an end.

Indeed, I wonder if this "unbundling" trend will ever hit the universities, which (I would guess) currently involve massive cross-subsidization from departments that are economically profitable to those that aren't. The only reason that universities can impose such unbundling right now is because they have a monopoly on an important credential, and can therefore tell accounting majors that they will not receive the credential of an accounting degree unless they take a course in American literature along the way. But as universities become more expensive (Baumol's cost disease, perhaps?), there will be an incentive for people to gravitate to alternate credentialing systems that are cheaper because they don't cross-subsidize uneconomical departments. That assumes that the alternate credentialing systems will be credible, of course, and this is all very speculative on my part, but we'll see.
* * *

Carr ends the book with questions about whether it will "change the way we think" as we "come to rely ever more heavily on the Internet's vast storehouse of information as an extension of or even a substitute for our own memory"? "As we put ever more intelligence into the Web, will we, individually, become more intelligent, or less so?" (p. 225). In a McLuhan-esque passage that I enjoyed, Carr says:
The medium is not only the message. The medium is the mind. It shapes what we see and how we see it. The printed page, the dominant information medium of the past 500 years, molded our thinking through, as Neil Postman has written, 'its emphasis on logic, sequence, history, exposition, objectivity, detachment, and discipline.' The emphasis of the Internet, our new universal medium, is altogether different. It stresses immediacy, simultaneity, contingency, subjectivity, disposability, and, above all, speed. The Net provides no incentive to stop and think deeply about anything . . . It's easier, as Kelly says, 'to Google something a second or third time rather than remember it ourselves.' On the Internet, we seem impelled to glide across the slick surface of data, as we make our rushed passage from link to link. (p. 227).
* * *

A few final tidbits:

1. I liked a "bootleggers and Baptists" story that Carr told about the adoption of alternating current as the standard for electricity. At the time, Edison opposed alternating current. Says Carr, "Edison, still convinced of the superiority of his own direct-current system, tried to magnify those fears by launching a grisly public relations campaign aimed at having high-power AC systems banned. Teaming up with an electrician . . ., he helped stage a series of public executions of animals, including dogs, cows, and horses, using the current from AC dynamos. He even convinced the New York legislature to purchase an AC generate . . . for use in the electrocution of death-row prisoners." (p. 39).

2. The one pet peeve I had about the book -- at least in the pre-publication version -- is the endnoting system, or rather, the almost complete lack of one. I check endnotes all the time when reading a book. Here, however, there are no numbers in the text signaling where an endnote occurs. And the endnotes themselves are listed only under "Chapter 1," "Chapter 2," and so on -- no reference to the textual page numbers (e.g., "Endnotes for pages 1-22), and no numbers corresponding to the endnotes. Instead, the endnotes are identified only by quotes from the text (e.g., "With less than ten people: . . . "). Thus, if I'm reading along in the text and wonder where a certain quotation or fact came from, I have to: 1) flip to the endnotes for that chapter; and then 2) read most or all of the endnotes for that chapter to see if anything corresponds to the text I had been reading (or at least read until the point where nothing seems familiar, and I realize that the endnotes correspond to a point in the chapter that I haven't reached yet).

I don't understand why publishers do this -- the only reason to have notes is to make it more convenient for the reader to check up on certain points. Why do publishers so far out of their way to make the reader jump through needless hoops in order to figure out what endnote goes with what bit of text?
UPDATE: I'm informed by Mr. Carr that the publication version will identify the endnotes by the page numbers from the text.

3. Also there was no index.

** My pre-publication copy is subtitled, "Our New Digital Destiny."

Sunday, December 02, 2007

Fake Advertisement

Inspired by Dr. Boli, here's a fake ad:
Christmas Card Blocking Now Available

Ever gotten a Christmas card on Dec. 24, and then panicked when you suddenly realized that it was from a mere acquaintance or even an enemy that you did not put on your own Christmas card list? Even if you reciprocate, the card will arrive so late that the other person will know you never intended to send them a card in the first place.

The Postal Service's new Christmas Card Blocker Service ensures that you'll never suffer that feeling again. Just provide the Postal Service with your own Christmas card list by Dec. 1, and then if anyone not on your list tries to send you a Christmas card, it will be blocked and returned to the sender stamped "Addressee Unknown."

This handy service is available for a one-time fee of only $19.95.