A Little Knowledge

Ezra Klein links to the video of lunatic CNBC business reporter Rick Santelli going on what appears to be an entirely sincere rant about the stimulus package and the Obama administration's plans to help distressed homeowners. As Ezra notes, "Santelli sells himself as a sort of financial sector Howard Beale: He's mad as hell, and he's not going to take it anymore. The problem is he's on the trading floor of a stock exchange surrounded by the very masters of the universe who started this mess."

At one point in the video, the conversation turns to idea of helping homeowners in danger of foreclosure refinance at lower rates. Santelli is having none of this, despite the fact that higher-than-projected mortgage default rates are what led to the rapid devaluation of allegedly investment-grade securities held by large financial institutions whose staggering blindness to such risk led to insolvency, frozen credit, a great recession (if we're lucky) and the end of the financial world as we know it. 

At that moment, the trading floor guy to Santelli's left turns and says "How about we all stop paying our mortgage? It's a moral hazard!"

The concept of "moral hazard" has itself become a hazard, I think. Like a lot of ideas rooted in economics, it's one of those insights into human behavior that isn't entirely obvious until somebody explains it to you, and then it seems like new window onto the world. Wikipedia defines it as "the prospect that a party insulated from risk may behave differently from the way it would behave if it were fully exposed to the risk." It's theoretically applicable to all kinds of situations: If drivers feel like airbags will protect them in a crash, they'll drive more recklessly, reducing the net injury-reducing benefits of airbags. If you forgive someone's debt, they'll borrow more recklessly, increasing the chance of future default. And so on. It's a useful perspective, something worth keeping in mind.

The problem is that "moral hazard" has for many people been elevated to the level of F=MA as an iron-clad law of the universe and as such is used to justify all kinds of cruel and spurious positions. It's aparently why we didn't bail out Lehman Brothers, which in retrospect looks like a catastrophically bad call. It's poorly understood by many as a catch-all justification for never offering any kind of helping hand. Empirically speaking, there's less evidence to support it than one might think--developing countries whose debt is forgiven don't actually have a horrible track record of subsequent reckless borrowing, for example. Air bags don't actually turn lots of safe drivers into reckless ones, because it still really sucks to crash your car. 

There's an educational lesson in all of this: picking up conceptual bits and pieces here and there without understanding the larger theory and complexities thereof can leave you worse off than if you'd never picked up the bits and pieces to begin with, because you think you understand more when you really understand less. 

The Sugar

If you drive away from the center of Washington, DC on East Capitol Street, around RFK Stadium and across the Anacostia River, you eventually come to the intersection with Benning Road, and a Denny's. For the last decade, the pancake emporium has been the only-sit down restaurant in Ward 7 and as such a symbol of the struggle to bring economic development and a decent life to the mostly low-income DC residents east of the river. This is one of the real dillemmas of urban development: people in low-income neighborhoods like Wards 7 & 8 may have little money per-capita, but they still have a fair amount of money collectively, since there are a lot people living there. And so businesses locate there to serve that market. But they tend not to be very good businesses from a quality-of-life standpoint; if you drive up and down Benning Road you see a lot signs for businesses that conveniently sell multiple things that are bad for you in the same place, e.g. "Checks Cashed, Instant Tax Refund" or "Lottery / Beer / Liquor" or  "Burritos, Chinese Take-Out, Fried Shrimp." 

The city of DC has been taking steps in recent years to try and fix this, as with a new government-subsidized shopping complex in Ward 8 that recently opened up, complete with a Giant supermarket, a Wachovia bank, and other sorts of businesses that one might find in the more affluent parts of the city and surrounding suburbs. And directly adjacent to the complex there's a brand-new IHOP that was filled with customers at 10:30 this morning, all eating at the first sit-down restaurant the area has had in a long time. 

But there's another business thriving in the various run-down strip malls east of the river: dialysis centers. Wards 7 & 8 appear to have been struck by the diabetes epidemic that is afflicting communities nationwide. And the only two sit-down restaurants east of the river, parking lots full because these are the only options the free market provides, are in the business of selling their customers liquid sugar. 

All of which is to say that it's complicated, this business of understanding and managing the intersection of market forces, private capital, consumer choice, public infrastructure, and multiple concerns of health, employment, and quality of life. And there are implications for public education, particularly as the public, non-profit, and private education sectors increasingly co-mingle. You'll see see more from us on this topic in the coming months. 

Love Your Children, Go To Jail

Via Eduwonk by way of DFER, the story of Yolanda Hill, a Rochester mother of five who has been shackled and thrown in prison for enrolling her children in a good school system:

Greece [school district] officials hired a private investigator to look into Hill's claim that her children lived with their grandmother. According to his report, over four months this school year, Hill was seen driving her kids each morning from her home on Morrill Street [in an adjacent disrtict] to her mother's home, where they would board buses for various Greece schools. The school district says that the education provided for the children due to the filing of false paperwork was worth $28,000.

This makes perfect sense when we start with a society that's unusually and increasingly stratified by income, with residential patterns to match, and say "Hey, let's draw lines around our gated enclaves of privilege and create school districts that look exactly the same!" In a decent society that takes educational opportunity seriously, it's utterly insane. 

This also raises some important points about "parental involvement," the lack of which is frequently cited as a reason to not expect too much from schools that enroll many poor and minority children. Ideally, it would be great if parents were able to invest a lot of time in helping their children learn. But if you're a single parent who didn't get a very good education when you were in school, and you have five children, and you're struggling to put food on the table by holding down multiple low-paying jobs (if you're lucky, given skyrocketing unemployment) then the best thing you can do for your kids--the best way to be parentally involved--isn't to spend three hours a night helping with homework or bake cupcakes for the PTA but to get your children into a good school, a school that has the resources and staff to give your children what you can't. 

See here for some ideas about how to fix the financial side of things and here for a look at breaking down barriers between districts.  

Questionable Odds

Nate Silver became justly famous for making the transition from sabermetrics to election predictioneering, but color me skeptical of this New York magazine article that includes his take on the upcoming Oscars. His statistical model, it says:

...involved building a huge database of the past 30 years of Oscar history. Categories included genre, MPAA classification, the release date, opening-weekend box office (adjusted for inflation), and whether the film won any other awards. We also looked at whether being nominated in one category predicts success in another. For example, is someone more likely to win Best Actress if her film has also been nominated for Best Picture? (Yes!) But the greatest predictor (80 percent of what you need to know) is other awards earned that year, particularly from peers (the Directors Guild Awards, for instance, reliably foretells Best Picture). Genre matters a lot (the Academy has an aversion to comedy); MPAA and release date don’t at all. A film’s average user rating on IMDb (the Internet Movie Database) is sometimes a predictor of success; box grosses rarely are.

So that's the six major Oscars (the four acting awards plus picture and director) with five nominees per category multiplied by 30 years multiplied by, what, 10 categories of data? 9,000 discrete pieces of information, total? That's not a huge database, that's a medium-sized Excel spreadsheet. Which may explain the improbable odds, e.g. that Slumdog Millionaire has a 99.0 percent chance of winning Best Picture while The Curious Case of Benjamin Button has a 0.0 percent chance.  That's just goofy; anyone who's watched the Oscars faithfully through the years, and yes I admit to this personal shame, knows that the academy is more fickle and unpredictable than that. Anne Hathaway has a 0.0 percent chance of winning Best Actress? Because Oscar voters are famously averse to giving awards to beautiful young actresses? Taraji P. Henderson is four times likelier than Viola Davis to win Best Supporting Actress, because Benjamin Button was nominated (but has no chance of winning) and Doubt wasn't (event though it produced three other acting nominations)? I know journalists are averse to numbers but this is an article that desperately needs a few sentences explaining the concept of "standard error." 

Anyone who wants to give me 50-1 odds on Button and/or Hathaway, or 5-1 on Viola Davis, I can be reached at kevincarey1@gmail.com.  Don't worry, it's a sure bet, Nate Silver's predictive models are foolproof. 

Reassurance Needed

In addition to fully funding NCLB, the stimulus bill includes a gargantuan $54 billion fiscal stabilization fund for education. In many ways this money is best understood as not education-related at all, but simply a politically palatable way for the federal government to prevent pro-cyclical state and local budget cuts that would accelerate the current economic death spiral. If you have to choose public employees to not lay off, in other words, teachers are (delusions of magical poverty-driven school reform notwithstanding) a pretty good choice. They're not "government bureaucrats," they often play an outsized role in their local communities, there are a lot of them, most people feel pretty good about spending public money on education, etc.

Most of the fund is in a big general pot. States have to apply for this money, which is mostly a formality since the funds will be disbursed using a pre-established population-based formula, and there is exactly no chance whatsoever that state applications will be turned down. Given the gravity of the economic situation, even something like Illinois voters amending their state constitution to anoint Rod Blagojevich governor-for-life shouldn't be disqualifying. Heck, bribe money probably gets into the economic bloodstream as fast as anything.

Nonetheless, there are a few policy-related items in the stabilization fund section of the law. For example, in submitting the sure-to-be-accepted applications, state have to provide "assurances" of the following:

• They will give K-12 and higher education (calculated separately) at least as much money in 2009, 2010, and 2011 as they did in 2006.
• They will establish longitudinal data systems to link up various student, school, university and teacher records. 
• They will improve their standards and tests. 
• They will support struggling schools
• They will, and here I quote directly, "comply with the requirements of paragraphs (3)(C)(ix) and (6) of section 1111(b) of the ESEA (20 U.S.C. 6311(b)) and section 612(a)(16) of the IDEA (20 U.S.C. 1412(a)(16)) related to the inclusion of children with disabilities and limited English proficient students in State assessments, the development of valid and reliable assessments for those students, and the provision of accommodations that enable their participation in State assessments" as well as "comply with section 1111(b)(8)(C) of the ESEA (20 U.S.C. 6311(b)(8)(C)) in order to address inequities in the distribution of highly qualified teachers between high- and low-poverty schools, and to ensure that low-income and minority children are not taught at higher rates than other children by inexperienced, unqualified, or out-of-field teachers."

In other words, in order to get the stimulus money, states have to assure the Department of Education that they will comply with federal laws that were enacted over seven years ago. Readers unfamiliar with the actual nature of federalism in the United of States of America might assume that compliance with the law is one of those things that are just kind of assumed and don't require additional bribes and written reassurances, but in the real world implementation of federal statute is much more a matter of negotiation. States have systematically chosen to ignore the above provisions with no real consequences, which unfortunately serves to highlight the generally toothless nature of the "assurance" process, in that it has less weight than the explicit statutes states have contravened. If history is any guide, the states will, in putting their applications together, cut and paste the above text out of the law, insert the words "We will" in front of each section, cash the checks, and be done. I would love to be wrong about this. 

Much more promising is the $5 billion "State Incentive Grant" portion of the stabilization fund. That's still a lot of money, even in the present depressingly grand scheme of things. Because the states are broke, they will all apply for the money, and there will be a lot of pressure on Secretary of Education Duncan to disburse the funds in the much same way as the big pot of stabilization money: proportional to size, so everyone gets their "fair" share, and subject only to a set of vague and infinitely malleable reassurances. But as I read the statute, he has a great deal of discretion to be much more inventive and forward-looking in deciding who gets how much money and establishing bright lines in terms of eligibility and use of funds. Done right, the process could be the launching pad for a terrific multi-year agenda. 

Beyond the Bubble (cont.)

Below, Chad highlighted my new Education Sector report on technology and the future of student assessment. In the report, I show how technology can help to both deepen and broaden assessment practice--by assessing more comprehensively and by assessing new skills and concepts.

Beyond the Bubble, of course, refers to the multiple choice question types that dominate NCLB-mandated state assessments.* But, it also refers to breaking through the static nature of the current policy debate around testing. You can hear me talk more about the current debate and discuss the implications of the report in this online interview.


*Disclaimer: Multiple choice question types can be an important part of a nutritious and well-balanced assessment diet.

Stalled at Launch

In K-12 education we have long debates about the purposes of public schooling, whether the focus should be on the basics or 21st century skills or if that's even a useful distinction, the best ways to assess student learning, how to contruct accountability systems that do more harm than good, etc. But underyling that discussion is a general consensus that it's possible to arrive at a reasonably accurate estimate of what students know, and that it's useful to compare students who attend one educational institution to students who attend other educational institution in this regard--or, at the very least, make the information public in a way that the people who attend the institutions and pay for the institutions can understand. In higher education, by contrast, no such agreement exists and indeed there are plenty of folks out there who find the whole idea puzzling and inappropriate. As such, higher education finds itself in a difficult spot: desperate for more public resources, but unable to marshall any convinving evidence of what those resources will produce in terms of how much college students learn. This is the subject of my new column in the Chronicle of Higher Education. 

Beyond the Bubble

During the 2008 presidential campaign, candidate Barack Obama frequently made comments like this one from April, where he said too much time was spent, "preparing students for tests that do not provide any valuable, timely feedback on how to improve a student's learning. Creativity has been drained from classrooms as too many teachers are forced to teach fill-in-the-bubble tests." This is a good sentiment, but it's an unfortunate reality that our current accountability system is reliant on such instruments. In a new Education Sector report, Bill Tucker looks beyond the bubble:

Students today are growing up in a world overflowing with a variety of high-tech tools, from computers and video games to increasingly sophisticated mobile devices. And unlike adults, these students don't have to adjust to the information age—it will be all they've ever known. Their schools are gradually following suit, integrating a range of technologies both in and outside of the classroom for instructional use. But there's one day a year when laptops power down and students' mobile computing devices fall silent, a day when most schools across the country revert to an era when whiteboards were blackboards, and iPhones were just a twinkle in some techie's eye—testing day....Still, the convergence of powerful new computer technologies and important new developments in cognitive science hold out the prospect of a new generation of student testing that could contribute to significant improvements in teaching and learning in the nation's classrooms.

Read the full report here.

That's Settled

The recently-enacted stimulus bill includes $13 billion in extra funding for Title I of the No Child Left Behind Act. Since Title I currently receives a little over $12 billion per year and the maximum amount authorized under the law is $25 billion, by my count NCLB is now "fully funded" and I assume those persons and organizations who have cited the lack of such funding as their principle objection to NCLB will now be foursquare in favor of holding all schools accountable for student performance via standardized tests.