The Education Gadfly Weekly

Henry Braun, Irwin Kirsch, and Kentaro Yamamoto
Teachers College Record
Forthcoming 2011

We’ve tried paying students to show up, behave, and perform but what about cash incentives to try harder on a no-stakes exam? The authors of this forthcoming article (available online, but to be published next year) put this to the test, literally, on the twelfth-grade NAEP reading assessment, a grade level whose scores are thought to be depressed by senioritis. (Seventeen-year-olds are thought to be savvy enough to know that this test doesn’t count, for them or their teachers.) Two-thousand-six-hundred seniors in fifty-nine schools from seven states were divided into three groups: The first received no money; the second was given fixed incentives of $20 before they took the test; and the third was given a conditional incentive of $5 before the test and an additional $15 for correct responses on each of two randomly selected questions for a maximum payout of $35. No group knew about the others or that it would receive incentives before test day. Researchers then administered one block of previously released twelfth-grade NAEP reading questions. (The actual NAEP tests use four blocks.) While control-group students scored an average of 289.2 points, fixed incentive students averaged 3.4 points higher, and conditional-incentive takers 5.5 points higher. This makes sense, and is no trivial amount; 5 points is about a quarter of the black-white achievement gap on twelfth-grade NAEP. This study indicates that U.S. twelfth graders know more than NAEP would have us believe. But it doesn’t account for the more troublesome fact that twelfth-grade NAEP scores have remained flat for decades. Read it for a small fee here.

Paul Hill and Marguerite Roza
Center on Reinventing Public Education
July 2010

These eleven pages do something quite useful: Explain the disconnect between education inputs and outputs in the language of economics. The affliction is Baumol’s cost disease, or “the tendency of labor-intensive organizations to become more expensive over time but not any more productive…(defined as the quantity of product per dollar expended).” Take a string quartet that “produces the same music from the time it is first assembled until the players retire.” Then consider that the education sector solves problems by adding more (money, teachers, etc.)—“the string quartet both gets wage and benefit increases and adds enough new members to become a sextet.” This is a structural problem, the authors explain, and one that better teachers and competition from independent operators (e.g., charter, private, and voucher-receiving schools) cannot (at least not alone) fix. But other industries have cured Baumol’s—or at least controlled its effects—which leaves hope for education. Some strategies are more applicable (or even already in use) than others: deregulation, which allows in new firms and new efficiencies (e.g., alternative certification routes), and information technology, which streamlines capacity, extending the touch of workers so that companies need fewer of them (e.g., virtual education). Others are further off, such as production process innovation, wherein tasks become specialized by competency and paid accordingly (e.g., the medical model: doctors supported by descending chain of residents, interns, nurses, etc.). The bottom line is that not tackling Baumol’s in education will result in more hiring freezes, layoffs, school day furloughs, and wage and benefits cuts, not fewer—and that, because the system is not currently structured to adopt achievement-raising efficiencies in response to financial difficulties, cuts will do damage unless we can change the model. Read it here.

Edward Crowe
Center for American Progress
July 2010

One of Secretary Duncan's primary concerns with teacher preparation when he went on the offensive last fall was that here was simply no quality control. Not only do these programs know next to nothing about their graduates, but ones that produce sub-par teachers see no consequences for doing so. In this paper, Edward Crowe lays out the myriad problems afflicting teacher licensure laws and how preparation programs are regulated by states (mostly) and the federal government (less so). For example, every state has their own set of licensure rules, which are informed by some subset of nearly 1,100 licensure tests of various quality and content. This mishmash basically means that district personnel offices must guess about the rigor of candidates' preparation, and preparation programs get mixed signals from states about what is expected of their graduates. Crowe's accountability system would start with a teacher-effectiveness measure (based primarily on how much students are actually learning), which would be the centerpiece of a system tying that measure to the alma mater of the students' instructor. Then, teachers and school leaders would complete feedback surveys about individual programs, the results of which, alongside teacher persistence rates and effectiveness numbers, would be publicly published for every preparation program. And finally, licensure tests would be streamlined and test cut scores and pass rate policies (i.e., how many students must pass for a program to stay open) uniformed. We've established that teacher quality is the number one determinant to student success, yet only three states (Florida, Louisiana, and Texas) use student-achievement data in teacher prep program evaluations. (Those numbers should rise as some stronger Race to the Top proposals are funded.) This paper offers one way of doing so. Read it here.

Paul Bambrick-Santoyo
Jossey-Bass
2010

“Data-driven instruction” is such a buzz phrase in education these days that a 300-page tome on the topic might seem like overkill. Yet this book turns out to be relevant, fresh, and accessible. It explicates the data techniques that have helped Bambrick-Santoyo and his colleagues at Uncommon Schools, a major and well-regarded network of charters, produce phenomenal results with some of America’s least privileged kids. The core insight is the need for interim assessments that enable teachers and school leaders to make mid-course corrections when warranted. It also offers case studies of schools that made dramatic achievement gains after installing comprehensive data management systems. The reform community has for years jabbered about the need for more “formative” assessments; this book is a guide to creating, using, and making the most of them. The author goes so far as to suggest that assessments should essentially determine standards: “Rather than have each teacher choose a level of rigor in response to vaguely written standards, the effective data-driven school leader or teacher works to create challenging interim assessments that set a high bar for student achievement.” One hopes that the advent of new assessments aligned with the Common Core standards will make this task easier and more efficient. Add this to your list of worthy 2010 how-to manuals (see here and here) from untraditional education providers. Purchase a copy here.

Gadfly’s not sure what to make of Secretary Duncan’s comments earlier this week in a big speech to the National Urban League. According to a press release from the Department, he proclaimed that “We will ensure that all schools—public, private and charter—serve the kids most in need…That is also something you told us was important. We heard you loud and clear, we are responding and these schools will be held accountable.” Set aside the Secretary’s assumption that charter schools aren’t serving “the kids most in need.” What on earth is he planning to do to “ensure” that private schools serve needy kids? How is he going to hold them “accountable”? Accountable to whom? Most don’t get public funds, and the administration's refusal to back the D.C. voucher program signals that they don't want private schools to get public funds--not even to educate needy, poor, black kids. The Administration’s thinking on civil rights has already been muddled; now it’s downright mystifying.

“Secretary Arne Duncan’s Remarks at the National Urban League Centennial Conference,” July 27, 2010

George Parker put his name on the dotted line. And thus, a city which previously paid, in the words of the Washington Post, “lip service” to teacher quality put its money where its mouth is. Or rather, Chancellor Michelle Rhee did, when she fired 165 teachers for ineffectiveness (plus 76 for licensure issues) and put another 700 or so on notice for being “minimally effective.” That rating comes from D.C.’s new teacher evaluation system, IMPACT, the use of which the union agreed in the newly ratified contract. Perhaps the union’s promise to file a grievance is just for show—or maybe not—but it can’t seriously have thought that Rhee was bluffing. WaPo has it exactly right when it observes that “a better use of [the union’s] time might be to work with Ms. Rhee to improve the performance of the 737 teachers in danger of losing their jobs next year.”

"Giving lousy teachers the boot," by William McGurn, Wall Street Journal, July 27, 2010

"Rhee dismisses 241 D.C. teachers; union vows to contest firings," by Bill Turque, Washington Post, July 24, 2010

“Opinion: The D.C. teacher firings,” Washington Post, July 25, 2010

We often find ourselves in the other corner of the policy ring from AFT President Randi Weingarten. So it is with a mixture of cheer and frustration that we discover Ms. Weingarten to be a talented crafter of haiku, sanely ambitious (“just enough to make a difference”), and notably tireless (“what is sleep?” she queries when asked how much she gets). That’s what we learned, at least, from her responses to Politico’s “Answer This” series. She’s even witty—she favors her legs over other body parts because she has to “walk a tight rope most of the time.” But we best her in one category: Humor. No jokes grace this (allegedly) off the cuff interview, because by her own admission, she is “far too serious.” We may not always agree, but we’ll tip our hats to a formidable, albeit humorless, opponent. 

“Answer This: Randi Weingarten” by Patrick Gavin, Politico, July 26, 2010

We confess. Mike and I were partly wrong last week: More than half a dozen conservatives have misgivings about the “Common Core” standards and the tests to follow. The number is up to at least eight and, since conservatives tend to get excited by the sight of red (red meat, red blood, red states, etc.), every time we wave this scarlet flag in front of them we can expect more of them to charge us. Perhaps including the piece you’re now reading.

So far, though, we’ve been nicked but not gored by their horns—and we cheerfully concede that critics have several legitimate concerns. Yes, it would have been better if the voluntary move by states to develop and consider adopting common standards hadn’t been entangled in a competition for federal money. Yes, it would be better if more of that same federal money weren’t paying for development of new assessment systems to accompany the standards. Yes, it would have been lots better if President Obama had never hinted at harnessing national standards to future Title I funding. Yes, the long-term governance of the standards and tests remains to be worked out.

But good grief, folks, do you really want to preserve the meager academic expectations, crummy tests, and weak-kneed accountability arrangements that currently drive—or fail to drive—K-12 education across most of this broad land? Are you so risk averse and change resistant as to see no merit in trying to do this differently in the future?

It’s true, as multiple bloggers have noted, that I spent part of 1997 itemizing the flaws in Bill Clinton’s plan for the federal government to create and administer a national testing system. And like practically everybody else (save for the progressive educators who drafted them), I didn’t like much about the federally-induced “national standards” that had emerged during the Bush I administration earlier that decade.

But many things have changed in thirteen years. Four deserve to be noted.

First, and most important, U.S. education has made only the slimmest achievement gains—none of these at the high-school level—and graduation rates have stayed limp, as more and more countries surpass us on more and more measures on this fast-flattening and ever-more-competitive planet.

Second, despite multiple rounds of asking (and Uncle Sam bribing) states to come up with rigorous academic standards on their own, few have done so. Those few are swell, but most are simply dismal—a “C” average on the latest Fordham ratings. And they’re ridiculously uneven from place to place. Modern countries don’t do this to their kids.

Third, much as I wish otherwise, conservatives’ preferred alternative education-reform strategies haven’t gained the traction or scale that advocates (myself included) hoped for, nor have they delivered reliably better academic results. Yes, the principle has largely been accepted that kids need not necessarily attend the district school in their neighborhood. Yet you can count the voucher programs on your fingers. And charter-school enrollments, while respectably up, don’t amount to more than 3 percent of all kids. The parent marketplace isn’t causing bad schools to close. (Only Catholic schools, many of them fine, seem to be closing.) One can keep beating this drum—and you’ll find more and more people snapping their fingers in time with the beat—but, mostly for political reasons that aren’t going away, it hasn’t produced a lot of marching.

Fourth, the main sources of resistance to change in American education aren’t conservatives (hard as some of the latter are trying!). They’re education interest groups, starting but not ending with the teacher unions. They still wield much clout—see previous point—but they’re weaker today than at any time in my memory, no doubt because they’re beset on more issues on more fronts by more forces. To give credit where it’s due, a contributor has been the unexpected emergence of the Obama administration as a source of reform pressure and the schism that’s emerged within the Democratic Party over education issues. (This is also the main reason that today there’s no serious GOP education platform. Except for vouchers, just about all the traditional “conservative” education enthusiasms have gone mainstream.)

So yes, I’ve partly changed my mind about national standards and tests. I’m mindful of the risks and unknowns that lie ahead. I’m not totally satisfied with the Common Core. (Our raters gave it honors grades but not straight As). It troubles me that we’re so narrowly focused on just two subjects within the school curriculum. I’ve no idea what “cut scores” will be established for the forthcoming tests nor whether colleges and employers will take them seriously. I’m alarmed that one of the new assessment consortia doesn’t seem serious about accountability. I’m wary of what Congress will do to the Common Core when it finally gets around to reauthorizing NCLB. I’m nervous about the administration’s political backbone as electoral stakes rise. I’m skeptical about the stick-to-it-iveness of states that pledge their troth to Common Core but are rejected at the Race to the Top altar. (This may get clear fast. On Tuesday, a dozen states that had already adopted the new standards—more than one third of all adopters—were omitted from Secretary Duncan’s list of RTT finalists.)

But what’s the point of just fretting and biting my nails and issuing cries of alarum? The education status quo sucks, to put it bluntly. Conserving it is no fit work for conservatives. In most of the country, they—we—should demand something better.*

That’s basically where I was in 1997 when I concluded that Bill Clinton’s federally-dominated national testing plan wasn’t better. No, I haven’t totally changed my mind. I still believe what I wrote then in concluding the Weekly Standard piece that some of my conservative pals have been selectively quoting other portions of:

If national testing is headed that way, the country would be better off without it. Congress should apply the brakes before a wreck occurs. Then maybe—just maybe—let a different driver take a turn at the wheel. If a fully independent version of the [National Assessment Governing] [B]oard I chaired in the 1980s were put in full charge, the risk of crashing would be reduced. Alternatively, the whole idea might be privatized, turned into a commercial (or philanthropic) testing program that picks up Clinton's basic concept but with no government entanglement or federal funds. We still need a means to compare achievement across state borders. But it's worth doing only if the twin tar babies can be avoided. The one thing indisputably worse than no national tests would be bad national tests.

We don’t yet know about the tests. But we know about the standards. They’re good. They’re not perfect, but they’re better than what we’ve been using across most of America. The proper work of conservatives (and others) who care about the nation’s future is to help them succeed, not grump about what might go wrong.

^*Massachusetts is a different story. Conservatives there were not wrong to press for retention of a set of solid standards and tests that was yielding good results. The Deval Patrick-dominated state-board-of-education decision to start afresh with Common Core was a decision to substitute a system that doesn’t yet exist for one that’s served the Commonwealth well.

Paul Hill and Marguerite Roza
Center on Reinventing Public Education
July 2010

These eleven pages do something quite useful: Explain the disconnect between education inputs and outputs in the language of economics. The affliction is Baumol’s cost disease, or “the tendency of labor-intensive organizations to become more expensive over time but not any more productive…(defined as the quantity of product per dollar expended).” Take a string quartet that “produces the same music from the time it is first assembled until the players retire.” Then consider that the education sector solves problems by adding more (money, teachers, etc.)—“the string quartet both gets wage and benefit increases and adds enough new members to become a sextet.” This is a structural problem, the authors explain, and one that better teachers and competition from independent operators (e.g., charter, private, and voucher-receiving schools) cannot (at least not alone) fix. But other industries have cured Baumol’s—or at least controlled its effects—which leaves hope for education. Some strategies are more applicable (or even already in use) than others: deregulation, which allows in new firms and new efficiencies (e.g., alternative certification routes), and information technology, which streamlines capacity, extending the touch of workers so that companies need fewer of them (e.g., virtual education). Others are further off, such as production process innovation, wherein tasks become specialized by competency and paid accordingly (e.g., the medical model: doctors supported by descending chain of residents, interns, nurses, etc.). The bottom line is that not tackling Baumol’s in education will result in more hiring freezes, layoffs, school day furloughs, and wage and benefits cuts, not fewer—and that, because the system is not currently structured to adopt achievement-raising efficiencies in response to financial difficulties, cuts will do damage unless we can change the model. Read it here.

Edward Crowe
Center for American Progress
July 2010

One of Secretary Duncan's primary concerns with teacher preparation when he went on the offensive last fall was that here was simply no quality control. Not only do these programs know next to nothing about their graduates, but ones that produce sub-par teachers see no consequences for doing so. In this paper, Edward Crowe lays out the myriad problems afflicting teacher licensure laws and how preparation programs are regulated by states (mostly) and the federal government (less so). For example, every state has their own set of licensure rules, which are informed by some subset of nearly 1,100 licensure tests of various quality and content. This mishmash basically means that district personnel offices must guess about the rigor of candidates' preparation, and preparation programs get mixed signals from states about what is expected of their graduates. Crowe's accountability system would start with a teacher-effectiveness measure (based primarily on how much students are actually learning), which would be the centerpiece of a system tying that measure to the alma mater of the students' instructor. Then, teachers and school leaders would complete feedback surveys about individual programs, the results of which, alongside teacher persistence rates and effectiveness numbers, would be publicly published for every preparation program. And finally, licensure tests would be streamlined and test cut scores and pass rate policies (i.e., how many students must pass for a program to stay open) uniformed. We've established that teacher quality is the number one determinant to student success, yet only three states (Florida, Louisiana, and Texas) use student-achievement data in teacher prep program evaluations. (Those numbers should rise as some stronger Race to the Top proposals are funded.) This paper offers one way of doing so. Read it here.

Henry Braun, Irwin Kirsch, and Kentaro Yamamoto
Teachers College Record
Forthcoming 2011

We’ve tried paying students to show up, behave, and perform but what about cash incentives to try harder on a no-stakes exam? The authors of this forthcoming article (available online, but to be published next year) put this to the test, literally, on the twelfth-grade NAEP reading assessment, a grade level whose scores are thought to be depressed by senioritis. (Seventeen-year-olds are thought to be savvy enough to know that this test doesn’t count, for them or their teachers.) Two-thousand-six-hundred seniors in fifty-nine schools from seven states were divided into three groups: The first received no money; the second was given fixed incentives of $20 before they took the test; and the third was given a conditional incentive of $5 before the test and an additional $15 for correct responses on each of two randomly selected questions for a maximum payout of $35. No group knew about the others or that it would receive incentives before test day. Researchers then administered one block of previously released twelfth-grade NAEP reading questions. (The actual NAEP tests use four blocks.) While control-group students scored an average of 289.2 points, fixed incentive students averaged 3.4 points higher, and conditional-incentive takers 5.5 points higher. This makes sense, and is no trivial amount; 5 points is about a quarter of the black-white achievement gap on twelfth-grade NAEP. This study indicates that U.S. twelfth graders know more than NAEP would have us believe. But it doesn’t account for the more troublesome fact that twelfth-grade NAEP scores have remained flat for decades. Read it for a small fee here.

Paul Bambrick-Santoyo
Jossey-Bass
2010

“Data-driven instruction” is such a buzz phrase in education these days that a 300-page tome on the topic might seem like overkill. Yet this book turns out to be relevant, fresh, and accessible. It explicates the data techniques that have helped Bambrick-Santoyo and his colleagues at Uncommon Schools, a major and well-regarded network of charters, produce phenomenal results with some of America’s least privileged kids. The core insight is the need for interim assessments that enable teachers and school leaders to make mid-course corrections when warranted. It also offers case studies of schools that made dramatic achievement gains after installing comprehensive data management systems. The reform community has for years jabbered about the need for more “formative” assessments; this book is a guide to creating, using, and making the most of them. The author goes so far as to suggest that assessments should essentially determine standards: “Rather than have each teacher choose a level of rigor in response to vaguely written standards, the effective data-driven school leader or teacher works to create challenging interim assessments that set a high bar for student achievement.” One hopes that the advent of new assessments aligned with the Common Core standards will make this task easier and more efficient. Add this to your list of worthy 2010 how-to manuals (see here and here) from untraditional education providers. Purchase a copy here.