Do We Need a Basic Rewrite of No Child Left Behind?

John Merrow:

Any talk of abandoning No Child Left Behind is foolish because NCLB is the continuation of a long trail of federal education legislation that traces back to the Elementary and Secondary Education Act of 1965.

Congress and the next Administration must do something, but what? That's the question posed to a remarkable roster of deep thinkers and activists.

Can NCLB be fixed? If so, what changes must be made? How wholesale must they be?

What good has NCLB done in its short history? What harm has it done?

Its supporters say that it has forced schools to�finally�pay attention to certain groups of children who have been all but ignored. By requiring that all identifiable groups of a certain size make what is called 'adequate yearly progress,' NCLB has held schools' feet to the fire.

Critics point out that the law is riddled with loopholes, and that alone has created contempt for the law. States and districts have wiggled out of many of the law's provisions�by changing the size of the subgroups, for example, rendering 'results' virtually meaningless.

Supporters say NCLB forces school districts to pay attention to the credentials of the teachers it hires�finally. No longer can districts put a warm body in front of classrooms, thanks to NCLB.

Critics say the law has dumbed down school by focusing almost entirely on reading and math. Gone, or greatly diminished, are art, music, science and so on. Supporters point out that, if kids cannot read and compute, their futures are bleak.

I anticipate an honest exchange of ideas over the next three days.

Diane Ravitch:

Every poll I have seen says that the overwhelming majority of teachers in the United States believe that No Child Left Behind is a failure. Any speaker who appears before a gathering of teachers will hear their negative appraisal of the law. Teachers know that the curriculum has been narrowed, that the time available for the study of the arts, history, science, geography, and every other non-tested subject has been reduced. They know that the law's laser-like focus on test scores in reading and mathematics has led to constant test-preparation, which has replaced instruction. When time devoted to testing the basic skills crowds out every other kind of learning, this is not good education.

If the people who are required to implement NCLB on a daily basis see its negative consequences and doubt its value, the rest of us should pay attention.

Diane Ravitch:

In fact, there is little to commend NCLB. This federal effort to micromanage the nation's classrooms has not worked. The test scores in states are increasing, but gains on the federally-sponsored national examinations called the National Assessment of Education Progress have not kept pace with the scores reported by the states. An examination of NAEP trends shows that achievement in reading and math increased faster before NCLB was passed than in the years since it was adopted.

The way the law is structured, schools and districts get credit for advancing students from below "proficient" to "proficient," so they can safely ignore the smartest children who are already proficient (and "proficient" is broadly defined by the states, in most states it is the equivalent of what the national test calls "basic"). Thus the law implicitly encourages teachers to disregard the most capable students who easily have jumped the bar to proficiency. We pay a heavy social price by disregarding the students who are likely to be our future leaders, scientists, and innovators.

The sanctions in the law are ineffective. Few students are taking advantage of the choice provision; great schools do not have lots of empty seats, and most parents don't want to send their kids cross-town to another school. There is no guarantee that conversion of a low-performing public school into a charter school or a privately managed school will lead to overall gains. Indeed, neither of these strategies has been proven effective by experience.

My own preference would be for Congress to authorize national testing (� la NAEP), based on coherent curriculum standards, but without stakes or sanctions. The federal role should be to provide accurate information about student performance. It should be left to states and districts to devise sanctions and reforms. These jurisdictions are closer to the schools and likelier to come up with workable reforms. If states and localities don't want to improve their schools, then we are in deeper trouble as a nation than any law passed by Congress can fix.

Chester E. Finn, Jr.:

NCLB needs a major makeover but I surely concur that it isn't going away. It should be turned upside down. Instead of being "loose regarding ends and tight regarding means", as every management expert recognizes, it would be far more effective if it were tight as to ends and loose as to means. In other words, the proper federal role is to set the academic standards (or have them set via some suitable national or multi-state entity), administer the assessments and report the results--but then leave it to states, districts, schools and educators to determine how best to work to attain those standards and on what timetable. NCLB has it backwards, leaving it to individual states to set their own standards and give tests, but then being highly prescriptive regarding timelines and what to do with schools that don't measure up. That creates perverse incentives for states to set low standards, game the system, and do as little as possible to comply with the prescribed federal remedies.

Jerry Wartgow:

NCLB was conceived and drafted by a strong bipartisan coalition of bright and well-intentioned people who shared the noble goal of improving public education and raising achievement for all students through implementation of the theories of standards-based education reform. It was also a perfect illustration of one of the lessons that we should have learned from prior attempts to improve education through legislation. That is to "beware of the unintended consequences of well-intentioned people".

While NCLB was very successful in focusing the nation�s attention on setting high expectations, improving student performance and closing the achievement gap, one of the unintended consequences of the law was that the regulations necessary to implement this very complicated piece of legislation mired the schools in a bureaucratic nightmare that placed far too much emphasis on testing and compliance reporting.

The purpose and goals of NCLB remain as compelling and important as they were when adopted nearly six years ago and should be a moral imperative for the nation. Therefore, rather than throwing it out we should work to make it a less bureaucratic, slimmed down, healthier version of the law that includes provisions for capacity building among those who will ultimately be responsible for its implementation.

Thomas Rogers:

With apologies to Dickens: �It was the best of laws, it was the worst of laws.�

The civil rights community considers NCLB to be one of its high water marks�for the first time in the long history of school desegregation lawsuits, a law was passed that requires a remedy without first finding negative intent. In other words, it is enough to observe the performance characteristics of subgroups of children (differentiated either by race or disability status) and to require schools to address the disparity in outcomes.

Moreover, the law has changed the dialogue in education from process accountability to one of outcome accountability. Most of the tomes of education law adopted by states like New York require adherence to process�presuming adherence results in achievement. Now we�re moving from �I taught it� to �they learned it�.

Unfortunately, there�s less agreement about whether we�re measuring the right outcomes and whether we�re measuring them well. Most states� NCLB tests focus on basic memorization and computation, not the kinds of rich problem solving and cultural competency skills that seem more likely to boost students� fortunes in a globalized economy.

Moreover, the tests are administered as summative snapshots and largely de-contextualized. We cannot readily see the challenges students bring with them to school because of poverty or disability, nor can we readily measure significant progress improving their achievement�the only thing really under schools� control, and therefore the most important accountability measure.

Ryan Hill:

The philosophical mind-set shift that NCLB forced upon our country is not trivial and its importance should not be underestimated. Until NCLB, it was understood�not hypothesized, but understood�that students who are low-income, of color, or classified as special education students simply could not be expected to learn at the same levels as high-income white kids. NCLB said that it�s possible for our kids, of all income levels, races, ethnicities, or classifications, to achieve at a high level. This is a critical message to send to schools and to the general population.

The problems with NCLB, however, should also not be underestimated. First, there is the oft-mentioned problem of state tests that vary dramatically in their difficulty. In �Knowledge is Power Program� schools (KIPPs), this inability to compare scores on tests in California to scores in New York makes us resort to taking a national norm-referenced test in addition to the state tests; for our purposes�researching best practices, accountability, etc.�the state tests are virtually useless. Presumably, the Department of Education has similar aims, and NCLB�s mandates provide them with similar problems.

Ryan Hill:

The other major problems with NCLB are a) that it�s completely unrealistic, and b) it focuses too much on jumping over bureaucratic hurdles and not enough on objective outcomes. We all know it�s unrealistic to say that, in 2014, all students in America will pass a rigorous state test. Most states, capitalizing on the first problem I mentioned above, deal with this by simply making their state test easier to pass. There is real utility and inspiration in an audacious goal, but if Kennedy had proclaimed in 1962 that by the end of the decade we�d be traveling to Pluto, NASA would�ve thrown their hands up and accepted their inevitable failure. Many districts, with little chance of making Adequate Yearly Progress (AYP), have done the same.

As for the bureaucratic hurdles, nothing in NCLB is better-intended or makes less sense than the Highly Qualified Teacher provisions. Rather than attempting to discern whether a teacher is actually effective, HQT enforcement has simply meant that teachers have to take more classes, have a more specific educational background, and jump through more administrative hoops in order to get into a classroom. Some of the best teachers I know are people who have taken an alternate route to teaching, whether through Teach for America or another source, and though these teachers have not always been �highly qualified�, they have been highly effective. Certification needs to bring congruence to the terms �effective� and �qualified�; HQT does not do this.

On the solutions front, Checker is right. A national test is critical, and the prescribing of means (as opposed to enforcement of ends) should be dropped. If schools could get to an agreed-upon proficiency standard with no teachers who went through the certification process, we should celebrate that just as we�d celebrate it in a group that�s 100% highly qualified. This is the reason a lot of charter schools like ours are having success in closing the achievement gap: fewer restrictions on the front end, more accountability on the back end. If our principals see something that works in another school, they are able to move quickly in emulating it. District schools, with more prescription of their behavior, are in many cases much slower to adopt innovations that help kids. NCLB should be changed to mimic the spirit of flexibility and accountability that characterizes most charter laws.

Randi Weingarten:

John Merrow is right: Helping all kids achieve, particularly kids at risk, was always the main goal of federal education law. NCLB correctly set high standards, but it over-emphasized testing and sanctions at the expense of helping all kids achieve.

What we need is a federal education law to help poor and minority children get a well-rounded education that includes the arts and physical fitness, emphasizes critical thinking, and teaches the value of active citizenship. Instead, NCLB slammed the schoolhouse door on much of what makes up modern civilization and replaced it with multiple choice questions.

So let's focus on what we need to do now. Let's put in place a federal education law that, unlike NCLB, provides space and opportunity for children to be taught a well-rounded curriculum, with standards and accountability that support that curriculum. At the same time, let's address outside factors like nutrition and healthcare that affect a child's ability to reach her full educational potential.

Chester E. Finn, Jr.:

NCLB is all about reading and math skills. That's only part of the curriculum, obviously, but to date is the only part that federal law seeks to "enforce". I'm all in favor of a broad, rich, liberal-arts style curriculum for all kids (see �Beyond the Basics: Achieving a Liberal Education for All Children,� for example) but am wary of Uncle Sam trying to say what that should be. Moreover, many teachers and principals have commented that "if they can't read, they're not going to learn much history or literature", i.e. that NCLB's priorities are sound. Can't we count on states and districts and principals and teachers to add the rest of the curriculum?

Sara Mead:

I�m sympathetic to Checker�s concerns that NCLB is too �tight on means, loose on ends,� rather than vice versa. That�s one of the reasons New America has joined with the Thomas B. Fordham Foundation in advocating high-quality voluntary national standards.

But given Checker�s very justified concerns that states have responded to NCLB by lowering standards, gaming the system, and avoiding compliance, I�m curious why he thinks giving states, districts, and educators more flexibility in how they identify and deal with underperforming schools is the way to go here.

It�s not as if we haven�t seen what happens when states and school districts have greater flexibility in how they identify and intervene in poor performing schools: That was the case under the 1994 Improving America�s Schools Act that preceded NCLB, and states largely used that flexibility to avoid identifying low-performing schools, or doing anything meaningful to help them improve. NCLB tightened up standards for AYP, the standard used to identify schools in need of improvement, as well as specifying sanctions for them, largely in response to the failures of states to implement the spirit of the 1994 law. Similarly, where states and districts have flexibility in intervening in schools under the current law, they�re doing everything in their power to do nothing about persistently poorly performing schools. Just look at how states have dealt with schools identified for restructuring under NCLB (these are schools that fail to make AYP for more than 5 consecutive years. In only a teensy fraction of cases have states and districts taken real action to reconstitute the school, close and reopen, convert to a charter school, etc. In most cases they�ve taken the law�s �other� option�a loophole that allows them to basically keep doing what they were already doing.

Certainly, there is a need for refinement in both AYP and NCLB�s mandated interventions in schools that fail to make AYP. The discussion draft put forward by the House Education and Labor Committee last year took some positive steps on this front, particularly in terms of differentiating consequences for different types of failures to meet standards. But handing it over to the states and districts is not the answer.

At the same time, the federal government needs to be much more active in building capacity at the state and district level to intervene effectively in poorly performing schools, as well as being much more aggressive in supporting meaningful research, development, and innovation that will generate solutions to radically improve the performance of these schools. I hope to discuss this further�and hear from others�later in today�s discussion.

Deb White:

Obviously the intention behind NCLB was positive; however the law itself is entirely punitive. Rather than focusing on successful models which struggling schools can emulate, all of the emphasis is on avoiding failure. This results in "teaching to the test", lowering standards on state tests, inaccurate reporting of drop out rates and "cheating" by states, schools and individual teachers. Accountability is essential, but you cannot punish schools into being successful. Success must be inspired.

We need to focus on what works. Small class sizes and small schools where caring adults know each student as an individual, curricula that are well designed and meaningful to students and experienced teachers who can mentor and encourage new teachers. All of the financial resources that are going into testing, testing and testing are resources that are not available to improve schools.

We know what works�unfortunately test scores are easier to explain to the public on the nightly news.

Sol Stern:

I must disagree with some of my friends that it's NCLB's fault that the curriculum has been narrowed and that subjects like history, geography and science are being neglected. It's true that some states and districts are gaming the system, dumbing down the tests, spending an inordinate amount of class time on test prep, and even tolerating cheating. But there is nothing inherent in the federal law that produces such destructive effects and almost always it's the state and district education officials that are behaving irresponsibly and often for political motives.

Nowhere is that more clear than in New York State and New York City, the nation's biggest school district. The state education department has been dumbing down its tests in grades 3-8, and even more so on the high school Regents exams, to maintain the illusion of significant progress in academic achievement. As I have pointed out in City Journal, there are some districts in the state that recorded proficiency rates of close to 100% on this year's math tests. This Lake Wobegon effect is an embarrassment, to be sure, but why blame this on George Bush or the Democratic authors of NCLB? Meanwhile New York City's education department is trying to boost tests scores by any means possible (including bribing children, teachers and principals) in order to advance the political ambitions and the historical legacy of Mayor Michael Bloomberg. Let's not let these state and city officials off the hook for their politicization of education.

On the other hand, there are districts in the country that have accepted the challenge of boosting the math and reading skills of disadvantaged kids in the early grades, without resorting to perverting the entire accountability system based on tests. We should honor those districts, while discussing ways of closing the loopholes in the current law and requiring total transparency for the districts and states that are trying to game the system.

Christopher Cerf:

For all its flaws, NCLB transformed educational opportunity, especially for urban children. Other than at the extreme fringes of the debate, few would disagree with the simple and related precepts that: 1) we care about how much all children are actually learning, as opposed to other input-based considerations; 2) that we are going to track and be transparent about such learning, not only in the aggregate, but by subgroup; 3) that we are going to operate on the principle that all children, regardless of birth circumstances, can learn; 4) that we hold schools and districts accountable for student progress, and; 5) that there are real consequences for failing to discharge that accountability.

Transformative system change�and who could deny at this point that anything less is required for urban K-12 public education�begins with culture change. In my view NCLB accomplished that in abundance. Indeed, I attribute much of the whining, the calls for amnesty, the overhyping of the many serious problems in the statute, to precisely that. If we are honest, serious accountability for student learning�while much ballyhooed�historically has been anathema in public education. Elaborate bulwarks have been erected over the years to deflect accountability away from districts, schools, and educators alike. It is of no surprise to me, at least, that the forced imposition of such accountability has yielded the resistance it has. Moreover, as the report of the Koret Task Force noted, signed among others by Diane Ravitch and Checker Finn: NCLB's basic principles of "accountability, transparency, and choice are fundamentally different from the traditional guidelines of public education ... and there is strong evidence that the kinds of tough accountability measures adopted by NCLB have been raising student achievement in the states."

That said, I agree that the Act is in need of serious repair. I'll start with two. Checker and others are right that a national test is critical (I'd go with national standards myself) if we are to avoid the shameless race to the bottom many states are now pursuing. Second, by the time the noble idea of ensuring a "highly qualified teacher" (HQT) in every class room emerged from the legislative sausage-maker, it had become a largely meaningless, bureaucratic box-checking exercise. Again, somewhat along the lines Koret recommended, let's recast HQT as "highly effective teacher" and build from there. It is high time we stop measuring teachers by credentials and pathway-tests that research now tells us convincingly are not predictive of student learning.

Diane Ravitch:

I never like to disagree with Sol Stern, but I will suggest here that the narrowing of the curriculum is the inevitable result of an accountability system that tests only reading and math (science is tested, but it doesn't count), and that threatens to sanction states unless their students are 100% proficient by 2014.

If the fed says the state must reach 100% proficiency in math and reading by 2014, then several things will happen: Districts will focus inordinate amounts of time on reading and math; will engage in constant test preparation; will figure out how to game the system, such as by closing low-performing schools and re-opening them with a new name, starting the sanctions clock over; and doing whatever else must be done to meet the goal. That is why we are very fortunate to have NAEP as an external monitor. And NAEP, as I said in an earlier post, shows very meager gains these past few years, especially when compared to the previous five years.

Ryan Hill:

Sol�s and Deb�s comments further illustrate the need for national standards and a national test. Sol�s comments show us that even when states or districts do make dramatic improvements, many will view them suspiciously regardless of whether the gains are legitimate, as I believe they are in NYC, or gamed, as they are in many other places. A mandated national test would settle this in ways that NAEP doesn�t, because state tests have varying degrees of alignment with NAEP.

As for Deb�s comments, as a teacher I can attest to the fact that the smaller a class is, the easier it is to teach. As a principal, however, I also know that it is harder to find 30 teachers who can expertly teach 20 kids in a class than it is to find 20 teachers who can expertly teach a class with 30 students. The major class-size studies I�ve seen draw conflicting conclusions, and part of the problem is that we don�t have a common measure by which we can judge class-size initiatives.

Christopher Cerf:

Sol, showing uncharacteristic restraint, I'll not rise to the bait and respond to your well rehearsed critique of the Mayor and the Chancellor. I must, however, state for the record that your hyperbolic suggestion that people are being bribed (merit pay for teachers and principals and a pilot student motivation program?) is not a characterization we would endorse. Perhaps we can agree to refer readers who are not yet exhausted by the exchange to our e-debate on Eduwonk a few weeks ago.

John Merrow:

Let's not lose sight of the central issue: what to do with NCLB. Is the country moving inexorably toward "Common Standards," and, if so, is this happening in response to NCLB? What exactly should the federal role in K-12 education be?

And I urge others to weigh in, because this is clearly not just a New York City issue.

Thanks�

Jerry Wartgow:

I agree with Checker and other participants that the current system is upside down. In the law�s original draft the National Assessment of Educational Progress was to serve as a uniform benchmark for measuring progress. Unfortunately this plank fell victim to the political compromises necessary to get the bill passed. We have learned from the experience of the past five years that national standards and a national test would be far more valuable in measuring success and much less bureaucratically burdensome on the states and schools than the current system. The challenge is to devise standards and a test that measure the entire educational experience, not just those that can be scored by a computer. Einstein reminds us that �Not everything that counts can be counted and not everything that can be counted counts�.

Sara Mead:

Chris makes an excellent point about NCLB�s �highly qualified� teacher definition�While clearly a necessary step to address the prevalence of out-of-field teaching in high-poverty, high-minority schools, HQT doesn�t come near ensuring that teachers are actually effective.

But we�ve been bogged down for too long in a debate over whether or not test scores are an appropriate way to measure teacher effectiveness. It�s time to move beyond that debate to look at validated, reliable measures that actually observe what teachers are doing in the classroom, determine whether or not they�re engaging in practices demonstrated by research to improve student learning, and provide feedback to help them improve. The Classroom Assessment Scoring System (CLASS) developed by researchers at the University of Virginia is one such model which can currently be used to assess teaching in pre-k through third grade classrooms. The next version of Elementary and Secondary Education Act (ESEA) should encourage states and school districts to adopt CLASS and related models when available, and also fund research and innovation to catalyze the development of new, validated, reliable observational systems to replace current toothless teacher evaluations.

Charles Kolb:

As I have for much of the last quarter-century, I find myself once again in agreement with Checker. I also share many of the concerns raised by Diane.

On balance I find that NCLB has been a positive development for many of the reasons listed by Chris Cerf. But why has it taken our country this long to get to where we are today? And is our debate/discussion today really penetrating to average citizens in a fashion that can generate and reinforce a true culture of learning (not just passing tests) in this country?

�A Nation At Risk� is 25 years old this year, and next year marks the 20th anniversary of the Charlottesville Summit between George H.W. Bush and the nation�s governors. Out of that summit grew the six national education goals with standards around five key content areas (English, science, math, history, and geography). We were going to meet all of those goals by � 2000. We had the National Education Goals Panel with all of its elaborate, detailed reporting (which put most people to sleep, unlike Bill Bennett�s flawed but otherwise galvanizing �Wall Chart�), and we even had a debate in the early 1990s about having voluntary national tests/testing, as opposed to a mandatory federal test. Where we are today in NCLB is the result of many of these past efforts, and that�s why I see the glass as more than half full.

But I remain frustrated at the pace of change, what Sol refers to as the politicization of education, and the ongoing efforts to resist real performance accountability that benefits our underachieving youth, those in the middle, and our overachievers.

Why is reform so difficult, and what can change this dynamic, especially as we contemplate NCLB reforms?

Gerald Tirozzi:

We are continually reminded that the world is rapidly changing and that educators must adapt their curricula and pedagogy to prepare students for jobs that do not yet exist. Global society, international competitiveness, technology-driven workplace�such phrases abound and shape our educational environment and the mission of our schools.

The future calls us to prepare students who are imaginative, creative, entrepreneurial, and who have the capacity for "high touch" abilities such as compassion, personal rapport, social interaction, and caring and helping others; yet, ironically, NCLB forces us to judge our schools on minimal technical proficiencies. The vacuous victory statements based on low level proficiencies in reading and mathematics are driving our schools in a direction diametrically opposite of where the future is leading.

Our test-obsessed education laws force educators to address only the skills that are being tested and to prepare our students, as one educator has noted, "for a life of tests" rather than "for the tests of life."

Deb White:

Let�s talk about testing. I teach in a small school district with approximately 180 students per grade. We test in the 5th, 8th and 11th grades; however, students actually test every year from 5th grade on and must demonstrate proficiency by the 8th and 11th. I will focus on the 11th grade. Every year we are testing different students who just happen to be in the 11th grade. Students move in and out at the rate of about 20% per year�which is probably less than the national average. However, combine that with a small sample size, just a few students performing higher or lower than average, and results from one year to the next are essentially meaningless. We need to test students against the curriculum rather than against unrelated groups of students.

More importantly, we need to test the same students over time and look for improvement. Let me give a simplistic example�a student with a 1st grade reading level moves into a school in the 5th grade. By the end of the year the student is reading at the 4th grade level. A huge improvement and one to celebrate. BUT that student is not proficient, not reading at the 5th grade level! According to NCLB that student is a failure. If this is a state with "high stakes testing" like Texas, that student will not move on. I have a friend who teaches 5th grade at a Title 1 school in San Antonio. She has had students in her classroom who have been held back for 3 years. At what point does that "failure" label so discourage that student that they just give up and drop out?

It is important to remember that desperate people (and organizations) take desperate measures. Cheating occurs when there is no avenue to succeed legitimately. Teachers who are evaluated by their test results are in a no-win situation. There are so many factors beyond the control of both teachers and schools. There is no career more important than teaching, but also no career where the person held accountable for the outcome (or product) has less ability to control the variables affecting that outcome. A teacher is totally accountable for whoever shows up.

Philip Howard:

Perhaps NCLB should be radically simplified, imposing a uniform testing regime, without all the mandates and consequences that have transformed educators into idiot savants. NCLB suffers all the flaws of central planning because that's what it is.

Children aren't widgets, as Deb and others remind us. With information from national uniform testing, more nuanced judgments can be made about success of schools, taking into account all factors. This information can be a powerful management tool, but those judgments can't be prescribed in advance.

Deb White:

Some concrete suggestions for improving NCLB�all focusing on the positive:

1. Fund National Board Certification bonuses at the federal level rather than leave it up to states. Some states reward teachers for going through the process and some do not. (Currently approximately 2% of teachers nationwide are certified. The process takes 200-400 hours and $2500 to complete. The first time success rate is about 33%�so it is obviously rigorous.) The process itself is highly introspective and results in even those considered master teachers to say their teaching improved dramatically.

2. Test for student improvement over time. (Compare students to their previous performance�are they still on track and improving?)

3. Use evaluation tools in addition to high stakes testing. Performance Assessments, portfolios, etc. Maybe a national test for minimum standards and then additional supporting materials designed "locally" for everything above that? Testing, like a snapshot, does not always give a complete picture.

4. Modify AYP requirements so that as schools approach 85% proficiency, improvement may level off as long as there is no statistically significant drop. (The idea of 100% proficiency is ridiculous. How can testing be both "highly rigorous" and "attainable by all students"? Definitely a contradiction.) Currently a school that moves from 35% to 40% is successful, but a school that has leveled off at 85% and is not improving is a failure.v

5. Increase flexibility to allow for trade classes and school-to-work options. Not all students are college bound.

6. On the other hand, stop shortchanging the high achieving kids. As school district resources are focused on remediating the students who might not pass the test, high achievers are largely ignored. A bored, gifted kid may be just as likely to drop out as a frustrated "failing" kid.

Randi Weingarten:

Some would imagine a dystopian pre-NCLB world in which educators didn't care whether students were learning, hid the test scores of poor and minority students, and assumed poor children couldn't learn.

Yes, it's true that some of NCLB's flaws have been "overhyped." That's what happens when a law is fundamentally flawed. But Chris Cerf and others may be setting up a straw man and using NCLB to knock it down. They credit NCLB with virtues that good educators have always possessed: a deep concern about their students and a desire to provide opportunities for every child. Educators not only believe that disadvantaged children can learn, they make it happen every day in their classrooms. They need some help to make it happen�whether that help is a good curriculum, good professional development, individual help for kids, or some of what we proposed this month in the community schools initiative. That's what we mean when we say let's build capacity, replicate what works and level the playing field.

Meanwhile, all the folks that talk about teachers should actually teach one class a day, or a week with the same kids, and try to achieve the same objectives I often hear about. Then they will see what it takes to convert their rhetoric to reality.

All of us have to take more responsibility, but that also means bottom to top accountability, not just top to bottom.

And, as far as Ryan Hill's comments go, I love when people talk about how research doesn't prove class size matters. If that is the case, why do so many charter schools, private schools and successful public schools lower class size as a means to differentiate instruction and ensure kids are not anonymous?

Chester E. Finn, Jr.:

Instead of ganging up to clobber NCLB, let's at least acknowledge that it's done a better job than anything before it of beaming sunshine down on the academic performance of every public school in the land and equipping parents, educators and state/local policymakers across America, as well as journalists and community leaders, with valuable information about how kids are doing in their own and other schools. Moreover, it hasn't settled for school-wide averages but has also reported results by grade and sub-group. This is a vast trove of crucial data that can be mined for diagnosis, school improvement and school choice in the places that still matter most, i.e. everywhere outside the Beltway. There's plenty about NCLB that needs fixing but this law has also made a historic contribution.

Christopher Cerf:

Some good ideas brewing. I am always leery, however, when the inevitable problems with tests, which undoubtedly include some cheating and some anti-educational test prep (as distinct from teaching the content on the test) leads to the conclusion that a regime predicated on high stakes tests must be irremediably flawed. Common assessments, whether old fashioned chapter tests or newfangled criterion-referenced tests, are critical to sound instruction. They allow good teachers to differentiate instruction, adjust pedagogy, and candidly assess what is working and what is not. Moreover, they provide some way to form a relative judgment about how students are progressing compared with others across a large population. Sometimes I try to imagine the world as the more radical "anti-testing" crowd would have it. I am confident it is not a place where children are attaining the knowledge and skills we can all agree are necessary to have a reasonable chance at competing in today's world. NCLB got it right in holding schools accountable for getting all students at least to a minimal level of proficiency on objective tests. Now the real work begins: 1) improving the tests; 2) norming them to national standards or keying off a single national test; 3) teacher development around sound teaching of content rather than test-taking skills, to name a few. But let's not fall into the trap of believing that the whole enterprise is beyond repair.

Diane Ravitch:

OK, so NCLB is historic. No doubt about it. Never before has the federal government reached so deeply into each and every public school in the nation. There was a fundamental error, however, in allowing states to define their own standards and write their own tests. As a result of this error, the public does not have the information that Checker speaks about; instead, in most states, the public gets a wildly inflated picture of student performance. Let us not praise those states that falsely claim that 80% or more of their students are "proficient," when the states define "proficiency" as hopping across a very low, very basic bar.

And let us not forget that accountability, however it is defined, may have perverse consequences. On this score, I commend to you Richard Rothstein's compelling narrative "Holding Accountability to Account," which describes how accountability works in a variety of fields and how it distorts goals and leads to gaming the system.

We really must do better or we will harm the kids and drive out many of the best teachers.

Ryan Hill:

Perhaps not all those schools are basing their decisions on research. My point, however, was not about the wisdom of lowering class size, it was about the difficulty of making informed decisions without a strong enough data set. I pointed to the conflicting studies on class size as an example of that, and as an example of one of the benefits of administering a national test. Whether class size reduction is a good idea is another important topic that the folks at NewTalk should consider for another time.

Having taught in both large, urban, district schools and in high-performing charter schools, I�m with you, Randi, on the optimism and the dedication of educators throughout the country. I�ve seen them in every setting, and the success of our schools depend on them. And among those dedicated practitioners, as in any field, there are outstanding educators who we can identify and mimic if we have the right measuring sticks; folks who successfully overcome even the lack of help and guidance that characterizes many big schools like the one in which I used to teach. As you said, there are also ways that schools can assist other teachers in becoming even more effective, and a common measuring stick similarly allows us to identify those that do this well, in order to spread their practices to other schools.

To improve the data set, we need a national test, and we also need to follow Deb�s suggestions � tracking cohorts longitudinally, adding more nuanced appraisals of student performance, and focusing on scale scores in addition to proficiency targets (which would reduce NCLB�s negative impact on gifted students), are all approaches that would give us a better sense of which are the most exceptional teachers and schools.

Gerald Tirozzi:

I think we need to return to John Merrow�s comment about common standards and the federal role in K-12 education.

If Congress truly wants to drive an education agenda for higher academic standards, it must not be content with a �tinkering� reauthorization of NCLB. If the goal of Congress is to be competitive in a rapidly growing global economy and address the inequality that exists for low income and minority students, it should shake the law at its foundation to institute national standards and a national test in reading and math. I fully appreciate the myriad and complex issues related to national standards and testing, and I expect that such a movement would be impalpable to policy makers (both Democrats and Republicans). But if Congress persists in its quest to have a significant federal role in public education, it must comprehend the major shortcomings of a national accountability system that allows for 50 different definitions of success. The reality is�absent a movement to national standards�in 2014, all 50 states will steer their vessels to Lake Wobegon to celebrate their students� universal proficiency in reading and math.

We as a nation have no aversion to national standards; we set them for everything from food to cars to toys. Why not national standards for all students in reading and math? There are, of course, a number of problems that national standards and assessments won�t solve. The mere establishment of standards and tests won�t lead to attainment. Dollars need to accompany the mandates, and we continue to hope against historical precedent. But at least we�ll know that, come 2014, whatever success we�re celebrating is genuine success.

Sol Stern:

I agree with John Merrow that this is "not just a New York City issue." What the New York case does demonstrate, however, is that it is impossible to have a fair national accountability system if the people in control of school districts and state education departments�and who are supposed to be held politically accountable for the performance of the schools as measured by test results�are the same people who have control of creating the tests and reporting on and interpreting the results to the public. One reason that we can have a rational democratic debate about the performance of our national economy and the degree of accountability of elected officials for that performance is that we can all agree that the basic economic data put out by the federal government is reliable. The section in the US Labor Department that releases employment figures is insulated from politics. Thus we can trust the basic economic data and then go on to have our political arguments about the best policies to follow. That confidence is almost entirely missing from the discussions around NCLB and its accountability system. Until we have a system of completely independent agencies (insulated from politics) to validate the reliability of test scores there will always be the suspicion that reports of education progress will turn out to be a chimera. As I said in my earlier debate with Chris elsewhere, and that he refers the other panelists to, "what our experience in New York proves is that politically ambitious mayors and school leaders will try to use their power as keepers of the data to advance their personal agendas at the expense of the public�s need to know the truth about student academic performance." This undermines public confidence in NCLB and feeds the cynicism of commentators like Charles Murray who deny the possibility of narrowing achievement gaps. Eventually this will turn into a disaster for the nation�s education reform movements.

Deb White:

A quick aside on the "highly qualified" teacher concept. To be honest�a lack of content is not usually what makes a poor teacher with the possible exception of upper level high school classes. (Please don't ask me to teach calculus!) Anyone capable of graduating from college with a teaching degree can learn the content in 9th grade science. A good teacher is the person who is passionate and engages and excites students. Master teachers need time to mentor and encourage new teachers, principals need to get out from under their paperwork and observe classes and give constructive feedback, subject area or grade level teachers need time to work together to develop activities and share their best ideas. Time is money and all the money that is being spent on testing and paper pushing is not being spent on teaching and learning.

Ultimately�teaching and learning is an inexact science�not easily evaluated. Like great art�you know it when you see it.

Sara Mead:

To Checker�s earlier point: absolutely. Disaggregated data, and disaggregated accountability, are essential to advancing equity for disadvantaged kids, and the reauthorization must preserve them. But I�m concerned many proposals to give states added flexibility in identifying or intervening in low-performing schools would undermine accountability for how schools serve all student subgroups.

John Merrow:

So, Deb, is it that we test too much or that the tests are crummy, or both? And let me throw into the mix Education Sector's finding�that we spend 15 cents of every $100 education dollars on NCLB testing. I know from conversations with the folks who make kitty litter, flea powder and other Hartz pet products that it spends at least 10 times that much testing its products.

Arthur Levine:

I spent much of the day on an airplane and just read through all of today�s postings in succession. I was surprised at the commonality of views the group shares, given the broad diversity of our politics and positions. Most of the writers were supportive of the purposes of NCLB and wrote positively of seeking an education system which is outcome-based, accountability-driven and inclusive of all children. These are not apple pie areas for agreement. The criticisms dealt largely with the familiar and widely agreed upon short-comings of NCLB such as AYP, highly qualified teachers, dumbing down testing, the diversity of standards, the narrowing of the curriculum and the like. Most called for revision rather than elimination of NCLB.

I take Randi�s observation regarding the dystopian view of schooling as very important. I don�t believe it is necessary to say that schools have failed to make the case for NCLB. Our entire world is changing-economically, demographically, technologically and globally. We live in a time which Checker alluded to. Information economies focus on common outcomes and industrial societies emphasize common processes. America�s shift to an information economy meant learning replaced teaching as the principal concern of schooling. This meant standards-based education, assessment and accountability. The information economy also made it essential that students have higher skill and knowledge levels than ever before. Moreover, this became true for all children if they were to be able to get decent jobs. Our schools, like every other social institution in America, were created for an earlier time and the challenge before us is to refit�government, healthcare, media and schools�for a new era. None should be accused of having failed, but each needs to change. NCLB recognized these new realities for the world of education, but sought to address them in a very imperfect fashion.

On outcomes, I can�t imagine federal action or national standards. But in recent conversation with governors, more and more seem to be talking about common standards across states and would be eager to join such an effort.

Frederick Hess:

Like most others engaged in this discussion, I�m of two minds about NCLB. I have great respect for its soaring ambition and its ability to emphasize a culture in which achievement and outcomes are central to the public debate (Let�s remember that, less than a decade ago, such a focus was far more controversial). Moreover, as Checker has noted, the copious sunlight that NCLB has cast on achievement is an unmitigated good. Of course, this is little more than ESEA and IASA had sought in 1994�and one regrets the foot-dragging and inertia that made the statutory overkill of NCLB necessary.

More significantly, though, I�m troubled by the manner in which NCLB�s authors took the �reinventing government� playbook and deployed it in ways that showed little regard for the lessons or limits of public policy. In place of tough, neutral scoring and sensible goals, they adopted heroic goals coupled with easily manipulated standards, testing, and scoring machinery. In lieu of combining increased outcome accountability with increased operational flexibility, NCLB largely kept the old regulatory framework in place and even imposed an array of new awkward compromises (can we spell �HQT�?). Rather than judging school or district quality (e.g. how much students are learning in the subjects of interest during the academic year), adequate yearly progress and the remedy cascade are oriented on where students stand at a particular point in time�with all the attendant problems of misaligned incentives.

The result is a law that embraces the language of �outcome-based accountability� but that looks much like its ancestral Great Society legislation�larded with new mandates, awkwardly conceived testing requirements, and an ambition that casually ignores the constraints of federalism, statute, or bureaucratic competence. Indeed, the incentives for states to dress up outcomes and for schools and districts to carefully follow rules rather than necessarily deliver real results has arguably led to a budding new compliance mindset.

Consequently, I see a strong case for setting clear and coherent metrics for measuring achievement, demanding that states identify particularly low-performing schools and identify strategies for addressing them, and reshaping NCLB so as to encourage new providers and delivery systems. This will likely mean moving away from the SEA/LEA framework that worked well as a mechanism for pushing out dollars, but not so well as a means for addressing mediocre schools or pioneering more effective solutions. Would this constitute a radical rewrite of NCLB? I�m not sure. I guess it depends on what one regards as the core precepts of the law. But either way, I do fear that standing by the law�s �bright-line� principles as currently understood has a real chance to not only unravel NCLB but to set back a quarter-century�s worth of thoughtful efforts to bring schools out of an era of input regulation and promote meaningful educational accountability.

Deb White:

I think that for many kids, a multiple choice test is not the best way for them to demonstrate proficiency. In Wyoming we have created performance assessments at the state level for each grade level and subject area. Each subject area picks two to give during the semester. So, for example, all students in Physical Science complete a final exam and the same two performance assessments that every other PS student in the school completes. This assesses the student in relation to the curriculum and to some extent to the teacher. The performance assessments are graded on a uniform rubric.

This is obviously more time consuming than a multiple choice test and in fact some of the performance assessments take multiple days. However, they are interesting and assess multiple real life skills. An example would be in Physical Science the students must complete a consumer report type project where they select two brands of a product, identify the five essential qualities of the product (i.e., absorbency for paper towels), design a controlled experiment to evaluate each quality, gather data, create tables and graphs and draw conclusions using a decision matrix. Is this a "better" assessment tool? YES! Is it easy and fast? NO.

All students do not learn in the same way and all students cannot be evaluated in the same way. In fact�we don't want to produce students who are all alike.

DAY 2

John Merrow:

Good morning on Day Two (less than 48 hours to solve all the problems!!)

Many of you seem to support 'National Testing,' and I wonder if we could explore that a bit more. Doesn't national testing assume the existence of national standards? And how far are we from that?

I am struck by the wisdom of Achieve, Eli Broad and others who talk about 'Common Standards,� perhaps recognizing that 'national' and 'federal' are widely confused concepts and red flags to many Americans.

But who sets the common standards? Are they evolving before our eyes, thanks to the work of Achieve, the New England group known (unfortunately) as NECAP, and others?

If, say, 35 or 40 states adopt common standards and tests, will that be sufficient to push the rest to follow?

Is the federal role to enable, i.e., provide the money to develop demanding standards and tests?

I look forward to the day.

Chester E. Finn, Jr.:

Multi-state standards are surely evolving before our eyes�Achieve, the CCSSO and others are hard at work on this�and at least one version is apt to evolve from an improved version of the "American Diploma Project" standards that are then "backward mapped" from high-school graduation to earlier grade levels. (Others may do likewise. It wouldn't be a bad thing to have several such projects.) I'm bullish about this approach and hope many states join in. I hope it also leads to multi-state tests that are aligned with the multi-state standards. This isn't quite the same as "national standards" but�assuming the resulting standards and assessments are sound�far better than today's motley array of individual state standards and tests. What is the federal role here (if any)? Extremely limited, I believe. Maybe grant (or even earmark!) dollars to underwrite these efforts. Maybe some favorable treatment (a lighter NCLB touch, say, or some extra money) for states that participate in them. Separate from all that, the feds should continue NAEP as the chief "external auditor" of academic performance by kids in states that do and don't join these voluntary multi-state ventures. But I don't know anybody who really wants the Secretary of Education or the U.S. Congress to write the standards for American schools.

Thomas Rogers:

The problem is not that we�re teaching to the tests, it�s that we don�t have tests worth teaching to.

Perhaps that�s an overstatement. However, even in a state that gets top marks for our system of standards and assessments, the tests are largely measures of basic computation and memorization skills, basic literacy, and some common content. Contrary to others, I believe de facto national standards are coming�in the form of the increasing prominence of NAEP as an external validation, and the obvious tenet that tests are only valid when they�re aligned to curriculum and standards. Therefore, I agree with Chester Finn, that we should be bullish on the prospects for richer standards leading to better tests.

Globalization and the knowledge revolution have almost overnight reprioritized skills for the high-wage workers we hope our children become. Our standards have been slow to keep up, too focused on rote, and unable to measure problem solving, creativity, innovation, and aesthetic skills. The richer assessments Deb White describes panic policy-makers whose legal concerns force them to prioritize technical validity over authenticity. Right now, comfortable validity margins can only be achieved by narrow, right/wrong answer tests. But how many of the problems we regularly solve as adults already have right/wrong answers lurking in the back of a textbook?

Hobbled by these considerations, the states are a poor laboratory for assessment innovation. Thus the obvious federal role is to invest deeply to research and develop the next generation of assessments. They in turn will drive practice, teacher preparation and professional development, and accountability.

Jerry Wartgow:

It was encouraging to see the support for �Common Standards� during yesterday�s discussion. And as difficult as it might be for this idea to become politically palatable in Washington, setting the standards is the easy part of the equation. The more difficult part is building the capacity that will be necessary to attain them. We�ve been down this road before. We adopted National Education Goals in 1990 with overwhelming agreement that by the year 2000, US students will be first in the world in math and science achievement; the high school graduation rate will increase to at least 90%; all students will leave grades 4, 8 and 12 having demonstrated competency in a wide range of subject matter. Not only did we not attain these goals, we lost ground on many, and they could be reissued today with no changes and would likely again be supported with the same level of enthusiasm exhibited 18 years ago. With the same type of hype, we adopted NCLB in 2002 with the goal of having every state bring 100% of its students to proficiency in designated subject areas by 2013-14. It is clear that if current trends continue we will fall far short on this goal as well. If we do establish national standards I�d be interested in suggestions for what else might be done this time around to prevent history from repeating itself.

Sara Mead:

Jerry is absolutely right about the need to build capacity. NCLB, and really the Improving America�s Schools Act (IASA) before it, asked state departments of education to do a whole bunch of new things, such as implementing high-quality assessments and overseeing and supporting intervention in chronically low-performing schools, for which they were totally unequipped. We need to radically rethink the structure and functions of state departments of education so that they can become partners with districts and schools in improving student achievement, not the compliance monitoring and check-cutting agencies they are now. The impetus for this needs to come primarily from the states themselves and from private philanthropy, however�although the federal Department of Education could help by shifting some of its own capacity in ways that model this better�it also needs to change.

What we really need the feds to do on the capacity side, however, is to be much more aggressive in supporting high-quality research and development, as well as catalyzing innovation to develop solutions to the educational problems that NCLB has laid bare. If we�re going to educate students for the demands of the new economy, we need much better tools, curricula, and instructional strategies, as well as ways to use technology to get efficiencies in education�but right now we�re not investing in the research and innovation needed to develop those tools.

Sol Stern:

Before we get to national or "multi state" standards, we ought to make sure that NAEP really does become what Checker describes as the "external auditor" of state tests. For a number of reasons, that's not the case now. The public is clueless about NAEP and state and district officials can use that lack of public awareness to gloss over the gap between what their own tests show about student achievement and what NAEP reveals. At the very least we should support Secretary Spellings's proposal to require state and local education authorities to post NAEP results when they release their own test scores.

Arthur Levine:

I want to raise a subject which has not been discussed so far�the impact of NCLB on universities, teacher education, and teachers. There is a fairly heated debate about whether teaching is a profession (requiring substantial education before entering a classroom like medicine) or a craft (learned largely on the job like journalism). NCLB has tipped the balance with the �highly qualified teacher� definition in favor of the craft position. This has produced an outpouring of alternative non-university-based paths to enter the teaching profession. The good news is that the competition has spurred action on the part of a number of universities to improve their teacher education programs. The bad news is that the teachers being produced by alternative routes are being hired principally by high-need schools where children need the most skilled and best prepared teachers. The university people charge that the research shows their programs are better. The alternatives counter that theirs are superior. In point of fact, a comprehensive examination of the research, which is often poor, shows no compelling evidence that one is better than the other.

The programs tend to be mirror images of one another�alternatives emphasizing clinical instruction and by expert practitioners. The traditional routes focus on theoretical instruction by academics. What would be desirable is a fusion of the approaches. To me, the definition of a highly qualified teacher is inadequate. I would prefer to have my children taught by teachers who know about child development and how children learn, how to prepare a curriculum, different pedagogical techniques, assessment, classroom management, etc. I would propose that skill and knowledge levels for teachers be raised to incorporate this under the banner of �highly qualified teachers.� I would suggest government (state or federal) funding for university-based and non-university-based teacher education, with compelling plans, to achieve those standards. In turn, states should review all of their teacher education programs (educational administration programs too) and close those that don�t meet these standards.

Frederick Hess:

I'm following up here on Sara Mead's 11:12 a.m. post, and want to push a bit further than Sara did.

There's a tendency (one that Sara admirably avoided in her post) to discuss "capacity building" when it comes to state departments of education or school districts as if it were a self-executing matter of applying dollars and knowledge. We too often tend to skip past the questions of incentives or institutional design in order to call for additional "investment" and expertise. In fact, I'm not aware of any industry where there are dozens or even scores of highly effective firms with the ability to turn around low-performing organizations.

To my mind, this suggests that we need to not merely devote more dollars to state departments and reconfigure them, but to ask whether the LEA/SEA framework that was the backbone of an ESEA focused on pushing dollars out to districts and schools is really the architecture equipped for a performance-based system. I'd argue that experience in many other sectors, where that kind of expertise and talent is concentrated in no more than a handful of outfits, suggests that we will need to embrace a new model that relies much more heavily on new providers.

Indeed, I would suggest the same question of rethinking incentives, institutional design, and the capabilities of existing entities will prove critical when it comes to teacher preparation, or to ensuring the testing and assessment industry is prepared for a future that many contributors are here envisioning.

Sol Stern:

Arthur Levine is right to raise the issue of how we define "highly qualified teacher" and therefore how our teachers are being trained in the universities and alternative programs. But let's get real about this. Prospective teachers attending the most prestigious ed schools in the land, including Columbia�s Teachers College, are likely to get a huge dose of Paolo Freire, Jonathan Kozol and William Ayers (American Education Research Association�s [AERA] Vice President for Curriculum)�but will hardly have heard of Diane Ravitch and E.D. Hirsch. They will know all about multiculturalism and diversity (the left wing versions) but nothing about the science of reading instruction. And for good measure they will have to demonstrate a "disposition" to social justice, but will remain clueless about how to manage a classroom in a typical inner city school. And if you want to discover why 2/3 of inner city, minority kids can't read proficiently at fourth grade the best place to start is the steady diet of whole language and balanced literacy their teachers received at their ed schools.

Randi Weingarten:

It's great that Achieve has been able to find a way to move toward national standards by working with the states and moving the consensus outwards, rather than starting at the top and moving down. Their work shows it's doable and I'd like to see more of it, more states, more subjects.

Standards, however, are only a beginning. If America's children are to attain these higher standards, focusing on the classroom is the essential next step. To reach high standards, there must be a high-quality curriculum, solid professional development, and fair, objective assessments. We also need to recognize that focusing solely on testing will not lead to better schools. Most important, teachers, who, after all, have the day-to-day responsibility of instructing our students, must be involved in creating and delivering the high-quality curricula, assessments and professional development they need to help students meet these new high standards.

Standards are about what kids should know and be able to do, and the reason to have a common and rich set is to ensure that we don't leave any child behind, due to the neighborhood in which he or she grows up. But the bigger point is not to let the debate delay our efforts to simultaneously provide kids the education they deserve.

We all know we need to set high standards, establish a strong curriculum, develop high-quality assessments and take other steps to help children learn. All of us�teachers, students, parents, government, school officials, community leaders � also have to take on greater, shared responsibility for the success of our children.

Thomas Rogers:

Earlier the discussants noted that �Highly effective� (outcome measure) is more important than �highly qualified� (input/process measure), suggested means for measuring effectiveness (CLASS) and programs for improving effectiveness (national board certification), and we�ve even discussed the preparation pipeline through universities and other routes.

What�s next is to raise two important points. First, there are some 3 million teachers/aides already in place with training and skills that produce our current results (both good and bad). If we�re to redefine the �good� (i.e. recast new standards, assessments and curricula) and improve the �bad�, we�ll need to retool that workforce at scale, not wait for its eventual replacement.

Second is to recognize that our distribution of effective teachers is currently uneven. �Effectiveness� is strongly correlated with longevity in the first 5 years of teaching. Yet new teachers and alternative route teachers (who are by design�as in Teach for America�often planning just 2 years teaching) are concentrated in the most challenging schools.

Given our apparent consensus about the importance of good teaching. These two concerns should be among our top priorities.

Arthur Levine:

Sol, this sounds just dreadful. I would be grateful if you shared the empirical evidence documenting these claims. I just completed a five year study of schools of education, piles of empirical data, which ended up producing three reports on education schools. They were quite critical, though they did not produce the same findings you presented. I am not suggesting that education schools be maintained as they are. Rather I think they need to be reformed. I think our choices are to invest in accountability-driven education school improvement or create alternatives to the education schools. While it is always appealing and easier to create something new, I think it makes more sense economically to try an experiment to fix education schools, beginning with say 50-100 schools, because education schools are now producing over 90% of the nation�s teachers. In contrast, the alternatives are hothouses producing less than 10% of teachers. On average, they cost significantly more than traditional education schools. Investing in both would not be terrible. For the short term, I am urging we put the bigger bet on education schools. We are trying to do this at Woodrow Wilson right now, to turn around education schools to meet standards the Foundation set.

Sol Stern:

Arthur, this is dreadful. As for the evidence, you surely must recall the study done several years ago by David Steiner (now Hunter College ed-school dean) and Susan Rozen surveying the syllabi of the basic �foundations of education� and �methods� courses in 16 of the nation�s most prestigious ed schools. The mainstays of these courses were works by Paolo Freire, Henry Giroux and Jonathan Kozol. For the methods courses, Bill Ayers�s To Teach: The Journey of a Teacher tops the bestseller list. Neither list included advocates of a knowledge-based and politically neutral curriculum, such as E. D. Hirsch Jr. or Diane Ravitch. Look at your own Teachers College: Lots of Lucy Calkins and Bill Ayers, but nothing on scientifically based reading research. As for the hold that social justice teaching has on the ed school professorate, consider that two of the recent past presidents of AERA, William Tate and Gloria Ladson Billings, are strong advocates of this pedagogical approach. In fact professors Ladson-Billings and Tate edited a book proposing social justice teaching in the public schools (including in math classes) published by Teachers College Press.

Ryan Hill:

Arthur Levine hit the nail on the head. National standards and a national test, regardless of how they are developed, will not by themselves fix everything. They will make it possible, however, to figure out what�s working�especially if they�re coupled with a well-funded federal R&D effort. The next step is to evaluate our training and selection methods, because teacher quality is critical. Tracking the results of the graduates of various ed schools would go a long way to helping us determine what works in educating the educators, and what doesn�t. I grew up playing basketball, and it would�ve been hard to learn how to shoot a jump shot if I couldn�t tell whether the ball went in or not.

Sol is wrong to assert that kids can�t read because of whole language and other efforts in curriculum and instruction. A great teacher can take any curriculum and get his or her kids to learn, and a poor teacher can have the world�s greatest curriculum and see no gains at all. Every teacher knows this from experience; in my second year in the classroom, I got much better results with a lower-tracked class than I had in my first year, and this had nothing to do with the textbooks I received (none in either year). This is not to say that curriculum isn�t important�it is�but it isn�t make-or-break.

What is make-or-break is teacher education. The work Arthur is doing at the Woodrow Wilson school is groundbreaking, and there are a number of other new initiatives around the country that offer a similar combination of up-front instruction, intensive mentoring, and an easing into the full-time teacher responsibility. We�re working on just such a program in Newark, and a number of other charter schools like MATCH in Boston have been pioneering in how they�ve built their on-ramps to teaching. One of the most ambitious of such efforts is the Teacher YOU program that combines the efforts of KIPP, Achievement First, Uncommon Schools, and Hunter College in NYC. These programs treat teaching as both profession and craft.

Deb White:

I just got off a plane and have been reading today's posts. My perspective is so different from the majority of you that it took a while to digest the info. A couple of thoughts from the classroom�not necessarily in a prioritized list.

Although I don't have the opportunity to observe many other classrooms, I spend lots of time talking to smart high school kids and they are brutally honest about their teachers. They know who knows the content but doesn't care about them and they know which teachers are there as a "job". (People tend to teach because they love their content area or they love working with kids.) I sponsor a number of programs where I haul high achieving students all over the country. We have lots of time to talk on school buses and planes.

1. Luckily for all of us�different teachers connect with different kids. Different learning styles, teaching styles, personalities, personal history and subject area all play a role. Some of the least likely individuals end up being the most inspirational and successful teachers.

2. The track a teacher takes to the classroom is not nearly as important as the absolute belief that their being there is going to make a difference. Passion and enthusiasm count as much as content knowledge in most cases. Career or craft�on any given day it could be either. I do know that the first few years are more difficult than most people can imagine.

3. We MUST include alternate forms of assessment even though they may be more subjective and therefore scarier to statisticians. (This also makes them more difficult to cheat on.) Some of the least prepared students I have ever sent off to college were great test takers. They could blow the top off of any standardized or final exam, but they could not solve a problem to save their lives. Life is about "figuring out what the problem is and then, how to solve it". Your mechanic, banker, electrical engineer and doctor do it every day. We have to help kids get there. That is the job of schools.

4. We need to free up some of the schools that have proven they can be successful to be successful and keep innovating and moving forward. When a school reaches 80-85% proficiency, require that they hold steady, and then let them free up some resources to try new things without the risk of being labeled a failure in the local paper. Not a great way to encourage an evolution in education. (Once a school has hit 80-85% proficiency, they probably can't go much higher based on all of the other variables affecting the students.)

5. CLASS SIZE MATTERS! Maybe it does not affect test scores because if you have a huge class you do fewer projects and more multiple choice tests just due to the load of grading. However, the kids do not learn as much or enjoy the class as fully. Having a positive experience in, say, physical science will determine if a kid decides to takes physics or chemistry. Small classes allow teachers to know their students as people, encourage class discussions where student interests can come forward, and results in more time spent learning because less time is wasted on discipline.

6. Every student needs to be able to read, write and do math. If you had a dinner�these would be the utensils. All the other subjects are the food. You can't eat the food if you can't use a fork and you can't learn physics without algebra.

7. As NCLB is re-written, include teachers and students.

Arthur Levine:

Ryan, your point that teacher preparation needs to be taught as craft and profession is excellent. Sol, I have been very critical of the research done in education schools and the field of education in general. I often show students examples of very bad research. The study you cited is one I almost always use. It is an excellent example of ideology trumping rigorous research methodology.

John Merrow:

Your moderator weighing in with props for Randi Weingarten, who wrote in part "....to ensure that we don't leave any child behind." What I am cheering about is her use of the active voice, which takes responsibility.

The very name of the Bush-Miller-Kennedy legislation, "No Child Left Behind," reveals its basic weakness. That inert language neither takes nor assumes responsibility. However, if we begin with active language, we are making a promise: "We will do all we can to make sure that we do not leave children behind." What follows from that, for me at least, is a commitment for a national debate about what we mean by 'behind,' what we mean by the goals of education, and so on.

I suspect it was not an accident that, when the lawmakers ripped off the Children's Defense Fund, they shifted from active to passive.

It's time for a change of voice, of attitude, of expectations, of commitment...

Deb White:

Randi also makes the very valid point that communities, local or national, must step up and take some responsibility for the education of our children. We all know "it takes a village" and schools cannot be successful in a vacuum.

Ryan Hill:

[Editor�s Note: This posting responds to a Reader Comment by Grant Wiggins of Authentic Education]

Grant, you're right. Those of us who see the good in NCLB sound a bit like airplane designers circa 1910 would've sounded sitting around and griping about how stupid the Wright brothers were. Sure, we want to build a better airplane, but NCLB was a breakthrough moment in education, and without the first fixed-wing aircraft, we'd never have gotten to the 747. Now we need to figure out what education's 747 looks like.

Randi Weingarten:

We've had a good dialogue about what's wrong with NCLB, whether it's awful or bad or not so bad, but I hope we can shift to a more important discussion about possible solutions.

One I've proposed is to use federal education law to promote community schools�schools that serve the neediest children by bringing together under one roof all the services and activities they and their families need.

In places like Baltimore, Brooklyn, Chicago, Minneapolis, and Philadelphia, community schools are open all day and offer after-school and evening recreational activities and homework assistance. High school students at community schools can sign up for morning, afternoon or evening classes.

Community schools include child care and dental, medical and counseling clinics, or other services the community needs. For example, they might offer neighborhood residents English language instruction, GED programs or legal assistance.

In cities where mayors control the schools, they could�and should�consider community schools as a way to use their power to integrate services on behalf of kids.

Sol Stern:

Arthur, you asked me for "empirical evidence" to support my claim that the ed schools were offering prospective students ideology rather than lessons on what works in the classroom. In reply I cited a study by one of your ed school colleagues, David Steiner, showing that foundations and methods classes in ed schools were dominated by left wing ideologues like Kozol, Freire, Giroux and Ayers. You now say that Dean Steiner's study is based on "ideology" rather than rigorous research, but you offer not a single shred of evidence�"empirical" or otherwise�to support your statement. Are you saying that E. D. Hirsch gets equal treatment with Paolo Freire in the ed schools, and particularly your old school, Teachers College? Please enlighten us further.

To the other members of the symposium who might think this is all irrelevant to the question at hand, I would say please pay attention to what our future teachers are being taught in the ed schools. If their professors ignore all the scientific research on the best methods of teaching reading in the early grades, for example, and instead train future teachers in methods like whole language that have never been proven to work, then much of the rest of the discussion we have been having here will prove to be totally inconsequential. On that note, Ryan, your assertion that classroom teaching methodologies don't matter is contradicted by three decades of scientific research on reading conducted by the National Institutes of Health. To ignore this rigorous, peer reviewed research, to assert that "good" teachers can use any methodology and succeed, is creating a handicap in the schools that not even the best federal education law can overcome.

Frederick Hess:

Actually, I�d take some issue with that, Ryan. Like the moon landing of �69, the Wright Brothers feat was a breakthrough accomplishment of engineering and execution. That�s quite different from �breakthrough� legislation notable for its ambition and intention to change the culture. Model Cities, Aid to Dependent Children (later AFDC), and ethanol subsidies were all one-time breakthrough statutes�that didn�t mean their design principles were necessarily sound or worth building upon.

Instead, the case for radically rethinking NCLB is that one can its admire its goals and munificent transparency while thinking it represents shaky foundation for future efforts. It�s not merely a question of improving aerodynamics and design, but asking whether the law was crafted in a fashion that�s likely to support sustained progress towards its popular goals. On that count, I�ve severe concerns�some of which I�ve previously flagged.

John Merrow:

We've covered a lot of ground today, mostly focused but with a few side trips. Reading through the comments, I find a lot of agreement. (and I encourage you all to read the comments from readers)

I keep thinking of Reagan's law, "trust....but verify." How does that apply to education? Should the new President call a 'time out' and have a genuine summit (not like 1989) where thoughtful people can argue about the purposes of public education?

Early on Checker argued for turning the law on its head: set the broad goals and ends and let states figure out how to get there. That's fine, but what would the law look like? Is it time to talk again about 'opportunity standards'

What I hope to see tomorrow are some concrete proposals. Would you have the federal government sanctioning schools and districts? Should the next legislation and its regulations cover AYP and HQT, et cetera, or has the experience of the past years proven that Washington cannot run public education?

And thanks for your enthusiasm.

Sol Stern:

Randi: Spending on New York City schools has increased from $12.7 billion when Mayor Bloomberg took over, to $21 billion for the current fiscal year budget. That's the biggest spending increase in the history of the city schools, and possibly in the country. Yet the city's NAEP scores haven't improved on three out of four of the benchmark tests. So now you want to spend billions more on an asserted but untested causal connection between better health and social services and higher academic achievement? And what makes you think that the same officials you have rightly indicted for creating a flawed accountability system for the schools will do a better job creating school health clinics?

Arthur Levine:

The problem is the lack of rigorous research on the topic of educating teachers. We could continue to debate this subject, but instead I bet we could agree on at least one thing. Education schools have serious problems. We probably disagree about the remedy. I would try to fix them and others might want to close them.

Randi Weingarten:

Sol-I understand your skepticism, but you are missing my point. This is an investment, that should assist in helping kids succeed, since simply focusing exclusively on testing, sanctions and teacher quality I don't believe has proven to be the formula to help all our children get to the educational standards that NCLB was supposed to be all about. Just like a dollar invested in early childhood education saves much later, investing in the whole child will level the playing field for kids who are at risk.

Philip Howard:

With apologies for a late arrival at the discussion (I was traveling all day yesterday), I hope the panel here can focus on concrete solutions. Sol, this isn't supposed to be a shoot-out at the corral, or a debate to score points against other participants, but a civil discussion on an important subtopic--what to do about NCLB. In the interest of fairness, I offered Chris Cerf an opportunity to respond to what he considers inaccurate and inappropriate statements about the Mayor, the Chancellor and the work in NY generally. He has declined in the interest of allowing this important discussion about NCLB to proceed without any further distraction.

DAY 3

John Merrow:

It's the beginning of the final day of this conversation/debate, and I am asking all of you to offer short and quite specific suggestions for the next ESEA/NCLB.

As I read over what you have written, different perspectives jump out. Some suggest that NCLB has it backwards, but how could that be rectified?

There's disagreement about the proper federal role. If Washington should limit itself to collecting accurate information and supporting opportunities for the disadvantaged, how can it do that effectively?

If, as history suggests, states and school districts and schools cannot be trusted to educate all children, what can Washington do?

If you feel that we must have common (demanding) standards and common tests, what's Washington's role in enabling that to occur?

In short, what's the proper mix of carrots and sticks?

(And I have been assured that, from this point forward, the gatekeeper will not post any more personal attacks, so don't waste any energy on that.)

Jerry Wartgow:

John, after reflecting on the dialogue of the last two days I offer the following in response to your last post. I believe that it is appropriate, important and in the national interest for the federal government to promulgate common standards and assessment techniques for measuring progress toward meeting those standards. NCLB can be a vehicle to do that. On the other hand, I do not believe that it is appropriate or productive for the feds to establish unrealistic and unattainable goals (i.e. 100% of the students will be proficient . . . ) and then punish and pillory the schools for failure to attain them. NCLB in its current form is too big, too long and too complicated. The index alone consists of 28 internet pages with the Act itself comprising another 670 pages. It needs to be significantly slimmed down and pared back. Our discussion focused mostly on testing and teacher preparation but there are thousands of other bureaucratic requirements emanating from the small print that distract those who are in the trenches of school reform from their primary teaching and learning responsibilities.

Philip Howard:

One change almost everyone seems to agree upon�national standards for testing.

Federal sanctions are another matter. I subscribe to Diane Ravitch's view that sanctions promote obsessive behavior. Parents and educators care. Give them this uniform metric, and most will respond in productive ways.

I also subscribe to Chester Finn's observation about being "loose as to the means" of improving. Most NCLB hurdles and sanctions are too clunky and bureaucratic�for example, annual yearly progress doesn't take into account the context�that a brilliant student is unlikely to get more brilliant or a severely disabled child is unlikely to improve. Teacher "certification" is another example of a useful goal�to improve the quality of teachers�that is ruined by a bureaucratic standard. As with other professions, training is vitally important. But as studies of effective teachers repeatedly show�for example, the work by Prof. Philip Jackson or Sara Lightfoot�much of what makes a teacher effective is the personality and spirit of the particular person. Having a graduate degree doesn�t mean the teacher cares, or can relate to the students.

Randi Weingarten recently floated the idea of schools as community centers, an idea reminiscent of my hero Jane Jacobs. What is powerful about that is that focus on culture. Good schools, studies repeatedly show, all have a nurturing culture. Even the best teacher has a hard time being effective in a school where, instead of mutual respect for the mission of learning, the culture is characterized by disorder and defensiveness. Listening to teachers, as some have pointed out in the public comments, the heavy-handed sanctions of NCLB further undermine already-shaky school cultures.

As Chris Cerf points out, NCLB injected two important goals�testing and accountability. But I read through the 670 page statute this week, with some pain, and do not see the point of most of those bureaucratic provisions. I'd vote for focus on uniform standards, with tracking of individual students. I'd do away with most of the rest, and replace rigid accountability with affirmative outreach programs to deal with problem schools and districts.

Thomas Rogers:

1. My earlier suggestion for a Federal role included R&D for a new generation of tests of higher order thinking skills, and some yardstick (NAEP?) so the states can be compared. Like Ryan, I�m not against NCLB, rather I�m for its improved successor.

2. Similarly, since the strongest correlation to low achievement is poverty, remediating poverty�s effects should be our first place to start, not the last. I embrace Randi�s idea of co-locating existing children�s health programs at schools (like the Community Schools program), instead of leaving them scattered all over cities and towns.

3. As for accountability. Neither NCLB, nor, say, New York�s accountability system recognizes any responsibility for (or imposes sanctions on) either the state or federal government. Yet both significantly impact student outcomes�by controlling the amount and fairness of allocating education funding; the quality, timeliness, and usefulness of standards and tests; and societal supports for children in poverty. The reason why NCLB �sanctions� have worked so poorly is because the statistical effect size of societal and funding factors is far larger than the statistical effect size of what schools can control � hence the call for value-added measures and (regrettably) lowered standards in some states.

Educators embrace accountability, but only for things they control. NCLB�s fallacy is that schools alone can somehow overcome the deficits created by a society that has the second highest rate of childhood poverty in the OECD, or a state that has the worst segregation by race compounded by a funding system so inequitable, our highest court threw it out.

Reciprocal accountability would consist of a series of mutual promises�government would be responsible for creating the conditions to maximize schools� potential success; schools would deliver results. Consequences to schools would differ if states/feds failed to deliver their promises. Government�s progress would be measured on criteria like childhood conditions and educational systems alongside schools� progress on student achievement. Parents should see both their child�s grade, and their state�s grade simultaneously. If government promised to eliminate childhood poverty by 2014, I�ll bet every educator would be willing to be held accountable for universal proficiency by that same date./

Sara Mead:

One thing no one has mentioned at all in this conversation is NCLB funding. The vast majority of school funding comes from states and school districts, but with federal increases for NCLB in recent years, the feds are now up to providing nearly 10 percent of K-12 public school funding�and much more in some high-poverty school districts. Federal policymakers should be thinking about how we can improve the way that funding is distributed to best advance equity and NCLB�s goals, and also how that funding can be used to leverage changes in state and local distribution of funds.

Here are two ideas we have:

Strengthen Title I�s comparability provisions: Under NCLB, school districts must provide �comparable� services to Title I (higher poverty) and non-Title I (lower poverty) schools using state and local resources. Federal Title I funds are supposed to come in on top of that level playing field. But the comparability provision is riddled with loopholes: More significantly, when determining whether or not services at Title I and non-Title I schools are comparable, school districts don�t account for differences in teacher salaries based on years of experience. This loophole supports inequities in teacher distribution and a system in which school districts persistently spend much less on teachers in high-poverty schools than they do on those in low-poverty schools. Fixing the comparability provision, by requiring school districts to spend comparable amounts on teacher salaries in high- and low-poverty schools, including salary differences based on years of experience, would push school districts to address inequities in teacher distribution, spur innovation to attract good teachers to high-poverty schools, and result in reallocation of state and local revenue towards the low-income students NCLB seeks to serve.

Distribute a greater share of Title I funding through the Targeted and Education Finance Incentive Grant formulas: This would better target federal funding to the highest poverty schools that most need it.

Ryan Hill:

I very much agree with Philip�s points, and I�d like to bring up another area in which there is tremendous opportunity to improve our schools: leadership development. We�ve talked a lot about the current problems with teacher development and training, but most principals and assistant principals have received even less adequate training for a job that is just as complex and difficult as teaching is.

The authors of the book The Carrot Principle describe the �basic four� principles that I think provide a pretty good working definition of leadership: goal setting, communication, trust, and accountability. NCLB can certainly help with accountability, and to a limited extent, with goal setting, but if the managers and school leaders don�t use the tools provided by NCLB, well, it can actually undermine trust and pervert communication. In other words, accountability systems, like any tool, are not good enough; people must know how to use them.

No competent CEO would think that issuing edicts and meting out punishments from on high with no intermediary management support would be a good way to run a business. In fact, the closer a manager is to the point-of-service�in this case, the classroom�the more critical it is for him or her to be adept in the tasks of management. Unfortunately, very few of our principals and even fewer of our vice principals are well-trained in people leadership, so what happens is that the accountability measures of NCLB are used as if they are precise instruments, or as if they are the only instruments. Good managers would be able to give specificity and robustness to their teachers� goals; bad managers use the only hammer they can see to hit everything that looks like a nail.

Ryan Hill:

Through NCLB, our nation�s CEO of education gave us our strategic objective, however unrealistic: 100% proficiency. It is now the task of management to implement that. For that to happen, we must:

1.Get better at multi-faceted evaluation of teachers, both to figure out whose practices we should be mimicking and whom to move into managerial roles. Too often, teachers are �promoted� into management because of personal ties to the principal or another administrator, or because they don�t make waves. If we were better at evaluating teachers, we�d have a better sense of which teachers would make great managers.

2.Invest serious time and effort in the training of managers and principals. Top-performing organizations spend a lot of time training their leaders; we should do the same in education.

3.Build leadership pipelines wherein managers are not only provided training, but also are evaluated on how well they perform the tasks of management. A transparent leadership pipeline would serve the dual purposes of creating better managers and keeping top performers in education longer, as possible career paths would be more evident.

4.Loosen the bureaucratic reigns on principals who have been adequately trained so that they can actually apply the skills ensured by numbers 1-3.

Jerry Wartgow:

I am going to have to sign off shortly but want to thank NewTalk, and my colleagues in the crusade to improve student achievement, for the opportunity to participate in the dialogue. I would also offer one final observation. We must remember that even if we �fix� NCLB, it does not follow that this alone will translate into no child being left behind. The majority of research findings indicate that most schools could be held responsible for no more than 25% of the achievement of their students. George Will put it this way, �The intractable problem for schools is �9/91�: only 9 percent of the hours lived by young Americans between birth and their 18th birthdays is spent in school, and the other 91 percent � families, popular culture and the culture of the streets � often overwhelms what schools do.� Schools must do their part and be held accountable for their performance. But we are kidding ourselves if we think that they alone can provide a level of student achievement that will keep our nation competitive in the rapidly changing global environment. We need to also take a close look at what we can do to improve accountability and responsibility among those who have influence over our youth during the 91% of the time they are not in school. Maybe the topic for a future NewTalk?

Charles Kolb:

I have tried to follow this fascinating and, at times, intense conversation. What I�d like to ask is whether any of the participants envision a role for the American business community in what comes next for NCLB?

Deb White:

I think that between them, Jerry Wartgow, Tom Rogers and Ryan Hill have accurately identified the perspective of the classroom teacher. The vast majority of educators chose their career for all the right reasons�they have (or at least had) great intentions of changing the world. However, it is incredibly difficult to be held accountable day after day for things over which you have no control. I will happily be held accountable for getting a grade level of improvement out of any student in my classroom, but don't hold me accountable for how the student that moved in from out of state last week does on the test. ("Snapshot" type tests are especially tough on kids who have a lot of turmoil in their lives.)

Ryan Hill is correct as far as school administrators go. There are some great ones, but many have attained their position through not rocking the boat, longevity in the system or worse yet�have been "promoted" out of the classroom rather than be terminated.&mbsp; (They just did not get along very well with kids....) Sadly�many great teachers move into administration because it is the only way they can make more money. Once teachers top out on the pay scale�leaving the classroom is the only way up. Maybe a mechanism where master teachers could continue to receive pay raises would keep more of them in the classroom. (Or back to the idea of pay incentives for National Board Certification)

The question is�are we running toward being the best educators we can�or are we running away from failing? A visionary and risk-taking building or district administrator can move a district forward in amazing ways. Unfortunately, risk-taking is not rewarded in education and not many administrators need their job and don't want to fail in a big way in the public eye.

In defense of administrators�even the ones who would/could be a great asset to new teachers, end up spending the majority of their time pushing papers or sitting in meetings, rather than interacting with staff and students. This has gotten worse as more and more state and federal requirements have been added on.

Tom Rogers is right on target�educators embrace accountability for the things they can control.

Thomas Rogers:

To Ryan: Thanks for raising the issue of leadership development. I had been reluctant lest it seem self-serving. However, as baby-boomers retire, we are going through a generational turnover of leaders. More than two-thirds of New York�s superintendents are new � in the first 5 years of their first superintendency.

The job of a principal � instructional leader � differs significantly from that of teacher. Similarly, the superintendent has managerial, budgetary, personnel, and political (board and community relations) responsibilities for which there is little formal preparation. Certification and degree programs are a start, but professional organizations like ours spend much of our energies plugging the training holes left by these institutions � an endeavor which is largely voluntary and horribly underfunded.

To Charlie Kolb: I envision the business community:

1. Very much engaged in developing the definitions of the kinds of skills we want the next generation of workers to have, including the ability to �learn to keep learning�;

2. Helping keep up the pressure to close achievement gaps. Unless we increase high school graduation rates businesses will have a shortage of 14 million post-secondary educated workers by the year 2020.

3. Returning us to where Chester Finn started the discussion � with a focus on accountability for results and less proscription about process, which is essentially the discipline the market exerts on businesses. (Isn�t process innovation how businesses differentiate, compete, and improve?)

4. Pushing a national dialogue about what the successor model to the �factory school� looks like.

It�s difficult to envision NCLB�s successor partly because there is little consensus about what a post-agrarian, post-industrial revolution school model looks like � much less how to regulate it. But it seems certain that the curriculum must be different, technology must be completely integrated, and institutional boundaries must become more transparent. Even the length and structure of the day and year must be reexamined in a deliberate way, not ad hoc: it can take years to negotiate an extra few days in the traditional 180-day school year, but in response to sudden fuel price spikes, nearly 18% of schools have (or are considering) moving to 4-day weeks.

Diane Ravitch:

Regarding the unreliability of test scores as reported by states and districts, this is a serious issue, which the Fordham Foundation explored in its publication "The Proficiency Illusion." And, it is directly relevant to NCLB!

As I pointed out earlier, it was a huge mistake in the law to allow states to set their own standards, choose their own tests, decide for themselves what "proficiency" is. The result is that states are reporting ever higher test scores, ever higher proportions of students reaching "proficiency," even though NAEP shows that no such miracle has occurred.

It is not that states are dishonest, but that they have been instructed by the legislation that all students must be proficient by 2014 or the state and districts and schools will face increasingly onerous sanctions. So, of course, proficiency rates are rising steadily towards that unattainable goal. It is unattainable because no nation has ever reached it, nor has any state. For Congress to legislate an unattainable goal makes liars of everyone.

Checker earlier said that we should all thank NCLB for having promoted transparency and new data, but instead of transparency, we have a deluge of phony claims of progress. This does not serve the cause of education reform. And, if anything, the failure of NCLB to produce reliable and honest information about student progress � due to its dependence on states' self-reporting � makes the case for national standards and national tests.

Randi Weingarten:

I'm glad Sara brought up the subject of funding. I agree with her that the federal law should use its leverage to ensure that all children�especially those from disadvantaged backgrounds�have access to equal resources. The federal government could address funding inequities by rewarding states that target additional funds to the poorest districts to help them build capacity and provide the other supports we all know are needed to improve outcomes for these students.

Nation at Risk brought a sense of urgency to federal education policy. The first President Bush and President Clinton fought to ensure that education policies were built on the principles of the standards movement. And the current President Bush made it very clear that ESEA could leverage big changes at the state and local levels.

Unfortunately, NCLB's sanctions threaten to eclipse the original ESEA's role of providing resources and greater opportunities. ESEA's historical purpose�one we need to return to�is to help compensate for the disadvantages of poverty.

An another note, it's encouraging that Phil and I and others are on the same page when it comes to community schools. Supported by federal law, they also could offer a way to compensate for the disadvantages of poverty.

Philip Howard:

As this discussion draws toward a close, I would first like to say that I've been very interested in the critiques of NCLB. While this is hardly a perfectly diverse sample (the panel is a subset of those invited by NewTalk who happened to be available these days), the points of view nonetheless represent many different perspectives I believe that, directionally, almost all the panelists were headed in the same direction�make testing standards uniform across the country, and relieve schools of the bureaucratic pathologies that NCLB has fostered. The public comments have been fabulous, impassioned and reflecting the frustration with NCLB by those on the front lines.

What I've been less impressed with is the willingness of the panelists to sketch out how NCLB should be changed. National standards...and then what? Directionally, Diane and Checker and Ryan seem to be more interested in carrots than sticks. That sounds right to me. But someone has to come and say what doesn�t work and needs to be eliminated. AYP? I vote to eliminate it. Teacher certification? I vote to eliminate it�is that really what the federal government is good at? Complex formulas for funding? A bridge too far. Focus funding on affirmative programs to improve leadership, or provide resources to manage troubled students so they don't disrupt the learning of everyone else.

Law is not neutral, even if it seems innocuous. Law diverts the attention of educators towards compliance when they ought to be thinking about the students in front of them. For that reason alone, the elaborate regulatory structure of NCLB is a mistake, no matter how thoughtfully constructed. That's why it needs, in my view, a radical downsizing.

Diane Ravitch:

Amen, Philip. The federal government should do what it is best at: Gathering information and dispensing it without fear or favor.

Gerald Tirozzi:

I would like to take this opportunity to weigh in on two of the recent commentaries. Initially, I would like to strongly support Tom Rogers� comments regarding the importance of school leadership � in effect the school principal. We cannot underestimate the importance of this position, in that it is almost impossible to identify a highly successful school without identifying a highly successful principal in that school. In particular, a much greater emphasis must be made on the development, induction and sustained professional development for our nation�s principals � with a major focus on instructional leadership. Recently, we had been working closely with National Board for Professional Teaching Standards (NPBTS) to begin the process to develop a national certificate for school principals, which would be predicated on nationally established high standards to identify exemplars of the profession. These individuals overtime could become �beacons of excellence�, and could become mentors for new and aspiring school leaders. It is important to emphasize that such recognition would be awarded to experienced principals (minimum of three years) going through a rigorous certification progress.

It is also time to address the development of our nation�s assistant principals � keeping in mind that approximately 70 percent of all principals have had experience as assistant principals, and who for the most part -- especially in urban school districts -- are developed and rewarded for a skill set of behaviors which is not aligned with instructional leadership (discipline, cafeteria and hallway supervision, bus duty, parental and teacher complaints). In effect, individuals with such a skill set are ill-prepared when they assume the role of principal, success for which is more directly aligned to student achievement, curricula and professional development leadership and accountability for academic results. The irony is that we tend to celebrate �small� initiatives to develop future principals when the greatest potential to scale up large groups of future leaders exists in the ranks of the assistant principalship. A metaphor I like to use to make the latter point is that we tend to �celebrate victory gardens and never get to amber waves of grain�.

Secondly, I want to strongly support the observations of Diane Ravitch that allowing states to set their own standards and choose their own tests is a �huge mistake�. As I stated in remarks that I submitted to the dialogue on Tuesday, I feel very strongly that it is both timely and necessary to develop national standards � initially in the areas of reading, math and language arts � if, in fact, we are to have a true picture of student performance across the states. I would offer that any attempt to reauthorize NCLB will be futile in effort unless and until we reach a national agreement as to what all students should know and be able to do. I bring to your attention that our Board of Directors � 24 practicing principals � have unanimously adopted a position statement supporting national standards.

Christopher Cerf:

I have to admit, I'm in the "stick" camp�although happy to couple it with lots of carrots. I certainly join Diane and most others in this discussion in the view that real accountability for student outcomes risks perverse and counter-educational behavior, most notably dumbing down tests and misusing class time. But, it is important now and then to remind ourselves of two related points. First, for a variety of reasons (most of them unfortunate) there is a deeply ingrained cultural hostility to accountability for student learning in public education. Left to their own devices, schools and districts would quickly yield to powerful inertial forces that push educators and regulators to focus maniacally on "inputs," whether or not they are correlated with how well children read or how many actually graduate. Second, accountability for results is the single greatest bulwark I know against the apologists' view that the job simply cannot be done given the complex social challenges many urban children bring to the scant six hours a day they spend in school. Giving way to that impulse strikes me as phenomenally dangerous as it serves to excuse everyone in the system from any incentive to actually bring about measurable student learning. Accountability in the "stick" sense, in my judgment, is a critical counterweight to these forces. Moreover, any accountability system creates incentives for some bad behavior. Only in public education do we even entertain the possibility that this reality is a reason to banish consequential accountability altogether.

Mind experiment�If we:

1) Left the core of NCLB's sanctions (wrong word, but you get the point) in place, but rationalized the process by keying off the NAEP or a new equivalent based on national standards;

2) Invested heavily in a teacher competency model based on Danielson's or Pianta's work or the like�that is, competencies that in a research-based way are highly correlated with student learning ;

3) Developed non-threatening self-assessment tools for teachers and principals to develop those competencies through innovative delivery systems (a far cry from most of the ineffectual professional development we rely on today);

4) Rewarded districts that treated tenure as an earned privilege, rather than a routinely granted expectation, based on demonstrated evidence of student learning (via value-added analysis in conjunction with other softer measures), would we really be so quick to give up on NCLB?

How about if we layered in a tranche of new federal dollars to augment the salary of those teachers who justly earned tenure by this rigorous "actual-student-learning-based" measure and taught in high-needs schools? How about if we also threw out the highly diluted and largely credential-based HQT notion and replaced it with a measure based on teacher effectiveness?

My point: NCLB provides much to build on if we had the courage to take on some of the many other "untouchables." Until we do, there is a real risk of many angels dancing on the head of a pin.

Thanks for letting me participate.

Deb White:

1. Set attainable goals for continued improvement, but get rid of AYP and 100% proficient by 2014. Be realistic!

2. Improve principal training (assistant principals are a great place to start). They are the next wave of educational leaders, and we need leaders, not managers.

3. Set at least minimum National Standards.

4. Identify qualities good teachers have in common and work to develop teacher training programs which encourage and enhance those qualities.

5. Support and nurture new teachers. It is a difficult career for at least the first few years.

I have read so many excellent ideas in the past three days. The next generation of NCLB must be envisioned and created from the bottom up. Make sure that those who will need to implement it in the classroom buy in and believe in it. (To be honest, I believe that most educators would be so thrilled to get rid of the current version of NCLB that they will jump on board enthusiastically.)

Diane Ravitch is correct�by legislating the impossible we have all become liars.

John Merrow:

First of all, it's been a privilege to participate. So many thoughtful comments, much to think about.

Re: carrots and sticks, I have a few thoughts and opinions: 1) Although NCLB is seriously flawed, it HAS forced districts to pay attention to forgotten kids. Yes, districts and states have gamed the system, cheated, lied and otherwise sinned, but they have also had to pay real attention.

2) It's had an impact on teacher unions as well: A few weeks ago, I interviewed the President of the Washington (DC) Teachers Union, George Parker, for the NewsHour. He told me that two forces were making his union concentrate on student learning, which heretofore had not been a 'bread and butter issue' for them. He cited charter schools and NCLB. Charter schools are nearly as flawed as NCLB, but they have become, like NCLB, a force for change.

3) What would happen if those who have the responsibility for rewriting NCLB were to ask "who benefits from educational failure, from educational mediocrity?" Address that question honestly, and let the chips fall where they may. (hint: it's a long list.)

4) My hope is that the next President will call a summit meeting (20 years after the first one) for a serious debate about the goals and purposes of public education. That debate should inform the new legislation.

5) It's more likely, however, that we will continue to muddle along, tinkering as we go. If that happens, then before too long I predict that public education as we know it will simply collapse. That's what seems to be happening now. This fall another million students will 'go to school' online. Millions more will enroll in charter schools, middle college high schools, religious schools, or private schools.

Again, my thanks to you and to Philip Howard and NewTalk.

This piece originally appeared in NewTalk

This piece originally appeared in NewTalk