Most observers recognize that as government increasingly held schools and teachers accountable primarily for the math and reading test scores of their students, schools inevitably narrowed their curricula to minimize attention to other important educational outcomes, substituted test preparation and test taking skills for real learning, and even engaged in cheating to meet politically determined targets.

Some policymakers have recently attempted to address these problems by advocating accountability for “multiple measures.” Their reasoning has been that the corruption of education that results from a near-exclusive focus on basic skills in math and reading can be ameliorated if other indices can be added to accountability systems to supplement the math and reading test scores. This was the goal of the California bill, sponsored by liberal Democrats, and sent to Brown for signature.

But because other important outcomes of education – like character, inquisitiveness, citizenship, civic awareness, historical reasoning, scientific curiosity, good health habits – cannot be standardized like math and reading scores, proponents of “multiple measures,” like the California senators who crafted the bill, are left with adding indices like attendance rates, parent satisfaction, graduation rates, the number of students taking advanced placement courses, and the like. But this does little to divert schools’ obsession with math and reading test scores, since they remain the only academic outcomes that count.

As Brown observed, “adding more speedometers to a broken car won’t turn it into a high-performance machine.”

In his veto message, Brown recalled an aphorism of Albert Einstein: “Not everything that counts can be counted, and not everything that can be counted, counts.” The bill, Brown said, “nowhere mentions good character or love of learning. It does allude to student excitement and creativity, but does not take these qualities seriously because they can’t be placed in a data stream.”

Brown invited the legislature to work with him to devise a truly workable accountability system for education, one that relies on qualitative evaluations by “panels [that] visit schools, observe teachers, interview students, and examine student work.”

The Broader, Bolder Approach to Education campaign has advocated such a system, and described it in more detail in a statement issued by nationally prominent educators and policy experts. The system is also described in Grading Education: Getting Accountability Right. If the panels that Brown advocates are constituted with appropriate experts in curriculum and instruction, and include members of the public as observers, they have the potential to finally provide citizens with the ability to distinguish effective from ineffective schools in their state and communities.

It is encouraging that California State Senator Darrell Steinberg responded to Brown’s veto message with a willingness to work with him to design such an accountability system. Should they succeed, it could signal that some in the nation may finally be ready to turn away from a well-intentioned but destructive reduction of schooling to the standardized tests that can, at best, measure only a small aspect of education.

Last week, BBA issued a set of 6 key principles that its leaders believe should guide federal policymakers in Congress and the Administration as they consider the re-authorization of the Elementary and Secondary Education Act (now called “No Child Left Behind”). BBA leaders include Co-Chairs Helen Ladd, Pedro Noguera, and Thomas Payzant, and Advisory Council Co-Chairs Susan B. Neuman and Richard Rothstein. Brief biographies of these and other BBA leaders are on the BBA website. 

In brief, BBA:

    * Warns against attempts to identify the lowest performing schools primarily by their students’ test scores on basic math and reading skills.

    * Cautions that even attempts to measure student “growth” on such tests, although an improvement over student performance data currently in use, will still misidentify low-performing schools and could spur new forms of gaming.

    * Urges identification of schools needing improvement by qualitative observation and evaluation, as well as by student scores on higher quality tests, and notes that such complex accountability systems can be developed successfully only if states have wide latitude to experiment.

    * Calls for school accountability that creates incentives to deliver a broad and well-rounded curriculum, including the arts, science, history, health and physical education, and character development.

    * Applauds Administration support for programs that address health, family and community impediments to learning, but observes that school accountability policy cannot ignore the absence (or presence) of such programs.

    * Proposes an increase in real federal compensatory education funding for all eligible schools during the current fiscal crisis, not a reduction of such funding in favor of competitive grants.

Click here for the full text of these principles is at the BBA website. 

For more information on the BBA campaign:

Click here for BBA’s initial statement.

Click here for BBA’s full statement on accountability policy.

The EPI Education Program has issued other publications recently, covering issues such as teacher compensation, the Administration’s proposals for ESEA re-authorization and its “Race to the Top” competition.  Click here for all recent EPI publications on education.

Broader, Bolder Approach (BBA): http://www.boldapproach.org/

Full text of key principles: http://www.boldapproach.org/20100512_bba_key_points_esea_reauthorization.pdf

BBA’s initial statement: http://www.boldapproach.org/bold_approach_full_statement.pdf

BBA’s full statement on accountability policy: http://www.boldapproach.org/20090625-bba-accountability.pdf

Teacher compensation: http://www.epi.org/publications/book-redesigning_teacher_pay/

Proposals for ESEA re-authorization: http://www.epi.org/page/-/pdf/pm162.pdf

“Race to the Top” competition:  http://epi.3cdn.net/4835aafd6e80385004_5nm6bn6id.pdf

All recent EPI publications on education: http://www.epi.org/issues/category/education/

Q. A recent New York Times story about the Obama administration’s plans for school reform concludes with a school superintendent saying, “education will never be the same.” Do you agree with this assessment?

Ravitch: It’s a scary thought. I think public education itself is at risk. On the current course we are on, we will see thousands of public schools turned over to private entrepreneurs. We will see an explosion of privatization.

Q. Education reform and Obama’s Race to the Top grant program (a program for spurring school reform often through increased focus on teacher accountability and charter schools) are receiving a lot of press.  Do you feel these stories accurately describe the state of the nation’s schools and the reforms being proposed?

Ravitch: There is a problem with calling what is happening in schools “reform.” Some articles extol unproven ideas and lack any fairness or balance. My book cites a lot of research showing that charter schools don’t do any better on the whole than regular public schools. If they do not produce consistently better results, why are we investing billions of scarce public dollars to create many, many more of these (charter) schools? And why ignore their likely impact on regular public schools? Proponents claim that competition improves public schools but that certainly has not happened in Milwaukee, where there are vouchers, charters, and regular public schools, and all three sectors have low performance.

Q. The Times article also states that “What the reformers have come to believe matters most (in education) is good teachers.” What do you think?

Ravitch: Of course good teachers are incredibly important, but we won’t get more of them by attacking teachers and reducing their job to the singular goal of raising test scores. The Times article has implicit disdain for the people who work in schools – a suggestion that it’s only the people who went to Ivy League schools who know how to fix things.

To say “we need good teachers” – of course we need good teachers. But their idea of [how to identify and fire] bad teachers is based mainly on test scores. They are making the assumption that schools are overrun with bad teachers and I don’t think that’s true. To have teacher evaluations decided by politicians and legislatures is insane.

Q. Should teachers be held individually accountable?

Ravitch: I have a problem judging teachers by test scores. Test scores should surely be considered by principals as part of teachers’ evaluation, but legislators should not quantify this. They are not competent to do so.  There are many things that cause test scores to go up or down, and it’s not just the teacher.

Part of what is going on is to try to blame low performance on teachers instead of recognizing that poverty is the single greatest determinant of low scores. Not being a native English speaker and being homeless are also major factors.

Testing can have some part in this, but it is a matter of professional judgment by competent and experienced principals, not something that should be legislated or politically determined.

Q. Should teachers have guaranteed lifetime tenure?

Ravitch: Lifetime tenure does not exist. Tenure means the right to due process. But firing people because they cost more is a way to destroy a profession. If you go in for surgery, do you want an experienced surgeon, or a resident? Senior teachers are a valuable part of every school staff. New teachers need their help. If teachers are incompetent, they should be brought up on charges and removed. A far more important problem than removing teachers is teacher attrition: Half of those who enter teaching are gone within five years. Yet the “reformers” ignore this problem, which is largely due to poor working conditions.

Q. The New York Times piece does make the point that many charter schools perform badly. But it also uses the example of one charter school in Harlem that actually shares the same building with a public school, yet has a vastly different school environment and much better student performance. How can two schools that are not only in the same neighborhood, but in the very same building have such vastly different outcomes?

Ravitch: The important point here is that these two schools don’t have the same population. The public school has a much larger population of English-language learners, special education students, and students with disabilities. The only thing they share is the same building.

Q. And why would two schools in such close proximity have such different student populations?

Ravitch: Charter schools, when they are popular, have a lottery. But a lot of parents never even hear about the lottery. In New York City, there is a very large number of homeless students in the population, and very few enroll in charter schools. It is a selective effect. There is a big difference between entering a lottery for a charter school, and going to a public school, where they are required to accept you.

Charter schools can also quietly remove kids. Some of the charter schools that are very successful have a very high attrition rate. They may start with 100 students and end up with 50. Then they say how successful they are but don’t tell you about the students who didn’t make it. 

Ideally, charter schools should collaborate with public schools, not compete with them. Both sectors should coalesce around common goals, funding and sharing ways to help all students succeed.  

Q. What are some of the other problems you see with charter schools?

Ravitch: Some skim away the most motivated students, and many are concentrated in cities in minority neighborhoods. If you look at the nations in the world with the highest performing schools — Finland, Japan and Korea — they have not privatized public education. They have made it better.

Q. In your view, what are the school reforms that are needed?

Ravitch: The most important reform is to have a strong and coherent curriculum in the arts and science, in history and geography. Every high-performing nation has a strong curriculum. And as a society we have to act on the other problems, such as poverty and homelessness, which contribute to poor educational outcomes. We should not punish schools and teachers because they have a high number of kids who are poor or homeless or aren’t native English speakers. We have to do something to help those students have a better life.

Although the method for awarding Race to the Top grants has the patina of scientific precision and objectivity, the reality is that it chooses winners and allocates aid subjectively, with little scientific basis, according to the report, Let’s Do the Numbers, by William Peterson and Richard Rothstein. The winners of the first round of Race to the Top, Tennessee and Delaware, could easily have placed lower in the competition with only slight modifications to the arbitrary weights assigned to the competition’s different criteria.

Particularly at a time of severe state budget crises, when class size increases and teacher layoffs are a concern in many states, the report argues for using a simpler, fairer system for awarding grants in the second round of the Race to the Top competition in July. Such a system would ensure that states receive funding as long as they are committed to educational improvement.

Briefing paper

When it comes to education policy, President Obama is repeating the most grievous errors of his predecessor, charge a trio of venerable education policy analysts, including one — Diane Ravitch — best known for her past support of conservative positions on testing, accountability, and choice.

As Congress begins to rewrite No Child Left Behind (NCLB), the Obama administration has offered its own vision for how the revised law should look, including a focus on tougher academic standards and more flexibility for schools. But a growing chorus of critics contends that too many of the administration’s policies follow the same punitive cycle of high-stakes testing and accountability ushered in under the presidency of George W. Bush—and that these policies are actually hurting students.

Both President Obama and Education Secretary Arne Duncan have acknowledged the need for better standards and assessments to ensure that students graduate from high school ready for college or 21^st century careers. But critics of their approach toward education reform say it continues to rely on a flawed system of high-stakes exams and accountability measures that has narrowed the curriculum, fails to take into account the various social and economic factors that influence a child’s learning, and does a disservice to those students it purports to help most.

Rather than tinkering around the edges of NCLB, they say, policy makers should rethink the very assumptions that underlie the nation’s education law.

Such concerns over high-stakes testing and accountability aren’t new; they’ve existed since NCLB became law in 2002. But what is new is that the chorus of critics now includes some unlikely characters—including education historian Diane Ravitch, who worked in the Education Department under President George H. W. Bush and was a staunch supporter of the younger Bush’s policies as well.

Ravitch has a new book out called The Death and Life of the Great American School System: How Testing and Choice Are Undermining Education. Its thesis marks a radical departure from her earlier views on education—and in an interview with eSchool News, she explained what transpired to change her mind.

“This represents a big change for me, because for many years I have been associated with things like testing, accountability, charter schools, merit pay, et cetera,” Ravitch said. “But as I saw the evidence accumulating, I began to think … that I was wrong.”

She added: “The Obama administration, although it promised change when it came to office, in effect has picked up precisely the same themes as the George W. Bush administration, which are testing and choice—and I think we’re on the wrong track.”

Watch interview with Diane Ravitch

Ravitch was one of several education experts who spoke out last month during the American Association of School Administrators’ National Conference on  Education against the Obama administration’s continued reliance on high-stakes testing and accountability to drive school reform. Other critics of the president’s policies included Richard Rothstein, a research associate at the Economic Policy Institute and former education columnist for the New York Times, and David C. Berliner, Regents Professor of Education at Arizona State University.

Under the nation’s current accountability system, Ravitch said, “we’re only measuring what we can, and not what matters most.” As a result, she said, we’ve narrowed the curriculum to the exclusion of other important subjects by focusing primarily on making adequate yearly progress in reading and math.

“If you look at what is working in other successful nations, “it tends to be a far more holistic approach to schooling than what we are doing now” in the United States, she declared.

Testing can be effective when used for diagnostic purposes, “but when testing becomes the focus of high stakes for kids and for teachers and for administrators, it has very harmful consequences,” she said. “To judge a teacher or a student based on a test score is like judging a baseball player on one at-bat; Babe Ruth struck out a lot more than he homered.”

Ravitch, who is a research professor at New York University, said she has looked at the National Assessment of Educational Progress (NAEP) scores of both charter and public schools since 2003 and has concluded that charter schools don’t outperform public schools on average.

Charter schools are “skimming off the best kids in the poorest communities, and that’s why they get better results,” she said. “They’re not taking a proportional amount of English language learners, special-ed kids, homeless kids. … Sure, if you cherry-pick the kids, you get better results.”

There is a growing body of evidence to suggest that school choice isn’t going to improve education, she concluded—and, given that struggling schools lose the funding that accompanies students who opt out for other institutions, “it might actually be very harmful to public education.”

The ‘wrong policies’?

Speaking to school superintendents during the AASA conference, Duncan identified three principles that he said would guide the current administration’s approach toward rewriting NCLB: (1) higher standards, (2) rewarding excellence, and (3) a “smarter, tighter federal role” in ensuring that all students succeed.

He also sounded like someone who understood many of the law’s failings.

“I’ll always give credit to NCLB for exposing achievement gaps and advancing standards-based reform. But better than anyone, you know [the law’s] shortcomings,” Duncan told the assembled education leaders. “NCLB allows, even encourages, states to lower their standards. In too many classrooms, it encourages teachers to narrow the curriculum. It relies too much on bubble tests in a couple of subjects. It mislabels schools, even when they are showing progress on important measures.”

Although Duncan and Obama acknowledge NCLB’s problems, their approach does not go far enough in addressing these issues, critics argue.

“Both President Obama and Secretary Duncan talk about the narrowing of the curriculum,” Rothstein said in an interview with eSchool News. “But the policies they’re implementing … are all about improving the quality of math and reading tests. Now, there’s nothing wrong with improving the quality of math and reading tests—but if [that’s] all we do …, [then] we [will] continue, and even exacerbate, this distorted emphasis on only one part of the curriculum.”

Like Ravitch, Rothstein sees huge flaws in the administration’s approach to education reform. He’s part of a group of policy experts called “A Broader, Bolder Approach to Education”, which advocates a different kind of accountability system for the nation’s schools—one that doesn’t rely on test scores as the primary indicator of student achievement and doesn’t create incentives to narrow the curriculum.

The group is hoping to press
ure Congress and the administration “to abandon the failed, test-driven policies of the last decade,” he said.

Watch interview with Richard Rothstein

Federal education policy in the United States today “is driven by a climate of opinion that assumes our schools have been failing,” Rothstein said—and one that assumes teachers are inadequate and have low expectations for their students. Policy makers have come to this conclusion, he explained, by looking at the achievement gap between black and white students, which “hasn’t really budged very much.”

But what they fail to acknowledge is that there has been “phenomenal improvement in both black and white student achievement over the last generation,” he said. “Black student achievement has risen so much in the last 20 years that it’s now higher than what white student achievement was 20 years ago.”

That kind of phenomenal improvement is “not consistent with the story of teachers sitting on their rear ends, having low expectations of disadvantaged children,” he said—and schools could have closed the achievement gap long ago if white students hadn’t made their own vast improvements at the same time.

When we develop policies based on a flawed analysis of the data, “we necessarily develop the wrong policies,” Rothstein asserted, “and that’s what’s going on in Washington today.”

Besides narrowing the curriculum, the climate of high-stakes testing and accountability is detrimental to schools because “it takes no account of the fact that one of the primary drivers of student achievement is the social and economic conditions that children come to school with,” Rothstein said. “Again, both Obama and Duncan acknowledge this; they talk about it frequently—yet their current policy takes no account of this.”

The administration’s plan to stem the dropout rate, for example, seeks to identify the bottom 5 percent of high schools in the country to intervene and turn them around. Yet, given the social and economic conditions that most students from these schools face, “many of these schools are not low-performing at all,” Rothstein said. “They [actually] add great value to the performance of these children.”

Outputs vs. inputs

Arizona State’s Berliner agrees with Rothstein’s assertion that the federal testing and accountability policies that continue under President Obama fail to acknowledge the enormous influence of socio-economic factors on student achievement.

“We have people who say teachers are terrible, or teachers are great, and they’re taking nothing into account about the context those people are working in,” he said. “Out-of-school matters matter.”

Watch interview with David Berliner

Berliner said the U.S. has shifted from a focus on providing equality in the “inputs” of education—family environment, community conditions, and so on—during the Johnson administration to a focus on providing equality in the “outputs” of education (the achievement gap) under the Bush administration, and this approach continues under Obama.

“We stopped worrying about inputs,” he said—and yet, “what’s coming out of schools is still a function of those inputs.”

A key problem with the high-stakes testing approach to education, Berliner added, is that when people’s jobs are on the line, the people get corrupted—and so do the indicators. That’s why we’re seeing schools supply test answers to students, and states lower their standards, and so on.

“If I’m going to lose my job, and I can change a score—I have a family to support, I’m going to change the score,” he said. “If you put people into that position, you’ve messed up one of the goals we have for an American educator, which is to provide a moral compass for our youth—and you’ve set them up to do things that are not good.”

The Administration’s education policy feints in several positive directions. It hopes to inspire more young people to go to college, but recognizes that college is becoming prohibitively expensive for low- and middle-income families. So the budget reconciliation bill took back the college loan program from inefficient private lenders. The more than $60 billion in additional grants that will result — from eliminating bankers’ fees for issuing government-insured loans — will enable about 8 million students to attend college who might otherwise not be able to do so. It would be foolish to try to re-organize elementary and secondary education to make students “college-ready” if college itself becomes less affordable.

In another positive move, the stimulus bill of February 2009 filled holes in state education budgets, avoiding the need for states to layoff perhaps a third of a million teachers and other school employees. Yet the college loan and stimulus measures, while steps in the right direction, are inadequate. College costs are far outpacing the added financial help provided by a more efficient college loan program. California and Georgia, for example, recently announced tuition increases of over 30%. And one-time stimulus money cannot prevent growing layoffs of teachers and other school employees, resulting in increased class sizes and less ability for states to provide other programs that the Administration desires — such as more access to advanced placement classes for disadvantaged students and a broader curriculum that goes beyond math and reading. The economic crisis of public education will only get worse

— many states fund public education in part by property taxes, and with real estate values low, layoffs of educators will intensify as reduced tax collections work their way through state and local budgets. Many school districts anticipate that budgets may not recover until Fiscal Year 2013, or perhaps later.

Blueprint has some worthy goals
The Administration’s Blueprint for re-authorization of the Elementary and Secondary Education Act (ESEA) also reflects some positive goals. Recognizing that the existing ESEA (No Child Left Behind, or NCLB) has narrowed the curriculum, the Administration proposes funding some states to broaden their curricula to include elements of “a well-rounded education [to give students the capacity] to contribute as citizens in our democracy and to thrive in a global economy — from literacy to mathematics, science, and technology to history, civics, foreign languages, the arts, financial literacy, and other subjects.”

The Blueprint also recognizes that student success cannot depend on the regular school day alone, and so proposes to fund some states that improve “early learning outcomes,” that provide before-, after-school and summer programs, that provide “comprehensive supports to students and families through full-service community school models,” and that develop “Promise Neighborhoods” projects that implement “a continuum of effective community services, strong family supports, and comprehensive education reforms to improve the educational and life outcomes for children and youths in high-need communities, from birth through college and into careers.”

Competitive grants make no sense in budget crises
Yet these positive steps seem somewhat inconsistent with other aspects of the Blueprint and associated budget recommendations. Perhaps most troubling are the Blueprint’s extensive proposals to use ESEA funds for competitive grants to fund some states’ implementation of the policies described above. These might not be problematic if the competitive grants were being funded in normal economic times, with dollars supplementary to regular formula funding (where states get compensatory education — “Title I” — funds on a per child basis). But in the present period, with extensive layoffs of teachers and other school employees looming, it makes no sense to cut (in real, inflation adjusted terms) formula-driven funds in order to reward states that innovate in ways that suit the Administration’s policy preferences. A full employment program for grant-writers is no substitute for stable employment for educators. And what are states that don’t win the competitive grants, and thus have no way to avoid the layoffs, to do?

The practical reality is that in any economic climate, but especially the present one, no Congressman is going to allow schools in his or her district to layoff teachers so that other districts can get competitive funding. The result will be money spread around a little more evenly, but in a more haphazard way than Title I formula funds. A lot of money will be wasted on grant-writing and evaluation when it could have been spent on real educational services. And with political pressure as well as (if not more than) proposal quality determining the awards, little policy innovation will be accomplished.

Curriculum narrowing and gaming will continue with middle-class schools exempt
In some disturbing ways, the Blueprint continues core NCLB policies that the competitive grants purport to correct. The most important is the accountability system. NCLB did the most harm by placing such high stakes on annual tests of math and reading that it created incentives for schools to narrow the curriculum and to substitute gaming for education. The narrowing was both inter-subject and intra-subject. Inter-subject narrowing consisted of reducing time and attention paid to the arts, science, history, physical education and other non-tested subjects. Intra-subject narrowing consisted of overemphasizing the basic skills most likely to be tested, and ignoring more difficult-to-test aspects of the math and reading curriculum, such as mathematical reasoning, critical thinking, and appreciation of good literature. Both types of narrowing disproportionately occurred in schools serving disadvantaged children, for these were the schools most in danger of suffering sanctions for low test scores.

NCLB also created incentives for other gaming. The most notorious was the targeting of students whose basic skills performance fell just short of “proficient.” Schools could increase their ratings, and escape sanctions, by improving the scores of a small number of students just below this criterion, and ignoring both students who were far behind (and had no chance of passing the proficiency point) and those who were already proficient.

The Blueprint continues this system, although with a twist. Only the bottom 5 percent of schools (called “Challenge” schools) will suffer severe sanctions (such as firing or replacing the staff) as a result of low test scores. Although all schools will be required to continue to administer annual math and reading tests alone, the Administration suggests that the absence of sanctions for the top 95 percent of schools will give states the flexibility to design new accountability systems that may even include requiring schools to broaden their curricula.

The next-lowest 5 percent of schools will be placed on “Warning” status; these schools, too, will have all the perverse incentives of the Challenge schools. Still, though, this presumably leaves 90 percent of schools with the flexibility to broaden their curricula and avoid excessive test-prep activities and gaming.

But not really. What Blueprint enthusiasts have neglected to recognize is that the tests of math and reading have always been highly unreliable. If we consider, say, the bottom 20 percent of schools, it is only a matter of chance whether a school will test low enough in any year to be categorized as Challenged. Schools that are Challenged in one year have a high probability of being in a safer category the next, and vice-versa.

The result of all this is that schools serving a large proportion of minority students will behave under the Blueprint very similarly to how they behaved under NCLB. If the Administration succeeds in raising the proficiency cut points (now called “college and career ready” standards), even more students will be ignored because they are too far below the passing point to matter.

Schools with largely middle class populations will effectively be exempt. The political consequences of this could be dangerous: federal officials will no longer be besieged by middle class parents upset that their schools have turned into test-prep factories; only poor and minority communities will suffer the punitive hammer of federal policy.

Recognizing the impact of economic stress on achievement
The Blueprint’s accountability provisions are also curiously contradictory. The Administration proposes competitive grants to establish Promise Neighborhoods that coordinate a wide range of social, health, and community services with schools, and will reward states with funds to establish programs that nurture children from birth and serve disadvantaged children’s before, after-school, and summer needs. A reasonable inference is that the Administration believes (correctly) that such programs contribute to student achievement. As Secretary of Education Arne Duncan has stated: “It takes more than a school to educate a student. It takes a city that can provide support from the parks department, health services, law enforcement, social services, after-school programs, nonprofits, businesses, and churches.”

If this is the case, however, whether a school is “Challenged” or on “Warning” status depends not only on the quality of its instruction, but on whether its students come to school ready to learn, with the benefits of Promise-Neighborhoodstype services. Imagine two schools of equal quality. Test scores in the first may be higher because its students come to school after enrollment in high quality early childhood programs and from the cooperative support of the parks department, health services, law enforcement, social services, after-school programs, nonprofits, businesses, and churches. Test scores in the second may be lower solely because of the absence of these supports. The Blueprint proposes to transform the second school by replacing the principal and teachers, turning the school over to private management or other draconian measures. It would be more logical to leave the instructional staff intact (after all, an instructional staff of identical quality is able to get satisfactory results in the first school) and instead remedy the situation by providing the early childhood, after school, health and other services that produce higher achievement.

If the Blueprint is politically tone-deaf with regards to the feasibility of shifting formula funds to competitive grants in a deep economic recession, it is sociologically tone-deaf with regards to the goal of raising achievement goals in this period. Presently, a quarter of black young and middle-aged adults, of an age to have children in school, are either unemployed, or so discouraged about looking for work that they have dropped out of the labor force. During the course of the year, approximately 40 percent of black adults will be unemployed at one time or another. Schoolchildren from families in such circumstances will change schools more often because of housing instability, will more frequently come to school hungry, in poor health, and with behavioral problems arising from family stress. It would be a remarkable accomplishment for the achievement of disadvantaged children to remain stable for the duration of the economic crisis. Expectations of near-term improvement are breathtakingly removed from reality.

Growth can’t reasonably be measured
The Blueprint refers repeatedly to the desirability of (and presumably, in Challenge schools, the requirement for) measurement of “growth” in math and reading scores. Not only are principals and teachers to be evaluated, and teacher compensation systems to be redesigned to account for the “growth” of students, but even teacher training institutions are to be judged by their contribution to such “growth.” Sanctions for the lowest-performing schools, and also the award of competitive grants, depend on such measurement. We can assume that, to measure growth, the Blueprint does not envision requiring schools now to test math and reading twice yearly, once in September and again in June. But if not, it is hard to imagine how states are going to respond to this demand for growth measurement. NCLB insisted that annual tests must be given early enough in the year so that they could be scored in time to design interventions for the next school year. The Blueprint will need a similar requirement if Challenge schools are going to be “turned around” the following year. Yet if tests are given, say, in March, which teachers are to get credit for students’ gains, or lack of them

— the teacher who had the students from September to March, or the teacher who had the students from the previous March to the previous June? And how can such “growth” models account for learning, or lack of it, that takes place during the summer? One recent analysis calculated that over half of the schools in the lowest quintile of performance are mistakenly assigned to this low-performance category solely because of differences in children’s summer learning — or lack of it. Perhaps in the future we will develop a way around these problems, but we have not found one to date, and it seems precipitous for the Administration — especially one that claims to be committed to research and evidence-based policy — to load so many incentives onto a statistical technology for which there is no present prospect.

How equal is “comparable?”
Other Blueprint proposals also seem to have been poorly (or too hurriedly) thought-through. One that has received relatively little attention is a requirement for intra-district funding equalization: “Over time, districts will be required to ensure that their high-poverty schools receive state and local funding levels (for personnel and relevant non-personnel expenditures) comparable to those received by their low-poverty schools.” Surely the Administration must be aware that there is now a 40-year tangle of litigation i
n a majority of states that attempts to define precisely what finance equalization means. Does “comparable” mean that high-poverty schools (however they are defined) should receive the same per-pupil resources as low-poverty schools, or does comparability require an adjustment for the greater needs of students in high-poverty schools? If the latter, how much of an adjustment?

Further, the most important cause of intra-district funding inequality is the tendency in some places for teachers serving low-income children to have less seniority, on average, than teachers serving middle-class children. As a result, total salaries paid to teachers in high-poverty schools may be less than the total paid to the same number of teachers in low-poverty schools. A requirement for equalization is, in effect, a requirement that more senior, higher-salaried teachers be re-assigned from low-poverty to high-poverty schools. The Blueprint calls for re-assigning the most effective teachers to high-poverty schools where “they are needed the most.” A fiscal requirement that more senior teachers be re-assigned to high-poverty schools seriously limits the ability of districts to make assignments based on other criteria of effectiveness. Many influential policy advocates have argued that seniority has little to do with effectiveness, and have promoted programs that recruit young college graduates for brief stints as teachers — Teach for America being the best known of these. The Blueprint endorses such programs, saying that priority for winning competitive grants “may be given to programs that work to recruit and prepare high-performing college graduates or non-traditional candidates, such as military veterans or midcareer professionals” for high-needs schools and students. Such teacher recruits will necessarily start at the bottom of the salary schedule. It is hard to imagine how such a priority can be fulfilled at the same time that districts must transfer senior teachers to ensure that funding levels in high-poverty and low-poverty schools are comparable.

Utopian goals — and cynicism — continue
The Blueprint’s overall theme is that by 2020 all students should graduate from high school “College and Career Ready.” Administration officials have explained that this entails the ability to gain admission to an academic college program without having to take remedial courses. (The addition of “Career” to “College Ready” is meaningless, because what the Administration intends to convey is that some students may choose to pursue a non-college career, but would still have gained the qualifications to enter an academic college program if they wished.) This is, perhaps, the most disturbing aspect of the Blueprint. It indicates that the Administration may have learned little from the NCLB experience.

The most widely ridiculed of NCLB’s pretensions was that all students would be “proficient” at a challenging standard by 2014. Arne Duncan has called this goal “utopian” and nobody seriously thinks that the goal ever was achievable, or close to achievable. The normal variability of human ability ensures that no standard of proficiency can simultaneously be “challenging” for students at the top and the bottom of the normal ability distribution.

But aside from ridicule, NCLB’s adoption of this goal did great harm to public education. It created incentives for educators to lie to the public and claim that they could achieve something that they knew was unachievable. It created well-known incentives to “define down” proficiency, to make it possible for more students to pass themselves off as proficient. It engendered a culture of cynicism in public education, and it discredited public education in the broader community, as it became apparent that school leaders could not deliver what they were promising. Any institution that sets an impossible goal runs the risk of such cynicism and loss of legitimacy.

The goal of all students college-ready by 2020 is just as fanciful as the goal of all students proficient by 2014. Today, perhaps 20 percent of all youth graduate high school fully prepared for academic college. It should certainly be higher. Aspiring to make it higher is a worthy ambition. But basing policy on a promise, or even an expectation, that we will quintuple this rate in a mere decade is laughable.

Administration officials defend this goal by saying it is “aspirational,” unlike NCLB’s goal which was a mandate. As we have seen above, this is not accurate. Virtually all schools serving concentrations of disadvantaged children will be in danger of sanctions for failing to make progress towards this goal. For these schools, the same cynicism, the same false promises, the same gaming, will be stimulated as occurred with NCLB.

Even middle class schools will be harmed by this goal. Although they will not be closed or otherwise sanctioned for failing to have all students college-ready by 2020, they will still regularly have to report to the public that their performance is inadequate, regardless of its actual quality. In these cases, we can foresee the consequences. Just as it was easy to make most students proficient by making the definition of proficiency sufficiently minimal, so we can expect states (that control both higher and K-12 education systems) to start defining college readiness down. Promising to make all students college-ready by 2020 is, in effect, an attack on the quality of America’s institutions of higher education.

Defenders of this absurd goal aver that it is inspiring to have an aspirational goal, even if it cannot be met. This is true only if the goal is within reach, albeit with great effort. If the goal is entirely out of reach, then holding it up as an aspiration is corrupting.

Of course, every teacher should attempt to inspire every entering kindergartner, as that child proceeds through school, with encouragement that greater effort could lead to college and a professional career. But no sane teacher believes that this encouragement will be effective with every single child. When an /institution promises such universal success, it undermines its own legitimacy.

Back to the drawing board
We can hope that the Administration thinks further about its proposals, and revises them as they proceed through Congress. It is, in any event, virtually certain that the Blueprint will not be adopted in its present design by this Congress, and perhaps not even by the next.

This suggests an unintended benefit of the Blueprint. For the foreseeable future, Arne Duncan will continue to be responsible for administering NCLB. Having now gone on record that its provisions are seriously flawed and that compliance with them is doing American education great harm, the Secretary will have no coherent choice but to begin issuing wholesale waivers to states from compliance with the old law. If it accomplishes this much, the Blueprint will have done a great service.

(An earlier version of this analysis was also posted on the education discussion site of the National Journal.)

It is fair to say that I’ve asserted that “school reform efforts are distracting from more important social welfare goals,” but a little misleading when phrased in this way. A better summary of my concern is that “school reform efforts are distracting from more important environmental causes of low student achievement.”

It is conventional in education policy these days to say that the most important influence on student achievement is the quality of a teacher. This is only true if it is qualified as the most important school influence on achievement. Fifty years of social science research has confirmed, over and over again, that the best predictor of student achievement is not teacher quality or any other school influence, but the social and economic circumstances of the children.

Consider this: We are all aware that the national unemployment rate is an unacceptably high 10 percent or so. But the unemployment rate for black adults of an age likely to have school-aged children is now 15 percent. This number includes only those actively looking for work. In the current economy, many of the unemployed have concluded that looking for work is futile. Considering the officially unemployed, those who are so discouraged that they have dropped out of the labor force, and those who are working part-time because they can’t find full-time work, the unemployment rate for black parents of school-aged children is now 25 percent.

What are the consequences of this social disaster?

1. Mobility: Families become more mobile because they can no longer afford to keep up with rent or mortgage payments. They are in overcrowded housing, often doubling up with relatives in apartments that were already too small. Children have no quiet place to study or do homework. They switch schools more often, falling behind in the curriculum and losing the connection with teachers who know them well-enough to adapt instruction to the their individual strengths and weaknesses. Inner-city schools themselves are thrown into turmoil because classes must frequently be reconstituted as enrollment rises and falls with family mobility.

2. Hunger and Malnutrition: When more parents lose employment, their income plummets and food insecurity grows. More children come to school hungry and/or inadequately nourished and are less able to focus on schoolwork.

3. Stress: Families where parents are unemployed are under greater psychological stress. Parents, no matter how well-intentioned, become more arbitrary in their discipline, less supportive of their children. Children from families in such stress are more likely to act out in school, less able to progress academically.

4. Poor Health: Families where parents lose employment are also more likely to lose health insurance. Their children are less likely to get routine and preventive health care, and more likely to miss school days because of illness. They are less likely to get symptomatic treatment for illnesses like asthma,the most common cause of chronic school absenteeism. Andchildren with asthma, even when they attend school, are more likely to come irritable, having been up at night wheezing.

All these consequences of unacceptably high unemployment rates for black parents contribute to depressing student achievement for black children. It is obtuse to expect to narrow the black-white achievement gap in such circumstances. It is fanciful for national policymakers to pick this moment to raise their expectations for academic achievement from children of families in such stress.

It would inappropriately undermine the credibility of public education if, in such an economic climate, educators were blamed for their failure to raise black student achievement. Indeed, they should probably get credit for preventing black student achievement from falling further.

So when I say that school reform efforts are distracting from more important environmental causes of low student achievement, I do not suggest that we should ignore school improvement. We should improve teacher quality, reduce class sizes for the most disadvantaged young children, and improve curricula. But, in the current economic environment, these measures are likely to be overwhelmed by the impact of unemployment. Reducing unemployment is not only a “social welfare goal.” It is also an educational issue, likely as important, if not more so, than the school reforms now advocated by the education policy community. And educators should be outspoken in calling attention to this critical impediment and resist demands that they meet “utopian” achievement goals when environmental forces are so forcefully arrayed against them. Politically, educators should be a mobilized constituency for new job-creating economic stimulus measures, for increased aid to state and local governments to prevent teacher layoffs (resulting in growing class sizes), and for increased formula funding for compensatory education funds (proposed for real inflation-adjusted reductions in the administration’s “blueprint” for education funding).

As a second issue, let’s examine more carefully the “urban myth” that “low-income children would benefit hugely from being assigned to the best teachers.” A previous contributor to this symposium attributed this finding to “multiple econometric studies,” but there is less here than meets the eye.

What policymakers who’ve been told about these studies–few have actually read them–have concluded is that, if low-income students were assigned for five years in a row to teachers who were better than roughly 80 percent of other teachers, these low-income students would achieve at the level of typical middle-class students. In other words, they would close the “achievement gap.”

The Death and Life of the Great American School System (the new book by historian Diane Ravitch) dismisses this claim by asserting that it is not feasible to seek a policy that would move all teachers up to the level of effectiveness of the top quintile of teachers. Even for the average teacher, this would require a full standard deviation in improvement, and, for teachers who are below average, up to a two standard deviation improvement would be required. No social policy can aspire to such a degree of improvement. A more realistic expectation is that average teachers might, with the right guidance, be transformed into, say, sixtieth percentile teachers–something like a 0.3 standard deviation in improvement. Even this task is especially difficult because the “econometric studies” from which this enthusiasm is derived cannot identify a single characteristic of a “good teacher.” Their conclusions are only circular: The best teachers are teachers whose students have better achievement, and students will get better achievement if they are assigned to the best teachers.

But there is a bigger problem, arising from a misunderstanding by policy enthusiasts of the econometric findings. What these studies actually find is that, in any year, some teachers get achievement gains from students that are about 0.2 standard deviations better than those of students who have typical teachers. The policy enthusiasm for closing the achievement gap simply by improving teacher quality stems from multiplying this result by five and concluding that a student who had such teachers for five years in a row w
ould gain a full standard deviation (5 X 0.2) advantage. This assumes that the gains students make during the year in which they had a good teacher persist and can be added to in each subsequent year. But there is absolutely no evidence for such a proposition in the econometric studies or elsewhere. No econometrician has actually observed students, or data regarding such students, who have achieved such cumulative gains for five years in a row under the tutelage of superior teachers. The most reasonable speculation is that students who make big gains in a year will find it more and more difficult to make similarly big gains in subsequent years, even with the best teachers. The earlier gains are likely to be much easier. Would the decline in gain be less if the next teacher was also in the top quintile? Nobody knows. It is a topic on which there has been absolutely no research.

And finally, even if we could confirm that the effect of teacher quality on student achievement was as great as these misinterpretations of econometric studies suggest, we should remember that the studies themselves are based only on basic skills test scores of math alone or reading alone. As we have learned from the corruption stimulated by No Child Left Behind (and documented in Death and Life), teachers responsible for greater gains might be accomplishing this with more test preparation, and less quality instruction, than teachers responsible for smaller gains. Are high-quality teachers of math the same teachers as high-quality teachers of reading? Do teachers who do a great job of teaching math also do a great job of inspiring a love of the arts, of teaching history and civic responsibility, of developing students’ ability to solve conflicts peaceably? We have no answers to these questions? If we did, should we seek to replace teachers who only do an average job of teaching math, but a spectacular job of developing character?

Of course, some teachers are better than others, and consistently so. Figuring out what makes these teachers better, and whether these characteristics can be taught to other teachers, would be a useful policy initiative that could result in raising the level of disadvantaged students’ achievement. But better teachers are not so much better that improvement in teacher quality is a plausible way to make a serious dent in the achievement gap. It is what Death and Life says it is: a nonexistent silver bullet or miracle cure.

Diane Ravitch says that was the logic that initially led her to support the policies of No Child Left Behind, requiring all public schools to measure student performance through standardized tests.

In the eight years since the act took effect, however, Ravitch said she has come to understand that the policies and the punitive measures put in place for schools failing to meet proficiency requirements were as unrealistic as requiring all cities to become crime-free by a target date, and then shutting down police departments that failed to achieve that impossible goal.

Ravitch, an education historian and former Assistant Secretary of Education in the George H.W. Bush administration, explains in her new book, The Death and Life of the Great American School System, how her thoughts on testing, school choice, and teacher accountability evolved. On March 15, she spoke on a panel at EPI, How Testing and Choice are Undermining Education,  about the damage that has resulted from a heavy focus on standardized tests.

Because of the strong bipartisan desire to improve the country’s educational system, No Child Left Behind won the overwhelming support of Democrat and Republican lawmakers in 2001. But in addition to mandating annual testing in reading and math, the policy contains strong punitive measures that Ravitch now says has left “public education in great peril.” Schools where all students do not meet standards for test scores by 2014 could be closed, and with that deadline fast approaching, it appears that many schools will fall short. Ravitch cited one study projecting that close to 100% of all the elementary schools in California could be deemed failing schools. No Child Left Behind, she says, put in place a set of “totally utopian and unrealistic expectations.”

Nor is the large number of so-called failing schools the only problem. Other panelists  at the March 15 event—Carmel Martin, Assistant Secretary for Planning, Evaluation, and Policy Development at the Department of Education;  Bill Galston, Senior Fellow at the Brookings Institution and White House policy advisor on education in the Clinton administration; and Randi Weingarten, President of the American Federation of Teachers—each stressed that education had suffered from a practice of  teaching narrowly to math and reading tests, while neglecting other curricular areas such as the arts, history, science, and physical education. In addition, in an effort to declare more students “proficient” in math and reading, proficiency standards had frequently been watered down.

“I can get a graduation rate of 100% if you put me in charge of a school,” explained Ravitch. “But some of the graduates will be illiterate.”

 Panelists debated the best way to move forward from the current policy but agreed that future policies should focus less on sanctioning poorly performing schools than on rewarding outstanding performance and recognizing that not all schools could be held to the same set of expectations. Schools in communities with high poverty rates, said the panelists agreed, were particularly challenged.

“Any time we alleviate poverty, we will increase the odds of educational improvement,” said Ravitch.

In The Death and Life of the Great American School System, Diane Ravitch charges that the Obama administration’s enthusiasm for charter schools can undermine democratic values because, when more motivated parents choose charter schools, regular schools are left with a concentration of needier students. She mocks the administration’s claims that its policies are evidence-based, describing research showing that charters, on average, perform about the same as regular schools–probably because “abysmal” charter schools are balanced by excellent ones.

Ravitch may concede too much here, if the only basis for claiming that some charters are successful is the standardized basic-skill test scores that she also shows have corrupted American education. Disadvantaged children that No Child Left Behind (NCLB) purports to help are those most likely to have scores inflated artificially by incessant drills and training in test-taking. This pedagogy may raise scores, but it can also subvert interest in math and literature. Thus, while some charters may be successful, others may only seem successful because they are good at making test performance an end in itself.

Ravitch once supported narrow, test-based accountability, but, in her new book, she now repudiates such systems generally and NCLB in particular. Acknowledging past error is embarrassing and courageous, but, in fact, she goes too far in her mea culpa: Her denunciationof the curriculum narrowing that inevitably results from holding schools (and teachers) accountable only for students’ math and reading scores is fully consistent with a lifetime of stressing educators’ obligation to transmit a common culture with a curriculum that includes the arts, science, history, literature, foreign languages, geography, and civics. With her new book, she has returned to deeply held convictions, expressed in such past works as The Troubled Crusade, The Revisionists Reconsidered, and Left Back, her continued membership on the board of the Core Knowledge Foundation, and her promotion of programs like the Concord Review, which publishes original historical research of high school students. Ravitch’s support of NCLB’s test obsession was the aberration.

In the new book, Ravitch observes that NCLB creates incentives for schools to abandon historical knowledge and appreciation of literature, and that the law’s obsession with short-term test scores also pushes schools to teach reading poorly. While easily testable decoding skills are important, more important is the hunger for book knowledge stimulated by children’s literary imaginations or curiosity about historical events. NCLB has its pedagogical theory partly backward: Mechanical skills do create the capacity for understanding, but understanding also creates motivation to learn mechanical skills.

In the past, Ravitch sometimes underestimated the obstacles to implementing a universal high-quality curriculum, leaving her somewhat vulnerable to the seduction of NCLB’s promise that accountability alone would close the achievement gap. For example, in her book Left Back, she praised an 1893 report that recommended a classical academic education (including Latin, Greek, and modern languages) for every high school student and asserted that progressive educators of the twentieth century erred by abandoning this model. Yet, in 1893, only 4 percent of American adolescents attended high school. The report’s authors never imagined that their curriculum would ensure college readiness for sharecroppers’ children whose families were emancipated from slavery only 30 years before, for urban slum-dwelling children of illiterate Italian peasant-immigrants, or for the many adolescents in the remaining 96 percent whose exposure to the common culture we call “Western civilization” would be limited.

Ravitch corrects her oversight in the book. Good standards in all academic areas are needed, she says, along with a curriculum that implements these standards. Good schools also require standards of character and behavior appropriate to an effective learning environment. And, as she writes in this TNR symposium, we must also “pay attention to the health and well-being of students, so that they arrive in school ready to learn.” In Death and Life, she goes further. Regular schools, she says, “cannot be improved if we ignore the disadvantages associated with poverty that affect children’s ability to learn”:

Disadvantaged children need … preschool and medical care. They need small classes … and … extra learning time. Their families need … coordinated social services that help them … acquire necessary social and job skills, and … obtain jobs and housing. While the school itself cannot do these things, it should be part of a web of public and private agencies that buttress families.

NCLB’s proponents have asserted that poverty is correlated with low achievement mostly because educators confront disadvantaged children with the “soft bigotry of low expectations” and forget that “all children can learn.” But, in truth, even with the best curriculum, instruction, and teacher expectations, children who come to school with limited vocabularies and who are unfamiliar with books and the treasures of imagination they unlock will learn less, on average, than children who come to school prepared to learn. Children in poor health have difficulty taking advantage of good instruction, if for no other reason than their excessive absences. Children whose parents are struggling economically come to school with more stress than children who are secure. Children with unstable housing are more likely to switch schools and thus less likely to develop the inspiring relationships with teachers that Ravitch describes experiencing as a child.

Ravitch is not the only prominent supporter of test-based accountability who has reconsidered a prior stance. Others include Susan B. Neuman and Christopher Cross, who, like Ravitch, were former assistant secretaries of education in Republican administrations–Neuman was in charge of NCLB’s initial implementation, and Cross chaired a committee that drafted NCLB’s regulations. (All three were signatories, along with me, of the Broader, Bolder Approach to Education statement, issued nearly two years ago, that challenged NCLB supporters’ views on how to close the achievement gap and raise students’ test scores. I worked with Ravitch, and others, in drafting the statement. When it called for holding schools accountable by means of expert inspections, not tests alone–a theme that reappears in Death and Life–it was Ravitch who insisted we make clear that inspectors should assess whether children are gaining appropriate knowledge, not just a broader range of academic skills.)

Ravitch says she was converted because the “facts have changed,” but that’s not really the case. What has changed is the recent appreciation by her and her colleagues of how incentives to boost basic-skill test scores at the expense of all else inevitably corrupt education. In Death and Life, she describes the familiarity of sociologists, economists, and business theorists with a 1975 observation of Donald T. Campbell that such corruption erupts in any field where simple quantitative measures are substituted for careful evaluation. “Campbell’s Law” is expressed when cardiac surgeons, held accountable for rai
sing surgical survival rates, refuse to operate on the sickest patients most in need of intervention; when employment agencies held accountable for job placement numbers meet their quotas by placing the unemployed only in the most easily filled and lowest-skill jobs; when colleges send pre-completed applications to unqualified high school students because U.S. News & World Report ranks college selectivity by the number of applicants rejected; or when Wall Street traders take reckless risks because they are rewarded only for short-term, easily measured outcomes.

In the education world, some critics long opposed to test-based accountability have reacted churlishly to Ravitch’s late conversion. In a letter to The New York Times, Peter Sacks, author of a 1999 book denouncing the nation’s obsession with standardized testing, complained, “Dr. Ravitch had access to the same evidence that the rest of us had, but chose to ignore it. I won’t be the only scholar who will feel embittered by the whole Ravitch affair.” Yet Sacks and others like him presumably penned their critiques to convince people like Ravitch. It is curious that his own success should embitter him.

Any supporter of narrow, test-based accountability who claims to have an open mind should read Ravitch’s new book and consider the evidence that caused her to reconsider. If any are persuaded that they have erred, I suspect she would welcome their support and not be embittered that they clung to a dangerous panacea longer than she.