Chairman Miller, ranking member McKeon, members of the committee, thank you for the opportunity to testify today. My name is Kevin Carey; I am the research and policy manager of Education Sector, a national, independent nonpartisan education think tank. Because Education Sector does not take institutional positions on issues or proposed legislation, the views I express today are my own.

For the past two decades, Democrats and Republicans alike have pursued a goal that transcends party affiliation: ensuring that all students—regardless of race, economic background, disability or language—have equal access to a high-quality education that will prepare them for work and life. The discussion draft amendments to the No Child Left Behind Act recently put forth by this committee clearly seek to further that goal, and for this the committee should be commended.

The draft also seeks to address many of the criticisms that rightly have been leveled against No Child Left Behind since its enactment over five years ago. Policymaking is by nature an iterative process, and no one should be surprised that the experience of implementing No Child Left Behind has revealed new opportunities to make the law more effective and fair. The committee should again be commended for carefully listening to the voices of parents, educators, researchers, and advocates who have recommended ways to improve the law.

Some proposed changes are particularly worthy of mention.

By improving the "comparability" provisions guaranteeing that schools receiving Title I funds must first receive an equal share of state and local funds, the draft takes a very important step toward ensuring that low-income students receive their fair share of school resources. Research from the Center on Reinventing Public Education has shown that in some school districts, high-poverty schools receive nearly a million dollars less per year than low-poverty schools of similar size. This provision alone will go a long way toward ensuring that low-income students are not forced to attend schools that serve as a revolving door for inexperienced, under-paid teachers.

Similarly, by making Title II funding contingent on states taking steps to ensure that poor and minority students get their fair share of experienced, qualified, effective teachers, the committee is taking a bold but necessary step to ensure real educational equity for disadvantaged children. Research has shown that the quality of classroom teaching has a huge impact on student learning, particularly for at-risk children. But studies also show those same students are much less likely than others to be taught by the best instructors. It is a long-accepted principle that all children deserve equal access to education funding. These proposed amendments simply extend that resource-equity principle to the single most valuable resource schools have: their teachers.

And by eliminating the so-called High Objective Uniform State Standard of Evaluation (HOUSSE) exception to the guarantee that all students be taught by a highly-qualified teacher, the draft closes a loophole that many states have used to avoid addressing the fact that many students—disproportionately low-income and minority students—are taught by teachers without sufficient training or content knowledge in their field.

The Title II amendments also reflect the need to improve the overall quality of the nation's education workforce. We have a major human capital problem in education; without high-quality personnel who are properly supported, even the best-laid plans and accountability systems will fall short. Some of the committee’s recommendations, particularly relating to teacher pay, are bound to be controversial. But they are also important and long overdue. We cannot recruit the best and brightest into the classroom and expect them to excel and persist once they get there without taking every opportunity to recognize and reward excellence in education. By investing new resources in innovative programs designed to increase teaching excellence, the federal government can help leverage change throughout public education.

In no small part because of the law's emphasis on data and the important work of states and organizations like the Data Quality Campaign, states are now in a much better position to collect and use data than when NCLB was first enacted. By supporting the development of longitudinal education data systems—including, crucially, the ability to link student data to teacher data—the committee will help further that progress and build the capacity of states to develop new information tools for schools and educators. The better we understand our schools, the better we are able to improve them.

One of the most promising applications of these new data systems is the ability to measure the academic growth of individual students. By allowing states to use year-to-year student growth for accountability purposes, but also requiring that students be on a three-year trajectory toward proficiency, the committee has struck the right balance between addressing valid criticisms of NCLB's accountability requirements while maintaining the law's core commitment to common performance standards for all students regardless of race or income. Similarly, the use of a "performance index" can give states an incentive to focus on students across the achievement spectrum—as long as success at the high end doesn't unduly divert resources from students who struggle the most.

By giving states incentives to adopt more rigorous, nationally and internationally benchmarked achievement standards, and by calling for new investments in the quality of state tests, the committee will strengthen the standards and assessment foundation on which the entire accountability enterprise rests. By creating a "Graduation Promise Fund" and requiring more stringent accountability requirements for high school completion, the committee will push schools to improve the appallingly low graduation rates that plague our secondary schools.

By allowing schools to consider college-going rates in judging high school success, the committee will help bridge the great divide between the nation's systems of P-12 and higher education. This provision could be expanded further still. As a recent Education Sector report titled Reality Check: Tracking Graduates Beyond High School shows, states like Oklahoma and Florida have used longitudinal data systems to create new measures of high school success, such as the average college grade point average of a high school's graduates, the percent of graduates forced to take remedial courses in college, and the percent who obtain a good-paying job. If, as we all agree, the goal of high school is to prepare students to succeed in the workplace and further education, it's fair to take into account whether students actually do succeed in those areas when judging high school success. By allowing these measures to be incorporated into NCLB, leading states would be rewarded for innovation, while others would have an incentive to invest in their information infrastructure.

The draft also limits the ability of states to use various statistical loopholes to reduce pressure on local schools and districts to improve. By disallowing the exclusion of sub-groups of students larger than 30, and by limiting statistical "confidence intervals" to the 95-percent level, the draft improves the law's focus on closing achievement gaps for disadvantaged students, and helps ensure that when targets are set for school improvement, schools actually have to meet them.

There are also areas where I believe this draft can be significantly improved to ensure that the law is clear, transparent, and focused on helping the students who need help the most.

By adding options like growth models, the committee recognizes that accountability systems need to account for the nuance and complexity inherent in an enterprise like public schooling. But complexity comes at a potentially high cost to both the integrity and transparency of the accountability system. Indeed, striking a balance between complexity, integrity, and transparency is probably the single most difficult task the committee faces. There is a danger that in seeking to address every criticism of NCLB, the committee will make the law's accountability provisions so complex that many new opportunities will emerge to exploit the law’s intricacies to undermine its core principles. There is also a risk that the law will become so inscrutable that it will cease to function as an effective engine of change.

As you know, the subgroup size and confidence interval loopholes closed by the discussion draft weren't originally part of NCLB. They were invented by state departments of education. A clear pattern has emerged during NCLB’s implementation: some states—not all, but some—have exploited their flexibility under the law to undermine the law's fundamental principles. I was a state education official before moving to Washington, D.C., and I believe many of these actions are born of good intentions—ensuring that hard-working educators aren't unfairly tarred as low-performing.

But by opening a series of statistical safety valves in the AYP system, and by looking for every opportunity to push back the day when underperforming schools are required to do what must be done on behalf of disadvantaged students, these states have greatly undermined the law’s effectiveness. As of today, some states have still identified less than one percent of their school districts as "in need of improvement," an amount that defies both the intent of Congress and plain common sense. As a recent Education Sector report called Hot Air: How States Inflate Their Progress Under NCLB shows, this unfortunate trend of stretching the letter of the law to subvert its spirit extends to many other NCLB provisions, including those governing teacher qualifications, graduation rates, and school safety.

States truly are, in the words of Supreme Court Justice Louis Brandeis, "laboratories of democracy"—but too often that creativity and energy has been used to develop new ways to ease pressure on schools to improve, rather than to find new ways to help them improve. And the U.S. Department of Education's oversight in this area has been inconsistent at best. There is no reason to believe these patterns won't continue with the next version of No Child Left Behind. Therefore, the committee should make new accountability options as clear, precise, and limited as possible.

Too much complexity can also undermine the process of school improvement itself. The idea behind standards and accountability is simple: Identify a problem, then focus resources and attention on fixing the problem. But when we pile system upon system and measure upon measure, it becomes difficult—if not impossible—for parents and educators to know why a given school is labeled a success or a failure. This will breed mistrust of the entire system. And if educators don't know why they're falling short, it will be very difficult for them to determine how to improve.

For these reasons, the committee should limit multiple measures to a small number of reliable, high-quality assessments that are accessible to all students statewide, and ensure that performance goals in these areas are high. It should also limit the extent to which success on these indicators can mitigate failure in the foundational subjects of reading and math. The percent of a school's annual measurable objective attributable to measures other than reading and math achievement should not be increased from the levels established in the discussion draft. This will balance the laudable goal of avoiding "curriculum narrowing" with the need to ensure that all students are proficient in the core subjects on which success in all other areas depends.

The proposed local assessment pilot project deserves particular attention. I commend the committee for working to forge a greater connection between the local educators who work directly with children and the process by which those educators are judged. Unless teachers believe assessment is reliable, accurate, and fair, accountability will never work as intended.

But purely local accountability is ultimately indistinguishable from no accountability at all. Everyone works better when they know someone else is paying attention to how well they work. It's unreasonable to expect schools to judge themselves objectively when the consequences of that judgment can be significant. Local assessments thus have the potential to undermine NCLB's core promise of equal education standards for all, perhaps the most important civil rights goal of our time.

For these reasons, I recommend that the committee reduce the number of states eligible for the local assessment pilot project from 15 to five, and that the committee ensure that data from state assessments continue to constitute the large majority of information used in determining adequate yearly progress.

In creating a new distinction between "High Priority" and "Priority" schools, the draft sensibly focuses scarce resources and attention on the schools in greatest need of help. But because the distinction between the two levels is primarily a function of the number of student subgroups who miss academic goals, there is a danger that significant, persistent achievement gaps for disadvantaged students will be allowed to endure. I recommend that the committee maintain the two levels of "High Priority" and "Priority" schools, but also ensure that a school cannot be identified at the less-severe "Priority" level if large achievement gaps persist for a student subgroup that constitutes a significant percentage of the school population—even if only one subgroup is falling behind. I also recommend eliminating the proposed "alternative process" for identifying "High Priority" schools; such a process will create needless complexity and opens up new avenues to circumvent the law's goals.   

For any accountability system to work for English language learners, states and districts must be able to do three things: accurately identify ELL students, provide quality instruction for language proficiency and academic content, and administer appropriate assessments that reliably measure the effects of this instruction. States are struggling with all three. The proposal to provide additional resources and attention to state capacity-building for the development of quality instructional practices for ELLs and the development of appropriate and valid assessments is important. And in extending the timeframe for using native-language tests to assess ELLs, the committee recognized that to accurately measure the academic knowledge and achievement of these students, we must use tests those students can read and comprehend.

However, requiring states to develop native-language tests for every language that represents 10 percent or more of the state's ELL population is onerous and, absent native instruction, will not ensure more accurate measurement of learning for a significant portion of the ELL population. The main priority should be investing new resources in developing psychometrically reliable and valid ELL assessments. States and districts do not currently have the expertise and capacity to do so without additional support. And as an Education Sector report titled Margins of Error: The Testing Industry in the No Child Left Behind Era has shown, the testing industry is currently hard-pressed to meet this and many other assessment challenges. The committee was also correct in requiring the improvement of state data collection on ELLs. As it stands, states and districts are simply not collecting reliable data on this population, nor are they collecting data in the same way. Without good information, we cannot expect any true measure of accountability for these children.

When Congress passed the No Child Left Behind Act almost six years ago, it renewed the historic promise of the Elementary and Secondary Education Act to give disadvantaged students a fighting chance to succeed in a world that increasingly values education above all else. To achieve this goal, Congress enacted a number of bold but necessary reforms. These actions were controversial, to say the least. But they were also the right thing to do.

The committee's first priority should be to sustain that commitment while enacting a new set of needed reforms for the years to come. This draft is a positive step in that direction, and my colleagues at Education Sector and I look forward to being of assistance in making it stronger still.

To learn more about this committee hearing visit the House Committee on Education and Labor Web site.