Journal Issue: The Next Generation of Antipoverty Policies Volume 17 Number 2 Fall 2007
The Federal Role
From one perspective, improving the education of children living in poverty is straightforward. Policymakers should define clearly the skills and knowledge students should master at each grade level. Schools should be run by school principals who know how to recruit and support effective teachers and provide them the tools to do this work. Schools should attract and support experienced, skilled teachers committed to working together over an extended period to continuously improve instruction. School staff should monitor the learning of every student, intervene rapidly at the first sign that a student is not making good progress toward mastering critical skills, and provide alternatives when conventional pedagogies are not effective. And the school day and school year should be long enough that students can have extra time to acquire critical skills if they need it.
But few children living in poverty attend such schools. Instead, they typically attend schools where leadership is weak, many teachers lack critical skills, instruction is inconsistent, and learning problems are left unattended. A great many disadvantaged children thus leave school without the skills they need to earn a decent living and to provide for their own children.
The reasons why disadvantaged children typically receive a poor education are numerous and interrelated. Housing patterns leave poor children, who have especially great learning needs, concentrated in particular schools and school districts.6 Precarious and uncertain city budgets prevent urban districts from hiring skilled teachers in a timely manner. Difficult working conditions, combined with seniority provisions of collectively bargained labor agreements, leave low-performing schools with the least teaching talent.
During the past fifteen years virtually every state in the country has adopted standardsbased educational reforms, often called testbased accountability, as its primary strategy for improving public education. Although details vary greatly from state to state, all standards- based educational reforms include three components. The first is the standards themselves: content standards that specify what students should know and be able to do, performance standards that describe how students should demonstrate their knowledge and skills, and assessments that measure the extent to which students meet performance standards. The second component is incentives to encourage educators and students to devote the time and energy needed to meet the performance standards.7 The final component is teachers who have the knowledge, skills, and resources to prepare all students to meet the performance standards.
The third component—building capacity to deliver consistently high-quality instruction to all children—has been the most neglected part of standards-based reforms in most states. By themselves, the first two components— standards and incentives—will not improve student performance. Teachers must know how to achieve the mandated outcomes. But in most schools serving high concentrations of poor children teachers lack the requisite skills and knowledge and have few opportunities to acquire them. Building teaching capacity is thus as critical to improving student outcomes as appropriate standards and incentives are.
Since the passage of the Elementary and Secondary Education Act of 1965 (ESEA), the federal government has provided funding to improve the education of economically disadvantaged children. Title I of this law has been the nation’s primary compensatory education program, distributing funds to school districts on the basis of a formula that weights heavily the number of children living in poverty.8 The most recent reauthorization of ESEA, the No Child Left Behind Act of 2001, marked a significant change in the federal role in public K–12 education. The new law requires states to test annually the reading and mathematics skills of all public school students in grades three through eight. It also specifies that all schools are expected to make adequate yearly progress (AYP) toward ensuring that all groups of students, including groups defined by race or ethnicity and poverty, reach proficiency within twelve years (by 2014). School districts and schools that fail to demonstrate adequate yearly progress for all groups of students are subject to corrective actions that can ultimately include the replacement of staff and school reconstitution.
One strength of NCLB is that it draws attention to the academic skills of children from low-income families, children of color, children whose first language is not English, and children with disabilities—groups that historically have not been well served by American schools. The importance of creating incentives for schools to pay attention to these often forgotten groups cannot be overestimated. It is the primary reason that some civil rights groups have supported NCLB.
At the same time, several provisions of the new law create perverse incentives for states and for educators. One source of perverse incentives is the fact that the adequate yearly progress requirements are well beyond the reach of even the states that have made the most progress in improving students’ reading and math skills. North Carolina, for example, made the greatest gain between 1990 and 2000 in the share of students who score proficient or above on the eighth-grade NAEP mathematics test. If North Carolina were able to sustain this top-ranking rate of progress, almost 60 percent of its eighth graders would earn scores of proficient or above by 2014—a remarkable accomplishment, but well short of the required 100 percent.9
A related problem is that the accountability system has only two categories: schools that make adequate yearly progress and those that do not. Thus a school in which a few students in one ethnic group in a single grade fail to make adequate yearly progress is not distinguished from a school in which all ethnic groups at every grade level fail to do so.
It makes sense to have ambitious performance goals. But an accountability system with unrealistically high goals will not improve public education. As educators become increasingly aware that even consistently good teaching will not allow their schools to satisfy adequate yearly progress requirements, their behavior will become increasingly dysfunctional and contrary to the interests of children. Skilled teachers, for example, will be likely to leave schools serving high concentrations of poor children. And some teachers will focus instruction unduly on test preparation. 10 In addition, in a system in which a great many schools, even those that have made real progress, are labeled “underperforming,” it may be difficult to identify the schools most in need of intervention.
A second source of perverse incentives in NCLB is the fact that states are allowed to choose their own tests and their own minimum scores for achieving proficiency. This latitude, combined with the pressure of having to meet adequate yearly progress requirements, encourages states to make their tests relatively undemanding and to set low minimum scores. A look at how students perform on state-mandated tests and on the National Assessment of Educational Progress highlights this problem. In 2003, 77 percent of fourth graders in Alabama scored “proficient” on the state Reading-Language Arts exam, but only 22 percent scored “proficient” on the NAEP fourth-grade examination.11
A third weakness of NCLB concerns high school graduation rates. Although the law requires states to include graduation rates in setting adequate yearly progress goals for high schools, it does not specify how they must do this. In interpreting the law, the U.S. Department of Education has allowed states to measure graduation rates in a variety of ways and to set their own goals for improving those rates. Moreover, there is no requirement that goals be met for subgroups of students, defined by race and poverty. As a result, high school graduation rates, one of the most important indicators of school districts’ success in serving students, play almost no role in the NCLB accountability system. One indication of the need to improve graduation rates is that among twenty-two industrialized democracies for which on-time high school graduation rates are available, the United States now ranks nineteenth.12
In summary, test-based accountability systems introduced by states and the federal government have had an impact on what happens in American schools, especially those that serve high concentrations of poor children. The challenge now is to revise NCLB and state accountability systems so that children living in poverty make greater gains. The federal government can help make standards- based reforms a success by strengthening the foundation on which they rest, namely, accountability, incentives to serve poor children, and the teaching capacity to serve poor children.
Before turning to my specific recommendations, I want to make clear that they are based on the presumption that the United States will retain its basic governance structure for education. Local communities, operating within boundaries set by states, will make most of the decisions that determine the day-to-day school experiences of children. They will hire teachers and administrators, choose curricula, set the length of the school day and year, and invest in improving the quality of instruction. Individual communities are responsible for educating all students living within their geographic boundaries; they have no responsibility to educate students in neighboring communities. States, in their evolving role, will create content standards, choose tests to measure students’ mastery of these standards, determine what requirements college graduates must meet to obtain a teaching license, and fund a significant share of local school spending.
The assumption about governance structure is important because the current structure limits the ability of federal policies to improve the education of poor children. If the governance of American public K–12 education were more centralized, the options for federal policy would be quite different and closer to those that some other industrialized countries have adopted. They might include national content standards and national assessments, a single set of training and licensing requirements for all teachers, assignment of teachers to particular schools in geographic areas encompassing many communities, opportunities for students to attend a wide variety of schools located in nearby communities, and a common strategy for identifying students who are not making good academic progress and for consistently applying intensive intervention strategies.
In an important sense, the governance structure of American public education is evolving. States play larger roles in determining curricular and testing requirements than they did thirty years ago. NCLB marks a larger federal role. Nonetheless, local control remains a central tenet of the educational governance structure.