In May, the U.S. House of Representatives began hearings on the federal No Child Left Behind Act, with an eye to reauthorizing the legislation next year. Given the enormous gaps in achievement between income and racial groups, the central question is how the federal government can better support states in reaching the law’s lofty goal: making all public school children, poor and wealthy, black and white, “proficient” in reading and math by 2014.
As Congress considers revising the law, it should draw lessons from a small but growing number of communities that are making substantial progress toward achieving the law’s goals. In order to reduce achievement gaps, some school districts—Wake County, N.C.; San Francisco; and La Crosse, Wis., among them—are consciously seeking to integrate students by socioeconomic status. These districts know that while educators have figured out how to make isolated schools with concentrations of poverty work, no one knows how to make high-poverty schools work on a systemwide basis. Indeed, a recently published study by Douglas N. Harris of Florida State University found that middle-class schools are 22 times more likely to be consistently high-performing than are high-poverty schools.
Research has long found that a given student will perform better in a middle-class school than in a high-poverty school. The highly regarded Coleman Report of the 1960s found that, after the influence of the family, the socioeconomic status of a school is the single most important determinant of a student’s academic success. The basic findings of the report—including that all children do better in middle-class schools—have been affirmed again and again in the research literature. (“Race Report’s Influence Felt 40 Years Later,” this issue.) In 2005, for example, University of California, Santa Barbara, professor Russell W. Rumberger and his colleague Gregory J. Palardy of the University of Georgia found that a school’s socioeconomic status had as much impact on the achievement growth of high school students as a student’s individual economic status. Scores from the 2000 National Assessment of Educational Progress among 4th grade students in math indicate that low-income students in more-affluent schools score at levels a year and a half more advanced than low-income students in high-poverty schools.
Almost everything that educators talk about as desirable in a school—high standards, good teachers, active parents, adequate resources, a safe and orderly environment, a stable student and teacher population—is more likely to be found in middle-class schools than in schools with high concentrations of poverty. The student populations of high-poverty schools are more likely to act out; their parents are more likely to move from one school zone to another during the school year, and are less active in the PTA; and their teachers have lower test scores, less experience, and lower expectations for students than do teachers in middle-class schools.
Drawing on these findings, the Wake County district, which includes Raleigh, has adopted a policy that no school have more than 40 percent of its students eligible for free and reduced-price lunches, or more than 25 percent performing below grade level. The program is working well, and Wake County’s low-income students are doing substantially better than low-income students in other large, urban North Carolina districts with concentrated poverty. On the state’s 2005 high school end-of-course exams, 63.8 percent of Wake County’s low-income students passed, compared with passing rates for low-income students of 47.8 percent in Mecklenburg County, 47.9 percent in Guilford, 51.8 percent in Forsyth, and 48.7 percent in Durham. Meanwhile, Wake County’s middle-class students are achieving at very high levels—86.8 percent pass the end-of-course exams—and there is no evidence that they are being harmed academically by economic mixing.
In theory, the No Child Left Behind Act dovetails with Wake County’s plan by allowing students in failing high-poverty schools to transfer to better (usually more affluent) public schools. Philosophically, the federal law directly challenges the idea that, through choice of housing, parents may “purchase” the right to send their children to a public school where all the children come from privileged backgrounds. In addition, by mandating that districts pay transportation costs for transferring students, the No Child Left Behind law rejects the argument that spending dollars on busing is a wasteful diversion of money from the classroom.
Four years of experience under the No Child Left Behind Act have tested this theory, however, and problems have emerged that undercut the promise of new opportunities under the federal legislation. A December 2004 study by the U.S. Government Accountability Office found that, of the roughly 3.3 million students in Title I schools who were eligible to transfer in the 2003-04 school year, only 31,500 (less than 1 percent) did so.
The low levels of student transfers have been linked primarily to two key limitations in the law. First, there are strong incentives for middle-class, high-performing schools to refuse to take in low-income transfer students. As the University of Virginia’s James E. Ryan notes, a receiving school that takes in low-income students faces a risk that the influx is likely initially to depress aggregate school scores, increasing the chances that the receiving school will itself be declared “in need of improvement.” Second, the number of good receiving schools available in certain high-poverty districts is very small. In Chicago, for example, the GAO found that 19,000 students applied for transfers in the 2003-04 school year, but the vast majority (almost 18,000) were unable to transfer because of a lack of space in higher-performing schools.
With a few critical changes, however, the No Child Left Behind Act could translate its theoretical potential into actual support for socioeconomic school integration.
First, steps should be taken to change the incentives so that it is easier for students to transfer out of failing schools and into succeeding ones. To reduce the current disincentive for receiving schools, there is a growing consensus across the political spectrum in favor of providing a grace period to ensure that schools are not judged on the basis of performance by transfer students who have just entered. The law further should bestow affirmative financial bonuses to receiving middle-class schools, to provide an incentive to spur such schools to recruit transfer students actively. The No Child Left Behind law’s system of testing, accountability, and sanctions is built around the concept that incentives matter. That insight must be extended to the student-transfer process as well.
Second, where individual districts lack the capacity to offer room at better-performing public schools, the law should provide a requirement that interdistrict public school choice options be made available. If receiving districts are given the right financial incentives, political opposition to interdistrict choice can be muted. In Michigan, school districts actually compete to attract interdistrict transfers because state funds travel with students to receiving districts.
The threat of private school vouchers is also likely to strengthen political support for interdistrict public school choice. Today, there is widespread agreement that limiting transfers to other schools within certain urban jurisdictions is unworkable, because there simply are not enough good schools into which students can transfer. The primary conservative answer to this dilemma is to open up choice to private schools. While vouchers are, for several reasons, a bad proposal on the merits, the threat of vouchers may push advocates of public schools (including teachers’ unions) to endorse interdistrict public school choice as a superior alternative.
Politically, if the battle comes down to interdistrict public school choice vs. publicly funded private school vouchers, history suggests that public school choice will prevail. An estimated 300,000 to 500,000 students cross district lines every day to attend public schools in another district—more than 10 times the number who use publicly funded vouchers for private schools.
While educators have figured out how to make isolated schools with concentrations of poverty work, no one knows how to make high-poverty schools work on a systemwide basis.
Interdistrict choice is already receiving support from some surprising quarters. The Thomas B. Fordham Foundation’s Chester E. Finn Jr. supports strengthening the No Child Left Behind Act’s interdistrict-transfer provisions. And the Bush administration’s fiscal 2007 budget request to Congress calls for funneling funds for voluntary public school choice into pilot programs specifically aimed at interdistrict choice.
Third, to balance the flow of students and money under an interdistrict public school choice program, federal funding for magnet schools in urban areas should be substantially increased. Allowing students to move from bad schools to good schools will, in the short term, usually mean transferring from high-poverty city schools to middle-class suburban schools, but many districts (including Wake County in North Carolina) have achieved socioeconomic integration by simultaneously attracting middle-class children to urban schools through magnet programs. Experience suggests that given the right program, magnet schools can attract middle-class suburban students to schools located in some of the toughest urban neighborhoods. In Hartford, Conn., for example, a Montessori magnet school, located near boarded-up buildings, has a long waiting list of white, middle-class suburban children because the program offered at the end of the bus ride is excellent.
Fourth, the No Child Left Behind Act should include a targeted-research component. The law’s transfer program offers an important opportunity to study the effects of socioeconomic integration on student achievement. The achievement of students who remain in failing, high-poverty schools should be tracked compared with that of students who transfer to higher-achieving middle-class schools. It is also important to measure the effect of the changing economic composition on the achievement of all students in the receiving middle-class schools.
At the end of the day, separate but equal schools for rich and poor have never worked well. If the twin goals of the No Child Left Behind Act are to be taken seriously—to raise overall achievement and narrow the achievement gaps between groups—the law should be amended to encourage what research has long found to be the single most promising step for raising the achievement of low-income students: allowing them to attend high-quality, middle-class public schools. The theoretical and philosophical underpinnings for this policy are already in place under the federal law. Now it is time to move from theory to practice.
A version of this article appeared in the June 21, 2006 edition of Education Week as Helping Children Move From Bad Schools To Good Ones