The Next Kind of Integration

July 20, 2008
The Next Kind of Integration

By EMILY BAZELON
In June of last year, a conservative majority of the Supreme Court, in a 5-to-4 decision, declared the racial-integration efforts of two school districts unconstitutional. Seattle and Louisville, Ky., could no longer assign students to schools based on their race, Chief Justice John Roberts wrote in his lead opinion in Meredith v. Jefferson County School Board (and its companion case, Parents Involved in Community Schools v. Seattle School District No. 1). Justice Stephen Breyer sounded a sad and grim note of dissent. Pointing out that the court was rejecting student-assignment plans that the districts had designed to stave off de facto resegregation, Breyer wrote that “to invalidate the plans under review is to threaten the promise of Brown.” By invoking Brown v. Board of Education, the court’s landmark 1954 civil rights ruling, Breyer accused the majority of abandoning a touchstone in the country’s efforts to overcome racial division. “This is a decision that the court and the nation will come to regret,” he concluded.

Breyer’s warning, along with even more dire predictions from civil rights groups, helped place the court’s ruling at the center of the liberal indictment of the Roberts court. In Louisville, too, the court’s verdict met with resentment. Last fall, I asked Pat Todd, the assignment director for the school district of Jefferson County, which encompasses Louisville and its suburbs, whether any good could come of the ruling. She shook her head so hard that strands of blond hair loosened from her bun. “No,” she said with uncharacteristic exasperation, “we’re already doing what we should be.”

Todd was referring to Louisville’s success in distributing black and white students, which it does more evenly than any district in the country with a comparable black student population; almost every school is between 15 and 50 percent African-American. The district’s combination of school choice, busing and magnet programs has brought general, if not uniform, acceptance — rather than white flight and disaffection, the legacy of desegregation in cities like Boston and Kansas City, Mo. The student population, which now numbers nearly 100,000, has held steady at about 35 percent black and 55 percent white, along with a small and growing number of Hispanics and Asians.

With its decision in Meredith, the court was forcing Louisville to rethink the way it would assign elementary-school students and, in the process, to confront some tricky questions. Is the purpose of integration simply to mix students of different colors for the sake of equity or to foster greater familiarity and comfort among the races? Should integration necessarily translate into concrete gains like greater achievement for all students? If so, is mixing students by race the most effective mechanism for attaining it?

In Louisville, the achievement gap between whites and blacks is 20 percentage points at many grade levels. For Todd and her team, whatever their reservations about the decision in Meredith, coming up with an alternative assignment plan was an opportunity to think about a new kind of integration and what it might accomplish. In Louisville, integration would no longer focus solely on race but also on the barriers of class, of advantage and disadvantage. Other cities have been thinking along these lines. In the wake of the Supreme Court’s decision, four other districts — Des Moines, Burlington, Vt., Omaha and Beaumont, Tex. — announced a switch to class-based integration. Seattle, too, is discussing setting aside 5 to 15 percent of the spots (a relatively small percentage) in desired high schools for low-income students. Some of the plans go into effect this fall; others, including Louisville’s, begin a year from September.

The chief justice didn’t address the idea of class-based integration in his opinion. But Justice Anthony Kennedy did, in a separate concurrence. And because Kennedy cast the fifth vote for the majority, his view controls the law. Though he agreed with Roberts that public school districts should not make school assignments based on the race of individual students, he added that the court’s ruling “should not prevent school districts from continuing the important work of bringing together students of different racial, ethnic and economic backgrounds.”

How were schools to do this? Around the country, school-district lawyers studied Kennedy’s opinion and came to a rough consensus. In its amicus brief before the court, the Bush administration cited socioeconomic integration as a “race neutral” alternative to race-based assignment plans. Kennedy picked up on this, and no other justice wrote to contradict him. As a result, the school-district lawyers concluded that districts could assign an individual child to a school based on any kind of socioeconomic measure they chose — income, assets, parental education attainment. Districts could also be “race conscious,” according to Kennedy, when they drew school boundaries, chose sites for new schools and directed money to particular programs. But in these situations, they would usually be limited to taking into account the racial composition of a neighborhood rather than the race of an individual student.

In terms of the court’s jurisprudence, this is a major change. Race has been the organizing principle of integration since Brown v. Board of Education. At the time of the court’s ruling in Meredith, hundreds of districts were pursuing some sort of racial integration, with or without a court order, while only a few dozen at most were trying any form of socioeconomic integration. Over the years, racial integration has proved to have tangible benefits. Amy Stuart Wells, an education professor at Columbia Teachers College, has found that going to school with substantial numbers of white students helped black students to form cross-racial friendships and, by giving them access to white social networks, eventually to find work in jobs higher up the economic ladder.

However important these gains are, they are long-term and cannot be easily or quickly assessed. And increasingly, schools are held to a standard of immediately measurable outcomes. The No Child Left Behind Act, signed into law in 2002, demands student test scores that climb ever upward, with a mandate for all students to be proficient in reading and math by 2014. Test scores may not be the best way to assess the quality of a teacher or a school, but the pressure to improve scores, whatever its shortcomings, is itself on the rise. And if high test scores are the goal, it turns out, class-based integration may be the more effective tool.

Researchers have been demonstrating this result since 1966, when Congress asked James S. Coleman, a Johns Hopkins sociologist, to deliver a report on why the achievement of black students lagged far behind that of white ones. The expected answer was that more than a decade after Brown, black kids were still often going to inferior schools with small budgets. But Coleman found that the varying amount of money spent on schools didn’t account for the achievement gap. Instead, the greater poverty of black families did. When high concentrations of poor kids went to school together, Coleman reported, all the students at the school tended to learn less.

How much less was later quantified. The Harvard sociologist Christopher Jencks reanalyzed Coleman’s data in the 1970s and concluded that poor black sixth-graders in majority middle-class schools were 20 months ahead of poor black sixth-graders in majority low-income schools. The statistics for poor white students were similar. In the last 40 years, Coleman’s findings, known informally as the Coleman Report, have been confirmed again and again. Most recently, in a 2006 study, Douglas Harris, an economist at the University of Wisconsin, found that when more than half the students were low-income, only 1.1 percent of schools consistently performed at a “high” level (defined as two years of scores in the top third of the U.S. Department of Education’s national achievement database in two grades and in two subjects: English and math). By contrast, 24.2 percent of schools that are majority middle-class met Harris’s standard.

There are, of course, determined urban educators who have proved that select schools filled with poor and minority students can thrive — in the right circumstances, with the right teachers and programs. But consistently good education at schools with such student bodies remains the rare exception. The powerful effect of the socioeconomic makeup of a student body on academic achievement has become “one of the most consistent findings in research on education,” Gary Orfield, a U.C.L.A. education professor, and Susan Eaton, a research director at Harvard Law, wrote in their 1996 book, “Dismantling Desegregation.”

Most researchers think that this result is brought about by the advantages that middle-class students bring with them. Richard Kahlenberg of the Century Foundation lays them out in his 2001 book, “All Together Now”: more high-level classes, more parent volunteers and peers who on average have twice the vocabulary and half the behavioral problems of poor students. And, especially, more good teachers. Harris, the economist, says that poor minority students still don’t have comparable access to effective teachers, measured by preparation and experience. The question, then, is whether a plan that integrates a district by class as well as by race will help win for all its schools the kind of teaching that tends to be linked to achievement. “The evidence indicates that it would,” Harris says.

Ronald Ferguson, an economist at the Kennedy School of Government at Harvard, is less persuaded. His research highlights the nagging persistence of a racial achievement gap in well-off suburbs. “What happens with the achievement gap in a place like Louisville,” he says, “will depend on how vigilant their leaders are to make sure high-quality instruction is delivered across the board.” Such teaching is more likely in a school with a critical mass of middle-class parents, he concedes. But he stresses that to reap the benefits, poor kids have to be evenly distributed among classrooms and not just grouped together in the lowest tracks. “To the degree a district takes the kids who struggle the most academically and spreads them across different classrooms, they’re making teachers’ work more doable,” he says. “And that may be the biggest effect.”

Once they started looking for them, Todd and her colleagues saw the effects of class division and poverty in the Jefferson County schools. Thorough racial desegregation had not, it seemed, led to thorough class desegregation. At 40 of 90 elementary schools in the district, 75 percent or more of the students came from low-income homes. And the effects of these high concentrations of poverty were striking: poor students in Louisville, black and white, fared worse when they attended schools filled with other poor kids. In elementary school, 61 percent of poor students at mostly low-income schools scored proficient in reading, compared with 71 percent of poor students at majority-middle-class schools. For math, the comparative proficiency rates were 52 percent to 63 percent. Because black students were disproportionately poor, they were more likely to attend high-poverty schools, and this was contributing to the district’s pronounced black-white achievement gap.

Todd and her planners wanted to tackle the problem, she says, but they were mindful of going too far in their efforts and losing the support of parents. In other districts — including Cincinnati, Evanston, Ill., Bibb County, Ga., and Madison, Wis. — the reaction to the Supreme Court’s ruling had been to move to dismantle racial-integration programs. Todd and other school officials didn’t want integration redefined to turn into no integration all. To get a handle on a new plan, Todd turned to an heir of James Coleman: the researcher John Powell.

In the 1960s, Powell was one of the only African-American students in his advanced high-school classes in Detroit; when he became the class valedictorian, a teacher told him he wasn’t the smartest student. He now directs the Kirwan Institute for the Study of Race and Ethnicity at Ohio State University, and he says he still thinks that race is a category with singular power. But he also appreciates the stark effects of segregating poor kids. “Ever since the Coleman Report, we’ve seen that there’s a high correlation between good schools and schools that are integrated socioeconomically as well as racially,” he says. “I think everyone agrees that what we need are more good schools.”

In Louisville, Powell lent his expertise to Todd and her team. They came up with a computer-generated map that shows what Powell defines as the district’s areas of “low opportunity.” Todd, who is 61 and taught every grade in the Louisville schools before becoming an administrator, went over the map with me one day last December. The map used two different measures of class to identify Jefferson County’s areas of disadvantage: income level and the educational attainment of adults. (To gauge disadvantage, districts embarking on class-based integration often use who among their students receives free or reduced lunch; Powell, however, contends that this is a relatively crude measure.) Using census data, Todd’s team identified the zones in the district in which households fall below the average income and education levels, with fewer adults who have finished high school or gone to college or beyond. Finally, the team added one more factor: a higher-than-average number of minorities, almost all of them African-Americans or Hispanics.

The map’s class-plus-race formula revealed a major partition. One region, which Todd’s team called Geographic Area A, is a mermaid-shaped swath of blue, with its head in Louisville’s West End, just south of the Ohio River, and its tail to the south. The region encompasses the parts of the district with a higher-than-average minority population, lower-than-average median income and lower-than-average adult educational attainment. In Geographic Area A live about 30 percent of Jefferson County’s students. The rest of the county, colored yellow, included everyone else — the better off, better educated and whiter Geographic Area B.

What if the district were to use this map as a guide for school integration? Instead of maintaining each school as no less than 15 percent and no more than 50 percent black, Todd’s team could propose that each school have no less than 15 percent and no more than 50 percent of students from Geographic Area A. By distributing students from the district’s residential zones of disadvantage, the new plan would integrate the schools by class. There would no longer be 40 elementary schools with heavily poor-student populations. There could potentially be no such schools.

Given the presumed boost to test scores resulting from distributing poor students more widely, you might wonder why Todd’s team retained race as an admissions factor at all. To answer this, it’s worth considering the country’s existing examples of purely class-based integration. The best known is in Wake County, N.C. With 134,000 students, the Wake County school district ranks 19 among the country’s 20 largest, spanning 800 square miles that include bleak tracts in the city of Raleigh, mansion-filled suburban cul-de-sacs and rural roads ending in the fresh earth of a new subdivision. The student population is about half white, one-quarter African-American and one-quarter Hispanic, Asian and multiracial. The district voluntarily pursued race-based integration in the 1980s and ’90s. In 2000, after the U.S. Court of Appeals for the Fourth Circuit began to frown on the use of race in student assignment — a harbinger of the Supreme Court’s stance last year — the district began assigning kids to schools based on the income level of the geographic zone they lived in. The aim was to balance the schools so that no more than 40 percent of the students at each one come from a low-income area. (This year, the district added another goal: to have no more than 25 percent of students at any one school for whom English is a second language.)

Wake County adopted class-based integration with the hard-nosed goal of raising test scores. The strategy was simple: no poor schools, no bad schools. And indeed, the district has posted striking improvements in the test scores of black and low-income students: in 1995, only 40 percent of the black students in Wake County in the third through eighth grades scored at grade level in state reading tests; by last year, the rate had almost doubled, to 82.5 percent. Statewide scores for black students also got better over the same time period, but not by as much. Wake County’s numbers improve as students get older: 92 percent of all eighth graders read at or above grade level, including about 85 percent of black students and about 80 percent of low-income students. (Math scores are lower, following a statewide trend that reflects a change in the grading scale.) The district has achieved these results even as the share of low-income students over all has increased from about 30 percent a decade ago to about 40 percent today.

But the lessons of Wake County, Powell and Todd argue, don’t apply everywhere. “In different districts, you have different geographic patterns,” Powell says. “So you need different integration models to shop around.” To begin with, Louisville is less affluent — more than 60 percent of its elementary school students receive free or reduced lunches, compared with Wake County’s 40 percent. In Wake County, the vast majority of the poor students are black and Hispanic, and so mixing kids by class tightly correlates to mixing them by race. But in Jefferson County, more than a third of the kids who receive free or reduced lunches are white. As a result, redistributing students by class alone might still isolate them by race.

This is a limitation of class-based integration that holds true elsewhere. The city of San Francisco, for instance, has undergone substantial racial resegregation since retooling its diversity plan to emphasize socioeconomic factors. Even in Wake County, the fraction of students in racially segregated schools has climbed a bit over the last decade, from 25 percent to 32 percent. A 2006 paper by the education researchers Sean Reardon, John T. Yun and Michal Kurlaender crunched census data across the country and concluded that “given the extent of residential racial segregation in the United States, it is unlikely that race-neutral income-integration policies will significantly reduce school racial segregation, although there is reason to believe that such policies are likely to have other beneficial effects on schooling.”

Many big cities have a different problem. Simple demographics dictate that they can’t really integrate their schools at all, by either race or class. Consider the numbers for Detroit (74 percent low-income students; 91 percent black), Los Angeles (77 percent low-income; 85 percent black and Hispanic), New York City (74 percent; 63 percent), Washington (64 percent; 93 percent), Philadelphia (71 percent; 79 percent), Chicago (74 percent; 88 percent) and Boston (71 percent; 76 percent). In theory, big cities can diversify their schools by class and race by persuading many more middle-class and white parents to choose public school over private school or by combining forces with the well-heeled suburbs that surround them. But short of those developments, big cities are stuck. “The options have shrunk,” says Tom Payzant, a former superintendent of schools in Boston.

Notably, there are a good many districts that have evaded this predicament. They are particularly found in the South, in part because of a historical accident. Because it was predominantly rural for longer, the South has more countywide school districts than the North. An unintended consequence was to ease the way to integration. Instead of city schools filled with poor black and Hispanic kids separated from a burgeoning ring of suburban districts stocked with affluent whites (and in some places, Asians), one district controls student assignment for the region.

Even in school districts with a mix of students of different races and income levels, however, there is no one-size-fits-all approach to socioeconomic integration, as underscored by the differences between Wake County and Jefferson County. Wake County’s demographics entail that mixing kids by class, on its own, produces a fair degree of racial integration. Jefferson County’s demographics don’t necessarily work this way. And so civil rights lawyers suggest that districts configured like Jefferson County should continue to pursue racial diversity directly. They point to cities like Berkeley, Calif., which has an assignment plan that primarily relies on socioeconomics, but like Geography Area A also factors in the racial composition of a neighborhood to guard against resegregation along racial lines. “It’s not either-or,” says Anurima Bhargava, an education lawyer at the NAACP Legal Defense Fund.

In addition, there’s a tacit liberal constitutional agenda at work in hybrid class-race approaches to integration: better to test Kennedy’s opinion, with its support for the drawing of “race conscious” school boundaries, than to retreat further than is in fact required. “For Kennedy, there are ways of taking race into account,” John Powell says. “It’s just the method that’s in question. How do you do it? We need to find out what’s still permitted.” He also points out that African-Americans are more likely than whites to be poor over generations — a bigger hurdle than a short stint in a low-income bracket.

The continuing attention to race aligns with the internal politics of Louisville and its suburbs. Many of today’s parents grew up there and tend to remember and care about overcoming their county’s Jim Crow legacy. In 1975, when a federal judge first ordered the city and its suburbs to desegregate, the Ku Klux Klan demonstrated, and the next day about 150 white protestors attacked eight school buses filled with black students. “We had tough times here when the buses burned,” says Ann Elmore, a black member of the Jefferson County School Board. “We can still include race as a factor in our plan, and let me say I think it’s important that we do.”

Elsewhere in the United States, it is too soon to tell how the politics of class-based integration (Wake County) or class-plus-race (Jefferson County) will play out. Richard Kahlenberg makes the case for shifting integration policies primarily or solely to being class-based over the next decade or two. What’s fair, he asks, about giving a spot in a coveted magnet program to the son of a South Asian college professor or an African-American politician over the daughter of a white waitress? Over time, such injustices threaten to sour white parents on the whole diversity enterprise, whereas giving poor kids a boost, whatever their color, is far less controversial. Polls at the time of the Supreme Court’s 2003 decision in Grutter v. Bollinger, which concerned affirmative action at public universities, showed public support running 2 to 1 for giving poorer kids a leg up in going to college, as opposed to 2 to 1 against race-based preferences. In her majority opinion in the case, Justice Sandra Day O’Connor famously said she thought that racial preferences would continue only for another 25 years. Barack Obama has said, looking ahead to his daughters’ college applications, that they don’t deserve an admissions break — an acknowledgment that the mix of race, affirmative action and privilege is a complicated one.

To catch on nationwide, however, class-based integration would have to generate momentum that it has so far lacked. In his State of the Union address in January, President Bush urged action “to help liberate poor children trapped in failing public schools.” And yet a provision in the No Child Left Behind Act that theoretically allows students to transfer depends on the availability of open spaces elsewhere and has barely been utilized. The administration may have advocated class-based integration to the Supreme Court, but Bush officials haven’t used their signature education law to make it happen.

If Congress were to revise No Child Left Behind to encourage more transfers of poor students to middle-class schools, would poor students drag down their better-off peers? In the end, the prospects of class-based integration will probably rise or fall on the answer to this question. Socioeconomic integration may be good for the have-nots, but if the haves think their kids are paying too great a price, they will kill it off at the polls. Richard Kahlenberg argues that the key is to ensure there is a solidly middle-class majority at as many schools as possible. That majority will then set the tone, he argues. Kahlenberg says that more research is needed to pin down the percentage of middle-class kids that a school needs to have to serve all its students well. Maybe a school can go as high as 50 percent low-income without losing ground. Or maybe it’s telling that in Wake County, a proposal to increase the ceiling for low-income students from 40 percent to 50 percent died a swift death last fall after concerted protest.

Whatever the exact answer, there is some support for the view that schools can handle a substantial fraction of poor students without sacrificing performance. In Wake County, test scores of middle-class students have risen since instituting income-based integration. Additionally, Kahlenberg points out that middle-class students are generally less influenced by a school’s environment because they tend to learn more at home, and that the achievement of white students has not declined in specific schools that experienced racial (and thus some class) desegregation.

Would schools need to track students by ability to protect middle-class students, who are more often higher-achieving than their low-income peers? Perhaps not. In a 2006 longitudinal study of an accelerated middle-school math program in Nassau County, N.Y., which grouped students heterogeneously, the authors found that students at all achievement levels, as well as minority and low-income students, were more likely than the students in tracked classes to take advanced math in high school. In addition, the kids who came into the program as math whizzes performed as well as other top-achievers in homogenous classes.

This study underscores Ronald Ferguson’s point about the value of seating students of different backgrounds and abilities in class together, as opposed to tracking them. Still, it’s worth noting that less than 15 percent of the students studied in Nassau County were low-income. So the math study doesn’t tell us what happens to the high-achieving middle-class kids when close to half of their classmates aren’t as well off.

At the end of February, Todd started showing the map of mermaid-shaped Geographic Area A, which she hoped to use to implement the new assignment system, to the parents of Jefferson County. Todd would start her presentation with quotes from Justice Kennedy and from Justice Breyer’s dissent; she especially wanted to remind her audiences of the sentiment Breyer expressed by quoting former Justice Thurgood Marshall: “Unless our children begin to learn together, there is little hope that our people will ever learn to live together.”

Todd’s first stop was at a forum sponsored jointly by the Urban League and the N.A.A.C.P., groups associated with Louisville’s black establishment. Most of their members supported the school district, but some clergy members who worked with the city’s black youth spoke against it. The Rev. John Carter, associate minister at Green Street Baptist Church, pointed to the district’s black-white achievement gap and called for a return to neighborhood schools and an earlier era of black self-reliance.

As more forums followed in high-school auditoriums across the county, white parents asked a different question: How would the new assignment plan affect their kids? Would they be forced to switch schools in second, third or fourth grade? “We like the diversity,” a white parent named Niki Noe told me the next morning at her son’s elementary school, St. Matthews. “But if we have to go to Chenoweth” — a school with lower test scores — “we’ll pull out and go to private school.”

That’s a serious threat to the district’s well-being, but one that Todd anticipated. She designed a grandfather clause for kids like Noe’s, so that the new assignments would apply almost entirely to new students. Meanwhile, at every meeting, Todd polled parents on whether they cared about maintaining diverse schools. The University of Kentucky also conducted a telephone survey with 654 parents of elementary schoolers. In April, Todd called me, elated and relieved, with the results: 88 percent of parents supported enrollment guidelines “to ensure that students learn with students from different races and backgrounds.” Todd said she had dropped Breyer’s dissent in Meredith from her presentation; she was no longer feeling frustrated with the court. “It’s been a personal emotional trek, but I think we’ve come out better for it,” she said in May.

Carter, the proponent of black-self reliance, was feeling more at ease, too. He had come to see the virtue of mixing kids by income level. “Once I did the research, I was pretty impressed by the economic part of it,” he said. Carter had taken note of the district’s data showing that a switch to neighborhood schools, as he had first advocated, would mean that median household income would range from a high of more than $100,000 at the wealthiest school to about $8,300 at the poorest. A split between rich students and poor schools, he agreed, was the wrong path.

It is, of course, the path taken by most of the country. And yet at the end of May, the Jefferson County School Board voted unanimously to make Geographic Area A the basis for integrating elementary schools for the 2009 school year, a new chapter in the district’s history. As the schools shift to the new class-plus-race formula, the district will closely watch the test scores of black students and poor students, hoping for an upsurge, and those of middle-class students, hoping to see achievement hold steady. And if they do, maybe the court’s decision in Meredith will come to seem less like a cause for regret and more like an unexpected opportunity.

Emily Bazelon is a senior editor at Slate who writes frequently about legal issues. Her last article for the magazine was about autistic girls.


Copyright 2008 The New York Times Company