School Segregation Redux

W hile public schools were continuously desegregated from the 1950s to the 1980s, the past 12 years has seen a rapid retreat from these efforts as federal courts terminated major and successful desegregation orders. In the 1990s, U.S. Supreme Court rulings in cases such as Board of Education of Oklahoma City v. Dowell and Freeman v. Pitts made it easier for school districts to be declared "unified" or desegregated. In the last 7 years, in the wake of these decisions, nearly 50 districts across the country have had their court-ordered desegregation plans abolished.

A study released by the Harvard Civil Rights Project in January illustrates how federal court rulings have contributed to the resegregation of public schools across the nation. "A Multiracial Society with Segregated Schools: Are We Losing the Dream?" examines research on the impact of desegregation and describes patterns of racial enrollment and segregation in U.S. public schools at the national, regional, state, and district levels based on the latest data from the U.S. Department of Education's National Center for Educational Statistics (the report is available at www. civilrightsproject.havard.edu).

Common myths about school desegregation—such as it was a good idea that didn't work, that it increased "white flight," or didn't solve education educational problems—are not supported by the enormous amount of research on the effects of desegregation. The report's authors—Erica Frankenberg, Chungmei Lee, and Gary Orfield—summarize the research on desegregation into three general findings:

• Segregated schools have much higher concentrations of poverty and other problems and much lower average test scores, levels of teacher qualifications, and advanced courses. With few exceptions, separate schools are still unequal schools. Ending desegregation tends to produce a rapid increase of such schools within a district and more qualified teachers tend to leave these segregated schools.
• In systems with desegregation plans, particularly those areas with substantial white enrollment, minority students tend to transfer to better schools and learn more, though a racial "achievement" gap remains. Going to desegregated schools improves students' chances for a desegregated future life, for going to and succeeding in college, and living and working in interracial settings.
• When teachers create positive academic interactions in racially diverse schools, the benefits of desegregated schools increase substantially.

In addition, the author's point to more recent research that shows educational and civic benefits of desegregation for all racial groups. For example, in Louisville-Jefferson County, Kentucky—the largest urban area in what the report claims is the nation's most integrated state—both black and white students report very positive results on a range of questions on educational and social outcomes. Ninety-three of white juniors and ninety-five percent of black African Americans said they are comfortable working with students of other races on group projects. Even higher percentages of white and black students said they were comfortable in classes learning about each others' cultures.

Despite the educational and social successes of desegregation, federal court rulings combined with the failure of the federal government to fund desegregation assistance programs for over two decades have created conditions for, indeed encouraged, the resegregation of public schools.

The Civil Rights Project report highlights the rapid racial transformation of U.S. schools. Since 1968, black student enrollment has increased nearly 30 percent and Latino student enrollment is up 283 percent. In contrast, public school enrollment of whites is down 17 percent. In every region of the country the school population has become less white and schools in the South and West have the highest concentrations of black and Latino students (and these regions are approaching student populations where whites are in the minority). There are now six states where white students are a minority of the enrolled school population: California, Hawai'i, Louisiana, Mississippi, New Mexico, and Texas. Schools in the Northeast and Midwest still have large white majorities.

The Harvard study reports that, on average, white, black, and Latino students all attend schools in which the majority of the student body is composed of students of their own race. Whites are now the most segregated group in public schools—attending schools that on average are 80 percent white. In contrast, the average Asian student attends the most integrated schools (although Asian students still attend schools that are on average 22 percent Asian). Native American students attend schools, on average, in which half the student body is white and slightly less than one-third of students are Native American. Native American students have the lowest exposure to black students among all racial groups.

White students are attending majority white schools at time when minority students make up nearly one-half of the public school enrollment. During the 1990s, the proportion of black students in majority white schools decreased by 13 percent—a level lower than any year since 1968.

There are only two states that have not shown an increase in black segregation in recent years and these states—Michigan and New Jersey—are highly segregated and showed virtually no change. States with large increases in segregation (such as Florida, Missouri, and North Carolina) are home to school districts that had long-running desegregation orders terminated in the 1990s.

Over the past two decades in Kentucky, there has been nearly a 10 percent decrease in the percentage of white students in schools attended by blacks. Despite this decrease in integration, the Harvard report notes Kentucky has had the highest level of black-white exposure in schools since 1980. This is largely the result of consolidation of city and county school systems in metro Louisville, which remain under a desegregation plan.

The Harvard study also identifies the importance of the relationship between racial segregation and poverty. High poverty schools have been shown to increase educational inequality for students because of a lack of resources and qualified teachers as well as low parental involvement and high teacher turnover rates. (There are nearly 200,000 noncertified teachers now, mostly in schools serving poor, minority, and immigrant children.) Almost half of the students in schools attended by the average black or Latino student are poor or very poor, while less than 20 percent of students in schools attended by the average white student is classified as poor. A substantial number of public schools that are virtually all non-white, what the study's authors label "apartheid" schools, have emerged in recent years. These schools educate a quarter of the students in the Midwest and Northeast and are often schools plagued by substantial poverty, social, and health problems.

Teaching For Whites Only?

I n addition to the racial segregation of students, there is a serious race gap between teachers (86 percent of whom are white) and the nearly 50 percent of students who are minorities.

Courts are largely responsible for the resegregation of students, but state and federal legislation has become a serious barrier to increasing diversity of the teachers in public schools, compounding the deleterious effects of resegregated schools. This legislation, in particular the No Child Left Behind Act, relies on standardized tests to improve education and teacher quality.

There is overwhelming evidence that standardized tests are primarily measures of race and class, rather than educational achievement of public school students. These findings are consistent with what we know about college-admissions and teacher licensure tests, which contribute to educational inequality by denying education, scholarships, and access to the teaching profession to minority students, thereby sustaining the race gap between teachers and students in schools.

ACT college admissions test scores, for example, are directly related to family income (the richer the students' parents are, the higher the average scores across income groups) and race (whites outscore all groups when factors such as course work, grades, and family income are equal). The ACT also does a poor job of predicting the college performance of minority students—explaining only 7 percent of the difference in first-semester college grades of black students. Despite its inaccuracies and biases, ACT scores are often used to determine entrance into colleges and for allocation of scholarships. The SAT, which is a direct descendent of the racist anti-immigrant Army Mental Tests of the 1920s, is also plagued by biases that are effective in eliminating promising low-income and minority students from college classrooms.

ACT or SAT test scores above a specified level are required for admission to most teacher education programs. As a result of biases in both these tests large numbers of potential minority teachers are being excluded from opportunities to become classroom teachers. A detailed study of the impact of standardized tests on the teacher candidate pool in Florida indicated that test score requirements eliminated 80 percent of black and 61 percent of Latino applicants to teacher education programs, but only 37 percent of whites.

There is also a long history of cultural bias on teacher licensure tests, which are typically taken upon exit from teacher education programs. A recent National Research Council report on teacher tests concludes that raising cut-off scores on these tests will reduce racial diversity in the teaching profession without improving quality. The differences in average scores among racial/ethnic groups on teacher licensure tests are similar to the differences found among these groups on college admission tests, showing substantial disparities between the passing rates of white and minority test takers.

Most importantly, the NRC found that these tests do not predict who will become effective teachers. The NRC concluded that by their design and as currently used tests like the PRAXIS—the most widely- used teacher licensure test—fall short in their use as accountability tools, as levers for improving teacher preparation, and encourage erroneous conclusions about the quality of teacher preparation. Still, over 40 states rely on standardized tests for teacher licensure.

Efforts to improve learning and teacher quality rest on a misguided use of standardized tests. Rather than improving learning or increasing teacher quality, the latest research indicates that an emphasis on testing results actually lowers student academic performance, increases dropout rates, and serves as a barrier to diversifying the teaching profession with improving teacher quality. A recent study by Arizona State University researchers showed that in states that have adopted high-stakes exams there has been a decline in student performance on independent measures of achievement, such as the National Assessment of Educational Progress (aka "The Nation's Report Card").

What is To Be Done?

A s the authors of the Harvard study note, segregation is a failed educational policy that produces deeply unequal education and a polarized society. So too is test-driven educational reform. Clearly the struggle for civil rights continues and desegregated schools are an important achievement that must be preserved, but school desegregation is not a panacea.

Frankenberg and her colleagues at the Civil Rights Project offer a basic policy framework that they say is needed to increase integration in U.S. public schools. The framework includes principles such as: (1) explicit recognition of integrated education as a basic education goal and judicial recognition that integrated education is a compelling educational interest in our society; (2) a resistance to terminating desegregation plans; and (3) in cases where schools districts are forbidden to continue its desegregation plan by a federal court, that consideration should be given to efforts to keep diversity by social and economic desegregation.

There is a mountain of evidence documenting the deleterious effects of high-stakes tests on teaching, learning, and society. Many of the backers of these tests are aware of the problems and nonetheless remain committed to their use as a tool to regulate knowledge in schools and universities; to sort students by race and class; and limit access of minorities to the teaching profession. Increasing numbers of students, parents, and educators are pushing back against educational "reform" efforts that divide students and teachers along racial, ethnic, and class lines. The Rouge Forum (www.rougeforum.org), the Whole Schooling Consortium (www.coe.wayne.Edu/Com- munityBuilding/WSC.html), and the Coalition for Commonsense in Education (www.free.freespeech. org/ccse) are three examples of grassroots groups working for more inclusive schools and classrooms; organizing across the barriers of race, class, ability; and acknowledging that schools remain a pivotal, if not the most important, battleground of political and economic interests in the U.S. today.

E. Wayne Ross is a distinguished University Scholar at the University of Louisville and co-editor of Workplace: The Journal for Academic Labor and Cultural Logic.