Econ Summaries

EDUCATION AND URBAN SOCIETY / November 2000Goldring, Smrekar / PURSUIT OF RACIAL BALANCE

MAGNET SCHOOLS AND THE PURSUIT OF RACIAL BALANCE

ELLEN GOLDRING CLAIRE SMREKAR Vanderbilt University

Magnet schools are being established in more and more school systems in an attempt to promote racial diversity, improve scholastic standards, and pro- vide a range of programs to satisfy individual talents and interests. Magnet schools, sometimes referred to as “alternative schools” or “schools of choice,” are public schools that provide incentives to parents and students through specialized curricular themes or instructional methods. The term magnet gained popularity in the 1970s when policy makers were designing desegregation plans in an effort to make them more attractive to parents, edu- cators, and students. Magnet schools provide school districts with an alterna- tive to mandatory reassignment and forced busing by providing a choice for parents among several school options with each offering a different set of dis- tinctive course offerings or instructional formats. In magnet schools, enroll- ments are often managed to ensure a racially balanced student population.

Since 1975, when federal courts accepted magnet schools as a method of desegregation (see Morgan v. Kerrigan, 1976), their number has increased dramatically. Indeed, between 1982 and 1991, the number of individual schools offering magnet programs nearly doubled to more than 2,400, and the number of students enrolled in these programs almost tripled. By the 1991-1992 school year, more than 1.2 million students were enrolled in mag- net schools in 230 school districts (Yu & Taylor, 1997).

Magnet schools are typically established in urban school districts with large student enrollments (i.e., more than 10,000). According to the U.S. Department of Education, 53% of large urban school districts include magnet

AUTHORS’ NOTE: This paper is adapted from our recently published book School Choice in Urban America (1999, Teacher’s College Press) and from “Social Class Isolation and Racial Diversity in Magnet Schools,” a paper presented at a conference sponsored by The National Center for the Student of Privatization in Education, Teachers College, Columbia University, May 2000.

EDUCATION AND URBAN SOCIETY, Vol. 33 No. 1, November 2000 17-35 © 2000 Corwin Press, Inc.

17

school programs as part of their desegregation plans compared with only 10% of suburban districts. More than half of all magnet programs are located in low socioeconomic districts (Levine, 1997). Although they can involve all grade levels, more than half of the nation’s magnet programs serve elemen- tary school students; only 20% of magnets are located at the high school level (Levine, 1997; Yu & Taylor, 1997). The most common type of magnet school is one that emphasizes a particular subject matter, such as math and science, computers and technology, or a foreign language. Other programs offer a unique instructional approach, such as Montessori or Paideia.

Magnet school programs are extremely popular, as measured by the fact that more than 75% of all districts with magnet schools have a greater demand for student slots than they can fill. Half of these districts maintain long waiting lists (Blank, Levine, & Steel, 1996). With this level of demand, and with a total of more than 15% of all students in magnet districts already enrolled in magnet schools, school districts must limit entry into the special- ized programs. Most accomplish this through an admissions process that uses a lottery format, and others rely on a first-come, first-served arrangement. Only about one third of all magnet programs use a selective admissions pol- icy, and these usually involve either a minimum test score requirement or, as in a performing arts magnet, admission may be based on performance in an audition.

In many instances, districts have supported magnet schools with a consid- erable investment of resources. On average, expenditures per student are 10% higher in districts with magnets. Almost three fourths of magnet programs have additional staffing allowances as well. Some magnet programs are funded through state desegregation funds, but most are funded under 2-year grants through the federal Magnet Schools Assistance Program (MSAP). These funds are made available to districts that are either implementing mag- net schools voluntarily or that are acting under court-ordered desegregation. The MSAP serves a critical role in magnet school creation and expansion efforts nationwide. The program provides about $110 million annually to support magnet school programs. Between 1985 and 1993, about $750 mil- lion in MSAP funds were allocated to 117 different school districts (Steel & Eaton, 1996).

This article explores magnet schools and racial diversity. We begin with a review of the research on magnet schools in terms of understanding the value and impact of magnet schools as a tool for reducing racial segregation. The analyses differentiate between within-school and districtwide outcomes. How effective are magnets in reducing racial isolation? What accounts for these differential effects?

18 EDUCATION AND URBAN SOCIETY / November 2000

We follow this macro-level analysis with findings from our 3-year study of magnets in two major urban school districts: St. Louis and Cincinnati. We conclude this article with troubling indications that the post-busing era of desegregation and litigation signals a heavy reliance on magnet schools and parental choice without the commitment to diversity goals that marked ear- lier decades of social and educational reform.

MAGNET SCHOOLS AND DESEGREGATION

There are numerous evaluations of local school magnet plans that suggest a very complex set of conclusions regarding the utility of magnet schools in achieving racial desegregation. This is expected; districts vary largely in terms of the nature of their magnet school plans (such as types and numbers of options), transportation availability, and overall district enrollment pat- terns. Obviously, districts with larger proportions of minority students will find it harder to achieve racial desegregation irrespective of the type of deseg- regation options in place. Interestingly, these are precisely the districts most likely to have magnets as central components of their desegregation plans. Specifically, 78% of students in districts with magnets are in large urban dis- tricts, 66% of students are in high-minority districts, and 47% of students are in low-income districts (Steel & Levine, 1994). It follows that, as Plank, Schiller, Schneider, and Coleman (1992) report, choosing a magnet school for Black middle school students is likely to racially segregate them from Whites, whereas Whites tend to integrate magnet schools. The effects of magnet schools will be very sensitive to the specific arrangements under which these schools operate.

An evaluation of the Charlotte-Mecklenburg magnet program in 1996 seemed to point to great success. After 3 years of operation, the magnet schools at all levels had racially mixed student bodies ranging from about 50% White to 44% Black students in elementary schools, and 55% White and 38% Black students in the magnet high schools. Of the newly admitted stu- dents to magnet schools, 40.4% were Black. Nonmagnet schools also remained racially balanced (Charlotte-Mecklenburg Schools, 1994). Other positive results emerged from evaluations of Montclair, New Jersey and New York State. Blank (1996) reviews these positive evaluation reports from Montclair, New Jersey and New York State to “attribute significant and sus- tained improvement in desegregation to the implementation of magnet schools” (p. 35).

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 19

In contrast, in Kansas City, a far larger and more complex environment, magnet schools did not have the intended effect on dramatically changing racial balance patterns. Minority enrollments from 1985 to 1993 remained a steady 73% to 75% of all students, and magnet schools did not meet their desegregation goals of 40/60 (Morrison, 1992). The district was largely unsuccessful in attracting nonminority students into its schools. Another example of an unsuccessful magnet school program is Prince George’s County, Maryland. Although a majority of students (72%) attended schools within new racial balance guidelines, these guidelines were so variable, being defined by the racial composition of the school system, that students often went to minority-isolated schools (Eaton & Crutcher, 1996).

Beyond individual district and local site evaluations, one of the major sources of information about the effect of magnet schools on school desegre- gation efforts is an evaluation of the MSAP (Steel & Eaton, 1996). The MSAP provides federal dollars to “support the elimination, reduction, or pre- vention of minority isolation in elementary and secondary schools with sub- stantial portions of minority group students” (Steel & Eaton, 1996, p. i) . The data for the evaluation were collected from 119 districts and 1,043 magnet schools following grantees from 1989 until 1991. Data regarding desegrega- tion in magnet schools were available from 615 schools (58% of the total school and 93% of the districts).

Schools participating in MSAP declared one of three possible desegrega- tion goals for their magnet schools. The first goal, to reduce isolation, per- tains to minority-isolated schools (i.e., those with more than 50% minority enrollments): “To reduce minority isolation is to reduce the percentage of minority students in a minority-isolated school” (Steel & Eaton, 1996, p. 20 ). To eliminate minority isolation, a second goal is to reduce a school’s percent- age of minority students to less than 50% of enrollment, whereas the third goal, to prevent minority enrollment, is to keep a school’s minority enroll- ment from rising to more than 50% of enrollment.

Table 1 presents the number of schools that specified and obtained each desegregation goal from the sample of 615 schools. The majority (85%) of schools (n = 529) stated a desegregation goal of reducing minority isolation; that is, they wanted to either reduce their absolute numbers of minority enrollments or prevent the minority enrollment at the particular magnet school from rising faster than the average districtwide rate. Less than half (44% of magnet schools) were successful in meeting this objective. On aver- age, of those schools meeting this goal, minority enrollment was decreased by 1.5%, with a range from 55% to .1%. Even less successful were schools that targeted the elimination of minority isolation. Although only a relatively small number of schools (2%; n = 12) attempted to eliminate minority

20 EDUCATION AND URBAN SOCIETY / November 2000

isolation, only 4 schools were successful in meeting this goal. The third and most successful desegregation goal met by magnet schools was the preven- tion of minority isolation. Although relatively few magnet schools slated this goal (13%; n = 182), of those schools, 72% (n = 60) were successful in pre- venting minority enrollment from rising to more than 50%. On average, minority enrollment was at 39% for schools successful in meeting this objec- tive. It should be noted that these results were relatively similar with one exception, regardless of whether the desegregation plan involving magnet schools was mandatory (court-ordered) or voluntary. Elimination of minority isolation was achieved only in mandatory desegregation plans. (A new fed- eral study of MSAP grantees is currently underway.)

Further analysis of the MSAP data reveals the importance of understand- ing the particular district context of any magnet school plan for evaluating the effectiveness of magnet schools in achieving racial desegregation. The MSAP data indicate that magnet school desegregation is greatly dependent on the type of magnet program, the overall minority enrollment in the district, changes and trends in minority school enrollments, and the type of desegre- gation plan. Specifically, Steel and Eaton (1996) report that those individual magnet schools with high proportions of minority students relative to their districts were more likely to meet desegregation goals. In contrast, magnet schools that had “higher overall proportions of minority students initially and/or were experiencing higher rates of growth in minority enrollment lev- els were less likely to meet their objectives” (Steel & Eaton, 1996, p. 43). In addition, “dedicated” or whole-school magnet programs (in which all stu- dents must choose the school) were the most successful in meeting desegre- gation goals when compared with programs in schools or mixed models of magnets and attendance zone magnets.

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 21

Table 1

Specified and Obtained Desegregation Goals From Magnet School Assistance Program

Specified Obtained

Percentage Number Percentage Number

Reduce isolation 85 529 44 228 Eliminate isolation 2 12 33 4 Prevent isolation 13 82 73 60 Total 100 615 47 292

SOURCE: Steel and Eaton (1996).

How can we interpret these results? Do these findings suggest success, failure, or mere uncertainty? To fully assess the effect of magnet schools on school desegregation, we argue that it is crucial to understand the local con- text of any given magnet school plan. A crucial question that must be consid- ered is “Compared with what?” In other words, what else is happening in the school district? Are other schools becoming racially isolated? Is overall minority enrollment in the district rising? We will turn to these questions by providing an in-depth view of two large urban district magnet school plans.

METHOD

Schools were chosen to participate in the study based on the following cri- teria: (a) the participating school included a fourth grade and a fifth grade and (b) the fourth grade was not the entry grade. These criteria were selected to increase the likelihood that each school would have a relatively substantial population of fifth-grade students who had been enrolled in the school for more than 1 year prior to the fall of 1993 (or the 1993-1994 school year) and whose parents or guardians would therefore be relatively familiar with the school.

To reduce possible response bias, the initial sample frame was further screened based on information provided by the central office, and schools were eliminated based on the following additional criteria:

• Fourth- and fifth-grade classes assigned to the school were not actually attend- ing that school in the 1992-1993 or 1993-1994 school year for any reason, such as redistricting or renovation projects.

• The school was a receiver of students reassigned for the same kinds of reasons, such as renovation or closure of their zoned school.

• The school added or dropped a program within a 2-year period prior to the 1993-1994 school year, resulting in a substantial change in the composition of the student body.

Out of 54 schools in the Cincinnati sample frame, 20 were selected for inclusion in the final study sample: 10 magnet schools and 10 nonmagnet schools. In addition, magnets that were not full (or dedicated) were elimi- nated from the sample. (That is, schools-within-schools were excluded, as were magnets composed of a mixture of zoned and choice students.) As a result, 17 schools were eliminated. Next, of the 15 remaining magnet schools, 5 were eliminated on the basis of information provided by the central

22 EDUCATION AND URBAN SOCIETY / November 2000

office (during the summer of 1993) that raised the possibility of significant response bias at these schools, such as major programmatic changes. This announcement generated significant negative parental reaction to the pro- posed changes at these schools. Thus, these 5 schools were ruled out of the final sample, leaving 10 magnets in the sample. Also, 1 of the 10 remaining magnet schools in the sample dropped out of the study during the school year. Thus, the final magnet sample contained 9 schools, including 2 Montessori magnets, 2 Paideia magnets, 3 schools with a foreign language theme, 1 fun- damental academy (emphasizing traditional curricular themes and instruc- tional approaches), and 1 school having a mathematics and science curricular emphasis.

Twenty-two nonmagnet schools were included in the Cincinnati sample frame. Of these, 10 were selected for the final study sample by pair-matching them with the 10 selected magnet schools on the basis of the racial composi- tion of the student body (using African American percentage).

In St. Louis, the initial sample frame included 66 schools. Five were excluded because fourth and fifth graders were not actually in attendance, and 4 were excluded because they received large numbers of reassigned stu- dents, leaving an adjusted sample frame of 57 schools. Of these, 26 were selected for inclusion in the study.

All 10 elementary magnet schools in the St. Louis sample frame were selected. Of the 36 nonintegrated schools, 8 were randomly selected for inclusion in the study. Of the 11 integrated schools in the sample frame, 10 were initially selected by pair-matching them on racial balance (using total African American percentage) with the 10 St. Louis magnet schools in the study sample. The principal of one of the schools selected declined to partici- pate, citing the excessive paperwork that would be involved with both this project and the school’s selection for participation in a mandatory statewide assessment program that was about to begin. The one remaining integrated nonmagnet school was then chosen to make up the sample. However, after the first series of meetings with principals in early September 1993, 2 more schools had to be dropped from the sample. Therefore, 8 integrated non- magnet schools remained in the final study sample.

DATA COLLECTION

An anonymous questionnaire was distributed to all fifth-grade parents and to all nonadministrative certified staff in each school in the sample. Teacher questionnaires were distributed either in their school mailboxes or during a faculty meeting.

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 23

The response rate in Cincinnati was 62.1% (n = 730) for the parent ques- tionnaire and 67.6% (n = 417) for the teacher questionnaire. The percentages of responses from African American and White parents from nonmagnet and magnet schools were equal. The response rate in St. Louis was 67.4% (n = 953) for parents and 70.6% (n = 553) for teachers.

QUALITATIVE MULTIPLE-CASE STUDIES

This article also includes qualitative case studies of four magnet schools (one Paideia and one math-science magnet in St. Louis and two Basic Acad- emy magnet schools in Cincinnati) that focus on the context of school choice, the nature of school communities, and patterns of family-school interactions. Semistructured interviews were conducted with the principal, counselor, and teachers (including a cross-section from both lower and upper primary lev- els) at each of the four sites. Interviews were also conducted with 12 to 14 sets of parents from each of three schools (two in Cincinnati and one in St. Louis). Parents were selected randomly from a stratified sample across race (two cat- egories: African American and White) and social class (as indicated by eligi- bility for the federal free lunch program). School records and parent data cards provided demographic information indicating parents’ race, occupa- tion, and city address. This information was used to select a sample of parents consistent with the socioeconomic and racial composition of the total popula- tion of school families.

Interviews with school staff were conducted at the school site; parents were interviewed in their homes. The interview sessions lasted an average of 90 minutes. All interviews were audiotaped, with participants’ permission, and transcribed verbatim. In addition to interviews, an array of school docu- ments (including brochures, enrollment applications, letters, newsletters, handbooks, and meeting minutes) was collected and analyzed.

CASE STUDY: RACIAL BALANCE IN URBAN MAGNET SCHOOLS

This section examines magnet school programs in two major school dis- tricts—Cincinnati and St. Louis—to understand the effect of well-estab- lished, robust magnet school programs on racial balance in elementary schools located in large urban districts. We relied on extensive surveys with parents and teachers and intensive case studies of four magnet schools. First, we will describe the two urban school districts in our study.

24 EDUCATION AND URBAN SOCIETY / November 2000

CINCINNATI

During the 1993-1994 school year, the Cincinnati Public School District operated 61 elementary schools, 8 junior high or middle schools, 10 second- ary schools, and 7 special schools. Magnet (or what the Cincinnati system calls alternative) program choices were offered to students at all grade levels (K-12).

In the Cincinnati system, magnet programs are differentiated by curricu- lum or special interest areas as well as by instructional approach (e.g., Mon- tessori, Paideia). Magnets in the alternative program are also differentiated by enrollment structure and program coverage. The Cincinnati system uses four types of structures. First, “full” or dedicated magnets enroll students strictly on the basis of a formal application and admissions process (described below) and provide alternative instruction to all students enrolled at the school site. In the second type, mixed magnets provide alternative instruction to all students enrolled at the school but enroll a combination of neighborhood and zoned students (because a percentage of the enrollment is reserved for zoned students) and students who have formally applied to the school but live outside the school’s attendance zone (citywide application zone). In the third type of structure, schools-within-schools are programmat- ically distinct components of a neighborhood school and provide alternative instruction only to those students who are enrolled in the magnet component based on their selection through the district’s alternative school application process. Finally, with the fourth structure, mixed schools-within-schools are a special version of schools-within-schools. They are organized in an exist- ing neighborhood school and reserve a percentage of their enrollment capac- ity for zoned children in addition to children living outside the attendance area. (Our study focused solely on the dedicated/full magnet schools).

At the time of our study, acceptance into magnet programs was based pri- marily on the application date (first-come, first-served) and race. Transporta- tion is provided for students in Grades K through 8 who live more than a mile from their alternative school. Transportation is provided for all students in Grades 9 through 12.

Total district enrollment in 1993-1994 was approximately 51,000 stu- dents (66% African American, 32% White, 2% other). The system enrolled 46% of its students in magnet programs in the 1993-1994 school year. Of those enrolled in magnet schools, 61.7% were African American. More than 43% of the district’s African American students were enrolled in magnet pro- grams in the 1993-1994 school year.

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 25

ST. LOUIS

The St. Louis Public School District operates a total of 104 schools, including 73 elementary schools, 21 middle schools, 10 high schools, and 7 special schools. The St. Louis City District has 26 full-time and 2 part-time magnet programs in the city.

The district operates the following three different types of schools under the terms of its desegregation plan: (a) magnet schools, (b) nonintegrated nonmagnet schools that are 98% African American and located in predomi- nantly African American neighborhoods, and (c) integrated nonmagnet schools in or near “naturally integrated” or transitional neighborhoods achieved by busing.1

Total enrollment in St. Louis City Schools is approximately 36,091, of whom 78% are African American. Any student who lives in St. Louis City (and White students who live in the 16 participating suburban county districts that are part of the voluntary interdistrict transfer program) may apply to magnet schools. Assignments to magnets are made on the basis of a general lottery held in the spring. In 1993-1994, the district enrolled 10,087 students in city magnets: African American students comprised 58% of total magnet enrollment, and Whites comprised 42%. Approximately 15% of the city’s African American students are enrolled in city magnets, whereas 40% of the city’s White students attend city magnets.

Beyond the districtwide data, what do magnets look like in Cincinnati and St. Louis? The following are brief descriptive sketches of our case study sites.

Greenwood Paideia (Cincinnati). Greenwood Paideia enrolls 378 stu- dents in Grades K through 6 and is located near an industrial park in a racially mixed, middle-class section of the city about 20 minutes from downtown Cincinnati. Approximately 95% of the students are bused to Greenwood from neighborhoods across the city. The student population is 52% African American and 48% White. Of the students at Greenwood, 45% qualify for free lunch.

Mathematics and Science Academy of Cincinnati (MaSAC). MaSAC enrolls 575 students in Grades K through 6 and is located in a working-class, predominantly White neighborhood on the western edge of the city. Approxi- mately 83% of the students are bused to MaSAC from areas across the city. The school population is 51% African American and 49% White. Of the stu- dents at MaSAC, 70% qualify for free lunch.

26 EDUCATION AND URBAN SOCIETY / November 2000

Overbrook Basic Academy (St. Louis). Student enrollment at Overbrook rests at 253 students and includes preschool through Grade 5. The student population is 60% African American and 40% White. More than 90% of the students who attend Overbrook are bused in from different neighborhoods across the city and county. Of the students at Overbrook, 64% quality for free lunch.

Viking Basic Academy (St. Louis). Viking includes Grades K through 5 and has a student enrollment of 298. The student population is 51% African American, 45% White, and 4% other. More than 90% of the students are bused from various neighborhoods across the city and county. Of the students at Viking, 68% qualify for free lunch.

MAGNET SCHOOLS AS A TOOL FOR RACIAL BALANCE

Without compromise, St. Louis and Cincinnati use magnet schools to effectively create racially balanced schools in their respective school dis- tricts. On average, African Americans comprise about 60% of magnet school enrollments in St. Louis and about 50% in Cincinnati, with White students making up the difference (see Table 2).

The racial balance of the schools in St. Louis that participated in our study is consistent with the court decree. The racial balance of the 10 magnet schools in our study ranged from 62% to 51% African American students, whereas the integrated nonmagnet schools ranged from 88% to 26% African American students. The racial balance of the Cincinnati magnet schools that participated in our study ranged from 57% to 46% African American, whereas

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 27

Table 2

Enrollment of African American Students in Different Types of St. Louis Public Schools

Percentage Number of Students

Magnet schools 15.0 6,646 Interdistrict transfer program 28.5 12,593 Integrated neighborhood schools 15.9 7,009 Nonintegrated neighborhood schools 40.6 17,915 Total 100.0 44,163

NOTE: Information taken from the Report From the Civic Progress Task Force on Desegregation of the St. Louis Public School System, Part I, December 1996.

the nonmagnet schools ranged from a high of 85% African American to a low of 30% African American.

It is important to note that in St. Louis, where African American students comprise 78% of the school-age population, only 15% of the city’s African American students are enrolled in city magnets. The large proportion of Afri- can American students in the city means that a significant and disproportion- ate number of African American students who apply for admission to a mag- net school are placed on a waiting list. A larger percentage of African American students in St. Louis choose suburban schools over magnets through the interdistrict transfer plan (see Wells & Crain, 1997).

In summary, although the issue of racial desegregation tends to be muted by more vocal claims among educators regarding magnet program stability, excellence, and instructional innovation, the evidence clearly indicates that the court-ordered desegregation guidelines from which these magnet school programs originated have been efficiently and explicitly addressed (Taylor & Yu, 1999). Are these efforts to maintain racial balance in urban school set- tings meaningful to the parents and students involved in these reform efforts?

THE PERCEIVED VALUE OF RACIALLY INTEGRATED SCHOOLS

Magnet school teachers in both cities described with insight and emotion the compelling rationale behind their school charter. Although some regard authentic racial understanding and integration “beyond us,” belonging more appropriately to the values modeled in family life and in community arrange- ments, the teachers expressed widespread regard for the racial balance repre- sented in magnet classrooms and school corridors. Sarah Grant, a third-grade teacher from Greenwood Paideia, punctuated the point.

I believe in integrated schools. I believe in as many kids together from as many different backgrounds as possible. I think that is the richest education the kids are getting when they are going to school with so many different cultures. I think that is a very important thing for all kids, kids from different socio-eco- nomic backgrounds, too.2

Teachers at Overbrook Academy in St. Louis voiced similar sentiments about the value of racially integrated schools and were equally impressed with the results of their racially balanced environments. Although not a perfect solu- tion to racially segregated neighborhoods—a prominent feature of residen- tial life in St. Louis, as it is in numerous U.S. cities (Orfield, Bachmeier, James, & Eitle, 1997)—many view the central aims of the desegregation plan and the strategy of using magnet schools as a viable and worthy enterprise.

28 EDUCATION AND URBAN SOCIETY / November 2000

Bill Rogers, a 24-year veteran teacher at Overbrook, explained this viewpoint in the following:

You can look out the window . . . and see black and white children playing together, which is why we were invented, and on that level for the children, it works. Now, of course when they get on the buses and go back to separate neighborhoods, it is hard to stay friends for life, but that is a community prob- lem; it is not something we can solve in the schools.

Charlie Allen, the gym teacher at MaSAC for 12 years, concurred:

I wish all the neighborhoods were racially and economically integrated so chil- dren could grow up in a diverse culture because when they get out of school, they are going to have to deal with all kinds of other people, rich and poor, black and white. And that is what schools have to provide. Neighborhoods don’t do that. So you go to an alternative and you wind up busing.

Many teachers spoke specifically to the value of children exposed to other cultures, the importance of “learning that they are just the same as you and me,” and the goal of “kids learning to get along better” across racial/ethnic groups. They observed with pride the pervasive patterns of children mixing across racial groups socially and academically in the classroom and out on the playground.

THE COST OF INTEGRATION

Most teachers were quick to point out that the racial integration in magnets comes with a cost; when school systems transport children far away from their neighborhoods, the social distance between schools and families grows to reflect the geographical space that separates them. This point was made repeatedly, but the justification for these policies and the willingness to sacri- fice a sense of community for a manifest commitment to integration is a con- stant reminder to teachers of a collective goal tied to improved racial under- standing. Although there is a sense of doing battle with influences beyond their control, there is an unwavering belief that promoting racially integrated schools is an important contribution to larger efforts. As first-grade teacher Shanika Taylor at Viking Academy in St. Louis noted,

You need to learn the cultures of others. There are students who hear things at home and they never would know the difference. They believe that until they are around another race and then they discover that these people aren’t so bad

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 29

after all. It is that sort of thing that I think about when I think about doing away with the busing.

In a similar vein, sixth-grade teacher Bill Rogers at Overbrook lamented the cavalcade of buses—11 in all—that transport children for up to 1 hour away from school to their neighborhoods across St. Louis:

In a perfect world, I guess what they would have done is integrate the neighbor- hoods, then children could go to their neighborhood schools. But I wouldn’t know how to do that. Maybe if they could magnetize neighborhoods. Is it worth it?

Still, other teachers are not as convinced as Bill Rogers that busing, even under voluntary conditions in a magnet system, is worth the costs. These teachers are more critical of what they perceive to be unaltered attitudes about race, social structures that discriminate against African American stu- dents, and behaviors that suggest that nothing has changed when it comes to improved racial understanding and trust. Many teachers pointed to the expense, time, and safety of bus riders as central concerns. Other teachers pointed out that the bus schedules have eliminated after-school activities and have shattered the kinds of positive social interactions and stable relation- ships that activities such as scouting and sports engender. All are convinced that magnet schools offer only a temporary, insufficient, and inconvenient mechanism for racial integration. Consider the perspectives of two teachers from St. Louis. First, Mrs. Settle, who is White and whose Viking Academy uses 14 buses and 4 taxicabs for transporting students across the city.

I’m not sure busing is working. Unless you have blacks and whites living next to each other and getting used to lifestyles, you are not going to have any more communication than you do now. I think integration of neighborhoods is the only way we are going to get integration in the schools to work.

Ms. Alvins, who is African American and teaches at Overbrook Academy, was more pointed in her assessment, arguing that desegregation efforts un- fairly and disproportionately disadvantage African American students and their families (Morris, 1997). This critical perspective, although far less com- mon in the viewpoints collected here, is no less important.

Until housing patterns can be changed and people live together, like you know your neighbor and I know mine, there is no such thing as [racial integra- tion]. . . . I think we are worse off than we were before, particularly black

30 EDUCATION AND URBAN SOCIETY / November 2000

children. Once their schools were eliminated and they were bused into other areas, their culture, their heritage, their history, it was ignored. It still is. Even here, we have to fight for it.

PARENTS: “IT’S WORTH IT”

When we asked parents to weigh the cost of busing with the value of racially integrated schools, most parents—African American and White— echoed the majority of teachers in a belief that the goal of integration is worth it. Consider the following observations made by Ruby Fox, a White woman from Greenwood Paideia in Cincinnati:

I like the magnet program because it offers so much diversity. . . . There is such a diverse group of students. My kids are more tolerant of a lot of things that I don’t think they would have been tolerant of had we lived in a neighborhood that was primarily black or primarily white. I feel like they have been exposed to all cultures, not like just a Catholic school.

The issues of racism and intolerance born out of ignorance and isolation seem to coalesce for most parents into a perspective that considers school integra- tion at the elementary school level an important starting point practically and symbolically. Althea Robinson from Greenwood spoke to this point with force and candor in the following:

Let me tell you why I didn’t think that much of my neighborhood school. With me being black, all your home base is if you are black or white, they are going to have all the majority of those people in it. I wanted my children to be among a multiculture of people. That is the way I was raised. I wanted my son to know about little white kids, Jewish kids, German kids, Indian kids—that is what kind of friends he has now. You find that you have to bump your kids out into the world. I love my black people dearly, but I also know that when you get out of school, you are dealing with multicultures.

Chontay Parks, a parent of two sons at Overbrook Academy in St. Louis, said she was proud to recently learn that the city high school from which she grad- uated 11 years ago is now racially integrated. To see this overcrowded, com- prehensive high school transform itself from a predominantly African Amer- ican school with high suspension rates to a magnet school with a high rate of academic success gave Mrs. Parks, in her words, “a good feeling.” As a resi- dent of a predominantly African American section of St. Louis (north St. Louis), she is convinced that the social values represented in the magnet school

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 31

model are as vital as the academic opportunities. Echoing the views of other parents we interviewed, Mrs. Parks was unequivocal in her defense of busing as a vehicle for promoting racial balance.

They get to mix with children of different cultures. They learn about different cultures. We live in the north city and (the magnet school) doesn’t keep them so isolated. They get to learn and be around other people so in the workforce, they won’t be so green and dumbfounded. I really like it.

Almost all the parents we talked to expressed some concern about enduring patterns of racial and social isolation in their cities and suggested that magnet schools are a way of ensuring that, early on, children learn to view other youngsters of different ethnic and racial groups as the same as themselves. This heightened sensitivity to ethnic diversity among schoolchildren was ref- erenced repeatedly by parents as a means of ensuring that, as one father from Overbrook put it, “people can get along” later in life.

If nothing else, this exposure to other children from different racial groups signals an element of the public school system of which parents and their children—selected as they are based on racial criteria—are a vital part. At the same time, it is important to underscore the muted emphasis on race evi- denced in district rhetoric and in conversation with participants. Although responses were typically vocal and detailed, the issues of race and ethnicity were addressed only after we prompted and probed parents, school officials, and teachers; race regularly followed discussions that were focused on the general goodness of magnets, the safety and security of the students, and the satisfaction of the faculty. We suggest that, in addition to racial balance and multicultural awareness, magnet schools represent a separate set of values and profitable investments vested in the currency of parental choice.

RACIAL BALANCE IN THE POST-BUSING ERA

We conclude this article by amplifying the significance of competing interests at this particular time in the history of school desegregation. Several large urban school districts in the nation stand at the crossroads of sweeping changes in desegregation policy in the post-busing era, including St. Louis. In a series of major federal court rulings (see Capacchione v. Charlotte- Mecklenburg Board of Education, [1999]; Eisenberg v. Montgomery County Public Schools, [1999]; Tuttle v. Arlington County School Board, [1999];

32 EDUCATION AND URBAN SOCIETY / November 2000

Wessman v. Gittens, [1999]), school district efforts to maintain admission policies that are designed to promote and ensure racial diversity in magnet schools have been repudiated. The 1971 U.S. Supreme Court landmark rul- ing in Swann v. Charlotte-Mecklenburg (1971) that authorized school offi- cials to take race into account in admission policies “to prepare students to live in a pluralistic society” now appears irrelevant, subordinated to individ- ual rights guaranteed under the equal protection clause of the Fourteenth Amendment. Under the precedent established in a 1995 Supreme Court rul- ing (in a case that involved a federal program that awarded a percentage of construction contracts to minority-owned construction companies; see Adarand, 1995), race-conscious programs that involve promoting diversity through strategies construed to involve racial balancing are constitutionally suspect and are subject to strict scrutiny. This elevated constitutional bar includes a two-pronged test that compels districts to prove that their racial classification scheme “furthers a compelling state interest” and is narrowly tailored (see Adarand, 1995). Applying these recent court decisions to a legal standard, it appears that only those districts currently under court order to remedy the effects of past discriminatory practices can clearly and defini- tively meet this compelling state interest. Consequently, unless school dis- tricts are currently under court order to remedy the effects of past racial dis- crimination in their systems, magnet school admissions policies must be race neutral. The concept of narrowly tailored deepens the complexity and legal rigidity of the new rules. Narrowly tailored involves such factors as (a) the efficacy of the race conscious program, (b) the flexibility of the program, (c) the relationship of the numerical goal to the relevant population, (d) the bur- den of the policy on innocent third parties (e.g., White students), and (e) the duration of the program.

In one of the most recent and historically significant rulings, the district court judge in the Charlotte-Mecklenburg case wrote that he “accepts that children may derive benefits from encounters with students from different races . . . ,” but in race-conscious magnet school admissions programs, “chil- dren are not viewed as individual students but as cogs in a social experimenta- tion machine.” Weighted or separate lotteries designed to yield racially diverse student populations that mirror the racial group averages for the entire school district are now construed to be unfair because such policies may deny special benefits—the unique curriculum offered by a magnet school program—to students based solely on their race (see Capacchione v. Charlotte-Mecklenburg Board of Educ., 1999). (The Charlotte-Mecklenburg case has been appealed to the U.S. Circuit Court of Appeals.) The Supreme

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 33

Court has yet to rule on whether promoting diversity constitutes a compelling state interest, although the Arlington case is currently before the Court await- ing action. Lower federal courts have simply sidestepped the issue by assum- ing that promoting diversity in schools may be a compelling state interest without so ruling (with the exception of Hopwood v. State of Texas, in which the Fifth Circuit Court of Appeals rejected diversity as a compelling state interest in a law school admissions policy that involved racial preferences for ethnic minorities).

The recent court ruling in the Boston Latin School (Wessman v. Gittens, 1999) case perhaps provides some insight on the shifting legal and political climate on racial diversity and the implications of this new era for magnet school policies. In rejecting the long held commitment to diversity in student admissions policies at Boston Latin, the First Circuit Court of Appeals wrote the following:

The potential for harmful consequences prevents us from succumbing to good intentions. The Policy is, at bottom, a mechanism for racial balancing—and placing our imprimatur on racial balancing risks setting a precedent that is both dangerous to our democratic ideals and almost always constitutionally forbidden.

With a chorus of city mayors, from Austin and Seattle to Boston and Nash- ville, calling for a return to neighborhood schools, there is strong suggestion that the ideal of diversity may be subjugated to an ideal of unencumbered choice and parental privilege in public education. These trends signal the need for a better understanding of new policies on race and schools, the social and political context of choice, and the consequences of these reform strate- gies for school systems in urban America and for the lives of educators, stu- dents, and their families.

NOTES

1. Under the provisions of a 1983 Federal court order, the St. Louis City Public School Dis- trict operates an interdistrict voluntary transfer program with 16 participating suburban districts that included magnet schools in the city. The interdistrict choice program allows parents to choose between schools inside the district and some schools outside the district to promote racial balance.

2. All names used in this article are pseudonyms.

34 EDUCATION AND URBAN SOCIETY / November 2000

REFERENCES

Adarand Constructors, Inc. v. Pena, Secretary of Transportation et al., 515 U.S. 200 (1995). Blank, S. (1996). The pull of magnets. The American School Board Journal, 183(9), 34-36. Blank, R., Levine, R., & Steel, L. (1996). After fifteen years: Magnet schools in urban education.

In B. Fuller, R. Elmore, & G. Orfield (Eds.), Who chooses? Who loses? Culture, institutions, and the unequal effects of school choice (pp. 154-172). New York: Teachers College Press.

Capacchione v. Charlotte-Mecklenburg Schools, No. 97-CV-482-P; 3:65-CV-1974-P (WDNC, September 1999).

Charlotte-Mecklenburg Public Schools. (1994). Magnet assistance program enrollment and participation report. Charlotte, NC: Author.

Eaton, S., & Crutcher, E. (1996). Magnets, media and mirages. In G. Orfield & S. Eaton (Eds.), Dismantling desegregation: The quiet reversal of Brown v. Board of Education (pp. 265- 289). New York: New Press.

Eisenberg v. Montgomery County Public Schools, 19F. Supp. 2d 449 (D. Md. 1998). Hopwood v. Texas, No. 95-5006L, 95F.3d 53 (5th Cir. 1996). Levine, R. (1997, April). Research on magnet schools and the context of school choice. Paper

presented at the Citizens’ Commission on Civil Rights Issues Forum: Magnet Schools and the Context of School Choice: Implications for Public Policy, Washington, DC.

Morris, J. (1997). Voluntary desegregation in St. Louis, Missouri: Impact on partnerships among schools, African-American families, and communities. Unpublished doctoral dissertation, Vanderbilt University, Nashville, TN.

Morrison, P. A. (1992). Forecasting enrollments during court-ordered desegregation. Santa Monica, CA: RAND.

Orfield, G., Bachmeier, M., James, D., & Eitle, T. (1997). Deepening segregation in American public schools. Cambridge, MA: Harvard Project on School Desegregation.

Plank, S., Schiller, K., Schneider, B., & Coleman, J. (1992, October). Choice in education: Some effects. Paper presented at the Economic Policy Institute symposium, Choice: What Role in American Education? Washington, DC.

Steel, L., & Eaton, M. (1996). Reducing, eliminating, and preventing minority isolation in Amer- ican schools: The impact of the Magnet Schools Assistance Program. Washington, DC: Department of Education.

Steel, L., & Levine, R. (1994). Educational innovation in multiracial context: The growth of magnet schools in American education. Palo Alto, CA: American Institute for Research.

Swann v. Charlotte-Mecklenberg Board of Education, 404 U.S. 811 (1971). Taylor, W., & Yu, C. (1999). The context of magnet schools: The policies and politics of desegre-

gation in Cincinnati and St. Louis. In C. Smrekar & E. Goldring (Eds.), School choice in urban American (pp. 15-25). New York: Teachers College Press.

Tuttle v. Arlington County School Board, No. 98-1604 (4th Cir. 1999). Wells, A., & Crain, R. (1997). Stepping over the color line. New Haven, CT: Yale University

Press. Wessman v. Gittens, 160 F.3d 790 (1st Cir. 1999). Yu, C., & Taylor, W. (1997). Difficult choices: Do magnet schools serve children in need? Wash-

ington, DC: Citizens’ Commission on Civil Rights.

Goldring, Smrekar / PURSUIT OF RACIAL BALANCE 35