The Legacy of Busing and Brown : How School Desegregation Experiences Shape Public Perceptions and Communities’ School Diversity Priorities in the Resegregation Era

Abstract

In a context that privileges neighborhood zoning and school choice over within-district busing, we examine urban residents’ perceptions of the benefits of racial diversity. We analyze public opinion trends by race and residents’ experience as a student in a district under court-ordered desegregation—Nashville, TN. We find racial differences regarding the values of proximity and school diversity that are moderated by experience as a school-age resident in a city under mandatory desegregation. We build upon literature showing long-term impacts of attending integrated schools, underscoring the influence of time, place, and experience in shaping perceptions of school diversity policies.

Keywords

school desegregation public opinion busing racial diversity

Introduction

While busing may have been an efficacious policy for creating racially desegregated schools and for cultivating the conditions for cross-racial social contact (Eaton, 2001; Wells & Crain, 1999), busing was unpopular (Pride, 2000; Pride & Woodward, 1995) and expensive both in terms of students’ time and district costs (Rossell et al., 2002). Few, if any, districts are considering returning to or initiating new crosstown busing plans reflective of the ones prevalent under court-ordered desegregation, even districts facing resegregation despite historical commitments to desegregation (Quick & Damante, 2016; Semuels, 2019). However, many districts (Taylor et al., 2019) are struggling to address racial and socioeconomic segregation across schools with interest in voluntary plans that address parental values and student needs, are legal under unitary status, and do not differentiate educational opportunities for students based on race and family socioeconomic status (Reardon et al., 2012).

This paper engages with multiple agendas of the field of urban education. Of the seven subject areas recommended for the field by Milner and Lomotey (2013), we address components of “Sociological Perspectives,” “Families and Communities,” and “Policy and Reform.” Specific areas of focus we engage with include societal factors shaping efforts to educate students in urban schools; the historical and contemporary roles of parents, families, and communities in shaping education in urban schools; and the influence of historical movements and policies on urban schools. We examine theories and practical influences of school desegregation policies to understand the historical influence on public opinion regarding the values of proximity and diversity in contemporary urban schools.

We examine the perspectives of residents of Nashville, Tennessee through an analysis of survey responses. This paper’s significance rests with the consideration of how court-ordered desegregation efforts have shaped residents’ perceptions of race and schooling. We examine the association of race and experience as a student in a district with mandatory busing and perspectives regarding the importance of proximity and the value of diversity in schools.

Key Literature and Theory

Brief Overview of Desegregation in the South

A historical legacy of school desegregation in the South informs this paper. The intersection of legal, demographic, and social contexts defines a history of schooling in this region that is significant and markedly different from school segregation and desegregation patterns in the West and North—regions that operated de facto segregated school systems without mandated separation. Anchored to civil rights movements that spread throughout the South in the late 1950s and early 1960s, the efforts toward school desegregation coalesced with efforts to desegregate various social institutions, civic enterprises, and commercial establishments across Southern states (Braddock, 2009; Johnson, 2019). Though the Brown decision in 1954 focused national attention on dismantling so-called “dual” schooling systems for Black and White students in the South, the ruling was met by a wall of intransigence and a pattern of insufficient enforcement. Enforcing desegregation would require passage of the Civil Rights Act of 1964, prohibiting discrimination in any school district receiving federal funds, coupled with the US Supreme Court decisions in Green (1968) and Swann (1971) that enabled civil rights groups and community leaders to move with power and authority to desegregate schools nationwide. By the late 1970s, implementation and enforcement of federal desegregation orders made school districts in the South the most materially integrated in the U.S. (Frankenberg et al., 2017).

Despite progress, some decisions limited the ability to achieve or maintain school diversity. Notably, for urban districts like Nashville surrounded by mostly White suburbs, Milliken v. Bradley (1974) limited the ability to achieve diversity by mixing students across school district boundaries. In the 1990s, multiple court cases contributed to the relaxation of court-ordered desegregation across many US districts. The Court’s rulings in Dowell (1991) and Freeman (1992) substantially halted and reversed the momentum and change that defined desegregation efforts in the South by ruling that court orders were meant to be temporary, local control was preferable when a district has made a “good faith effort” to desegregate, and that districts could be released from court order in a piecemeal fashion. Three years later, the courts further accelerated the end of court orders by ruling in Missouri v. Jenkins (1995) that district remedies are only required to bring the victims of past discriminatory acts to the point they would have occupied had the acts not occurred. For the past three decades, unitary status has fueled an accelerated path toward resegregated schools with high inter-district segregation levels and a reluctance of White families to return to urban public schools (Billingham, 2019; Reardon et al., 2012).

In the 2000s, districts’ policy options once no longer under court-ordered desegregation to maintain desegregated schools decreased. The 2007 PICS ruling ended schools’ use of an individual student’s race as a sole factor in assigning students to a particular school. Instead, school districts were instructed to consider what became known as the Kennedy Remedies. These included strategic site selection of new schools, drawing attendance zones in recognition of neighborhood demographics, allocating resources for special programs, recruiting students and faculty in a targeted fashion, and tracking enrollment, performance, and other statistics by race (Welner & Spindler, 2009).

The past 30 years paint a pattern of persistent and expansive multi-racial segregation in the South among Black, White, and Latinx students across the canvas of both urban and suburban school districts in this region. According to the Civil Rights Project (Frankenberg et al., 2017), more than one in three (35.8%) Black students attends an intensely segregated (>90% students of color) school in the South. The rise in concentrated isolation of Black students is noteworthy as compared to the period of active desegregation in the South when 23% of Black students attended an intensely segregated school. A precipitous decline in Black-White student exposure or inter-racial contact has followed, with clear and compelling implications for creating constructive cross-racial communities of respect and diversity in schools (Williams & Graham, 2019).

While there are many benefits of desegregation, the burdens have been disproportionately borne by Black students. Black students often rode school buses in younger grades and over greater distances than White students; entered schools where they often did not feel welcome and sometimes faced angry protesters (Karpinski, 2006; Pride & Woodard, 1985). Black students left schools with primarily same-race teachers as role models who believed in Black students’ potential to attend schools with more different race teachers holding lower expectations for Black students relative to White students (Carter, 2019; Woodward, 2011). During busing, districts moved students without providing adequate resources for White teachers to adjust to a diverse classroom or foster Black student development (Milner, 2018). Unfortunately, teachers have not only been under-prepared to teach diverse classes and many enter the classroom with racial biases, but there is also evidence that Black students tend to receive lower-quality instruction than White students and tend to face uneven tracking and discipline (Darling-Hammond, 2007; Lankford et al., 2002). These are just a few examples of how Black students bore greater desegregation burdens during and after court-ordered desegregation. This paper attempts to further discussion of the long-term influences of court-ordered desegregation and racial differences in schooling experiences under desegregation.

Priorities Regarding Racial Diversity and Schools: A Paradox and Potential Interest Convergence

Many White Americans purport to support the ideal of racial equality while refusing to support specific policies meant to aid in the attainment of racial equality, such as busing, affirmative action, or bilingual education (Bobo & Fox, 2003; Houvouras, 2001; Huddy & Sears, 1995; Sears et al., 2000). While White parents may claim to value school diversity for their children, they will not necessarily choose to send their child to a diverse school, perpetuating school segregation (Roda & Wells, 2013; Torres & Weissbourd, 2020). As Bobo and Fox (2003) explain, three main sets of theories are thought to explain this attitude paradox. First, social-psychological theories purport that symbolic racism, racial resentment, or subtle prejudice have replaced overt historical racism (Henry & Sears, 2002; Pettigrew & Meertens, 1995). These theories incorporate the ideas of color-blindness or new racism (Bonilla-Silva, 2003; Lewis, 2003). Thus, it is not a lack of racism or prejudice, but a new acceptable form of racism or prejudice that is more subtle and permits one to oppose policies directly intended to create equality in a system currently lacking it under the guise of fairness and neutrality.

Second are social structural theories that consider the role of competing group interests, such as realistic group conflict theory (Key, 1949; Levine & Campbell, 1972). These theories purport that the paradox is due to individuals identifying with their racial group and seeking to maintain or attain advantage over competing interests, and that this disposition cannot be changed through socialization (Bobo & Fox, 2003; Jackman & Muha, 1984). According to these theories, we would not expect changes associated with experience in desegregated schools.

Third, principled politics suggest that the paradox represents a desire for race-neutral policy rooted in ideals of fairness and individualism (Sniderman & Carmines, 1997), whereby White respondents are opposed to explicitly or even implicitly racial policies, with a preference for policies they do not see as privileging one race over another, even if in practice they do. According to these theories, there is potential for change through socialization, such as through experience with desegregation.

This paper empirically examines the racial attitude paradox as it examines respondents’ opinions both to a question about the inherent value of diversity in schools, which they may be more willing to state they value, and two specific policy options that could shape the reality of diversity in schools. They may be less likely to agree to the historically unliked use of busing that carries with it baggage from the experience of desegregation and the use of a diversity plan in students’ assignment to schools. Critically, this paper examines how experience as a student in a district under court-ordered desegregation shapes one’s views on the value of school diversity and practical diversity policy.

Long-Term Effects of Inter-Racial Schooling on Adults: A Conceptual Framework

Districts face practical challenges related to the attitude paradox for school diversity. The theory of interest convergence indicates potential policy options to increase school diversity if implemented in a way that also reflects the interests of White families. According to this theory, White families will support equity-oriented policies if they can be implemented while advancing their self-interests and without altering their status (Bell, 1980; Castagno & Lee, 2007; Lopez, 2003; Milner, 2017). Thus, if school diversity can be achieved while reflecting White priorities about proximity and neighborhood schools, this might present a promising option. Relatedly, social justice and equal educational opportunities are afforded when Black and White interests converge and when there are shared priorities across race. In this paper, we explore stated differences in racial priorities related to school diversity and proximity exploring whether there is convergence in interests related to diversity and proximity across and within race.

While students who attended public schools during court-ordered desegregation faced burdens, there is evidence of positive long-term effects of attending desegregated schools (Braddock, 2009; Ellison & Powers, 1994; Kurlaender & Yun, 2001; Mickelson et al., 2012; Sigelman et al., 1996; Stearns, 2010; Wells et al., 2008). These long-term effects are not limited to closing the Black-White student achievement gap, but also include an impact on adult inter-racial interaction and understanding of inter-racial differences.

Stearns (2010) found exposure to diverse racial groups in high school reduced racial isolation in the post-high school workplace. Formation of cross-racial friendships in school is also associated with acceptance among different groups and appreciation of cross-cultural differences (Chang et al., 2006). Wood and Sonleitner (1996) found that childhood interracial contact in schools and neighborhoods not only disconfirms negative racial stereotypes, but has a direct, significant effect on (lowering) levels of adult anti-Black prejudice. Hurtado (2006) found that students entering college from diverse high schools were more likely to view the world from the perspective of others and value racial and cultural diversity.

While the largest effects are found in cases where genuine inter-racial contact in schools leads to inter-racial friendships (Williams, 2010), there is evidence of long-term impacts of attendance at a desegregated school even when close interracial friendships did not occur. Johnson (2015) found Black students who attended desegregated schools experienced significantly increased educational and occupational attainment levels and lifetime earnings, improved adult health status, and a reduced likelihood of incarceration with no adverse effects for White students. Qualitative evidence suggests that graduates of desegregated high schools were better prepared for life in a racially diverse society, gained a deeper understanding of people of other backgrounds, and were more comfortable in interracial settings (Holme et al., 2005).

The interpersonal effects of increased inter-racial contact beg the question of whether adults who experienced desegregated schooling as children believe those experiences hold value well into adulthood, and whether adults believe that school diversity should remain a priority today. While this literature addresses the long-term impact of attending a racially diverse school, there has been less work examining how the experience of desegregated schooling as a child influences one’s opinion on the value of racial integration in schools for children today. Kimble (1980) used qualitative interviews to examine differing opinions on busing and desegregation attitudes among adults who participated in court-ordered desegregation programs and those who did not, finding little evidence of effects of busing on one’s adult desegregation attitudes.

This paper examines a larger, more contemporary sample to further unpack how attending a public or non-public school as a youth in a district under court-ordered desegregation shapes adult perspectives regarding race and racial diversity in schooling, and how this differs by respondent race. We examine both the direct associations between race and opinions on the values of proximity or the burden of busing and of school diversity and the association between being of school age during court-ordered desegregation and one’s values, as well as the potential moderation of racial differences in opinion by whether the respondent was of school-age during court-ordered desegregation. It is essential to clarify that this paper only looks at whether a respondent was of school age in the county while the county was under court-ordered desegregation; it does not look at whether the respondent attended an integrated school or the level of integration in the school attended. A district under court-ordered desegregation may not have had every student in integrated classrooms or even integrated schools, as private schools remained segregated. For our purposes, desegregation is limited to a policy interpretation, whereby exposure to school desegregation is limited in scope to the policy of court-ordered desegregation when the respondent was a student, whether or not they themselves experienced integration or school diversity. Our framework, in Figure 1, suggests that (and this paper tests if) racial differences in contemporary opinions on the values of school proximity and diversity as well as on the practical burden of busing and diversity-based school assignment policy could be moderated by one’s experience as a school-age resident of a county under court-ordered desegregation.

Figure 1.

A framework for contemporary differences contemporary school values by race and court-ordered desegregation.

Research Questions

We attempt to better understand the long-term influence of court-ordered school desegregation on contemporary public opinion regarding the value of diversity in schools and the burden of busing in an emerging urban city. Specifically, we ask:

To what extent are race and experience as a school-age resident in a county under mandatory desegregation associated with contemporary views on the value of neighborhood schools and the practical burden of busing?

To what extent are race and experience as a school-age resident in a county under mandatory desegregation associated with contemporary views on the value of diversity in schools and the practical use of a diversity policy?

Context, Data, and Methods

The Nashville Context

We focus on Nashville, TN as a notable case of a Southern city recognized for a series of peaceful demonstrations that galvanized the Civil Rights Movement and local desegregation efforts in the late 1950s and early 1960s (Erickson, 2016; Halberstam, 1998). Nashville fits within the typology of urban districts developed by Milner (2012), being characterized as an emerging urban district. These districts are characterized by the city’s size, with fewer than a million people in relatively large spaces and facing fewer complex challenges between schools and the surrounding communities than the largest metropolitan cities in the country. Nashville is also a racially and economically diverse city, with a majority non-White representation of students in public schools.

In 1957, Metro Nashville Public Schools (MNPS) began implementing a stepwise year-by-year court-ordered desegregation plan, centered on mandatory busing beginning with first grade (Erickson, 2016). Under the desegregation plan, many Metro students rode the bus for over an hour each way to school; most of these were Black students who lived in inner-city neighborhoods and were bussed to outer ring, White suburban areas of the district. The school district gained unitary status in 1998, ending nearly three decades of court-ordered crosstown busing, and adopted a system of “closer to home” student assignment and neighborhood-based school zones.

The new plan radically reduced or eliminated the formerly court-ordered crosstown busing patterns; these were replaced with new neighborhood-anchored zoned schools and a system of elementary-middle-high school cluster feeder patterns (Metropolitan Nashville Board of Education, 1998). The plan expanded school choice policies in the district to include new magnet schools. However, under unitary status, only a single, race-neutral lottery could be utilized, unlike the previous dual lottery system that guaranteed student enrollment reflecting district-wide percentages of Black and White students (Goldring, et al., 2006; Smrekar, 2009). In the end, the Nashville district and the committee tasked with fostering a consensual end to crosstown busing adopted what Erickson (2016) calls a “hope for growth rather than a pursuit of equality” (p. 302).

By 2014–2015, the district slipped out of racial balance into a pattern of resegregated schools, despite a policy adopted in 2011 calling for schools to reflect the district’s racial diversity (Metropolitan Nashville Public Schools [MNPS], 2016). The district’s demographic composition has changed markedly over the past 20 years; today, the district enrolls 85,688 students, 41% of whom are Black or African American, 26% Hispanic/Latinx, 28% White, and 4% Asian (MNPS, 2018). District enrollments reflect an out-of-balance racial composition in terms of county demographics for school-aged children. Nashville presents a particularly important site for research into potential long-term effects of court-ordered desegregation efforts and contemporary public opinion regarding the value of school diversity in emerging urban cities, given its significant role in the Civil Rights Movement and the post-Brown efforts to desegregate public schools.

Data

The data for this analysis came from a phone survey of Davidson County (Nashville) residents in 2015 (17 years after the end of mandatory busing) conducted by Public Policy Polling as a part of an NSF-funded study (Parcel et al., 2016). The larger study involved five southern metropolitan areas and compared public opinion regarding school desegregation in the southern US. The Nashville portion of the survey includes 1011 respondents resulting from four lists of phone numbers. Two lists of approximately 30,000 randomly selected landline numbers were contacted using Interactive Voice Response (IVR), one mixed-race and the second with households identified as Black; Black respondents were over-sampled for sub-group analyses despite trends of lower response rates. The third and fourth lists consisted of 6000 randomly selected cell phone numbers contacted by live operators, one mixed-race, and the second identified as Black. Numbers were contacted at least six times at different times of day and days of the week, with a goal of 800 respondents age 21 or above answering and completing the survey. Even with the over-sampling of Black numbers, the total sample in Nashville maintained a slight over-representation of White respondents as compared to the 2015 American Community Survey (ACS) estimates for Davidson County (with 67% White respondents as compared to 57% in ACS) and a slight under-representation of Black (23% compared to 27% in ACS) and other race respondents (10% compared to 16%). Probability weights were used in estimation models to better approximate the demographics from the 2015 ACS. We define urban residents in this study as respondents residing in Metropolitan Nashville with either landline or cell-phone service, and the survey has been designed and weighted to reflect the demographics of adult residents of Nashville.

This paper has four dependent variables. The first three were derived from subjects’ Likert style responses (“strongly agree,” “somewhat agree,” “neither agree nor disagree,” “somewhat disagree,” or “strongly disagree”) to the following statements:

“Children learn best when they attend school with children from their own neighborhoods.”

“School bus rides over 30 minutes long are too hard on young children.”

“Children learn best when they are in classrooms that are racially diverse.”

The final dependent variable was derived from the responses to the following question:

“The Metro Nashville Public Schools Diversity Plan describes diversity as ‘multi-faceted, including race/ethnicity, income, language, and disability.’ Should students be assigned to schools in Metro Nashville in ways that take into account the Metro Nashville Public Schools diversity Plan?”

Responses included “students should be assigned to schools in Metro Nashville in ways that take into account the diversity plan,” “students should not be assigned to schools in ways that take into account the diversity plan” or “not sure.” We recode 15 respondents with no response to this question as “not sure.”

We centered the dependent variables so that neither agree nor disagree or not sure are coded as 0. We have two main independent variables of interest. The first is race: for this variable, we used indicators for White, Black, or other. Our second independent variable of interest measures whether the respondent was of school-age during mandatory busing in Nashville and lived in Nashville at the time. For this measure, we restricted the sample to the 323 respondents who had lived in Nashville their entire lives, as we cannot be sure of the school desegregation policies in the place of residence for transitory residents, or if they had been in Nashville long enough to have experienced the city under court-ordered desegregation as a student. We also dropped one respondent under age 23, as they were too young to have participated in mandatory busing. Respondents age 63 and below were of school-age while the district was under court-ordered desegregation, while respondents age 64 and above were not of school-age while the district was under court-ordered desegregation. Therefore, respondents age 63 and below had an educational experience that was personally impacted by court-ordered desegregation, while their counterparts did not. This does not necessarily mean that they attended integrated or even necessarily attended desegregated schools, as we do not know which respondents attended private schools or the level of integration in the school the respondent attended. However, it does mean that they attended school in a system where the public schools were facing court-ordered desegregation and private schools experienced White flight.

Methods

We examined descriptive statistics of the mean, or proportion where appropriate, and standard deviation for all variables included in our models. We conducted Kruskal–Wallis equality of populations rank tests with Chi-squared with ties and a Dunn’s pairwise test for significance in difference in means in our dependent variables across racial subgroups. We conducted Wilcoxon Mann–Whitney tests for significance in ordinal dependent variables for our indicator of school age during mandatory desegregation.

As controls, we include indicators for identifying as liberal, conservative, or moderate/refusing response, having school-age children, having a child attending MNPS, gender, marital status, educational level (“completed some high school” “graduated from high school” “completed some college” “graduated from college” “post-graduate training or degree”), and an indicator for household income. The household income variable has some missingness; therefore, we used imputation for the missing values (21.5% of the sample, 217 respondents). We used single imputation for descriptive statistics and multiple imputation for the regression models.

We explored trends in the proportion of respondents within our subgroups by their responses to our survey questions of interest. We then ran ordinal logistic regressions for the responses with and without control variables in our full sample and our lifetime Nashville resident sample to detect an association between race or being of school age during mandatory busing. We converted the coefficients into odds ratios. We explored potential differential relationships in models with interaction terms for race and being school-age during court-ordered desegregation.

Results

Descriptive Results

Table 1 presents the means or proportions for each variable and their corresponding standard deviation. Our two dependent variables related to values of proximity are labeled “Better with Neighbors” and “Busing too Hard.” Both are centered Likert style variables ranging from ‒2 to 2 with positive responses showing agreement that students learn better with their neighbors and that busing is too hard on young children. Both samples had a mean response between 1.46 and 1.62 for these variables with a standard deviation of less than 1. This suggests that while there was variation in responses, on average, respondents agreed to the positive value of proximity and negative burden of busing.

Table 1.

Descriptive Statistics.

	Full sample		Always lived in Nashville
	Mean	SD	Mean	SD
Better with neighbors	1.57	0.83	1.58	0.91
Busing too hard	1.46	0.97	1.62	0.79
Better in diverse school	0.76	1.26	0.69	1.27
Assign with diversity plan	0.04	0.86	−0.02	0.85
Age	60.80	15.82	62.94	14.41
	Proportion	SD	Proportion	SD
During busing			0.49	0.50
White	0.67	0.47	0.58	0.49
Black	0.23	0.42	0.33	0.47
Other race	0.10	0.30	0.09	0.29
Married	0.54	0.50	0.53	0.50
Female	0.62	0.49	0.66	0.48
Kid in MNPS	0.14	0.35	0.14	0.34
Has kids	0.20	0.40	0.19	0.40
Under $25K	0.12	0.33	0.17	0.37
$25 to 50K	0.24	0.43	0.25	0.43
$50 to 75K	0.24	0.43	0.26	0.44
$75 to 100K	0.15	0.35	0.14	0.34
$100 to 150K	0.14	0.35	0.11	0.32
$150K+	0.11	0.31	0.08	0.27
Democrat	0.22	0.42	0.16	0.36
Republican	0.29	0.45	0.33	0.47
Some HS	0.04	0.19	0.07	0.25
High school	0.10	0.30	0.17	0.37
Some college	0.18	0.39	0.23	0.42
College	0.29	0.45	0.26	0.44
Post-graduate	0.39	0.49	0.27	0.45
N	1,011		321

Note. Always lived in Nashville represents respondents who lived in Davidson County their entire life. “Better with neighbors,” “busing too hard,” and “better in diverse school” are ordinal variables ranging from −2 to 2 with negative values disagreeing with the statement regarding kids and schools and positive values agreeing. “Assign with diversity plan” is an ordinal variable ranging from −1 to 1 with −1 being disagree and 1 agree. “During busing” is an indicator for if the respondent was of school age in Nashville during court-ordered desegregation. “Kids in MNPS” is an indicator for if the respondent currently has a child attending Metro Nashville Public Schools.

Our two dependent variables related to the value of diversity were also centered with “Better in Diverse School” being a Likert style variable ranging from ‒2 to 2 and “Assign with Diversity Plan,” ranging from ‒1 to 1. Negative values disagree that students learn better in racially diverse schools or that assignment to schools should consider the diversity plan, while positive values agree. The “Better in Diverse School” variable had a mean of 0.76 for the full sample (SD 1.26) and 0.69 for the always lived in Nashville sample (SD 1.27). This means that, on average, the respondents somewhat agreed that students learn better in diverse schools, but that there was a wide range of responses, including agreement and disagreement among the majority of respondents. The “Assign with Diversity Plan” variable is slightly above 0 with a Standard Deviation of about 0.86, suggesting an average response of neutrality with a wide range of positive and negative responses.

To descriptively analyze the racial differences in opinions on the value of proximity or school diversity, we analyzed mean responses by race in Table 2. This table presents the means for each of our dependent variables for each of our racial groups, with the Chi-squared value with ties from the Kruskal–Wallis equality of populations rank test. The rows with stars represent the group with which the means are statistically significantly different, according to Dunn’s pairwise comparison. For each of our dependent variables, we have statistically significant differences between White and Black respondents, with Black respondents on average less strongly agreeing that students learn better with their neighbors or that busing is too hard on young children, although on average, they do still agree, and with Black respondents on average more strongly agreeing that students learn better in diverse schools and tend to agree that students should be assigned to schools in a way that reflects the diversity plan.

Table 2.

Average Responses by Race with Kruskal–Wallis Significance Test.

	Full sample			Always lived in Nashville
	White	Black	Other	White	Black	Other
Better with neighbors	1.68	1.33	1.35	1.81	1.31	1.07
K–Wallis	χ² = 25.317***			χ² = 29.227***
White		***	**		***	***
Black	***			***
Other	**			***
Busing too hard	1.52	1.35	1.38	1.75	1.47	1.34
K–Wallis	χ² = 8.154*			χ² = 14.315***
White		**			***	*
Black	**			***
Other				*
Better in diverse school	0.63	1.20	0.61	0.41	1.19	0.66
K–Wallis	χ² = 42.838***			χ² = 28.428***
White		***			***
Black	***		***	***		*
Other		***			*
Assign with diversity plan	−0.06	0.44	−0.16	−0.18	0.33	−0.28
K–Wallis	χ² = 65.398***			χ² = 27.191***
White		***			***
Black	***		***	***		***
Other		***			***
N	674	236	101	186	106	29

Note. Mean response by race. Always lived in Nashville represents respondents who lived in Davidson County their entire life. “Better with neighbors,” “busing too hard,” and “better in diverse school” are ordinal variables ranging from −2 to 2 with negative values disagreeing with the statement regarding kids and schools and positive values agreeing. “Assign with diversity plan” is an ordinal variable ranging from −1 to 1 with −1 being disagree and 1 agree. K–Wallis represents the group χ² estimate for the Kruskal–Wallis equality of populations rank test with ties with 2 d.f. for differences between at least two groups. Stars for white, Black, and other matrices represent the level of statistical significance for paired samples by racial category of Dunn’s Pairwise comparison test to indicate which pairs are statistically significantly different.

p < .05. **p < .01. ***p < .001.

For our proximity related dependent variables, in the cases where other race respondents had statistically significant differences, they on average agreed less strongly that students learn better with their neighbors or that busing is too hard on young children than White respondents. For our diversity outcomes, we found statistically significant differences between other race and Black respondents, with other race respondents agreeing less strongly that students learn better in diverse schools and tending to disagree that assignment of students to schools should reflect the diversity plan.

We examined the descriptive differences in mean responses by one’s experience as a youth in a district with or without mandatory desegregation. Table 3 presents the mean response to our proximity and diversity questions, separated by whether or not the respondent was of school-age when Nashville had court-ordered desegregation. Due to the ordinal nature of the dependent variables, we used the Wilcoxon-Mann–Whitney test of significance, and only found statistically significant differences for the question of whether students learn better with their neighbors. We found that respondents of school age during mandatory busing in Nashville agreed less strongly that students learn better with their neighbors than respondents who were not of school age during mandatory busing.

Table 3.

Average Responses by Indicator for School Age During Mandatory Busing.

	Always lived in Nashville
	Not during busing	During busing
Better with neighbors	1.72**	1.43**
Busing too hard	1.63	1.61
Better in diverse school	0.66	0.73
Assign with diversity plan	−0.07	0.03
N	164	157

Note. Mean response by whether the respondent was school age in Nashville during court-ordered desegregation. Always lived in Nashville represents respondents who lived in Davidson County their entire life. “Better with neighbors,” “busing too hard,” and “better in diverse school” are ordinal variables ranging from −2 to 2 with negative values disagreeing with the statement regarding kids and schools and positive values agreeing. “Assign with diversity plan” is an ordinal variable ranging from −1 to 1 with −1 being disagree and 1 agree. Statistical significance reflects the Wilcoxon-Mann–Whitney test.

**p < .01.

Value of Neighborhood Schools and Burden of Busing Results: RQ1

Table 4 presents exponentiated (odds ratio) ordinal logistic regression models for our “Better with Neighbors” outcome. Models with non-exponentiated coefficients, including coefficients for all controls, are available upon request.

Table 4.

Children Learn Best When They Attend School with Children From Their Own Neighborhoods: Ordinal Logistic Regression Models on Value of Proximity or Neighborhood Schools.

	(1)	(2)	(3)	(4)	(5)	(6)
	Full sample	Full sample	Always lived in Nashville	Always lived in Nashville	Always lived in Nashville interaction	Always lived in Nashville interaction
White (omit)
Black	0.580* (0.158)	0.586 (0.170)	0.205** (0.104)	0.280* (0.168)
Other	0.371 (0.144)	0.395* (0.165)	0.114* (0.119)	0.205 (0.236)
Not during busing (omit)
During busing			0.143 (0.073)	0.549 (0.454)
White × not during busing (omit)
White × during busing					0.285 (0.204)	0.617 (0.599)
Black × not during busing					0.613 (0.518)	0.296 (0.374)
Black × during busing					0.053*** (0.034)	0.171 (0.177)
Other × not during busing					0.559 (0.516)	0.806 (1.061)
Other × during busing					0.029 (0.033)	0.114 (0.163)
Controls		×		×		×
Cut 1	0.018*** (0.007)	0.072*** (0.045)	0.003*** (0.002)	0.036 (0.080)	0.005*** (0.004)	0.039 (0.087)
Cut 2	0.028*** (0.009)	0.113*** (0.068)	0.004*** (0.002)	0.049 (0.112)	0.007*** (0.005)	0.054 (0.123)
Cut 3	0.105*** (0.021)	0.435 (0.267)	0.007*** (0.004)	0.089 (0.212)	0.012*** (0.008)	0.098 (0.232)
Cut 4	0.355*** (0.053)	1.586 (1.016)	0.024*** (0.014)	0.372 (0.922)	0.045*** (0.024)	0.412 (1.011)
Observations	1,011	1,011	321	321	321	321

Note. Exponentiated coefficients; Standard errors in second row. Ordinal dependent variable ranging from −2 to 2 with negative values disagreeing with the statement that children learn best when they attend school with children from their own neighborhoods and positive values agreeing. Always lived in Nashville represents respondents who lived in Davidson County their entire life. “During busing” is an indicator for respondents of school age in Nashville during court-ordered desegregation. Controls include indicators for married, female, having a child in MNPS, have a child, Democrat, Republican, income, and level of education. Omitted categories: not married, male, no child in MNPS, no child, moderate, $50 to 70K income, high school graduate.

p < .05. **p < .01. ***p < .001.

As Table 4 indicates, we found very few statistically significant results in our models that include our control variables for responses to whether children learn better with their neighbors, and some statistically significant results in our naïve models without controls. In our full sample, we found, net of all variables, other race respondents had a 60.5% reduction in odds (or an odds of 0.395 less) than White respondents of strongly agreeing versus responding differently. In our sample limited to lifelong Nashville residents, we also found that Black respondents had, net of all variables, a 72% reduction in odds (or an odds of 0.28 less) than White respondents of strongly agreeing versus responding differently. Our models without controls also indicate lower odds of agreement for Black respondents relative to White respondents. This suggests that we have limited evidence of differences by race, with White respondents being more likely to strongly agree that students learn better with their neighbors. Table 4 shows a statistically significant decrease in the odds of agreeing for Black respondents who were school-age during mandatory busing relative to White respondents who were not school age in the interaction model that does not include control variables. We do not have any statistically significant results for the indicator of being of school age during busing overall or in an interaction model when controls are included. This could be due to the association being tied to one of the control variables (however the only control variables with statistically significant results are indicators for having school-age children and having a child in MNPS), or could be due to the small sample size with many controls.

Table 5 contains the exponentiated results of the ordinal logistic regression models for responses to whether school bus rides over 30 minutes long are too hard on young children. In these models, we found statistically significant differences in the odds of strongly agreeing versus responding differently for Black respondents versus White respondents in our always lived in Nashville sample when controls are included, with Black respondents having a 92.6% reduction in odds compared with White respondents of responding strongly agree versus responding differently. This implies that while Black respondents on average still agree, their odds of strongly agreeing are much lower than White respondents holding all else equal.

Table 5.

School Bus Rides Over 30 Minutes Long Are Too Hard on Young Children: Ordinal Logistic Regression Models on Burden of Busing.

	(1)	(2)	(3)	(4)	(5)	(6)
	Full sample	Full sample	Always lived in Nashville	Always lived in Nashville	Always lived in Nashville interaction	Always lived in Nashville interaction
White (omit)
Black	0.594 (0.160)	0.565 (0.181)	0.150 (0.083)	0.074*** (0.051)
Other	0.602 (0.244)	0.655 (0.255)	0.089** (0.075)	0.078* (0.083)
Not during busing (omit)
During busing			0.914 (0.472)	3.832 (4.302)
White × not during busing (omit)
White × during busing					1.641 (1.388)	4.631 (6.137)
Black × not during busing					0.272 (0.219)	0.095* (0.093)
Black × during busing					0.192** (0.117)	0.305 (0.415)
Other × not during busing					0.384 (0.398)	0.100 (0.143)
Other × during busing					0.108* (0.097)	0.328 (0.545)
Control		×		×		×
Cut 1	0.037*** (0.010)	0.128** (0.081)	0.003*** (0.003)	0.134 (0.608)	0.004*** (0.005)	0.144 (0.657)
Cut 2	0.092*** (0.019)	0.333 (0.205)	0.007*** (0.005)	0.316 (1.333)	0.010*** (0.007)	0.341 (1.444)
Cut 3	0.181*** (0.031)	0.693 (0.423)	0.030*** (0.019)	1.708 (6.959)	0.044*** (0.025)	1.842 (7.524)
Cut 4	0.418*** (0.060)	1.747 (1.082)	0.075*** (0.040)	5.110 (20.823)	0.110*** (0.054)	5.504 (22.489)
Observations	1,011	1,011	321	321	321	321

Note. Exponentiated coefficients; Standard errors in second row. Ordinal dependent variable ranging from −2 to 2 with negative values disagreeing with the statement that school bus rides over 30 minutes long are too hard on young children and positive values agreeing. Always lived in Nashville represents respondents who lived in Davidson County their entire life. “During busing” is an indicator for respondents of school age in Nashville during court-ordered desegregation. Controls include indicators for married, female, having a child in MNPS, have a child, Democrat, Republican, income, and level of education. Omitted categories: not married, male, no child in MNPS, no child, moderate, $50 to 70K income, high school graduate.

p < .05. **p < .01. ***p < .001.

In our interaction model with controls, we found reduced odds of strongly agreeing for Black respondents who were not of school age during busing as compared to White respondents who were not of school age during busing. Black respondents who were of school age during busing only respond in statistically significantly different ways from White respondents who were not of school age during busing in our models without controls. This suggests that the experience of living in a district with mandatory busing when one is of school age may be associated with different contemporary views on the burden of busing by race once other factors are controlled for. However, we must interpret carefully due to the small sample size, as differences could reflect racial differences that transcend experience as a student in a county under court-ordered desegregation that are lost when the small sample is split across many variables.

Value of Diversity in Schools Results: RQ2

It is expected that experience as a student in a city with court-ordered desegregation might influence one’s perspectives on the value of diversity in schools and specific diversity policy. This section explores the responses to two questions related to diversity in schools using ordinal logistic regression. Table 6 presents exponentiated coefficients for models of the response to whether children learn best when they attend racially diverse schools. We find in both the full sample and the lifelong Nashville resident sample that Black respondents have increases in odds of 229.5% and 754.1% respectively compared with White respondents of responding strongly agree versus responding differently when controls are included, holding all else equal. The large magnitude of difference implies sizable differences regarding the value of diversity in schools by race. Other race respondents have even larger differences in odds from White respondents in our lifelong Nashville resident sample. However, we recognize that while these may be statistically significant, the sample sizes are small for other race respondents. While we do not have statistically significant findings for being school-age during mandatory busing, we find an interesting result in the interaction between race and busing. Black respondents who were of school age during mandatory busing have a 573% increase in odds of strongly agreeing that students learn better in diverse schools versus White respondents who were not of school age during busing when controls are included. This result by itself is not surprising, given the previously mentioned evidence of differences between Black and White respondents. However, it is interesting that we do not find statistically significant results for Black respondents who were not of school age during busing, as compared to White respondents who were not of school age during busing, including in the model without control variables. Therefore, we have evidence of the difference in contemporary perspectives on the value of school diversity associated with being of school age during busing that differs by race.

Table 6.

Children Learn Best When They Attend Schools That Are Racially Diverse: Ordinal Logistic Regression Models on Value of Diversity.

	(1)	(2)	(3)	(4)	(5)	(6)
	Full sample	Full sample	Always lived in Nashville	Always lived in Nashville	Always lived in Nashville interaction	Always lived in Nashville interaction
White (omit)
Black	2.393 (0.553)	2.295*** (0.574)	5.898 (2.282)	8.686*** (3.734)
Other	1.064 (0.545)	1.005 (0.472)	17.839*** (14.650)	23.608*** (22.544)
Not during busing (omit)
During busing			0.830 (0.293)	0.646 (0.458)
White × not during busing (omit)
White × during busing					0.562 (0.226)	0.532 (0.408)
Black × not during busing					2.673 (1.841)	3.058 (2.259)
Black × during busing					4.342** (1.955)	5.730* (4.970)
Other × not during busing					1.260 (0.790)	1.685 (1.362)
Other × during busing					18.587** (17.309)	20.274* (24.275)
Controls		×		×		×
Cut 1	0.135*** (0.025)	0.099*** (0.055)	0.175 (0.068)	0.044 (0.093)	0.130*** (0.058)	0.076 (0.162)
Cut 2	0.257*** (0.039)	0.192** (0.103)	0.309*** (0.100)	0.080 (0.169)	0.232*** (0.083)	0.140 (0.297)
Cut 3	0.818 (0.104)	0.655 (0.351)	1.289 (0.398)	0.383 (0.792)	0.987 (0.303)	0.680 (1.419)
Cut 4	2.200*** (0.298)	1.903 (1.031)	5.112 (1.627)	1.755 (3.630)	4.033*** (1.288)	3.286 (6.910)
Observations	1,011	1,011	321	321	321	321

Note. Exponentiated coefficients; Standard errors in second row. Ordinal dependent variable ranging from −2 to 2 with negative values disagreeing with the statement that Children learn best when are in classrooms that are racially diverse and positive values agreeing. Always lived in Nashville represents respondents who lived in Davidson County their entire life. “During busing” is an indicator for respondents of school age in Nashville during court-ordered desegregation. Controls include indicators for married, female, having a child in MNPS, have a child, Democrat, Republican, income, and level of education. Omitted categories: not married, male, no child in MNPS, no child, moderate, $50 to 70K income, high school graduate.

p < .05. **p < .01. ***p < .001.

The exponentiated ordinal logistic regression results for our second dependent variable related to the use of a diversity policy in schools is presented in Table 7. Again, we found statistically significant differences in odds for Black versus White respondents in the full sample and the lifetime Nashville resident sample with controls, with Black respondents having 248.3% and 291.6% higher odds of agreeing that students should be assigned to schools in ways that take into account the district’s diversity plan than White respondents, all else equal. Again, we found no statistically significant results based on the indicator for being of school-age during mandatory busing. However, we did find an interaction effect with a significant effect associated with Black respondents who were of school age during busing as compared to White respondents not of school-age during busing, but no significant results for Black respondents who were not of school age during busing as compared to White respondents who were not of school age during busing. One cannot prove a null relationship, but we detect significant differences by race and differences sorted across race and age during busing.

Table 7.

Should Students Be Assigned to Schools in Metro Nashville in Ways That Take into Account the Metro Nashville Public Schools Diversity Plan? Ordinal Logistic Regression Models on Value of Diversity.

	(1)	(2)	(3)	(4)	(5)	(6)
	Full sample	Full sample	Always lived in Nashville	Always lived in Nashville	Always lived in Nashville interaction	Always lived in Nashville interaction
White (omit)
Black	3.031*** (0.817)	2.483** (0.694)	3.598** (1.438)	2.916** (1.167)
Other	1.096 (0.386)	0.984 (0.337)	3.019 (1.885)	2.546 (1.777)
Not during busing (omit)
During busing			1.760 (0.583)	2.123 (1.384)
White × not during busing (omit)
White × during busing					1.369 (0.535)	1.478 (1.032)
Black × not during busing					2.335 (1.236)	1.212 (0.656)
Black × during busing					5.860*** (2.838)	6.347* (4.774)
Other × not during busing					0.523 (0.413)	0.658 (0.590)
Other × during busing					5.401* (3.761)	5.238 (5.085)
Controls		×		×		×
Cut 1	0.617*** (0.080)	0.109*** (0.071)	1.428 (0.376)	0.385 (0.860)	1.202 (0.318)	0.343 (0.767)
Cut 2	1.698*** (0.226)	0.326 (0.206)	3.983*** (1.097)	1.168 (2.577)	3.377*** (0.865)	1.057 (2.329)
Observations	1,011	1,011	321	321	321	321

Note. Exponentiated coefficients; Standard errors in second row. Ordinal dependent variable ranging from −1 to 1 with negative values disagreeing with the statement that Children should be assigned to schools in Metro Nashville in ways that take into account the Metro Nashville Public Schools diversity Plan and positive values agreeing. Always lived in Nashville represents respondents who lived in Davidson County their entire life. “During busing” is an indicator for respondents of school age in Nashville during court-ordered desegregation. Controls include indicators for married, female, having a child in MNPS, have a child, Democrat, Republican, income, and level of education. Omitted categories: not married, male, no child in MNPS, no child, moderate, $50 to 70K income, high school graduate.

p < .05. **p < .01. ***p < .001.

Discussion and Implications of Findings

Our findings regarding differences in opinions on the value of racial diversity and the value of proximity, including differences across racial groups with and without experience as a student in a county under court-ordered desegregation, contribute to the literature on the long-term impacts of school desegregation. As mentioned above, there is extant literature on long-term impacts of attending a racially diverse school (Braddock, 2009; Gamoran et al., 2016; Kurlaender & Yun, 2001; Mickelson et al., 2012). There is further evidence of long-term benefits in the cases where there is exposure to diverse racial groups or inter-race interactions (Ellison & Powers, 1994; Mickelson et al., 2012; Sigelman et al., 1996; Stearns, 2010). Therefore, our results showing that Black respondents who were of school age during court-mandated desegregation may be more likely to believe students learn better in diverse classrooms are consistent with the existing literature on racial attitudes. We find results for Black students who were of school age during court-ordered desegregation without accounting for whether the respondent indeed ended up attending a diverse school or having interracial contact. This is consistent with theories on the complexity of race and the fact that race is shaped by one’s experiences in a school system where schools can be race-shaping institutions (Leonardo, 2013; Lewis, 2003; O’Connor et al., 2007).

Related to racial attitudes, our evidence of the average White respondent’s neutrality or agreement that students learn better in racially diverse schools, while disagreeing that diversity should be considered in assignment to schools is consistent with the literature on the racial attitudes paradox. We do not find that being school age in a county under court-ordered desegregation significantly moderates White respondents’ assessment of the value of diversity or of the use of practical diversity policy.

Unfortunately, schools today remain segregated, and many of the tools previously available to districts to attain school integration are no longer available, as districts under unitary status are no longer allowed to use race-conscious student assignment plans. Districts interested in racial diversity have a limited number of available options, known as the Kennedy remedies. Now that Kennedy has retired, there may be some questions about these options’ viability and how the current courts will interpret them.

There has been limited success of using diversity indexes or using income diversity as a proxy for racial diversity. However, these measures have been less successful than race-conscious plans (McMillian et al., 2018; Reardon et al., 2006). Diversity targeted magnet schools have previously led to increased levels of school diversity. However, without the ability to implement racial quotas, they are less successful (Smrekar, 2009), and even when they do have increased diversity it is sometimes at the expense of lower diversity levels in traditional public schools (Harris, 2018). Seeing as transportation and convenience remain two of the most essential school choice criteria for parents, where possible, strategic siting decisions that place schools conveniently between racially segregated neighborhoods could be fruitful (Smrekar & Honey, 2015). However, any school assignment policies will have to balance the desire for diversity with the practical realities of families prioritizing proximity and the continued possibility of White flight to private schools, schools of choice, or surrounding counties.

District plans also must face the changing demographics of urban public schools and school choice’s continued growth. As Horsford (2019) argues, districts must consider what school integration means today, as many public districts become majority non-White. Our results suggest that the Black residents of urban districts are more likely to agree that students learn better in diverse schools; thus, as urban districts become less White, it might present an opportunity for more families to be open to diversity-focused assignment systems. Additionally, the gentrification in urban areas may present districts’ opportunity to re-draw attendance boundaries in ways that allow for greater diversity in schools without forcing students to cross the city. Our results show White respondents, on average, were more likely to agree that students learn better with neighbors than to agree that they learn better in diverse classrooms. Thus, if attendance boundaries can be drawn in gentrified neighborhoods in ways that prioritize proximity while also leading to greater school diversity, this could present an opportunity for racial interest convergence. On the other hand, this may only be a temporary solution as non-White families are pushed out of gentrified neighborhoods, and White families may be reluctant to send their children to existing majority non-White neighborhood schools (Pearman, 2020). Diversity plans must consider how to balance diversity with convenience and efficient transportation access.

Conclusion

This paper asks two critical questions related to urban education: To what extent race and experience is as a student in a city under mandatory desegregation associated with contemporary views on the value of neighborhood schools and the burden of busing? To what extent is race and experience as a student in a city under mandatory desegregation associated with contemporary views on the value of diversity in schools? We find racial differences regarding the value of proximity and regarding the value of school diversity. We also find that one’s experience as a school-age resident in a city under mandatory school desegregation via busing moderates the racial differences found between White and Black respondents. These results can inform both future research on the influence of historical urban education policies on the potential of contemporary policy for urban districts that value school diversity while hoping to appeal to resident values.

This study has several limitations, as it is limited to one geographic context, and we are unable to follow people who may have either grown up in Nashville and moved away or people who grew up somewhere else and moved to Nashville. One’s experience in a desegregated school may influence one’s likelihood of moving, and future work should attempt to follow these students. It would be useful to know whether respondents attended a desegregated school or participated in mandatory busing, rather than limiting our analysis to those of school-age during court-ordered desegregation. It would be useful to have a deeper understanding of the educational experience of the respondents. Our sample size is limited, and further research should be conducted with a larger sample to detect more significant differences. Finally, it would be useful to have a sample of respondents who were too young to participate in mandatory busing to compare their perspectives with those who were school-age during mandatory busing and with those who were too old to be school-age during mandatory busing.

While busing is disliked, the average response falls between neutrality and agreement that students learn best in diverse schools. This appreciation for racial diversity is maintained among those whom themselves were students in Nashville when under court-ordered desegregation, suggesting even those who had to deal with the difficulties associated with busing still see value in school diversity. Beyond that finding, racial differences in opinion regarding the value of racial diversity in schools are moderated by whether the respondent was of school-age during busing, with Black respondents who were of school age during busing having higher odds of valuing diversity than White respondents who were not school age, while Black respondents who were not school age do not. Unfortunately, while residents are more likely to believe diversity is vital for student learning than to disagree, the district has been unable to increase the number of schools that meet the district’s definition of diverse since this survey was taken. In 2018, only 45% of district schools were considered diverse, unchanged from 4 years prior (MNPS, 2018).

Urban education researchers should explore the inter-generational impact of desegregation and resegregation that emerge from this research, with an eye to how the context of race, class, and place matter across families’ lived experiences, expectations, and decision-making. What are the consequential effects of these cumulative experiences across contexts that are shaped by school choice and student assignment policies, the legal landscape, and shifting demography?

Researchers must further examine whether urban residents still value school racial diversity. If so, under what conditions and constraints? Can urban districts find solutions to balance openness to school diversity with the realities of residential segregation, a desire for neighborhood connections and schools, and contemporary limitations on race-based policies? This paper examines trends in urban residents’ opinions on these factors. It sheds light on how desegregation histories shape contemporary public opinion, but many questions remain for developing district policies for diverse schools that meet resident priorities and realities while respecting the distinct historical, educational experiences families and students faced.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Associative Research Programme - ANID PIA CIE160007.

ORCID iD

Ngaire Honey

Author Biographies

Ngaire Honey is an associate researcher at the Center for Advanced Studies in Educational Justice, Pontificia Universidad Católica de Chile. Her work focuses on policies that shape educational access, segregation, and equity in a context of school choice with uneven geographic distributions of students and resources.

Claire Smrekar is associate professor of Public Policy and Education at Peabody College, Vanderbilt University, and Editor of the Peabody Journal of Education. Her work focuses on the social context of education, with specific reference to school-community partnership practices, school choice programs, and desegregation policy.

References

Bell

D. A.

(1980). Brown v. Board of Education and the interest-convergence dilemma. Harvard Law Review, 93(3), 518–533.

Billingham

C. M.

(2019). Within-district racial segregation and the elusiveness of White student return to urban public schools. Urban Education, 54(2), 151–181. https://doi.org/10.1177/0042085915618713

Bobo

Fox

(2003). Race, racismo, and discrimination: Bridging problems, methods, and theory in social psychological research. Social Psychology Quarterly, 66(4), 319–332.

Bonilla-Silva

(2003). Racism without racists: Color-blind racism and the persistence of racial inequality in America. Rowman & Littlefield Publishers.

Braddock

J. H.

(2009). Looking back: The effects of court-ordered desegregation (Chapter 1). In Smrekar

Goldring

(Eds), From the courtroom to the classroom: The shifting landscape of school desegregation. Harvard Education Press.

Carter

P. L.

(2019). A shade less offensive: School integration as radical inclusion in the pursuit of educational equity. Paper presented at the 16th Annual AERA Brown Lecture. https://www.youtube.com/watch?v=Mvxt9Q3Mf8g&feature=youtu.be&ab_channel=AmericanEducationalResearchAssociation

Castagno

A. E.

Lee

S. J.

(2007). Native mascots and ethnic fraud in higher education: Using tribal critical race theory and the interest convergence principle as an analytic tool. Equity and Excellence in Education, 40(1), 3–13.

Chang

Denson

Saenz

Misa

(2006). The educational benefits of sustaining cross-racial interactions among undergraduates. Journal of Higher Education, 77(3), 430–455.

Darling-Hammond

(2007). Third annual Brown lecture in education research—The flat earth and how America’s commitment to equity will determine our future. Educational Researcher, 36(6), 318–334.

10.

Eaton

(2001). The other busing story. Yale University Press.

11.

Ellison

C. G.

Powers

D. A.

(1994). The contact hypothesis and racial attitudes among Black Americans. Social Science Quarterly, 75(2), 385–400.

12.

Erickson

(2016). Making the uneven metropolis. University of Chicago Press.

13.

Frankenberg

Siegel Hawley

Orfield

(2017). Southern schools more than a half-century after the civil rights revolution. The Civil Rights Project.

14.

Frankenberg

(1993). White women, race matters: The social construction of whiteness. University of Minnesota Press.

15.

Gamoran

Collares

A. C.

Barfels

(2016). Does racial isolation in school lead to long-term disadvantages? Labor market consequences of high school racial composition. American Journal of Sociology, 121(4), 1116–1167. https://doi.org/10.1086/683605

16.

Goldring

Cohen-Vogel

Smrekar

Taylor

(2006). Schooling closer to home: Desegregation policy and neighborhood contexts. American Journal of Education, 112(3), 335–362.

17.

Halberstam

(1998). The children. Random House.

18.

Harris

J. C.

(2018). Integrating urban schools in a modern context: Roadblocks and challenges with the use of magnet schools. Urban Education. Advance online publication. https://doi.org/10.1177/0042085918802616

19.

Henry

P. J.

Sears

D. O.

(2002). The symbolic racism 2000 scale. Political Psychology, 23(2), 253–283.

20.

Holme

J. J.

Wells

A. S.

Revilla

A. T.

(2005). Learning through experience: What graduates gain attending desegregated high schools. Equity and Excellence in Education, 38(1), 14–24. https://doi.org/10.1080/10665680590907846

21.

Horsford

S. D.

(2019). School integration in the new Jim Crow: Opportunity or oxymoron? Educational Policy, 33(1), 257–275. https://doi.org/10.1177/0895904818810526

22.

Houvouras

S. K.

(2001). The effects of demographic variables, ethnic prejudice and attitudes toward immigration on opposition to bilingual education. Hispanic Journal of Behavioral Sciences, 23(1), 136–152.

23.

Huddy

Sears

D. O.

(1995). Opposition to bilingual education: Prejudice or the defense of realistic interests? Social Psychology Quarterly, 58(2), 133–144.

24.

Hurtado

(2006, June). Diversity effects on student cognitive and social cognitive outcomes. White paper prepared for the American Sociological Association Spivack Workshop, Arlington, VA.

25.

Jackman

M. R.

Muha

M. J.

(1984). Education and intergroup attitudes: Moral enlightenment, superficial democratic commitment, or ideological refinement? American Sociological Review, 49(6), 751–769.

26.

Johnson

R. C.

(2015). Long-run impacts of school desegregation & school quality on adult attainments (NBER Working Paper 16664). Cambridge, MA: National Bureau of Economic Research.

27.

Johnson

R. C.

(2019). Children of the dream. Basic Books.

28.

Karpinski

C. F.

(2006). Bearing the burden of desegregation: Black principals and Brown. Urban Education, 41(3), 237–276. https://doi.org/10.1177/0042085905284961

29.

Key

V. O.

(1949). Southern politics in state and nation. Knopf.

30.

Kimble

C. E.

(1980). Factors affecting adults’ attitudes toward school desegregation. Journal of Social Psychology, 110(2), 211–218. https://doi.org/10.1080/00224545.1980.9924248

31.

Kurlaender

Yun

(2001). Is diversity a compelling educational interest? Evidence from Louisville. In Orfield

(Ed.), Diversity challenged: Evidence on the impact of affirmative action (pp. 111–141). Harvard Education.

32.

Lankford

Loeb

Wyckoff

(2002). Teacher sorting and the plight of urban schools: A descriptive analysis. Educational Evaluation and Policy Analysis, 24(1), 37–62.

33.

Leonardo

(2013). Race frameworks: A multidimensional theory of racism and education. Teachers College Press.

34.

Levine

R. A.

Campbell

D. T.

(1972). Ethnocentrism: Theories of conflict, ethnic attitudes, and group behavior. Wiley.

35.

Lewis

A. E.

(2003). Race in the schoolyard: Negotiating the color line in classrooms and communities. Rutgers University Press.

36.

Lopez

G. R.

(2003). Interest convergence and desegregation in the Ohio Valley. Journal of Negro Education, 72(3), 269–296.

37.

McMillian

M. M.

Fuller

Hill

Duch

Darity

W. A.

(2018). Can class-based substitute for race-based student assignment plans? Evidence from Wake County, North Carolina. Urban Education, 53(7), 843–874.

38.

Metropolitan Nashville Board of Education. (1998). School board minutes, June 23.

39.

Metropolitan Nashville Public Schools (MNPS). (2016). 2015–16 Annual diversity report. https://static1.squarespace.com/static/57752cbed1758e541bdeef6b/t/57927bc1579fb3fb9d306e2f/1469217732838/2015_16_Annual%2BDiversity%2BReport.pdf

40.

Metropolitan Nashville Public Schools (MNPS). (2018). 2018–19 Annual diversity report. https://static1.squarespace.com/static/57752cbed1758e541bdeef6b/t/5d82508adf9b752886feda52/1568821399351/2018-19+Annual+Diversity+Report+083019.pdf

41.

Mickelson

R. A.

Nkomo

Wimberly

G. L.

(2012). Integrated schooling, life course outcomes, and social cohesion in multiethnic societies. Review of Research in Education, 36(1), 197–238.

42.

Milner

H. R.

(2012). But what is urban education? Urban Education, 47(3), 556–561.

43.

Milner

H. R.

(2017). Opening commentary: The permanence of racism, critical race theory, and expanding analytic sites. Peabody Journal of Education, 92(3), 294–301.

44.

Milner

H. R.

(2018, October, 25). Disrupting punitive practices and policies: Rac(e)ing back to teaching, teacher preparation, and Brown. Paper presented at the 15th Annual AERA Brown Lecture. https://www.youtube.com/watch?v=wBoF5pFHtDM&t=1502s&ab_channel=AmericanEducationalResearchAssociation

45.

Milner

H. R.

Lomotey

(2013). Handbook of urban education. Routledge Handbooks.

46.

O’Connor

Lewis

Mueller

(2007). Researching “black” educational experiences and outcomes: Theoretical and methodological considerations. Educational Researcher, 36(9), 541–552. https://doi.org/10.3102/0013189X07312661

47.

Parcel

T. L.

Mickelson

R. A.

Smith

S. S.

(2016). Using NSF funds to study differences in school desegregation in five southern systems. Footnotes, 44(2), 5–6.

48.

Pearman

F. A.

(2020). Gentrification, geography, and the declining enrollment of neighborhood schools. Urban Education, 55(2), 183–215.

49.

Pettigrew

T. F.

Meertens

R. F.

(1995). Subtle and blatant prejudice in Western Europe. European Journal of Social Psychology, 25(1), 57–75.

50.

Pride

(2000). Public opinion and the end of busing: (Mis)perceptions of policy failure. The Sociological Quarterly, 41(2), 207–225. https://doi.org/10.1111/j.1533-8525.2000.tb00092.x

51.

Pride

Woodard

(1995). The burden of busing: The politics of desegregation in Nashville, Tennessee. University of Tennessee Press.

52.

Quick

Damante

(2016). Louisville, KY: A reflection on school integration. The Century Foundation. https://tcf.org/content/report/louisville-kentucky-reflection-school-integration/

53.

Reardon

Grewal

Kalogrides

Greenberg

(2012). Brown fades: The end of court-ordered school desegregation and the resegregation of American public schools. Journal of Policy Analysis and Management, 31(4), 876–904.

54.

Reardon

S. F.

Yun

J. T.

Kurlaender

(2006). Implications of income-based school assignment policies for racial school segregation. Educational Evaluation and Policy Analysis, 28(1), 49–75.

55.

Roda

Wells

A. S.

(2013). School choice policies and racial segregation: Where white parents’ good intentions, anxiety, and privilege collide. American Journal of Education, 119(2), 261–293.

56.

Rossell

Armor

Wahlberg

(2002). Chapter 1: Assessing the promise of Brown. In Rossell

Armor

Walberg

(Eds.), School desegregation in the 21st century. Greenwood Publishing.

57.

Sears

D. O.

Hetts

Sidanius

Bobo

(2000). Race in American politics: Framing the debates. In Sears

D. O.

Sidanius

Bobo

(Eds.), Racialized politics: The debate about racism in America. Chicago University Press.

58.

Semuels

(2019, April 11). The utter inadequacy of America’s efforts to desegregate schools. The Atlantic. https://www.theatlantic.com/education/archive/2019/04/boston-metco-program-school-desegregation/584224/

59.

Sigelman

Bledsoe

Welch

Combs

(1996). Making contact? Black-White social interaction in an urban setting. American Journal of Sociology, 101(5), 1306–1332.

60.

Smrekar

(2009). Beyond the tipping point: Issues of racial diversity in magnet schools following unitary status. Peabody Journal of Education, 84(2), 209–226. https://www-jstor-org-443.web.bisu.edu.cn/stable/25594837

61.

Smrekar

Honey

(2015). The desegregation aims and demographic contexts of magnet schools: How parents choose and why siting policies matter. Peabody Journal of Education, 90(1), 128–155.

62.

Sniderman

P. M.

Carmines

(1997). Reaching beyond race. Harvard University Press.

63.

Stearns

(2010). Long-term correlates of high school racial composition: Perpetuation theory reexamined. Teachers College Record, 112(6), 1654–1678.

64.

Taylor

Anderson

Frankenberg

(2019). School and residential segregation in school districts with voluntary integration policies. Peabody Journal of Education, 94(4), 371–387.

65.

Torres

Weissbourd

(2020). Do parents really want school integration? https://mcc.gse.harvard.edu/

66.

Wells

A. S.

Crain

R. L.

(1999). Stepping over the color line. Yale University Press.

67.

Wells

A. S.

Duran

White

(2008). Refusing to leave desegregation behind: From graduates of racially diverse schools to the Supreme Court. Teachers College Record, 110(12), 2532–2570.

68.

Welner

K. G.

Spindler

E. R.

(2009). The post-PICS picture: Examining school districts’ policy options for mitigating racial segregation. In Smrekar

Goldring

(Eds), From the courtroom to the classroom: The shifting landscape of school desegregation. Harvard Education Press.

69.

Williams

(2010). Through the eyes of friends: An investigation of school context and cross-racial friendships in racially mixed schools. Urban Education, 45(4), 480–505.

70.

Williams

S. M.

Graham

(2019). Cross-racial interactions in schools 65 years after Brown. Peabody Journal of Education, 94(5), 545–554.

71.

Wood

Sonleitner

(1996). The effect of childhood interracial contact on adult anti-black prejudice. International Journal of Intercultural Relations, 20(1), 1–17.

72.

Woodward

J. R.

(2011). How busing burdened Blacks: Critical race theory and busing for desegregation in Nashville-Davidson County. The Journal of Negro Education, 80(1), 22–32.