Choice without Inclusion?: Comparing the Intensity of Racial Segregation in Charters and Public Schools at the Local, State and National Levels

Abstract

We conduct descriptive and inferential analyses of publicly available Common Core of Data (CCD) to examine segregation at the local, state, and national levels. Nationally, we find that higher percentages of charter students of every race attend intensely segregated schools. The highest levels of racial isolation are at the primary level for public and middle level for charters. We find that double segregation by race and class is higher in charter schools. Charters are more likely to be segregated, even when controlling for local ethnoracial demographics. A majority of states have at least half of Blacks and a third of Latinx in intensely segregated charters. At the city level, we find that higher percentages of urban charter students were attending intensely segregated schools.

1. Introduction

In its landmark Brown v. Board of Education decision in 1954 [1], the United States Supreme Court powerfully concluded that in the field of public education, the doctrine of ‘separate but equal’ had no place. Further, “separate educational facilities,” Chief Justice Earl Warren wrote for a unanimous court “are inherently unequal.” It has been over sixty years since the Supreme Court ruled unanimously in Brown to abolish the separate-but-equal legal doctrine and Jim Crow segregation by race. Yet, since that time, courts have allowed de facto segregation to flourish [2] and, as a result, schools in the United States are more segregated than they were at the time of the Brown decision [3].

The resegregation of the United States, in contravention of Brown, has occurred as a result of judicial retrenchment, but also due to other factors such as lax executive enforcement and White flight [4]. Not incidentally, during the past two decades, schools in the United States have become increasingly segregated by race and class. According to the national data, nowhere is the problem more acute than in the nation’s charter schools [5]. While public schools have generally acknowledged the problem and have usually agreed to remedies to address segregation [6], some charter supporters have sought to downplay the issue, emphasizing the need to provide greater choice to low income and minority students as a means of achieving an educational equity in outcomes regardless of the racial composition of the school [7]. In fact, some charter advocates have suggested that racial segregation within schools is acceptable if that comes as a natural by-product of parental choice [8].

Established nearly a quarter-century ago, the first taxpayer supported, privately-operated charter schools were conceived of as learning laboratories that might inspire curricular innovation [9]. In the past decade, proponents have reimagined charter schools as institutions of learning dedicated to providing poor and disadvantaged students with greater access to a high-quality education [10]. These viewpoints mask the serious issues of inequity that remain outstanding, even after the Supreme Court first declared that segregated schools were inherently unequal. More than 60 years after Brown, research confirms that charter and public schools servicing predominately poor students of color still do so with reduced resources, less academic rigor, in the form of limited access to advanced coursework, and largely untrained or inexperienced teachers [11].

Purporting to address the educational opportunity gaps in the U.S., school choice proponents have linked market-based educational approaches to the legacy of the Civil Rights movement by framing their movement to foster “education choice” as the greatest Civil Rights issue of our time [12]. However, substantiation on the claims of academic excellence proffered by charter advocates is mixed [13,14,15]. Opponents have been quick to point out a number of flaws in the rhetoric including the high degree of segregation within such schools [16]. They see the charter movement as a betrayal of the Brown decision in abdicating, through privatization and private-control of education, an essential function of government to provide education to citizens as a public good [17]. Critics have also been disapproving of the way in which the proliferation of charters has redirected crucial funding away from traditional public schools while, in many cases, reproducing and perpetuating the same racial imbalance Brown sought to correct [11].

According to US Department of Education, charters currently makeup only a small percentage of U.S. schools, approximately 7% [18]. Prior research using national data has found that they are the most segregated of the nation’s schools, especially for Black and Latinx (We use Latinx as an attempt to decolonize the Spanish language and neutralize gender [19]) students [20]. Many of the nation’s charters can even be classified as “apartheid schools”—a term coined by UCLA Professor Gary Orfield for schools with a White student enrollment of 1 percent or less [21]. School choice supporters often point out that while neighborhood segregation is out of their control—although in some states charter schools can use neighborhood borders to fix enrollment—the reality is that most charter schools have not prioritized or experienced desegregation as a desired outcome [22]. While geography and residential segregation patterns contribute to the segregation in charter schools, in reality the schools with the most flexibility, hypothetically, to achieve significant diversity, have instead apparently chosen not to address the problem [23,24]. Are charters more segregated than public schools at the local, state and national levels? If so, does local demography explain why charter schools feature more racial isolation than public schools? The answers to these questions are clear in this study—national, state, and local data indicate that the charter industry has a segregation problem in the US and it is not simply explained away by locality or demography.

2. Literature

Charter schools have, since their inception, been billed as an opportunity to innovate and offer higher-quality educational opportunities in the United States. Initially conceived as a way to experiment with new forms of pedagogy while freed from the bureaucratic burdens of public schools, charters even gained significant support from teacher unions [25]. Fast forward to 2018, and the now nearly three-decades old history of charters reveals a more problematic story than originally envisioned. As discussed above, the Brown decision made clear that there was a public obligation to integrate schools and, despite our best efforts, both charter and public schools in the United States have not come to realize that goal [26].

2.1. Charter Schools and Ethnoracial Segregation

In the public discourse, there is a historical and contemporary debate on whether school choice options such as charters have fomented segregation along racial and class lines. And, if there is differential segregation, to what extent is it happening and, similarly, why it is happening. And while the integration problem is not isolated among charter schools, the predominance of the literature on charters suggests that they are more racially segregated than public schools [16,20,21,27,28,29]. For example, Frankenberg and Siegel-Hawley’s analysis of national data from 2008 found that charters located throughout the country and in large metropolitan areas were characterized as being “more racially isolated than traditional public schools” and “while examples of truly diverse charter schools exist, data show that these schools do not reflect broader charter trends” (2011, p. 45). Garcia (2010) likewise posited that “there is considerable evidence that charter schools may over represent minority students compared with [public] schools” (p. 33).

School choice proponents have long suggested that charters and other forms of school choice (e.g., school vouchers) do not cause or exacerbate segregation along racial and class lines; however, such claims have seemingly shifted in recent years [30]. For example, the American Legislative Exchange Council (ALEC), a coalition of legislators, businesses, and foundations that has consistently promoted private control and privatization model legislation (i.e., vouchers, charters) for education, acknowledged in 2015 that its support for school vouchers had, really, always been an effort to afford White affluent families the ability to self-segregate away from predominantly non-White schools [31]. The similarly conservative American Enterprise Institute (AEI) conducted a study of the entire universe of charter schools in the United States concluding that parents were self-segregating along racial and class lines but that such segregation was simply a result of a “well-functioning education market” that was responsive to the apparently different desires of schooling options based on race and class [8]. Swanson [32] summarized peer reviewed research on “the impact of charters on racial integration” conducted between 1994 and 2010 and found different levels of segregation between regions and concluded “it is important to recognize” that stratification in charters is due to “deliberate efforts” to create homogenous schools (p. 520).

Charter school lobbyist have concurred with the conservative think tanks, arguing that it doesn’t matter that charter schools, writ large, have not assuaged segregation. The National Alliance for Public Charter Schools has responded to segregations concerns with “so what?” while also suggesting that “in the end, parents’ and students’ opinions are the only ones that matter. And every year, more parents are choosing charter schools” [33]. In essence, school choice proponents are now arguing that charters are more segregated, but claim that segregation is okay because it’s a deliberate effort to “reach underserved students” and as long as that is what parents want through their exercising of free-market practices.

2.2. Charter Schools, Student Achievement, and Diversity

Frankenberg and Siegel-Hawley [20] suggested that much of the expansion of charters has overlooked the impact they have had on the magnitude of segregation due to claims that charters provide better outcomes when compared to their public school counterparts. Policymakers and school choice supporters, such as President Donald Trump and U.S. Secretary of Education Betsy DeVos, have argued that the crux of education reform strategies is leveraging school choice to improve educational outcomes for all students [34] rather than consideration of diversity and integration [35]. In fact, when considering the extant literature on school performance comparisons, the minority of charter schools, at best, provide minimal academic benefits whereas the majority underperform public schools [13,14,15].

The Center for Research on Education Outcomes (CREDO) produces probably the most cited non-peer reviewed research in policy conversations and the public discourse about charter schools. CREDO’s studies usually find that students in charter schools display slightly greater overall gains (typically in the tenths and the hundredths of a standard deviation) in performance than their peers in matched public schools [36]. However, there are methodological issues to consider when comparing public and charter school achievement. On the face of it, CREDO and other pro-charter research that compares the outcomes of students enrolled in a public school relative to their peers in charter schools ignore the student parsing effects of charters that introduce bias. Accordingly, school choice proponents are not able to account for the fact that there are differences between the opportunities, and choices, available to families who attend charter schools which makes comparison between public and charter schools achievement outcomes problematic. Frankenberg [37] pointed out that the ability to exercise “choice” in educational options varies quite a bit depending on quite a few factors. Namely among those factors are the geographic location of charters (and a parent’s access to transportation), the level and quality of information that a parent has about the available school choice options, and the level to which a charter school can, or can’t, provide the same necessary support structures such as special education and language accommodations that public schools are required to provide.

Rotberg [38] noted that because charter schools don’t typically produce appreciably better academic outcomes on average, more focus should be placed on informing policy decisions based on what we do know: that charters are more segregated. Rotberg does point out that small cases of integration exist in some charter schools that enact intentional policies towards those goals, but that the number of such charters pales in comparison to the vast majority of charters who ignore resulting segregation. Recently, there have been efforts—funded by the Walton Family Foundation, which has a long history of promoting market-based reforms throughout every level of educational policy—to promote charters as “diverse by design.” However, by their own measurement, charters that are intentionally diverse only make up about 2 percent of all charter schools [23].

Some studies have sought to explain away previous findings showing a lack of focus on diversity and a greater intensity of segregation in charter schools by reframing the conversation to focus on segregative growth instead of the magnitude of segregation [39]. Concurrently, market-based school choice proponents have sized upon that argument and also posited that because charters intentionally locate themselves in predominately non-White areas that questions about segregation should be asked by comparing the racial demographics of charter schools with the public schools that are located next to them [7]. However, it is important to note that finer grained geospatial analyses have also found—even when comparing schools that are located near each other—that charter schools are more segregated than nearby public schools [40]. Using Geographic Information Systems (GIS) software, Lubienski, Gulosino, and Weitzel [41] argued that charters may actually decide where to locate based on a goal of attracting higher achieving students while also maintaining lower costs. LaFleur [42] found a similar pattern of charter school placement among highest-need tracts in Chicago and noted that “charter school operators may have some disincentive to locate in a highest-need tract” (p. 11).

2.3. The Role of Charters in Education Reform

Furthermore, evidence in the field has suggested that the hyper-focus of market-based education reformers on locating charters in non-White communities has reinforced colonialistic ideologies [43] about Black and Brown students whose “no-excuses” charter schools often employ strict behaviorism and militaristic pedagogy that results in what Jim Horn [44] posited is a type of cultural eugenics. Reminiscent of Native American schools like the Carlisle Indian School that sought to “kill the Indian, save the man” [45], the isolation and concentration of non-Whites into charter schools that rely on punitive, arbitrary, and culturally insensitive practices [46] are able to redirect the onus of ending the effects of poverty—regardless of persisting inequality and segregation—on the shoulders of individual students through meritocracy; thereby justifying inequalities [47,48].

Given the penchant for charters to locate within Black and Brown urban neighborhoods as a method of “reforming” underfunded local public schools, and that one in eight Black students in the United States are now enrolled in charter schools [11]—it is important to revisit charter school data at the national, state, and local levels to provide a broad based understanding of demographics of students who attend charters and public schools. Examining data at the local and state level is also especially important because charters have focused their expansions in predominantly non-White regions and urban areas. New Orleans, by way of example, became nearly a 100 percent charter district following the aftermath of Hurricane Katrina in 2005. In what follows we provide national, state, and local analyses to illuminate the scale of the matter at each level.

3. Method

3.1. Descriptive Analyses

In this study, we conducted descriptive analyses of publicly available school-level Common Core Data (CCD), from the National Center for Education Statistics (NCES). The US Department of Education’s CCD is a comprehensive, annual, national database of all public elementary and secondary schools and districts. The wealth of information gathered in the CCD presents the opportunity to examine segregation at the local, state and national levels. The most recent data available at the time of writing was for 2015–2016.

We modeled our descriptive analysis tables and recoded the CCD into categorical variables in similar fashion to update and compliment earlier research examining segregation in charters and public schools previously conducted by Frankenberg, Siegel-Hawley, and Wang (2011). We began our work by recoding CCD variables to represent school-level proportions of White and non-White students. Utilizing the SPSS recode variable feature, we aggregated students in each school into categorical variables denoting school-level non-White majorities (i.e., 99%–100% non-White, 90%–90% non-White, etc.). We used [5,49] to define the term intense segregation as 90% or more non-White in a school.

We compared the intensity of racial segregation in charter and public schools at the local, state and national levels. In our first analysis, we compare the national percentage of public and charter school students in segregated non-White schools (N = 91,320) by race/ethnicity. In our second analysis, we aggregated data at the state-level (N = 43) using Excel pivot tables for White, Black and Latinx students to understand the proportion that are “intensely segregated” in each state’s charter and public schools. Notably, not all states were included in the analyses because charter school data is not available in all 50 states for 2015–2016, as eight states did not allow charter schools (Alabama, Kentucky, Montana, Nebraska, North Dakota, South Dakota, Vermont, and West Virginia). In our third analysis, we used Excel pivot tables to aggregate data for US cities with the largest number of students enrolled in charters (N = 50) to compare intense segregation at the local level for charter and public schools. In our fourth descriptive analysis, we examined segregation at the school-level using the CCD school level variable (high school, middle, elementary, and multilevel) and also created and used a majority non-White dichotomous variable in a crosstabs analysis to compare proportions of segregation in charter and public schools (N = 91,320). The final descriptive analyses were also undertaken using crosstabs to consider proportion of segregation for a dichotomous majority Free and Reduced Lunch (FRL) variable and a dichotomous majority non-White variable at the national level (N = 84,477).

3.2. Multivariate Linear Regression

We also conducted an inferential analysis to understand the segregation of students in U.S. schools. In a multivariate linear regression we calculated the relationship between a school-level percent White dependent variable tabulated from the 2015–2016 CCD for all public and charter schools and percent White in a geographic area and school type as the predictor variables. To include local ethnoracial demographics from the local level in the model, we used SPSS to match ZIP Code Tabulation Areas (ZCTAs) to the school location zip code in the CCD. ZCTAs are calculated by the US Census Bureau to be generalized areal representations of zip code areas. We included school type by using the dichotomous CCD variable denoting charter or public school (N = 89,189). Our basic linear regression model is,

Y_i = β₀ + β₁X₁ + β₂X₂ + ε

In summary, the following variables were included as independent variables in the multivariate linear regression: percent White at the local level and school type (charter and public). The dependent variable is percent White at the school-level.

Of note, in this study we refer to charters and public schools as different entities. Charter schools reimagine public education into private functionalist conceptions of schooling [50]. To that point, Miron and Nelson [50] argued that charters should not be understood as public schools. The traditional definition of public is the formalist definition, which focuses on issues of control and ownership. According to this definition, “a school (or other institution) is public if it is owned or controlled by citizens or their duly elected representatives” [50]. As a result, in this study we do not refer to charters as public. Charter schools, which are overseen privately, do not align with this conception of public, or necessarily primarily focus on the common good. Schooling is reclassified in market-based education reform as an individualistic commodity and then an inadvertent manifestation of the common public good.

4. Findings

4.1. National Level Analyses

The 2015–2016 CCD show that higher percentages of charter school students of every race attend intensely segregated schools (99%–100%) and less attend predominately White schools (0%–50%) than do their same-race peers in neighborhood public schools (see

Table 1. Percentage of Charter and Public School Students in Segregated Non-White Schools, by Race/Ethnicity, 2015–2016.

	Public				Charter
	99%–100% Non-White	90%–99% Non-White	50%–90% Non-White	0%–50% Non-White	99%–100% Non-White	90%–99% Non-White	50%–90% Non-White	0%–50% Non-White
Latinx	8%	33%	39%	20%	20%	36%	30%	13%
Black	11%	27%	41%	21%	39%	30%	21%	10%
White	0%	1%	18%	80%	0%	3%	24%	74%
Native	10%	15%	35%	40%	13%	16%	31%	40%
Asian	1%	18%	46%	34%	9%	16%	46%	30%

Source: 2015–2016 NCES Common Core of Data.

). Nationally, the higher levels of segregation for charter school students is particularly noticeable for Black students, who are more than three times as likely to attend racially isolated charter schools. Also, Latinx students are more than twice as likely to attend racially isolated charter schools when compared to neighborhood public schools.

Thirty percent of Black students in charter schools attended 90%–99% non-White schools in 2015–2016. Frankenberg et al. [5] found that the segregation of Black students has been increasing since 1990 across the country and reached its highest level in nearly four decades. Approximately ten years later, the segregation of Black students in charter schools continue to outpace neighborhood public schools. Charter schools enroll about 31% more in schools that are 90%–100% non-White (see Table 1). The percentage of Black charter students are in racially isolated non-White schools (69% in 2015–2016) has remained stubbornly high over the last two decades as Frankenberg et al. [5] relayed that 70% of Black students in charter schools were in 90%–100% non-White schools in 2000–2001.

Charter segregation has increased for Latinx students since 2000. In 2015–2016, 56% were in schools with 90% or more non-White compared to about half in 2007–2008. Further, more than half of charter and public school students from Black, Latinx and Asian American backgrounds attended predominantly non-White schools. Also, a higher percentage of Black, Latinx and Asian American students were in 50%–100% non-White charter schools than in predominantly non-White public schools (see Table 1). In the case of Black and Latinx, about 90% charter students were in segregated non-White schools. Notably, like their public school counterparts, Asian American charter school students were the least likely of all non-White students to be enrolled in segregated non-White schools.

Many charter students attended schools where 99% or more of the students were Non-White. About two-fifths of Black charter school students attended such extremely segregated non-White schools, a percentage which was the highest of any other ethnoracial group, and more than three times as high as Black students in public schools. Latinx charter students were more than twice as likely to be in these almost totally segregated non-White schools. Asian American students were considerably more likely to attend virtually all (99% or more) non-White charter schools than were their same race peers in public schools.

In the 2015–2016 CCD data, Native Americans were nearly equally segregated in charter and public schools. About 25% of Native American students were segregated in schools that were 90% or more non-White students compared to 29% in charter schools. About 60% of Native American students attended majority non-White schools in both the public and charter school sectors. Finally, 10% of Native American students were intensely segregated in public schools that were 90% or more non-White compared to 13% of charter school students.

4.2. State Level Analyses

An intriguing dimension of this higher segregation of students in charter schools was the extent to which the gap between charter and public manifests in different states relative to ethnoracial demographics (see

Table 2. Percentage Gap of Charter and Public School Students in Intensely Segregated Non-White Schools, by Race/Ethnicity and by State, 2015–2016.

	Latinx Gap		Black Gap		White Gap
Largest	MS	88%	CT	69%	MS	99%
	MO	79%	MN	59%	IL	39%
	CT	61%	MS	55%	TN	27%
	MN	56%	TN	54%	CT	18%
	IL	54%	DE	54%	NY	17%
Smallest	GA	−10%	AZ	−4%	VA	−1%
	VA	−10%	AK	−8%	NM	−2%
	KS	−11%	HI	−8%	NV	−3%
	HI	−16%	KS	−12%	CA	−4%
	NV	−21%	VA	−18%	HI	−19%

Source: 2015–2016 NCES Common Core of Data.

). For example, Connecticut and Mississippi were both in the top five largest gaps for White, Latinx and Black students (see

Table 3. Percentage of Charter and Public School Students in Intensely Segregated Non-White Schools, by Race/Ethnicity and by State, 2015–2016.

	Average White Share of Enrollment			90%–100% Non-White Charter School Enrollment Rate			90%–100% Non-White Public School Enrollment Rate
	Charter	Public	Charter-Public Difference	White	Black	Latinx	White	Black	Latinx
AK	63%	38%	24%	1%	10%	5%	0%	2%	7%
AR	45%	65%	−20%	0%	25%	6%	1%	35%	17%
AZ	42%	43%	−1%	3%	27%	42%	2%	23%	37%
CA	30%	27%	3%	7%	49%	57%	3%	55%	55%
CO	51%	56%	−5%	1%	22%	17%	1%	28%	33%
CT	14%	56%	−42%	1%	25%	19%	19%	94%	80%
DC	11%	12%	0%	7%	83%	73%	8%	90%	63%
DE	36%	45%	−9%	0%	7%	3%	2%	61%	31%
FL	33%	40%	−7%	2%	35%	31%	4%	41%	44%
GA	34%	41%	−7%	2%	45%	31%	1%	43%	21%
HI	23%	12%	11%	23%	28%	52%	4%	20%	36%
IA	54%	82%	−28%	0%	1%	2%	0%	0%	0%
ID	83%	76%	7%	0%	0%	0%	0%	0%	0%
IL	8%	56%	−48%	1%	55%	43%	40%	95%	97%
IN	35%	73%	−38%	0%	25%	11%	2%	66%	38%
KS	76%	71%	5%	0%	12%	11%	0%	0%	0%
LA	17%	45%	−28%	1%	33%	11%	4%	82%	48%
MA	33%	64%	−31%	1%	22%	25%	5%	58%	61%
MD	16%	40%	−24%	2%	53%	44%	8%	78%	36%
ME	94%	92%	2%	0%	0%	0%	0%	0%	0%
MI	35%	71%	−36%	0%	37%	10%	2%	75%	46%
MN	47%	70%	−23%	0%	14%	7%	1%	73%	63%
MO	14%	76%	−62%	0%	38%	7%	13%	82%	86%
MS	1%	42%	−41%	1%	45%	12%	100%	100%	100%
NC	53%	49%	4%	1%	22%	16%	0%	41%	31%
NH	89%	89%	0%	0%	0%	0%	0%	0%	0%
NJ	15%	51%	−35%	1%	44%	42%	13%	94%	84%
NM	29%	25%	5%	6%	16%	35%	4%	13%	34%
NV	41%	39%	3%	3%	26%	31%	0%	34%	10%
NY	7%	48%	−41%	2%	61%	54%	19%	91%	87%
OH	34%	74%	−40%	0%	32%	6%	3%	66%	22%
OK	33%	52%	−19%	0%	13%	9%	4%	52%	59%
OR	76%	66%	10%	0%	0%	1%	0%	19%	0%
PA	28%	72%	−44%	0%	38%	26%	2%	73%	65%
RI	30%	64%	−34%	2%	29%	44%	12%	66%	62%
SC	54%	48%	7%	1%	18%	6%	1%	30%	21%
TN	8%	69%	−61%	0%	39%	11%	27%	93%	56%
TX	17%	33%	−15%	4%	42%	53%	11%	69%	77%
UT	75%	75%	0%	0%	2%	2%	0%	2%	12%
VA	53%	54%	−1%	1%	18%	10%	0%	0%	0%
WA	31%	59%	−27%	0%	9%	15%	4%	35%	33%
WI	68%	75%	−6%	0%	41%	11%	1%	81%	54%

Source: 2015–2016 NCES Common Core of Data.

). Mississippi exhibited the largest gap for Latinx and White students while Connecticut was the topmost for Black students. Minnesota (Latinx and Black), Illinois (Latinx and White) and Tennessee (Black and White) showed the largest gaps between intensely segregated charter and public schools for two of the three groups.

At the state level, Virginia and Hawaii had the smallest gaps in terms of a heavy concentration. that favored charter schools—in those states public schools had heavier concentrations of segregation. Nevada (Latinx and White), Kansas (Latinx and Black) showed the smallest gaps between intensely segregated charter and public schools for two of the three groups. Notably, there are smaller gaps between intensely segregated White and non-White schools overall—only Hawaii exhibited a double digit gap (19%) signifying that public schools were more intensely segregated than charter schools.

4.3. City Level Analyses

Reflecting national and state trends, in the top 50 cities with the largest charter enrollments, higher percentages of charter students were attending segregated non-White schools than urban students in public schools (see

Table 4. Public and Charter School Students in Intensely Segregated Non-White Schools by City, 2015–2016.

City	Public	Charter	Charter-Public Difference
Cincinnati, OH	19.4%	88.7%	69.3%
Rochester, NY	22.8%	86.0%	63.2%
Minneapolis, MN	24.9%	81.5%	56.6%
Kansas City, MO	28.4%	84.9%	56.5%
Camden, NJ	48.8%	97.5%	48.7%
Nashville, TN	12.0%	52.2%	40.2%
Newark, NJ	44.4%	83.9%	39.5%
Brooklyn, NY	50.6%	88.8%	38.2%
Washington, DC	45.1%	78.9%	33.8%
New York, NY	59.3%	92.4%	33.1%
Austin, TX	32.2%	63.0%	30.8%
Chicago, IL	71.0%	98.9%	27.9%
San Antonio, TX	49.2%	76.1%	26.9%
Milwaukee, WI	61.3%	84.8%	23.5%
St Louis, MO	39.3%	62.2%	23.0%
San Francisco, CA	50.5%	73.3%	22.9%
Gary, IN	77.2%	100.0%	22.8%
Dallas, TX	68.7%	91.4%	22.7%
Baltimore, MD	49.1%	68.7%	19.7%
Chula Vista, CA	50.9%	70.0%	19.2%
Memphis, TN	81.0%	100.0%	19.0%
Cleveland, OH	29.7%	48.1%	18.4%
Denver, CO	42.7%	59.9%	17.3%
Oakland, CA	63.1%	80.2%	17.2%
El Paso, TX	83.3%	100.0%	16.7%
Philadelphia, PA	57.0%	72.9%	15.9%
San Jose, CA	55.8%	69.9%	14.1%
Baton Rouge, LA	61.1%	73.5%	12.4%
Saint Paul, MN	42.4%	53.0%	10.6%
Bronx, NY	90.2%	100.0%	9.8%
Brownsville, TX	90.5%	100.0%	9.5%
Indianapolis, IN	19.3%	28.0%	8.7%
San Diego, CA	34.7%	43.4%	8.7%
Houston, TX	73.1%	81.5%	8.4%
Sacramento, CA	39.8%	47.5%	7.7%
Southfield, MI	93.8%	100.0%	6.2%
Miami, FL	76.1%	79.5%	3.4%
Columbus, OH	29.9%	31.3%	1.4%
Detroit, MI	93.5%	93.7%	0.2%
Doral, FL	100.0%	100.0%	0.0%
Hialeah Gardens, FL	100.0%	100.0%	0.0%
Huntington Park, CA	100.0%	100.0%	0.0%
Los Angeles, CA	84.1%	82.5%	−1.6%
Tucson, AZ	30.3%	23.7%	−6.6%
Inglewood, CA	100.0%	91.9%	−8.1%
Stockton, CA	61.7%	46.1%	−15.5%
Phoenix, AZ	53.3%	37.7%	−15.6%
New Orleans, LA	93.7%	77.4%	−16.4%
Homestead, FL	90.1%	61.3%	−28.8%
Atlanta, GA	59.0%	18.5%	−40.6%

Source: 2015–2016 NCES Common Core of Data.

). The largest difference between the percent of students attending intensely segregated non-White schools were Cincinnati (69%), Rochester (63%), Minneapolis (57%), Kansas City (57%) and Camden (49%). The level of segregation in charter schools in the cities with the largest charter-public difference ranged from 81% to 98% of students in the charter sector compared to 19 to 49% of public school students attending intensely segregated schools.

Forty-two of the nation’s top 50 cities serving the most charter students had lower percentages of students intensely segregated in their public schools (see Table 4). However, there were 8 cities where CCD showed that public schools were more intensely segregated than charter schools. The choice of White students to attend charters relative to other urban areas may explain why public schools were more segregated and charter schools less segregated in some cities. About half of White students in Atlanta and Homestead, a quarter in Phoenix, and a fifth in Tucson attended charter schools. Other potential explanations could be that charter schools are serving as havens for students trying to avoid segregation in public schools. For example, Tucson Unified School District (TUSD) was placed under court ordered desegregation in 1976 and LAUSD is currently under a voluntary order with the US Department of Education from 2008 and 2013. Also, the difference could be explained by the preference of students in those cities to attend charter schools with either larger proportions of White or non-White students. Phoenix and Atlanta had the first and third, respectively, highest enrollments of White students in charters in the United States.

New Orleans appears to be a different case altogether. In 2003, the Louisiana legislature created the Recovery School District (RSD). With this law, schools that did not meet “minimum academic standards” were to be taken over by the state. Soon after Hurricane Katrina, in November 2005, the Louisiana legislature passed Act 35. The new law lowered the academic criteria that made a school eligible for charter takeover and empowered the state to reconstitute more than 100 “low performing” public schools. The RSD was given the vast majority of New Orleans public schools, leaving just a few schools to be run by the Orleans Parish School Board. Ten years after Katrina, about 97% of White, 91% of Black and 96% of Latinx students attend a charter school in the city of New Orleans. That meant only about 170 White students, 120 Latinx and 3500 Black students attended public schools in the city in 2015–2016 (results not shown). The few public schools that remained (i.e., Bethune, Franklin elem., Jackson, McDonogh and McMain) are intensely segregated and serve primarily Black students).

Several California cities in the CCD have a negative charter-public difference when comparing students attending intensely segregated non-White schools. Stockton is an interesting case because the schools there were required by court order to desegregate in 1974. The CCD shows that charters in Stockton are about 16% less intensely segregated than the public schools. While Orfield and Ee [3] found that Stockton is one of California’s least segregated metro areas, there are clearly remnants of the purposeful segregation in the city as 61.7% of students are still attending public schools that are intensely segregated non-White schools. Stockton is your classic impoverished, overwhelmingly minority school system. They found,

“The [W]hites and Asians [Americans] who remain in the Stockton school district on average attend schools with more than two-thirds combined enrollment of Latin[x], [B]lacks, and [Native Americans].”

[3]

Orfield and Ee [3] also relayed that the most segregated districts in California are located in the Los Angeles-Inland Empire Region—which explains why students attending charter schools in Inglewood (8%) and Los Angeles (1.6%) that are only minimally less segregated than public schools. One notable caveat, the data in Inglewood show the scale of the difference—only about 450 students attended non-intensely segregated charter schools.

Table 5. US Public and Charter Schools by Level and Majority Non-White.

			Primary	Middle	High	Multilevel	Total
Public	Majority Non-White	Count	21,613	5948	6557	2149	36,267
		% within Non-White	59.6%	16.4%	18.1%	5.9%	100.0%
		% within School level	45.8%	39.7%	37.7%	43.8%	42.9%
		% of Total	25.6%	7.0%	7.8%	2.5%	42.9%
	Majority White	Count	25,625	9026	10,822	2763	48,236
		% within Non-White	53.1%	18.7%	22.4%	5.7%	100.0%
		% within School level	54.2%	60.3%	62.3%	56.3%	57.1%
		% of Total	30.3%	10.7%	12.8%	3.3%	57.1%
	Total	Count	47,238	14,974	17,379	4912	84,503
		% within Non-White	55.9%	17.7%	20.6%	5.8%	100.0%
		% within School level	100.0%	100.0%	100.0%	100.0%	100.0%
		% of Total	55.9%	17.7%	20.6%	5.8%	100.0%
Charter	Majority Non-White	Count	2172	546	964	768	4450
		% within Non-White	48.8%	12.3%	21.7%	17.3%	100.0%
		% within School level	68.0%	76.5%	65.7%	53.3%	65.3%
		% of Total	31.9%	8.0%	14.1%	11.3%	65.3%
	Majority White	Count	1023	168	504	672	2367
		% within Non-White	43.2%	7.1%	21.3%	28.4%	100.0%
		% within School level	32.0%	23.5%	34.3%	46.7%	34.7%
		% of Total	15.0%	2.5%	7.4%	9.9%	34.7%
	Total	Count	3195	714	1468	1440	6817
		% within Non-White	46.9%	10.5%	21.5%	21.1%	100.0%
		% within School level	100.0%	100.0%	100.0%	100.0%	100.0%
		% of Total	46.9%	10.5%	21.5%	21.1%	100.0%
Total	Majority Non-White	Count	23,785	6494	7521	2917	40,717
		% within Non-White	58.4%	15.9%	18.5%	7.2%	100.0%
		% within School level	47.2%	41.4%	39.9%	45.9%	44.6%
		% of Total	26.0%	7.1%	8.2%	3.2%	44.6%
	Majority White	Count	26,648	9194	11,326	3435	50,603
		% within Non-White	52.7%	18.2%	22.4%	6.8%	100.0%
		% within School level	52.8%	58.6%	60.1%	54.1%	55.4%
		% of Total	29.2%	10.1%	12.4%	3.8%	55.4%
	Total	Count	50,433	15,688	18,847	6352	91,320
		% within Non-White	55.2%	17.2%	20.6%	7.0%	100.0%
		% within School level	100.0%	100.0%	100.0%	100.0%	100.0%
		% of Total	55.2%	17.2%	20.6%	7.0%	100.0%

Source: 2015–2016 NCES Common Core of Data.

shows the demographics of charter and public schools by level in US when non-White proportions are combined. As much as 43% of all public schools in US are majority non-White compared to 65% of charter schools. As a result, the majority of charter schools across the nation are majority non-White. For public schools, the highest levels of racial isolation (46%) in terms of majority non-White status are at the primary grades. At approximately 77%, the greatest racial isolation occurs at the middle school level for charter schools. Our analysis also shows that more public multilevel schools are majority non-White compared to charters as nearly half (44%) of multilevel public schools are majority non-White compared to 17% of charter schools.

The CCD also show that double segregation by race/ethnicity and FRL is still an issue in the United States for charter and public schools (see

Table 6. Doubly Segregated: Public and Charter by Majority Non-White and Majority Free/Reduced Lunch.

			Not Majority FRL	Majority FRL	Total
Public	Majority Non-White	Count	6345	26,919	33,264
		% within Non-White	19.1%	80.9%	100.0%
		% within FRL	17.7%	63.1%	42.4%
		% of Total	8.1%	34.3%	42.4%
	Majority White	Count	29,408	15,714	45,122
		% within Non-White	65.2%	34.8%	100.0%
		% within FRL	82.3%	36.9%	57.6%
		% of Total	37.5%	20.0%	57.6%
	Total	Count	35,753	42,633	78,386
		% within Non-White	45.6%	54.4%	100.0%
		% within FRL	100.0%	100.0%	100.0%
		% of Total	45.6%	54.4%	100.0%
Charter	Majority Non-White	Count	978	3008	3986
		% within Non-White	24.5%	75.5%	100.0%
		% within FRL	38.0%	85.5%	65.4%
		% of Total	16.1%	49.4%	65.4%
	Majority White	Count	1596	509	2105
		% within Non-White	75.8%	24.2%	100.0%
		% within FRL	62.0%	14.5%	34.6%
		% of Total	26.2%	8.4%	34.6%
	Total	Count	2574	3517	6091
		% within Non-White	42.3%	57.7%	100.0%
		% within FRL	100.0%	100.0%	100.0%
		% of Total	42.3%	57.7%	100.0%
Total	Majority Non-White	Count	7323	29,927	37,250
		% within Non-White	19.7%	80.3%	100.0%
		% within FRL	19.1%	64.8%	44.1%
		% of Total	8.7%	35.4%	44.1%
	Majority White	Count	31,004	16,223	47,227
		% within Non-White	65.6%	34.4%	100.0%
		% within FRL	80.9%	35.2%	55.9%
		% of Total	36.7%	19.2%	55.9%
	Total	Count	38,327	46,150	84,477
		% within Non-White	45.4%	54.6%	100.0%
		% within FRL	100.0%	100.0%	100.0%
		% of Total	45.4%	54.6%	100.0%

Source: 2015–2016 NCES Common Core of Data.

). Of public schools that are majority FRL, 34% (26,919) of them are also majority non-White. For charter schools that are majority non-White and FRL, 49% (3,008) of them are doubly segregated. By comparison, the difference between charter and public schools that are double segregated by majority non-White and majority FRL is 15%. As a result, the national incidence of double segregation by majority FRL and majority non-White is higher in charters than public schools which reaffirms earlier findings by Frankenberg et al. [5].

4.4. Geographic Area Analysis Considering Ethnoracial Demographics

Considering our descriptive analysis of the CCD, segregation at the local, state, and national levels is an issue in public and charter schools. Furthermore, the data show that segregation is particularly intense for charter schools. In response to similar findings from prior studies, one of the arguments commonly heard in the public discourse regarding charters and segregation is that charters are more segregated because they are located in neighborhoods that are more segregated [7,39]. To consider the validity of this argument, we concluded our analyses by conducting a multivariate linear regression for the national universe of charter and public schools. In our analysis we controlled for school type (charter and public) and aggregate race/ethnicity at the local level. The dependent variable is percent White at the school-level.

The coefficient for local percent White is 0.988 (p < 0.001;

Table 7. Linear Regression Analysis of Public and Charter School Race/Ethnicity School Segregation and Local Percent White.

Model	Unstandardized Coefficients		Standardized Coefficients
	B	Std. Error	Beta	T	Sig.
(Constant)	−17.559	0.214		−82.232	0.000
Percent White	0.988	0.003	0.770	356.372	0.000
Charter or Public	−1.255	.296	−0.009	−4.244	0.000

Model goodness of fit: R² = 0.596.

). So for every unit increase in percent White in a geographic area, there is a nearly one percentage point increase in percent White in a school predicted, holding school type constant. As would be expected, the more White a community is, the larger the percentage of White children in a school. For charter schools, controlling for local ethnoracial demographics, we expect a−1.255 percentage point decrease in the percent White in a school. Since the predicted percent White is 1.255 (p < 0.001; Table 7) percentage points lower in charter schools compared to public schools—considering all schools in the US—the data show that charters are more likely to be segregated than public schools even when controlling for local ethnoracial demographics.

5. Discussion

Now is a watershed moment for school choice as the current political context in Washington, D.C. finds school choice in a position of favor with the presidential regime. Initially, President Donald Trump proposed spending $20 billion on vouchers and charters during his campaign [51]. Although spending at that level has not yet been realized, it is well established that school choice and charter schools have the unequivocal support of President Donald Trump and Secretary of Education Betsy DeVos [52]. Additionally, not only has Secretary Betsy DeVos generally scaled back the U.S. Department of Education’s federal oversight of Civil Rights [53], but former Attorney General Jeff Sessions specifically sought to limit the federal role in diversifying and integrating K-12 [35].

Given the increased attention and focus on charter schools by President Donald Trump and Secretary of Education Betsy DeVos as an alternative to neighborhood public schools and the retrenchment of federal engagement in Civil Rights, diversity, and integration—it is important to analyze whether or not charter schools have continued to be more segregative than neighborhood public schools. In summary, our findings using national, state and local level analyses illustrate that students who attend charter schools are more likely to find themselves more racially isolated when compared to their public school counterparts. Nationally, we find that higher percentages of charter school students of every race attend intensely segregated schools and less predominately White schools than do their same-race peers in public schools. As much as 43% of all public schools in the US are majority non-White compared to 65% of charter schools. For public schools, the highest levels of racial isolation (46%) in terms of majority non-White status are at the primary grades. At approximately 77%, the greatest racial isolation occurs at the middle school level for charter schools. We also find that the national incidence of double segregation by majority Free and Reduced Lunch and majority non-White is higher in charters than public schools. A majority of states have at least half of their Black students and a third of states’ Latinx charter students are enrolled in intensely segregated schools. At the city level, we find higher percentages of charter students were attending intensely segregated schools than urban students enrolled in public schools. Additionally, considering the universe of all charter and public schools in the US—our multivariate regression shows that charters are more likely to be segregated than public schools even when controlling for local ethnoracial demographics.

As evidenced by Plessy v. Ferguson [54]—the decision the U.S. Supreme Court issued in 1896 that upheld the constitutionality of racial segregation laws for public facilities as long as the segregated facilities were equal in quality—racial and economic segregation are a mainstay in American history and our nation’s schools are no different. In fact, Joseph Oluwole and Preston Green [55] argued that Black and Latinx children are being educated in charter schools that are both segregated and unequal in a Plessy fashion. According to their argument, charter school students’ experience is similar to the one received by Black students in the aftermath of the Plessy case. If handled correctly, charter schools could have provided a tool for Black and Latinx children to attend diverse schools of excellence. However, their unregulated nature has enabled privately-managed entities to create segregated schools that also drain resources from neighborhood public school systems, thus creating a situation that is even worse than Plessy [55].

Our mired struggle to provide equitable and equal access to quality education for all children in the US continues to remain outside of our collective grasp. Despite our best efforts, and some progress in decades past, our nation’s schools continue to be segregated not by the de jure segregation of pre-Brown, but by de facto segregation of choice post-Brown. This educational segregation exists across all types of schools but, as has been explored here, charter schools continue to be more segregative. Market-based school choice models, such as charters, have reimagined education as an individualistic commodity. Accordingly, if we are to conceive of schooling and education as an individualistic enterprise guided by free-market ideology, how will we then achieve the common good of integration?

As noted in this paper, school choice and charter school advocates have in recent years acknowledged that charters have generally acknowledge that charter schools are more segregated. The problem, however, is that the segregation is shrugged off or even celebrated because of an unwavering belief that integration is not an aspiration; rather, a functioning market that expands segregation for the sake of an ideological commitment to education reform, privatization, and private-control have become the goals. If we are to achieve the common good of integration, we must acknowledge the research that finds that not only are school choice schemes, like charter schools, impeding that goal, but we must also continue to assess the ideology and public discourse that shrugs at the apparent reality of enhanced segregation in charter schools.