What is diverse enough? How “intentionally diverse” charter schools recruit and retain students

School choice has the potential to be a tool for desegregation, but research suggests that choice more often exacerbates segregation than remedies it. In the past several years, hundreds of ‘intentionally diverse’ charter schools have opened across the country, potentially countering the link between charter schools and segregation. Yet, these schools raise important questions about choice, segregation, and equity. For instance: how do leaders of diverse charter schools prioritize diversity in decisions about location, marketing, and recruitment? What are the implications of these diversity efforts for equity, especially within competitive and marketized educational contexts? We explore the concrete recruiting and marketing strategies schools used to build and retain their diverse communities, drawing on qualitative data from New Orleans, LA and Minneapolis-St. Paul, MN. We identify key strategies used by school leaders, but also note that many strategies were ad-hoc and experimental. Furthermore, we note that schools often did not articulate their goals for diversity, Education Policy Analysis Archives Vol. 26 No. 165 2 making them susceptible to external pressures that might refocus attention away from equity and diversity, or allow groups with more power to shape agendas within the school. Finally, we find that gentrification and widening economic inequities threatened schools’ efforts to recruit and maintain a diverse student body. We discuss implications for leaders of diverse charter schools and other leaders seeking to diversify their student bodies, as well as policymakers and charter authorizers.

In the past several years, hundreds of "intentionally diverse" charter schools have opened up across the country (Kahlenberg & Potter, 2012;Potter & Quick, 2018).These schools have a commitment to a diverse student body, seemingly countering established trends linking charter schools to increased segregation.Furthermore, they have emerged in a time where policymakers have backed away from school desegregation, and as court orders to desegregate schools have been lifted, leading to the re-segregation of schools (Reardon, Grewal, Kalogrides & Greenberg, 2012).While leaders of diverse charters may be committed to revitalizing this policy agenda, their commitment is also politically convenient for charter school supporters seeking to deflect attention away from growing criticisms about school choice and segregation (Wilson & Jabbar, 2017).Indeed, diverse charters have emerged in the national policy conversation at a time when groups like the NAACP and Black Lives Matter have called for a moratorium on charter schools until certain issues are resolved, including the role that charter schools play in exacerbating school segregation (Strauss, 2016).While these calls have generated backlash from the education reform community, they have also drawn attention to the "missed opportunity" that charter schools might have to recruit students across catchment areas and foster greater diversity.
The emergence of diverse charter schools thus offers one potential mechanism for expanding choice while keeping diversity at the forefront.The Century Foundation recently identified over 200 charter schools with a strong commitment to diversity, although not all of these schools achieved their diversity goals (Potter & Quick, 2018).Although diverse charters are currently a small segment of the charter sector, they have recently attracted philanthropic support, from funders such as the Walton Foundation.Despite the growth in diverse charter schools in the past few years, and some attention from media, policymakers, and philanthropists (Kahlenberg & Potter, 2012;Russo, 2013), there has been little systematic study of these schools.In particular, these schools raise important questions about choice, segregation, and equity.It is unclear whether charter school reforms, even when they emphasize diversity and equity, can alter longstanding patterns of inequality, given their roots in neoliberal ideas of choice and deregulation (Wells, Slayton, & Scott, 2002).Diverse charter schools may, as advocates argue, present an opportunity to use school choice to address historical patterns of segregation and unequal resources.But it is also possible that diverse charter schools might perpetuate racial and socioeconomic inequities, echoing concerns that choice often marginalizes low-income communities and communities of color (White, 2014), especially in efforts to attract middle-class families to urban public schools (Cucchiara, 2013).Against the backdrop of these concerns about choice and segregation, our study explores the following questions: How do intentionally diverse charter schools recruit and retain a diverse student body, especially in often highly segregated contexts?How is "diversity" defined and operationalized in such contexts, and amid pressures of marketization?And, finally, what are the implications of these diversity efforts for equity?
In this paper, we explore these questions through a closer look at the marketing practices adopted by diverse charters in their efforts to recruit and maintain a diverse student population.To do so, we draw on qualitative data from two different urban contexts: New Orleans, LA, and Minneapolis-St.Paul, MN.Comparing school leaders' practices across these different contexts increases our understanding of how diverse charter schools recruit and maintain a diverse student population.Furthermore, our work can yield insights regarding the conditions under which recruitment practices are equitable and advance diversity, as well as when they may reproduce inequities in access to schools.

Literature Review
We situate our study at the intersection of three related strands of research.First, we draw on research that examines charter schools and how they contribute to school segregation more broadly.Next, we draw on research examining charter schools' practices, which we argue are important in order to understand how patterns of segregation occur in the charter sector, and how inequalities are reproduced.Finally, we draw on a small, but growing, body of literature that examines how urban redevelopment policies intersect with efforts to diversify urban public schools, often creating new challenges for access and equity.

Charter Schools and Segregation
Researchers have found that school choice, and in particular charter schools, have exacerbated segregation in urban contexts (Bifulco & Ladd, 2007;Frankenberg, Siegel-Hawley, & Wang, 2010;Garcia, 2008;Ni, 2012;Marshall, 2017;Scott, 2005).Scholars generally agree that charter schools are more segregated than their traditional public school counterparts, but there is debate over why this is occurring, and how much it matters.One reason for increased segregation in schools of choice could be that parents prefer and select more segregated schools, or are constrained by geography or transportation (DeArmond, Jochim, & Lake, 2014;Marshall, 2017;Welner, 2013), which is consistent with other literature on parental choice (e.g., Holme, 2002;Margonis & Parker, 1995;Schneider & Bulkley, 2002;Scott, 2005).In addition, many urban charter schools have focused exclusively on poor students of color, especially the most prominent networks of "No Excuses" schools, such as KIPP (Golann, 2015).Other reasons for charter school segregation might include selective enrollment practices, intentional school siting in locations with higher median incomes or fewer students who receive free or reduced lunch, or counseling out students who are viewed as 'harder to serve' by school staff (Gulosino & d'Entremont, 2012;Henig & MacDonald, 2002;Jabbar, 2015;Jennings, 2010;Jessen, 2013;Koller & Welsch, 2014;Lubienski, Gulosino, & Weitzel, 2009;Weiler & Vogel, 2015).Research on whether schools select students is emerging, but it is difficult to document these behaviors, and to disentangle such practices from parent preferences that can also drive segregation.
Others have argued, from a diverse set of political perspectives, that while charter schools segregate students, this might not be problematic.For example, No Excuses charter schools like KIPP are highly segregated but also increase student test scores, and some have argued that their quality and programming can overcome the negative impacts of segregation (Whitehurst et al., 2016).In other words, they argue that separate can be equal.Other scholars and advocates have argued that 'diversity' is often premised on the problematic idea that low-income students or students of color must be exposed to middle-class or White families in order to succeed academically, rather than focusing on deeper underlying issues related to school quality and access to equitable resources (Asante, 1991;Binder, 2009).Institutions controlled and operated by people of color may be segregated, but they can also provide a sense of belonging and community that is not possible in integrated settings that adopt colorblind perspectives, ignoring the racial heritage and cultural backgrounds of students of color (Crosnoe, 2009;Schofield & Hausmann, 2004;Wilson, 2016).In this sense, diversity is not a "silver bullet," and its impacts are context-dependent.
Research specifically examining diverse charter schools is limited, with a handful of exceptions.In one study, researchers interviewed leaders at 21 diverse charter schools to understand their general practices in terms of curriculum, discipline, and parent engagement (Wohlstetter, Gonzales, & Wang, 2016;Wohlstetter & Wang, 2017).In a preliminary study, this same team has shown that diverse charters have positive impacts on student achievement (Wohlstetter et al., 2016).In New Orleans, Beabout and Boselovic (2016) interviewed 26 individuals involved in founding two community-based, intentionally diverse charter schools post-Katrina, tracing the history and challenges associated with this process.However, we know little about how school leaders in these schools conceptualize diversity, and the practices they use to achieve it.

Marketing and Recruitment Practices in Charter Schools
While the mechanisms contributing to charter school segregation may, in part, reflect parental choices, research also suggests that charter schools' mission, marketing, and recruitment strategies may play a role, particularly since many urban charters focus exclusively on low-income students of color (Lubienski, 2007).Charter schools can also exacerbate segregation through missions that attract or "cream-skim" middle-class families (Jennings, 2010;Jessen, 2012).In all these cases, marketing plays a critical role in the process through which school choice shapes school (and district) demographics.
As the number and concentration of charter schools grows across the US, and around the world, school leaders in all types of schools are increasingly tasked with marketing their schools to attract and retain students (DiMartino & Jessen, 2014, 2018;Lubienski, 2007;Lumby & Foskett, 1999;Lundahl & Olson, 2013;Oplatka & Hemsley-Brown, 2010).In systems of choice, school leaders are no longer more or less guaranteed students from traditional feeder pipelines; instead, they must often compete for students and adopt business-like strategies, such as marketing (Gewirtz et al., 1995;Lubienski, 2007).Schools are expected to differentiate themselves in a crowded marketplace of schools, yet there is a great deal of isomorphism in the charter sector, with many schools mimicking one another rather than adopting distinctive missions (Lubienski, 2003;Lubienski & Lee, 2016).Increased marketing has mixed implications for students and families.Marketing can provide information to parents, but it can also exclude parents by targeting more "desirable" families through the language used or symbols embedded in marketing materials (e.g., Jabbar, 2016;Lubienski, 2007;Welner, 2013).
Additional studies in this area have demonstrated how charter schools use broader marketing strategies to shape student enrollment (Johnsson & Lindgren, 2010;Lubienski, 2006;2007;Oplatka & Helmsley-Brown, 2010;Wilkins, 2012).Notably, Lubienski et al. (2009) found that schools in competitive markets increase their promotional efforts and tailor messaging to the unique competitive features of particular communities.In a study of one competitive, racialized market in Michigan, Lubienski (2007) argues that schools used marketing materials to target-and in effect, select-particular students and families, often along lines of race and class.Similarly, DiMartino and Jessen (2014) documented how small schools in New York City used marketing to create "boutique" or "niche" identities within an educational system, in ways that exacerbated segregation.At the same time, schools with "monopoly power," due to occupying a market niche, for example, may have little need to adopt a marketing approach (Foskett, 1998, p. 199).In both cases, school marketing takes shape against the racial and political contexts of different metropolitan areas, including debates about integration, gentrification, and urban development.

Charter Schools, Diversity, and Urban Development
There is a complex relationship between diversity, integration efforts, school choice, and equity.Even in contexts where voluntary choice programs seek to integrate schools for reasons of equity, policies and regulations can privilege the social, material, and cultural capital of some families over others (André-Bechely, 2004).This has been particularly problematic in cities undergoing "revitalization" efforts and gentrification, where school districts seek to lure White and middle-class families into highly segregated public school systems.In particular, scholars have drawn attention to the unique power dynamics within schools and districts when schools market to middle-class families (e.g., Cucchiara, 2013) and "when middle-class parents choose urban schools" (Posey-Maddox, 2014) and how these dynamics influence equity within and across schools.
School choice and desegregation policies are embedded in broader demographic trends related to housing, segregation, and urban 'revitalization,' or gentrification (Cucchiara, 2013;Lipman, 2011), and shaped by broader patterns of structural inequality, White supremacy, and power (Aggarwal, 2016;Buras, 2011).Because diverse charter schools operate-in ambiguous wayswithin these broader patterns, we argue that it is important to look more carefully at the particular strategies these schools employ in pursuing diversity goals.How, for instance, do these schools conceptualize what diversity means, why this goal is important, and which families' interests might be served by such goals?How do schools act on these commitments and beliefs, as they work to recruit and retain a diverse student body?In effect, our study explores how the practices of diverse charters intersect with debates about equity and access, investigating what groups benefit from diversity in practice versus in rhetoric.
We explore these questions in conversation with research that describes the reasons why schools in choice systems might shape their enrollment, in ways likely to exacerbate segregation.There are, however, critical gaps in this literature.First, prior work focuses on traditional charter schools, and it remains unclear how these patterns and practices apply to intentionally diverse schools, which have explicit missions around the racial and socioeconomic demographics of their schools.Second, the literature focuses on exclusionary tactics, rather than the potentially inclusionary tactics employed by diverse charter schools.At the same time, the niche missions of these diverse schools may further segment or stratify educational opportunities, perpetuating inequities.Understanding the strategies of diverse charter schools may help shed light on practices that can foster greater diversity in choice-based educational settings, as well as those that can reify existing segregation patterns, even in the name of "diversity."

Framework
To frame our study, we draw on existing definitions and conceptions of "diversity" that have been used to measure the racial and socioeconomic diversity of schools, as well as critical policy research that problematizes how the term "diversity" is employed in ways that evade considerations of equity, race and power.Our study focuses on both racial/ethnic and socioeconomic diversity, which are often the goals of intentionally diverse charter schools.However, we want to underscore that both the meaning of diversity and how to measure it remain subject to debate.Imprecise definitions leave schools considerable room to decide whether and how they might achieve their diversity goals.Our goal is not to pose a specific definition of diversity; rather, we are interested in how leaders of diverse charters understand what counts as 'diversity,' and how they sought to recruit and retain a diverse student body.
On a basic level, our study follows the working definitions developed within the diverse charter school community.Since 2013, over 100 schools, serving over 50,000 students have joined the "Diverse Charter Schools Coalition," a national advocacy and membership organization.This coalition emphasizes both racial and economic diversity but allows schools to self-nominate themselves for inclusion in the coalition.Researchers have also attempted to categorize this sector.In the most comprehensive review to date, Potter and Quick ( 2018) used a two-part criterion to determine what counted as a diverse charter school.They identified 125 charter schools whose mission statements had a "strong commitment" to diversity and that were successful in meeting diversity goals, as measured by having no more than 70% of any single racial/ethnic group in the school.However, they also identified dozens of schools that had strong commitment to diversity, but were not able to achieve diversity (based on their definition).This raises questions about the potential challenges school leaders may face in reaching their goals, as well as varying diversity goals held by different schools.
While we rely on these specific studies of diverse charters, it is important to note that studies of school diversity have a longer history and are closely connected to policy debates about the segregation, integration, and re-segregation of schools.In these contexts, scholars have developed different methods for measuring the segregation-and by extension, integration-of specific school, districts and metropolitan areas.Here, research often relies on exposure, isolation, dissimilarity, or divergence indices (Massey & Denton, 1988;Orfield, Siegel-Hawley, & Kucsera, 2014;Whitehurst, Reeves, & Rodrigue, 2016), which all attempt to measure the extent of segregation in schools.For example, exposure indices measure the percentage of White students in the average Black student's school, while isolation indices measure the percentage of Black students in the average Black school (Whitehurst et al., 2016).However, these measures do not capture how a school's demographics compare to its local context.Therefore, scholars have used dissimilarity measures that measure the extent to which school populations represent the demographics of their surrounding areas.However, these measures are often more informative when applied to a large geographic area (e.g., a metro area), and not as useful for particular schools (Orfield et al., 2014).Practically, however, these measures are rarely used outside of policy studies and legal decisions.Schools themselves have a variety of ad-hoc (and largely unnamed and unstudied) working definitions of diversity.
Questions of diversity also go beyond the composition of students enrolled in a school.Here, a long-standing body of research has explored the positive effects of diverse and integrated school environments on both academic and non-academic outcomes.This research suggests that integrated contexts may benefit both White students and students of color, in terms of developing positive intergroup attitudes, cultural flexibility, increased civic engagement, reduction in bias, greater access to informal networks, and improved critical thinking skills (Borman et al., 2004;Braddock, Crain, & McPartland, 1984;Carter, 2010;Crosnoe, 2005;Dawkins, 1994;Guryan, 2001;Mickelson, 2006;Schofield & Hausmann, 2004;Wells & Crain, 1994).
However, ostensibly diverse spaces can also be quite segregated, and can reinforce harmful or problematic practices for students of color (Carter, 2005;Gamoran, 1986;Mickelson, 2001;Oakes, 1985).This research highlights how diversity-as a concept-can often bypass concerns of equity, race and power.As other scholars have noted, "diversity" is a term that evades specificity.At once "more encompassing and concealing," it moves away from specific categories of race, class, and gender, as well as from concrete measures of inequality (Ahmed & Swan, 2006).Bell and Hartmann (2007) found, for instance, that wide agreement about the importance of diversity often masks contradictory and undeveloped conceptions of the term.Such "happy talk" of diversity can appear to recognize difference, without grappling with White normativity and systemic racial inequality.The fluidity and elasticity of the term-as we note in our analysis-can be problematic for diverse charters, as commitments to diversity can be revised and modified to fit changing circumstances and new competitive pressures.
Beyond these varying measures of what "counts" as diversity, schools and school leaders may define diversity by its potential value to students and families.While often echoing research on the benefits of learning in a diverse school environment, leaders sometimes frame diversity-and parents' choice of a diverse school-in terms of the benefits for White and middle-class families.
Here, our analysis draws on recent critical policy studies that explore the contested value of diversity for different families, particularly in contexts where 'diversity' can advance conflicting aims.Early studies of diversity-in higher education contexts, for instance-explored how difference is framed as an economic advantage, particularly for already advantaged students.As Anshuman Prasad notes, "By emphasizing the economic potential of so-called diversity advantage, difference has been largely transformed into an economic asset that can be acquired and actively managed" (2001,64).
In this same sense, "diversity" can often be used to express progressive values, while evading specific measures of inequity (Marvasti & McKinney, 2011).In doing so, measures of diversity risk furthering the interests of those already in power.For instance, studies of dual-language schools and bilingual education policy have highlighted the recent embrace of dual-language schools by White families (Valdez, Delevan, & Freire, 2016;Kelly, 2018).Here, diversity and access to a second language are framed as powerful assets in a competitive and precarious economy.But the goals of bilingual education can also be distorted to meet the needs of advantaged White and middle-class families, as opposed to focusing on the educational needs of emerging bilingual students.
Drawing on these critical studies of diversity, our analysis focuses on how the shifting and complex ways that school-level actors conceptualize diversity.Rather than imposing a particular, pre-existing definition of diversity, we explored the varied meanings and understandings of the concept across participants and schools.We asked, for instance, how school leaders understand what diversity means, why it might be valuable, and for who?Further, how have they measured whether they were successful in achieving those goals?In what ways might other pressures-to maintain enrollment, to satisfy the needs of more advantaged families, etc.-work against leaders' commitment to diversity?In effect, our study explores how school leaders define and operationalize diversity in segregated and marketized contexts.

Data and Methods
Our data come from case studies of an emerging group of diverse charters in two metropolitan areas with a large share of charter schools: New Orleans, Louisiana, and Minneapolis-St.Paul, Minnesota.These sites each have a significant number of diverse charter schools, yet they represent distinct policy environments, allowing for comparison.Minneapolis-St.Paul was initially selected because of its long history with charter schools, and the emphasis of a growing number of schools focused on particular racial and ethnic communities.In contrast, New Orleans's charter school market is relatively recent, but has rapidly expanded.In recent years, several intentionally diverse charter schools have opened in this metro area, in contrast to some of those in Minneapolis-St.Paul, which have been around for decades.These two sites thus represent important urban contexts for school choice, yet allow us to see how differences in the local contexts may shape the development of diverse schools in each metropolitan area, as well as differences in how schools market their schools and recruit families.
Diverse charters emerged for different reasons in each site.In New Orleans, they were founded as an alternative to the very segregated public school system, which is made up of more than 90% charter schools, and in which most students are African American and qualify for free and reduced lunch, despite a more diverse urban context (Wilson & Jabbar, 2016).In Minneapolis-St.Paul, diverse charters emerged in contrast to larger patterns of sorting and segregation, and in response to other schools that have focused on serving particular cultural and ethnic communities (Wilson & Jabbar, 2016).Diverse charter schools in this metropolitan area, albeit in different ways from New Orleans, still contrast themselves with wider patterns of segregation in other charter schools and the public school system.
While these studies were initially conducted separately, we have re-analyzed parts of our data together in this paper.Our studies initially had different theoretical lenses and areas of focus.In New Orleans, the study examined how school leaders perceived and responded to competitive pressures resulting from choice.In Minneapolis, the study examined a growing number of charter schools designed to serve particular racial, ethnic and linguistic communities.However, in both of our larger studies, we each separately documented a smaller outlying group of intentionally diverse charter schools.In learning about each other's' studies, we were curious about this similar phenomenon-a small, outlying group of intentionally diverse charter schools-in two very different policy contexts.Guided by similar comparative work that combined two qualitative studies (Cucchiara & Horvat, 2009;DiMartino & Jessen, 2014;Griffin & Reddick, 2011), we reanalyzed our original data in light of a new, collaboratively generated framework.In some cases, we conducted additional interviews with school leaders to explore our new research questions in more depth.We employ a meta-ethnographic method (e.g., Noblit & Hare, 1988), which allowed us to reinterpret our data in light of new conceptual frameworks and to test findings from one site against another, thus strengthening our claims by exploring a phenomenon-recruitment and marketing in diverse charter schools-across multiple sites.

Data Sources
Our data sources include formal and informal interviews (Patton, 1990), and data from school websites, charter applications, and annual reports.Across both sites, we conducted 26 interviews with school leaders of 13 diverse charter schools with explicit missions around diversity: five schools in New Orleans and eight in Minneapolis-St.Paul (the Twin Cities).See Table 1 for details about these schools and their missions.Each researcher has also conducted a larger, in-depth study in one of the sites (over 100 interviews with stakeholders in New Orleans, and surveys and document analysis; and over 80 interviews with stakeholders in Minneapolis-St.Paul, and document analysis and observations).We draw on this expertise to provide additional context for each site.Note: AMI=American Indian, API=Asian, Pacific Islander, HIS=Hispanic, BLK=Black, WHT=White, FRL=free or reduced lunch eligible, ELL=English Language Learner, SPED=Special Education Each group of schools is embedded in different policy contexts.The schools in New Orleans have been part of national conversation on education reform and school choice.After Katrina, the city moved to a majority-charter model, where now 90% of public-school students attend charters (Cowen Institute, 2015).New Orleans is also a segregated education system that does not reflect the demographics of the city.While the city is approximately 60% Black, New Orleans public schools are 85% Black and 83% economically disadvantaged.Most White students attend private schools.In this context, diverse charters seek to recruit a population that is more reflective of the city demographics, rather than the school system's demographics.
As the first state to pass charter school legislation, Minnesota has a mature marketplace of choice options, including charter schools, magnet schools and many intra-and inter-district choice options.Approximately 1 in 5 students attends a charter school in Minneapolis-St. Paul (National Association of Public Charter Schools, 2013).Compared to New Orleans, schools in this area enroll a wider range of racial and ethnic groups, yet there is evidence that choice has exacerbated segregation between racial and ethnic groups, as specific schools have come to enroll particular groups of students (Institute on Metropolitan Opportunity, 2013).

Data Analysis
First, we each independently analyzed data from our sites for themes related to diversity, marketing, and recruitment strategies.We had regular conversations to identify key themes that emerged from our data, noting similarities and differences in how these themes played out in each context (e.g., Cucchiara & Horvat, 2009).We drew on our broader understandings and study of the sites to explain these emerging differences.We used themes (e.g., conceptions of diversity, practices related to diversity) to organize our findings, then sought out relevant sub-themes through a constant-comparative method (Glaser & Strauss, 1967).These themes emerged through an iterative process where we combined relevant data from our separate contexts, and re-categorized data, where necessary, to explore common themes and points of difference across the two metropolitan areas.For example, because one study initially focused primarily on practices related to school competition, we recoded those data to identify instances where schools' practices related to fostering or maintaining diversity.In this way, we combined the underlying data from our respective studies, focusing on the concepts central to this paper (definitions and understandings of diversity, marketing and recruitment practices related to achieving or maintaining diversity.)To triangulate our findings, we drew on a range of sources, including interviews, schools' initial charter applications to the authorizer, and websites (Miles & Huberman, 2013).For example, if an interviewee described a particular approach for recruiting students, we could compare their responses to what was outlined in the charter school's application, which often included a recruitment plan.We focus on the interview data for this analysis, but the charter applications and websites provide additional insights into schools' marketing plans and missions and help to corroborate our findings.

Limitations
It is important to note key differences between the two contexts in terms of race and class demographics, history, major events (for example, Hurricane Katrina's effect on the reform environment), and immigration patterns; differences that limit easy comparisons across contexts.These districts, however, help to illuminate shared marketing and recruitment strategies across different contexts.There were also some differences in the original intentions of both research studies, as noted above.However, through our collaboration, we re-analyzed these data using common themes to address our key questions about the marketing and recruitment practices of diverse charter schools.

Recruiting and Marketing for Diversity
In this section, we summarize three major themes: how schools recruited for diversity, in the absence of policy incentives to do so; the practical recruitment and marketing strategies they employed; and challenges determining how much diversity was "enough" for their schools.

Recruiting for Diversity, Without Incentives
Across our sites, schools sought a diverse student body, but many of the schools were caught off-guard by how few low-income families and families of color were initially attracted to their schools.Some leaders noted that the low numbers of applications from these groups may have reflected that their schools had missions that appealed more directly to White middle-class families, creating challenges in attracting low-income families or families of color.In New Orleans, Ellison School1 was initially concerned about recruiting middle-class families to the school, given that so few of them attended public schools in the city.The leader held several "playground play dates," and attended festivals and events where middle-class families might attend.As a result of these efforts, he noted that, "We were well known among the middle-income."However, when the applications started to come in, they received 50 applications in the first day, "almost exclusively from middleclass and higher-income families, so we're like, 'uh-oh, we're in trouble'" (Principal, Ellison).The leader then ramped up recruitment of lower-income families, visiting Head Start centers and attending school fairs.
Several schools in Minneapolis-St.Paul also described how they had to work hard to recruit low-income students to the school in order to have a diverse community.In the absence of such efforts, they felt that would have enrolled much larger percentages of White and middle-class families, who may have been more likely to seek out choice schools, and schools with their particular missions.At Clark Academy, for instance, the founding director recalled that the school had intended to serve about 50-55% students from lower socio-economic (SES) backgrounds.However, the school opened with more White students than expected, and with only 35% of students qualifying for Free and Reduced Lunch (FRL).The school recalibrated, engaging in outreach over the next few years to recruit more students from disadvantaged backgrounds.Similarly, at Dewey, the school felt that its project-based and experiential learning mission tended to attract middleincome and White families.Therefore, it targeted all of its recruitment efforts in lower-income communities adjacent to the school.More advantaged families, the school director explained, were "already searching." While some schools engaged in strategic efforts to recruit low-income families and families of color, there were few incentives to do so in their local policy environments.In their contexts, schools were required to compete to attract a sufficient number of students.However, most of the schools in our sample had long and "healthy" waitlists of students; their recruitment efforts, in this sense, were not primarily motivated by the need to increase enrollment or any concerns about financial stability.But the schools-especially those with larger percentages of White, middle-class families-saw an absence of recruiting as a threat to their mission to serve a diverse group of students.In New Orleans, one school leader said, "though we don't have any problem getting applicants, we feel a moral obligation to go out and make sure we're reaching everyone.That we're not overlooking families who just hear 'international school' and think 'that's not available to me'" (Principal, Horizon).In this sense, the recruitment of diverse families was dependent on the effort, goodwill, and missions of individual schools.
In some cases, central policy requirements-like a mandated lottery for oversubscribed schools or common enrollment system-often worked against schools' efforts to diversify their population.In New Orleans, for example, schools were concerned about how they could maintain a diverse student body in the future, if they were to join OneApp, a centralized enrollment system.As one school leader noted, even though his school was "totally open admission," he believed the charter schools were fighting the OneApp because they would "get a random selection of students...which then destroys the whole concept of choice and theme schools and focused schools" (Principal, Baxter).The leader noted, "If we go to the One App...it would make it more difficult to maintain our diversity in this school."Similarly, at Ellison, the principal noted that they aimed to create quotas, whereby at a minimum, one-third of the seats would be reserved for low-income students via OneApp.In this case, retaining more discretion over their enrollment might have helped schools to maintain diversity.
In Minneapolis-St.Paul, by contrast, recruitment efforts were almost completely decentralized.They depended entirely on individual school and leader decision-making with little support for (or regulation of) such efforts.Recruitment efforts were equally dependent on the goodwill and efforts of schools.There were no requirements, and few incentives, to deliberately recruit low-income students to build a more diverse community.Indeed, many charter school leaders found that school-based lottery requirements (mandated by state charter school law) worked against their efforts to recruit low-income students.The director at Rustin School described their competitive lottery and the challenges of diversifying the pool of applicants before the lottery takes place.Since their intentionally small school only accepts a small number of Kindergarten students a year, relying on word-of-mouth and networks of existing families could result in a far less diverse entering class.As he noted, "We don't really need to do recruitment in terms of numbers, but we want to recruit, in terms of getting more diverse applicants." With this concern in mind, this director had tried to build stronger relationships with local Head Start programs and other community organizations.While holding spaces for those children would be "legally impossible," the director stressed the importance of "diversifying the pool of applicants, so that acceptances will be more diverse."Here, the director notes that their work to create a diverse school also occurs against a legal backdrop that has sharply curtailed how race-and even class-can be used as a factor in school admissions.Before becoming a charter, Rustin had been an independent school with greater control over admissions.One of the founders recalled, "In the old days, we would get someone in trouble, someone who really needed the school and we could say: we need to take this person, or we could say, this grade really needs more students of color, more boys.All that is hard to do now."Becoming a charter meant agreeing to hold an open lottery for available spaces.While state lottery requirements were designed to prevent schools from shaping their enrollment in inequitable ways, several schools noted that lotteries-in advantaging early choosers-worked against their efforts to diversify their student bodies.
Another school director in Minneapolis-St.Paul attempted to work around the lottery requirements by limiting the timeframe in which the school would accept applications.As he noted, more advantaged families are searching and applying early, giving them a natural advantage in lotteries.To counter this, he limited their school's application period, recruiting in low-income communities in advance and stressing the important upcoming deadline for applications.Once he is more certain of a diverse pool of pending applications, he opens up the application process to everyone.He described recruiting "just enough" so that the school doesn't always need a lottery, and can maintain a diverse enrollment.
In sum, the considerable work that these schools undertook to recruit diverse families was not rewarded or incentivized; in contrast, certain factors (the lottery system, limited time and resources, accountability pressures) often worked against recruiting for diversity.While there was a more systematic enrollment process in New Orleans, in case of the OneApp, there were also no built-in ways to prioritize diversity in the lottery, although this might be changing for some schools.Schools in New Orleans were concerned about how losing control of admissions might impact diversity.In addition, diverse schools in both contexts were working against the many incentives to serve higher-income and 'easier-to-educate' students, ones well documented in the literature.Yet, it is also important to note that these schools' commitment to diversity may have also been shaped by their competitive, marketized policy environments.Many schools noted that their missions were designed to appeal to White and middle-class families, or at least tended to draw those families to their schools.Diversity, in this sense, may also be a valuable and powerful market niche to attract new or broader audiences to charter schools.

Student Recruitment Strategies Used by Diverse Charter Schools
In the absence of incentives, schools used a range of specific strategies in order to recruit diverse communities.These strategies including targeted recruitment to communities; strategically locating in particular neighborhoods to maximize diversity; relying on parental and school networks and word of mouth; and managing their public image.
Targeted recruitment to minority communities.In order to recruit diverse families, school leaders engaged in targeted recruitment efforts.In New Orleans, as described earlier, school leaders at Horizon and Ellison targeted Head Start centers and other organizations that worked with low-income families.The school leader at Baxter also described visiting enrollment fairs and feeder schools, as well as more widespread outreach: "We send a representative, they do a presentation, they hand out packets, we go on radio, go on TV.We're very entrepreneurial about how we get students to come to our school." This type of targeted advertising helped schools to recruit a diverse student body.Schools in Minneapolis-St.Paul also reported targeted recruitment in specific communities.The Rustin School developed relationships with early childhood programs that served low-income students, community organizations in nearby neighborhoods that were predominately African-American and Latino, and a local housing project that served Native American families.The Dewey School targeted a nearby neighborhood with high numbers of African American and Hmong immigrant families.Clark Academy focused on having a presence in several diverse neighborhoods, participating in community festivals and employing neighborhood liaisons to recruit students.The two Montessori schools in the Twin Cities were each associated with independent Montessori daycares with missions focused on serving a diverse mix of families.The schools, in this sense, could recruit directly from an already diverse pool of families committed to their curricular mission.
Location decisions to maximize diversity.Studies have found that charter schools often strategically locate to shape their enrollment (e.g., Henig & MacDonald, 2002;Koller & Welsch, 2014;Lubienski et al., 2009).We also found that diverse charters used their schools' locations to meet their goals for diversity.Leaders underscored that location decisions must be made early on, and they shape enrollment in ways that are hard to predict or control.As the director of Harmony Academy (in the Twin Cities) recalled, "We really strived to be a racially and culturally diverse community of students, families and staff.We wanted an integrated school environment.But then the reality set in: you have to find a building, in a particular neighborhood."In this school's case, an early location decision meant that the school started with a high number (65-70%) of Asian immigrant families.This early location--and its original school community--came to characterize the school's identity and shape its subsequent work to recruit diverse families from other communities.
Location decisions are also dependent on available space and influenced by local policies about school facilities.In New Orleans, for example, there was a citywide plan to rebuild schools, but finding adequate facilities in the interim created challenges for schools seeking to target a particular geographic community.For example, Vista was in a temporary facility, but was "working hard to reach out to the community where the school will be."The uncertain nature of school location, and the timeline for construction, made it difficult for schools to use location as a strategy.At Ellison, the school ended up being located (temporarily) in a more affluent neighborhood, but this was not by choice: "We did not want to open up in [affluent neighborhood] but … I had three other handshake deals on three other locations… and they all fell through" (Principal, Ellison).He did not intend to put the school in a "higher income community" or in a "low socioeconomic, really high crime" area because, either way, he would not be able to sustain the diversity he sought.
In Minneapolis-St.Paul, schools were even more responsible for securing their own locations.Clark Academy described an early decision the board made between an identity as a "community school" or a "destination school."They "decided to be a destination school, knowing that there were real pros and cons to that decision....We looked for a place that was central, that many kids could come to."Clark located near a university, in a well-resourced neighborhood, but one close to many public transportation options and also near comparatively less resourced communities.In effect, they located on a border between neighborhoods.The Rustin School was also located on a border, near a major highway, that separated a neighborhood with more White residents from other neighborhoods with more African American, Latino, and Native American residents.Yet border crossing also posed challenges.As the director of the Dewey School noted, their school-located in a historically White neighborhood-had prioritized recruiting many families from a historically African American community nearby.He had learned, however, that many African Americans still associated the school's neighborhood with old "sundowner laws" and a history of racial oppression and violence.These histories posed challenges to recruiting African American families to the school.However, the school's first director, also African American, worked to "translate" the school's approach to curriculum to these families, which helped to diversify the school's enrollment from its opening, despite its location in a historically White and newly gentrifying neighborhood.
Indeed, one key challenge in using location as a tool for recruitment was the rapid gentrification of neighborhoods, particularly in New Orleans.This was particularly problematic in the neighborhood where Flynn Academy was located.As the leader of this school said, "[Neighborhood] is gentrifying quickly....Now, all of a sudden our base of families is not the same level of diversity as it used to be.What do we do?We want to be a neighborhood school, but we also recognize that we want diversity."The gentrification of neighborhoods has been described as a potential opportunity to bring White families into the public school system, but has a complicated relationship with school choice and school diversity (e.g., Cucchiara, 2013;Kimelberg & Billingham, 2013).Gentrification can create challenges for maintaining diverse student bodies as families of color are displaced out of those areas, or as policies to restructure public schools draw on racist discourses, marginalizing the families of color that policymakers claim they are targeting, and reproducing inequities (Buras, 2011;Cucchiara, 2013;Lipman, 2011).One driver of gentrification in New Orleans was the influx of reform leaders after Katrina, who now sought schools for their own children.One leader described these parents as an "untapped market" for diverse charters (Principal, Ellison).In these different ways, location and gentrification offered both opportunities and challenges to recruiting diverse families, in ways that were not always easy for schools to predict.
Using existing networks and word-of-mouth.Across both sites, the biggest source of recruiting was word of mouth and the use of existing networks, based on school leaders' or parents' connections.As the leader of Baxter recalled, "What's really working, believe it or not, more than any of our public relations budgets or anything else we do, is mouth to mouth, word to word, [and] people to people."At times, however, as a number of schools noted, this word-of-mouth recruiting worked against a school's efforts to be diverse.Several schools in Minneapolis-St.Paul noted that word-of-mouth recruitment tended to advantage White, middle-class applicants to their schools.But, in other cases, parents who were committed to the school's diverse mission could be powerful ambassadors to recruit other families likely to be interested in a diverse community.As the leader of New Orlean's Flynn Academy noted: We have really honest conversations with families about what we want as a student body so that when it comes time to bring students next year, that families can go out and say, 'Hey, you should come to this school.'Word of mouth in all camps will help us with that.
The director of Harmony Academy in Minneapolis-St.Paul, observing that their marketing strategy, "by and large…has been through word of mouth," also noted that their families were able to represent the school and its commitments: "We've garnered a certain reputation…that we really offer resources ...to address the challenges and obstacles our families face, that we have our cultural liaisons, social workers, a supportive community, proactive behavioral supports."While word-ofmouth strategies could result in recruiting more advantaged families, in at least a few diverse schools, networks of diverse families were also able to recruit other families around a school's mission and commitment to diversity.Related to this, a number of schools cautioned that who recruits is an important consideration, and highlighted the importance of having people from specific communities lead recruiting efforts in those communities.For example, Harmony Academy employs full-time "cultural liaisons" that-in addition to their many roles in the school-help to lead recruitment and marketing efforts in various communities.The school director said: When we do a school fair, we make sure that the people representing us...reflect the community and racial make-up of populations at the school….thereare just so many subtle messages that can be sent about what we value and what we prize.Even the images on Facebook, or the images in our monthly newsletter....I'm aware of how easy it is to let lenses and filters shape what we see and show, and that we need to work actively to ensure we are not perpetuating stereotypes.
For this director, questions of diversity shaped countless small decisions they made: from choosing photos to selecting award winners.Indeed, school leaders in marketized environments are increasingly asked to play more extensive roles in promoting their schools.These new roles come with weighty responsibilities, but little guidance.Across our two metropolitan areas, we observed that many of the recruitment and marketing decisions made by school leaders were more serendipitous or reactive, rather than purposeful or strategic.Few, if any, schools had developed formal recruitment or marketing plans, or any strategic initiatives around maintaining a diverse community.While many staff members were committed to diversity, most schools had not had conversations about what was diverse enough, or how they might reach specific goals.

Recruitment Challenges: What is Diverse Enough?
While focused on diversity, many schools had not considered their goals or targets for an ideally diverse community, nor had formulated specific strategies to reach those goals.One school in New Orleans had a clear idea of what it considered to be "diverse enough": "What you really want in a diverse school is a critical mass of kids who are quote unquote like you....You want at least 40% of kids to be low income, 40% to be at least middle or high income" (Principal, Ellison).He was working with the OneApp to build in a mechanism to ensure that at least one-third of their students would be low-income.However, such goals were less well-defined in many of the other schools we studied.When asked about the school's goals around diversity, the director of Rustin in Minneapolis-St.Paul noted some of these dynamics.As he said, "Over time, we've had a shift to a more White and middle-class population; and this honestly concerns some of the staff."He noted that staff wanted more diversity, "both racial and SES" to reflect the mission of the school.The school had "made a push to recruit in certain populations and neighborhoods," but hadn't "had any conversation about the ideal amount of diversity."In the absence of such goals, it is possible that a school's commitment to diversity may be challenged by particular circumstances or external forces.What happens, for instance, when a commitment to diversity comes into conflict with other goals a school may have (to a particular curriculum, or specific accountability goals)?How might a school preserve a commitment to diversity in the face of such challenges, and in the absence of incentives?While a full answer to this question is beyond the scope of this analysis, one example points to the challenges that schools may face in preserving a commitment to diversity.
Clark Academy, in Minneapolis-St.Paul, had prioritized a diverse population of students along lines of race, class, gender identity, and religion.The school's enrollment plan called for growth each year, but they experienced higher than expected enrollment numbers in their third year, as well as an increase in the number of students who needed special education services and additional academic support.These challenges resulted in a difficult year for the school, and led 45 families-mostly White and middle-class-to transfer out.Since then, the school has struggled to regain its projected enrollment.The school's director recalled these challenges, which occurred prior to her time at the school, noting that over-enrollment and a lack of resources, such as having only one special education teacher, contributed to the loss of many families.Since that challenging year, however, the school has reworked its structure, curriculum, and staffing models to better serve a diverse group of students.Central to these efforts was a new inclusion model for special education that "pushes in" special education teachers and support staff to facilitate learning for all students in a classroom setting.With these changes, enrollment has stabilized and even increased slightly.
The school's former director spoke of the contentious conversations in this difficult year, framing them as a challenge to the school's commitment to diversity.She recalled: With a diverse population comes more challenges....Along with some more privileged folks that wanted us to do things in a certain way.This group, of wealthy, all White, families....This was a defining moment in our work to be a diverse school.They really wanted us to kick some of our kids out, and we were not willing to do that.I mean, the questions were really about whose kind of school this would be and for who....We knew when we opened and started, that we had some people thinking that this was going to be their private school, but public…I mean, this school is for all kids.A diverse school that is a microcosm of society.
Here, the director highlights a potential tension when recruiting both middle-class White families, many of whom saw the school as a quasi-private school, as well as low-income students, including some who came to the school with specific needs and challenges.These recollections also illustrate how agreement about the value of diversity may evade harder conversations about racial justice, limited resources and the tensions that arise when schools try to serve families with potentially different needs and priorities.While many of the tensions centered on the school's capacity to serve special education students, a school board member noted that these conversations often implicitly invoked race; that is, middle-class families left due to "concerns about safety," concerns that often had "a racialized dimension."Such losses challenged the school's commitment to diversity.But, in many positive ways, these challenges were opportunities to reaffirm the school's mission, make curriculum "more culturally relevant," and adopt approaches designed for a diverse and inclusive community of students.While positive, in many ways, the school has faced continued challenges around recruiting.As the board member noted, "Enrollment has never shifted back around to what it was...it has been a constant struggle, attendance numbers are low."One of the challenges, as the former director explained, was that the school came to be seen as an "alternative school," rather than as a more "academically oriented" school.This created challenges for maintaining the diversity of their student body, as they now have had to work to recruit and retain White families.The current director said: It's a tricky thing to have a diverse community….Now that we are as balanced as we want to be, it will be a trick to stay there.A big part of that is making sure that we keep our percentages of White kids and non-free-and-reduced-lunch kids where it has to be.
As she noted, these new challenges will reshape the school's approach to recruitment and marketing, in ways that might ask the school to prioritize certain features or curriculum to appeal to more advantaged families.Likewise, the school now faces the kinds of incentives common to many schools of choice: reasons to recruit (at least some) more advantaged and easier to educate students.In this case, and in other instances, the schools in both metropolitan areas were upfront about the need-and their ability-to shape their enrollments through their missions, locations, and marketing strategies.While these efforts have largely been made in the service of recruiting diverse communities of students, their existence also reminds us that schools can just as easily recruit less diverse students.Here, our research echoes other studies (Jennings, 2010;Jessen, 2012) showing how charter and selective enrollment schools can subtly-and not so subtly-shape their enrollment; they can, in effect, choose their students.In the absence of incentives, maintaining diversity relies on the commitments of particular schools.These commitments, as the challenges at Clark illustrate, can sometimes be challenged.

Discussion & Conclusion
In both sites, intentionally diverse charter schools engaged in a range of targeted marketing and recruitment practices to achieve student body diversity, going above and beyond other charter schools in their areas.They targeted recruitment to particular minority communities by visiting Head Start centers or particular feeder schools serving underserved populations.They also fostered diversity through location decisions and used their existing networks, or "word of mouth," strategies to recruit students.However, relying on existing networks is likely to preserve a school's demographic balance rather than change it, so the initial makeup of the school matters for diversity.School leaders found that it was especially hard to recruit families of color and there were few incentives to do so.Enrollment fluctuations over time and demographic change in metropolitan or neighborhood contexts created new challenges for school diversity.Efforts to recruit a diverse student body thus depended on the goodwill and missions of individual schools, and many schools' strategies seemed more ad-hoc, rather than purposeful or strategic.
While some school leaders referred to the demographics of their local communities, they did not use any systematic measures of diversity to track or measure their progress towards their goals."Diversity" thus continued to evade specificity.Such elastic definitions perhaps gave leaders more flexibility in their recruitment and retention efforts, but also came under pressure, as circumstances changed.Here, a commitment to diversity-as an ideal-sometimes minimized harder conversations about race and inequity within school communities.Furthermore, it is unclear whether these strategies were effective in achieving their goals, an area that future researchers could assess.Generally, the haphazard nature of recruitment for diversity raises concerns about how central this priority might be when other pressing issues arise.Maintaining a diverse student body requires constant effort on the part of schools, and often must be maintained in the absence of any policy of market incentives to do so.This is a fragile foundation for integration efforts, and one that can easily be challenged by leadership turnover, internal tensions in school communities, and plain old inertia.

Implications for Policy and Practice
This study has several implications for practice.First, we found that school leaders set their criteria of a sufficiently diverse school in rather haphazard ways.Here, it may be valuable to support school leaders to more intentionally identify targets for enrollment, in conversation with school stakeholders.Authorizer-level interventions could help to address some of these challenges, such as requiring a plan for diversity in recruitment as part of the authorization process.We also found that there is an element of demographic inertia in schools, so the starting enrollment of a school becomes important in determining its trajectory.It establishes a baseline that is very difficult to change, especially given sibling preferences.This baseline also shapes the school's identity and the willingness that other parents might choose that school.This suggests that authorizers should pay greater attention to starting enrollment and initial recruitment plans in assessing charter applications.Location decisions were an early and key decision that significantly shaped enrollment.By strategically siting the school, either in an affluent neighborhood, an impoverished one, or one undergoing change, school leaders may be able to take advantage of parents' preferences for geography to help recruit some groups of families, giving them space to more strategically target other groups.Indeed, geographers might explore the patterns of diverse charter school locations, and the extent to which they locate in "border" areas or in gentrifying neighborhoods, as researchers have done for other types of charter schools (e.g., Lubienski et al., 2009).Transportation policies may also play a role.Part of the idea of schools of choice recruiting diverse groups is to go across existing lines between segregated neighborhoods.Putting this commitment into practice might involve a greater need for transportation funding for diverse charters, and requirements for transportation plans to be developed as part of the charter planning process.
We also found that the different policy contexts shaped school leaders' strategies in recruiting diverse student bodies.In particular, the centralized enrollment process in New Orleans created new challenges (and some new opportunities) for diverse charters.There may, however, be a potential conflict between fully open-enrollment charter schools and attracting a population that aligns with a diverse charter school's mission, especially if there is no emphasis on diversity in the local policy context and/or no preference or priority for diversity in the centralized enrollment system.In the case of Ellison, the leader was working to weight their lottery towards equity (by reserving a certain number of seats for low-income families).Several schools in Minneapolis-St.Paul (Clark, Rustin, Dewey) expressed the desire to be able to hold spaces, outside of the normal lottery process, to prioritize low-income families.Our findings suggest that policymakers should design systems that do not depend on individual schools' commitments to diversity, and ones that are not counterproductive for diversity.For example, lotteries, while aiming at equity, often advantage families who choose early, ones more likely to come from higher socioeconomic backgrounds.There are concerns about the transparency of existing lottery procedures, and whether they are truly random or fair (Dolle & Newman, 2008).There are procedures, however, that could introduce greater transparency and accountability in the enrollment practices of charter schools.Lotteries at the school level could, for example, hold seats outside the lottery system specifically for low-income students, if policies allowed that, or such priorities could be built into the OneApp or other centralized enrollment policies.

Conclusion
Overall, we find that intentionally diverse charter schools strive for more equitable choice and marketing practices, and in this sense, they hold promise for the charter sector.However, in other ways, diverse charters are reproducing inequities or struggling to maintain a diverse student body in the face of external pressures and lack of incentives.Without policy and regulation, including authorization requirements, weighted lotteries, and system-wide transportation interventions to support diversity missions, diverse charters will likely face challenges in sustaining their missions.These efforts to build diverse communities are also complicated by broader forces of gentrification and economic inequality.Against this backdrop, diverse schools will continue to struggle to make sure that their missions-many of which privilege racial equity-are not co-opted by more privileged families seeking diverse schools.Diversity, after all, is a term that evades specificity, sometimes sidestepping harder conversations about racial and economic inequality.Indeed, the fact that some of these schools had difficulty attracting low-income families and families of color may suggest that diversity has become a "good," which appeals to White, middle-class families, preparing their children for an increasingly competitive economic future (Reay et al., 2007), moving away from the civil rights goals of desegregation, which included redressing historic inequities.Moving forward, diverse charter schools-as a movement-may need to have some of these harder conversations about power, privilege and justice with their communities.

Table 1
Diverse Schools in Two Metro Areas