THREE PAPERS INVESTIGATING WHITE INDIVIDUALS’ OPINIONS AND ENROLLMENT PRACTICES CONCERNING DIVERSITY POLICIES IN SCHOOLS By Annie A. Hemphill A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Education Policy – Doctor of Philosophy 2023 ABSTRACT White individuals can play a key role in the realization of an education policy and can drastically impact how a policy is implemented because of their resources and the status that come with associating with the privileged racial group in the current racial hierarchy that exists in the United States. Several education policies, such as school desegregation and bilingual education, have goals of attaining education equity or targeting services to marginalized student groups, but these policy goals may not be realized when policies are implemented if White individuals’ exploit these policies. This dissertation is comprised of three papers that examine White individuals in different policy settings including policies that are designed to intentionally achieve a more diverse student body. In my first paper, I use data from a unique survey experiment conducted in November 2020 informed by interest convergence theory, I aim to understand how different policy frames regarding student benefits influence White individuals’ support for a policy that would increase the racial diversity of schools in their community. In my second paper, I conduct interviews with White parents at intentionally diverse charter schools throughout Denver to understand why they chose these schools and how they perceive the diversity of the schools’ student body. In my final paper, I employ a mixed methods approach to analyze which students are served by dual language immersion programs in North Carolina. I draw on various theoretical frameworks such as policy framing, colorblindness, diversity ideology, Whiteness as property, and interest convergence theory to illustrate how White individuals make sense of and relate to policies that attempt to integrate students from different backgrounds, while also challenging the initial frames of the policies so accommodate their own interests. These three papers draw on both qualitative and quantitative research methods to examine these behaviors. ACKNOWLEDGEMENTS I have many people I would like to thank that have helped me throughout this journey and through the writing of this dissertation. First, I would like to thank my dissertation committee members for their feedback and guidance on the many drafts of this paper that they looked over. I would especially like to thank my committee co-chairs, Dr. Josh Cowen and Dr. Sheneka Williams, for their constant support and encouragement throughout this project. Lastly, I could not have completed this if it wasn’t for my boyfriend, family, and friends who provided constant positive encouragement. I would like to thank the fellow grad students in my writing group and in the education policy program for being sounding boards and venting partners as we went through this process together. Finally, I would like to thank my boyfriend for helping me through the highs and lows and always being there for me. iii TABLE OF CONTENTS INTRODUCTION .......................................................................................................................... 1 PAPER 1: PUBLIC OPINIONS ON SCHOOL DESEGREGATION: A SURVEY EXPERIMENT USING POLICY FRAMES AND INTEREST CONVERGENCE THEORY .... 4 Introduction ......................................................................................................................... 4 Literature Review................................................................................................................ 8 Theoretical Frameworks ................................................................................................... 14 Study Contribution ............................................................................................................ 16 Data and Analysis ............................................................................................................. 17 Results ............................................................................................................................... 27 Discussion ......................................................................................................................... 36 Policy Implications and Future Research ......................................................................... 40 REFERENCES ................................................................................................................. 42 APPENDIX ....................................................................................................................... 51 PAPER 2: INTENTIONALLY DIVERSE CHARTER SCHOOLS: UNDERSTANDING WHITE PARENTS’ MOTIVATIONS FOR ENROLLING THEIR CHILDREN IN DIVERSE LEARNING ENVIRONMENTS.................................................................................................. 54 Introduction ....................................................................................................................... 54 Literature Review.............................................................................................................. 57 Theoretical Frameworks ................................................................................................... 63 The Denver Context .......................................................................................................... 67 Data and Methodology...................................................................................................... 69 Findings............................................................................................................................. 79 Discussion and Conclusion ............................................................................................... 89 REFERENCES ................................................................................................................. 92 APPENDIX ..................................................................................................................... 102 PAPER 3: DUAL LANGUAGE IMMERSION IN NORTH CAROLINA: A MIXED METHODS APPROACH EXAMINING CHANGES IN ACCESS TO DUAL LANGUAGE IMMERSION .............................................................................................................................. 105 Introduction ..................................................................................................................... 105 Background of Dual Language Immersion (DLI) programs .......................................... 107 Theoretical Frameworks ................................................................................................. 110 The North Carolina Policy Context ................................................................................ 113 Data and Methodology.................................................................................................... 116 Findings from Critical Policy Analysis........................................................................... 120 Growth and Changes in Access to DLI programs .......................................................... 125 Discussion ....................................................................................................................... 136 Conclusion ...................................................................................................................... 139 REFERENCES ............................................................................................................... 141 APPENDIX ..................................................................................................................... 147 iv CONCLUSION ........................................................................................................................... 153 v INTRODUCTION White individuals can play a key role in the realization of an education policy and can drastically impact how a policy is implemented because of their resources and the status that come with associating with the privileged racial group in the current racial hierarchy that exists in the United States. Several education policies, such as school desegregation and bilingual education, have goals of attaining education equity or targeting services to marginalized student groups, but these policy goals may not be realized when policies are implemented if White individuals’ exploit these policies. This dissertation is comprised of three papers that examine White individuals in different policy settings including policies that are designed to intentionally achieve a more diverse student body. I draw on various theoretical frameworks such as policy framing, colorblindness, diversity ideology, Whiteness as property, and interest convergence theory to illustrate how White individuals make sense of and relate to policies that attempt to integrate students from different backgrounds, while also challenging the initial frames of the policies so accommodate their own interests. These three papers draw on both qualitative and quantitative research methods to examine these behaviors. In my first paper, I use data from a unique survey experiment conducted in November 2020 informed by interest convergence theory, I aim to understand how different policy frames regarding student benefits influence White individuals’ support for a policy that would increase the racial diversity of schools in their community. I vary information about the types of benefit students receive and the student racial group who receives the benefit to understand if White individuals are more support of a policy after being exposed to information highlighting academic benefits and benefits for White students. I find that White individuals are the most 1 supportive of the policy when it is framed in race neutral language and the least supportive of the policy when White students are specifically mentioned. In my second paper, I conduct interviews with White parents at intentionally diverse charter schools throughout Denver to understand why they chose these schools and how they perceive the diversity of the schools’ student body. Drawing on the frameworks of colorblindness and diversity ideology, I find that while these schools are intentionally diverse, the diversity of the student body is the not key motivating factor that persuades White parents to ultimately enroll in these schools, with some saying that it was not considered at all. Additionally, White parents at these often schools view the diversity as something for their children to be consumed with the ultimate hope that the exposure will benefit their children in the future. Furthermore, White parents in my sample explained why the diversity at their charter schools was “acceptable” for them while also highlighting examples of schools with different student body makeups that they deemed as “unacceptable”. Many viewed diversity as something to be celebrated and sought after in some environments, but they could also be critical and dismissive of other economically and racially diverse schools. In my final paper, I employ a mixed methods approach to analyze which students are served by dual language immersion programs in North Carolina. I conduct a critical policy analysis of policy documents that pushed for an expansion of dual language immersion programs throughout the state and then compare school demographics in the schools that house dual language immersion programs before and after the policy release that prompted an expansion of the programs. Using the lenses of globalized human capital, Whiteness as property, and English hegemony I find that policy documents primarily rationalize the program expansion to generate a more globally competitive human capital while silencing folk bi and multilingual communities. 2 Additionally, I find that the early and late adopter DLI are comparable in student demographics except that late adopter DLI schools serve a significantly higher percentage of White students than early adopter schools. Together, these three papers add to multiple education policy literature canons and weave theories related to Critical Whiteness throughout to gain a better understanding of how White individuals view and use policies that have the potential of desegregating students. 3 PAPER 1: PUBLIC OPINIONS ON SCHOOL DESEGREGATION: A SURVEY EXPERIMENT USING POLICY FRAMES AND INTEREST CONVERGENCE THEORY Introduction School segregation continues to be a reality for the majority of students across the United States. Data from the U.S. Government’s Accountability Office indicates that over a third of K- 12 students attended predominantly same race/ethnicity schools in 2021 with 14% attending schools with a single race student body (Carillo & Salhotra, 2022). Numerous school districts, such as Seattle, Louisville/Jefferson County, and Detroit, have attempted to remedy school segregation by employing policies that manipulated student enrollment numbers through quotas, rezoning, or busing, to achieve a more racially and ethnically diverse student body (Kahlenberg, 2016; McDermott et al., 2015; Sandberg, n.d.; Holme & Finnigan, 2018). However, in all these cases, after the policy was implemented, it was overturned by the Supreme Court with White parents citing concerns about their children attending schools with lower performing students, typically students of color with lower incomes, fearing that the influence of these students would negatively impact their children’s learning outcomes (Holme & Finnigan, 2018; Sandberg, n.d.). One of the most recent cases of this occurred in Seattle when a group of parents, led by a White mother whose daughter attended school in Seattle, formed the group Parents Involved in Community Schools to advocate against the racial quotas that were used to assign students to public schools in the district (Holme & Finnigan, 2018; Sandberg, n.d.). These efforts resulted in the 2009 Supreme Court case Parents Involved in Community Schools v. Seattle School District No 1. which ruled to prohibit school districts from using student race as the determining factor when deciding student placement (Sandberg, n.d.). This ruling hampers schools’ and districts’ authority to consider a student’s race when instituting school desegregation programs. Thus, successful desegregation policies now must rely on fostering community support among White 4 parents and community members so that they encourage and voluntarily take part in such programs. This study experiments with policy frames to understand how the framing of desegregation policies can be used as a tool for garnering such support. White community members also have a history of exiting school districts, a trend termed White flight, when school desegregation plans gain traction. This also undermines desegregation efforts because the Supreme Court has restricted the ability to construct desegregation policies that desegregate schools across district lines. (Holme & Finnigan, 2018; Sandberg, n.d.). In 1973, the Supreme Court case Milliken v. Bradley, concerning the segregation of Detroit Public Schools, ruled that districts could not desegregate across school district boundaries, unless these lines were explicitly proven to be drawn to segregate by race. Forty-five years after the Milliken vs. Bradley case, Detroit Public Schools and the nearby suburban Grosse Pointe district remain two of the most segregated districts in the country (Balsa & Levin, 2019). Because of this ruling that restricts inter-district desegregation, when White community members move across district lines, the district can no longer include them in their desegregation efforts and districts become segregated across district boundaries. Therefore, school districts aiming to implement desegregation policies need to convince at least some of the White families to stay in the district and be a part of the desegregation effort. White citizens’ responses to policies aiming to further school integration are clearly an important component when it comes to feasibly implementing these plans. In addition, White community members’ responses to policies aiming to further diversify schools are an important consideration because of their privilege and its accompanying political clout. As James Scheurich explained, the privilege associated with being White “is akin to walking down the street with money being put into your pocket without your knowledge” 5 (Leonardo, 2013). This privilege can have major political consequences for White individuals and people of color because politicians consider White individuals’ priorities more often during political decision-making processes. In a study that analyzes the racial inequalities found in the local policies that are enacted across the United States, researchers find that local governments enact more policies that align with their White constituents’ ideals (Rhodes et al., 2020). This is even true in areas where White individuals ware the minority. Policy makers are listening to White voter’s opinions about public policies, particularly liberal leaning race-oriented policies such as school desegregation, and following suit, making White individuals a powerful group when it comes to gaining support for a policy (Rhodes et al., 2020). White individuals’ ability to wield power to influence the policy making and implementation processes, and their privileged position that benefits from the legal interpretation of the issue of school desegregation, are important factors to successfully desegregate schools in the current racial and legal environment. Interest convergence theory may offer a lens when considering how to gain White individuals’ support. Interest convergence theory argues that policies that advance for the minority group are only supported when they converge with the interest of the majority (Bell, 1980; Delgado & Stefancic, 2012). However, in the case of increasing racial diversity in schools, policies with this aim typically frame policy benefits by emphasizing the advantages for students of color (Schofield, 1981, 1995; Orfield, 2001; Wells et al. 2016). Because of this framing, White individuals may feel like they must make a choice between a diverse school or a “good” school because they are not clear how or if diversity benefits White students, leading to further resistance (Roda & Wells, 2013). Using interest convergence theory to inform the policy framing of racial diversity in schools by specifically highlighting to White citizens the benefits received by White students can help to further clarify how White individuals’ view racial 6 diversity in schools. Furthermore, framing the policy by emphasizing the academic versus nonacademic students benefits of school desegregation can challenge the preconceived belief held by White parents that diverse schools and “good” schools are separate entities (Roda & Wells, 2013). Testing different policy frames with interest convergence theory in mind could better inform the implementation of these policies in the future to garner support from White community members so that they do not attempt to block these policies and repeat the behaviors of the past. Using data from a national survey experiment, I test different policy frames for increasing racial diversity in schools by varying informational treatments about students’ benefits from attending desegregated schools. The policy frames vary the student target group and type of student benefits received from attending a racially diverse school to better understand how these framings influence White support for a hypothetical policy that would increase the racial diversity of schools in their community. This study is guided by three research questions: RQ1: How does exposing individuals to policy frames emphasizing the student benefits of school desegregation impact their support for increasing the racial diversity in schools in their community? RQ2: How does exposing White individuals to policy frames emphasizing White students’ benefits from desegregated schools’ impact White individuals’ support for increasing the racial diversity of schools in their community? RQ3: How does exposing White individuals to policy frames emphasizing the nonacademic vs. academic benefits of desegregated schools’ impact White individuals’ support for increasing the racial diversity of schools in their community? 7 In the following sections I will first review relevant literature on Critical Whiteness and how a school’s diversity is considered by White parents when they contemplate schooling options for their children. Next, I further explain interest convergence theory and how I used the theory to inform my survey experiment design. Third, I describe my survey experiment, the methods of analysis I employ to answer each research question, and the results. Lastly, I end with a discussion of the results and the policy implications of my findings. Literature Review Whiteness Race is a socially constructed idea in American society, but the fact that race is created by society does not negate the implications that are experienced by different racial groups (Anderson, 1999; Castagno, 2014). The construction of race has real, systemic, and lasting material consequences. Racial groups are defined as “those that have been categorized and singled out on the basis of presumed physical or cultural characteristics and that are subject to groups subordination and domination (Anderson, 1999, pg. 6). Racial categories are difficult to define because, historically, the have not been fixed in the United States, but are products of processes and experiences that have evolved to represent ever changing power dynamics. Critical Whiteness seeks to explore the ideologies of racial power imbalances that Whiteness seeks to maintain that have resulted in a society where White individuals hold the dominant and privileged position among the racial and ethnic groups in the United States (Castagno, 2014). Research examining Whiteness has used race as an ideological bakcdrop that shapes how individuals understand and react to policies that attempt to diversify schools (El-Haj 2006; Castagno, 2014; Gillborn 2008; Ladson-Billings and Tate 2006; Leonardo 2009; Vaught 2011). Acknowledging the power of Whiteness is key to understanding White individuals’ 8 priorities for their children’s’ schooling and their support for policies that aim to increase the racial diversity in schools. White Attitudes Towards Racial Diversity in Schools Research on White citizens’ opinions around school desegregation policies finds that White individuals have mixed and often conflicting attitudes surrounding racial diversity in schools. Several studies using data from public opinion polls report that White participants are overwhelmingly in favor of increasing diversity in schools, at least conceptually. For example, in a 2007 NORC poll, 95% of White participants support Black and White students attending the same school (Frankenberg & Jacobsen, 2011). In addition, a 2017 Phi Delta Kappan survey indicated that 48% of White participants feel that a racially and ethnically diverse student body is extremely/very important, and 70% of White participants indicate a preference for racially diverse schools. Nevertheless, White attitudes towards increasing racial diversity in schools are less supportive when desegregation policies are implemented in their own communities. Nearly every district that has implemented mandatory busing policies has experienced public pushback from local White residents (Carlson & Bell, 2021). Furthermore, almost no district released from a court mandated desegregation plan has voluntarily enacted a policy to continue to pursue school desegregation (Carlson & Bell, 2021). Other districts have voluntarily attempted to desegregate public schools by redrawing attendance zones and offering voluntary school choice programs, but they have also experienced significant resistance from White residents (Taylor & Parcel, 2015). Although polls indicate that White individuals generally support the idea of increasing the racial diversity in schools, their actions tell a different story, especially when these pro- 9 diversity sentiments are attached to a policy that will directly impact their community. This discrepancy may be explained by several different ideas. First, Fenno’s Paradox may apply to the disconnect between stated and real support for school desegregation. Fenno’s paradox examines the disconnect between the public’s support for their individual congressperson and their lack of support for Congress more generally (Harbridge & Malhotra, 2011). The disconnect between support for a general policy that increases racial diversity in schools and the resistance to a policy that would do so in a person’s specific district may demonstrate how individuals apply different standards of judgement to policies depending on the context. White individuals’ proximity to the policy proposed for their district may make them more critical of the policy because they see how they personally may be affected (Taylor & Parcel, 2015). For example, when Taylor & Parcel (2015) survey White individuals examining their support for a school integration policy, they find that while White liberals are more supportive of the policy as a broad concept then White moderates or conservatives, this difference becomes statistically insignificant when White individuals’ personal situations are more directly impacted by the policy, denoted by having a school aged child. A second but perhaps related possibility that explains this discrepancy of support comes from research revealing White parents’ contradictory opinions about school desegregation because of the multiple priorities that they consider when choosing a school for their child (Roda & Wells, 2013). Interview studies indicate that White parents do want to send their children to a racially diverse school. One such study finds that White parents think attending diverse schools will help their children develop nonacademic qualities which will prepare their children for the increasingly diverse twenty-first century (Wells et al., 2009). Other scholars find that White parents want their children to attend schools with a diverse student body because they feel it will 10 give their children more “real world” experiences (Byrne & De Tona, 2014; Hollingworth & Williams, 2010), and the necessary social skills to help them navigate racially and ethnically diverse contexts in the future (Hernandez, 2019; Kimelberg & Billingham, 2012). Additionally, scholars find that White students in more racially diverse schools have better critical thinking skills compared to White students in majority White schools (Wells et al., 2016) and a lower probability of dropping out of high school (Billings et al., 2014). Even though there are student benefits associated with racially diverse learning environments, and White parents report considering a school’s diversity when choosing a school, research indicates that many White parents prioritize a schools’ academic performance measures when choosing a school for their child. In a study that surveys 1,898 White parents and interviews a subsample of parents, researchers find that while White parents desire racially and socioeconomically diverse schools for their children, they ultimately choose schools based on their academic profiles, safety record, and location, pushing their consideration of the racial and socioeconomic makeup of the school towards the bottom of their list of priorities (Hernandez, 2019; Torres & Weissbourd, 2020). Furthermore, White parents living in urban areas report a dissatisfaction with their neighborhood schools because they believe those schools provide less physically and emotionally safe classroom environments citing concerns of their children being victims of bullying (Crozier et al., 2008). In addition, White parents feel that their urban neighborhood schools could stunt the academic achievement of their children because teachers may have to devote more of their time to meet the needs of lower achieving students. Specifically, parents cite concerns about schools in urban areas having more English Language Learners (ELLs) and students with special needs that may require more of the teacher’s time (Levine-Rasky, 2008; Vowden, 2012). 11 Although studies on survey research have cast substantial light on the question of the role of race in school preferences—and the present paper indeed draws on survey research for its empirical contribution—survey responses on the topic of race may be limited to some extent by the way parents choose to reveal underlying preferences to researchers. Another source of data on school preferences comes from enrollment applications in portfolio-style choice cities. For example, Lincove et al. (2018) use OneAppp application data to examine parents’ implicit school values and show that race plays a minor role in parents’ school preferences in New Orleans. Lincove et al. (2018b) also exploits OneApp application data to analyze the differences between a student’s school quality placement and their application preferences and finds that family’s first choice schools have better school letter grades, and more experienced teachers. In addition, they find that parents rank schools higher that demonstrate higher academic performance. Harris and Larson (2019) also study the preferences of parents in New Orleans using school choice application data and find that parents prefer schools with higher value-added scores, and expansive extracurricular offerings. Abdulkadiroglu et al. (2020) use New York City parent application data to examine parental preferences in relation to short and long term achievement outcomes and find that parents prefer schools with higher achieving students and schools with a more educated teacher workforce. Denice and Gross (2016) use Denver Public School’s centralized application data to determine how the supply of nearby schools shapes parents’ choices and finds that Hispanic, Black, and White parents all prefer schools with higher academic performance compared to their zoned public school. However, the supply of schools in the nearby vicinity differs amongst racial and ethnic groups with Hispanic and Black parents having less academically high achieving schools nearby. Glazerman and Dotter (2017) utilize rank ordered school choice applications from Washington D.C. and find that academic 12 performance is the most prioritized factor for parents. These studies indicate that a school’s academic performance, rather than the diversity of the student body, is consistently the most valued aspect for families who are selecting schools in choice rich environments. Therefore, while White parents claim to want to send their children to racially diverse schools, they may only do so if doing so does not force them to compromise on other factors that they deem are more essential (Evans, 2021; Roda & Wells, 2013) Research relying on surveys and choice application data indicate that parents prioritize a school’s academic reputation, but these studies may be limited and underestimate parents’ consideration of race when selecting a school because of social desirability bias. Social desirability bias, where White individuals respond in more politically correct, or colorbind (Bonilla-Silva, 2003) ways to avoid being perceiving as racist can muddy White individuals’ true opinions on increasing racial diversity in schools and how they consider the race of the student body when choosing a school for their child (Evans, 2021; Hailey, 2022; Billingham & Hunt, 2016). For example, Evans’ (2021) interview study in which they interview self-identifying progressive White parents with pro-diversity beliefs finds that these participants still evaluate schools with anti-Black stereotypes and ultimately avoid majority Black schools. Furthermore, survey experiments demonstrate how White individuals prefer White schools even when accounting for other school factors that research indicates are top priorities for parents. In Billingham & Hunt’s (2016) survey experiment analyzing the influence of a school’s racial composition on parent’s school choices, they find that even when factors that proxy for race are controlled for, such as safety, academics, and location, White parents ultimately prefer schools with a higher percentage of White students. Additionally, in Hailey’s (2022) similar survey experiment, she finds that a schools’ racial composition has a larger effect on White parents’ 13 choices compared to Black parents’ choices, with White families preferring Whiter schools, independent of school characteristics that correlated with race. Theoretical Frameworks Policy Framing and Policy Priming Framing theory can provide insight to disentangle the apparent contradiction in White individuals’ attitudes regarding racial diversity in schools. Scholars draw on framing theory to understand the relationship between issue description and survey responses to measure policy attitudes (Svallfors, 2012). Framing scholars test the effect of different policy frames by varying the descriptions of the policies in experimental settings (Chong & Druckman, 2007; Sniderman & Grobb, 1996). Scholarship on framing theory is based on psychological cognition theories (Chong & Druckman, 2007) that assert that individual choices and opinions are products of low- information, uncertainty, and/or ambivalence. Based on this logic, the measurement of attitudes is inherently predictable, with participants responding to survey items based on some predispositions to personality characteristics or routine behavior (Svallfors, 2012). Nevertheless, framings can disrupt this probabilistic aspect of attitude measurement because they can influence attitude formation by making new information available. The framing of policies can occur in several ways. First, frames can provide new information about the external environment in an attempt to alter one’s opinion. Frames can do this using emphasis policy framing effect, where the framing of a policy emphasizes a subset of details to focus the participant’s attention to said details when determining their level of support (Druckman, 2001). A second method utilizing policy frames occurs when the frames relay new information that accesses the biases that individuals already believe in a process known as priming (Svallfors, 2012). For an illustrative example of priming, Svallfors’ (2012) survey experiment finds that 14 when questions about support for welfare policies are connected with Black or immigrant target groups in their descriptions, an individual’s support for welfare significantly declines. Svallfors (2012) found that priming is important because it develops new connections between the preexisting beliefs about people in the target groups and the policy issues, prompting individuals to adjust their policy opinions. Interest Convergence Theory as a Policy Framing Lens When analyzing the ways that policy priming can specifically influence White individual’s perception and support for increasing racial diversity in schools, I rely on interest convergence theory to construct policy frames regarding student target groups. Interest convergence theory was first introduced by Bell in 1980 when he claimed that “the interest of blacks in achieving racial equality will be accommodated only when it converges with the interests of whites.” (Bell, 1980, p. 523). For interests to converge, it is often assumed that the dominant group must give something up (Lopez, 2003; Milner, 2008). This concept creates a competitive atmosphere where White individuals feel that White children may lose their access to “better” schools if people of color gain access, ultimately preventing a convergence of interests and perpetuating a zero-sum mentality. However, Aléman & Aléman (2010) clarify that interest convergence theory can be used as a conceptual tool to understand the adoption of past racial policies, or as a strategic tool to influence the public’s opinion on policies concerning race. Literature applying interest convergence theory to school desegregation has primarily utilized the theory as a conceptual tool to provide alternative rationales to explain the success of school desegregation rulings (Bell, 1980; Leigh, 2003). For example, Bell (1980) first applies interest convergence theory when conducting a historical analysis of events leading up to The Supreme Court’s Brown v. The 15 Board of Education decision. Bell (1980) argues that White Americans had much to gain from desegregating schools, and that these gains for White individuals, rather than White morality, were the motive behind the decision. While interest convergence theory has been used in historical analyses examining the adoption process of school desegregation policies, it has not been employed as a theoretical framework in conjunction with policy framing and priming to test public support for increasing racial diversity in schools. Education literature that utilizes interest convergence theory as a political tactic has employed the theory to study how it sways support for other education policies such as affirmative action (Castagno & Lee, 2007), teaching about racial bias in teacher education programs (Milner, 2008), and equitable school finance laws (Adamson, 2006). These authors view interest convergence theory as a glimmer of hope for social justice progress and an alternative means of negotiating group interests to gain White support for racial policies. Study Contribution After considering the literature on White individuals stated versus revealed preferences toward racial diversity in schools, it is clear that more research needs to be done that explores how White individuals make sense of racial diversity in schools. Overall, some White individuals claim to support the idea of racial diversity in schools but have conflicting views when they begin to consider it in terms of its relationship to academic and nonacademic outcomes (Roda & Wells, 2013). Some White individuals view racial diversity in schools in zero sum terms and assume that they must sacrifice academic rigor to have a more diverse learning environment (Roda & Wells, 2013). White parents who claim to value racially diverse schools continue to opt for White majority schools, even when the factors that they list as top 16 priorities, such as academics, safety, and location, are synonymous between White majority and minority majority schools (Billingham & Hunt, 2016, Hailey, 2022). Most theoretical frameworks that examine White individuals’ views do not consider the role of policy framing and its contribution to policy attitudes. Further evidence on adjustments to White individuals’ attitudes based on different frames and the priming of existing attitudes is crucial to understanding the nuances of these stances. Drawing on interest convergence theory to inform these policy frames in a survey experiment illustrates a unique approach that bridges these two theoretical canons to better understand White individuals’ preferences for racial diversity in schools. Druckman, et al. (2006) argues that survey experiments can be used to identify causal relationships, test theories, and/or inform policy by analyzing real-world situations. The information gleaned from this study accomplishes all three of these tasks by randomly assigning people to treatment groups and testing different policy frames about racial diversity. Moreover, the results produce insights for policy makers and schools who are attempting to increase the racial diversity of their schools by tapping into the preferences of White individuals when they are exposed to different types of information about student benefits in racially diverse learning environments. Data and Analysis To answer my research questions, I utilize data from an original survey experiment. In the following section I first outline how the aforementioned literature and theories inform my hypotheses for each research question and the survey design. Next, I explain my process for developing my survey experiment items and the provide further information regarding the survey 17 experiment design and administration. Lastly, I describe my analysis strategy for each research question. Applying Framing and Interest Convergence in a Survey Experiment To address my first research question, how does exposing individuals to policy frames emphasizing the student benefits of school desegregation impact their support for increasing the racial diversity in schools in their community, I first ask participants to rank their level of support regarding increasing racial diversity in schools with no additional information about students. I then employ emphasis policy framing by exposing individuals to one of six treatment groups that gives the participant information connecting a specific target group of students to a particular type of student benefit received from attending desegregated schools. After this exposure, I again ask them about their support for increasing racial diversity. I hypothesize that participants will be more supportive of increasing the racial diversity of schools after being exposed to this new information about student benefits in the treatment groups that reframes racial support in a positive light by emphasizing a benefit to students. Interest convergence theory and policy priming inform my second hypothesis in relation to research question two which asks, how does exposing White individuals to policy frames emphasizing White student’s benefits of desegregated schools’ impact White individuals’ support for increasing the racial diversity of schools in their community? I hypothesize that White individuals will be more supportive of increasing racial diversity in schools when they are exposed to policy frames where White students are the target group connected to the student benefits compared with when the target groups “students of color” or “all” students are connected to the same benefits. 18 Additionally, the mindsets revealed from literature on White parents’ competing priorities informs my third research question, how does exposing White individuals to policy frames emphasizing the nonacademic vs. academic benefits of desegregated schools’ impact White individuals’ support for increasing the racial diversity of schools in their community? If White citizens are exposed to information stating that students academically benefit from attending a racially diverse schools, could this defy their current ideas that diverse and academically rigorous environments occur in isolation? By altering the types of benefits students receive in the policy framing between a nonacademic benefit, which is commonly thought of as a benefit from attending a more diverse school, compared to an academic benefit, which challenges a parent’s preconceived ideas about racially diverse classrooms, I am introducing individuals to a new, competing policy frame that challenges their currently held beliefs (Rein & Schön, 1996). Changing existing beliefs involves individual reflection to identify existing beliefs, and a revision of current attitudes when new information is presented (Rein & Schön, 1996). Framing racial diversity in schools by emphasizing academic benefits, which many White individuals in previous studies claim to give up at the expense of enrolling their children in diverse schools, forces White individuals to reflect on their previously held beliefs and could potentially change their view of racially diverse schools. I hypothesize that White individuals will be more supportive of increasing racial diversity in schools when they are exposed to treatments that challenge the existing beliefs and tout the academic benefits of racially diverse schools compared to the nonacademic benefits. Survey Sample and Administration I leverage data from a nationally representative survey experiment conducted in November 2020. The survey was conducted through a public opinion survey platform, Lucid 19 Academia, and received a response rate of 44%, yielding a total of 1,529 complete observations. Table 1 shows the demographics of the sample compared to the general United States population according to the 2020 U.S. Census. The table indicates that my sample and the general population are comparable with the largest differences being the population of White participants, with my sample having a larger share of White participants than the U.S. population. However, this overrepresentation of White participants works in my favor since I am particularly interested in White participants’ responses to the various policy frames. 20 Table 1 Descriptive Statistics Table 1: Descriptive Statistics Sample U.S. 2020 Census and Pew Research n Mean/Prop. Freq. Mean/Prop. Gender 1524 Gender Male 47% 709 Male 49% Female 53% 810 Female 51% Age 1507 Age 18-34 30% 451 18-34 23% 35-54 31% 465 35-54 25% 55+ 39% 590 55+ 30% Race 1519 Race White 78% 1178 White 63% Black 8% 122 Black 12% Latinx 6% 94 Latinx American Indian/Alaskan 1% 15 American Indian/Alaskan 1% Asian 4% 66 Asian 6% Native Hawaiian 0% 4 Native Hawaiian 0% Other/ Two or more 3% 40 Other/ Two or more 18% Education 1521 Education No HS degree 2% 36 No HS degree 11% HS degree 23% 343 HS degree 26% Some College 23% 353 Some College 20% 2-yr dgr 12% 181 2-yr dgr 9% 4-yr dgr 24% 369 4-yr dgr 21% Post-grad 16% 239 Post-grad 14% Income 1514 Income Under 20K 19% 287 Under 20K 7% 20-40K 23% 353 20-40K 13% 40-60K 19% 284 40-60K 14% 60-80K 14% 212 60-100K 25% 80-120K 13% 199 100-150K 20% Over 120K 12% 179 Over 150K 22% Ideology 1567 Ideology Extremely Lib 9% 143 Extremely Lib Lib 14% 224 Lib Slightly Lib 10% 151 Slightly Lib Middle 35% 544 Middle Lean Cons 9% 142 Lean Cons Cons 13% 202 Cons Strong Cons 10% 161 Strong Cons Political ID 1570 Political ID Strong Dem 15% 231 Strong Dem Dem 16% 251 Dem 33% Lean Dem 9% 148 Lean Dem 49% Independent 27% 425 Independent 34% Lean Rep 9% 146 Lean Rep 44% Rep 11% 167 Rep 29% Strong Rep 13% 202 Strong Rep Region 1530 Region West 34% 522 West 24% SW 12% 177 South 38% Midwest 18% 281 Midwest 21% SE 24% 372 NE 17% NE 10% 158 Not in US 1% 20 21 The survey experiment began by first asking participants to record their level of support for education policies that would increase the racial diversity of schools in their community. This first item acts as a baseline question to understand participant’s initial level of support and allowed for within subject analysis comparing the level of support between baseline and treatment responses. After responding to the baseline item, participants were then randomly assigned to receive one of six policy frames, as summarized in Table 2, and were then asked again to record their level of support for implementing a policy that would increase the racial diversity of schools in their communities. Table 2 Randomly Assigned Table Treatment 2: Randomly Conditions Assigned Treatment Conditions Treatment N (White) Policy Frames 1,529 How do you feel about policies that would increase Baseline (1,178) the racial diversity of schools in your community? T1 White Research shows that White students achieve higher Higher Test 241 (190) test scores when they attend racially integrated Scores schools. T2 SoC Research shows that students of color students Higher Test 231 (189) achieve higher test scores when they attend racially Scores integrated schools. Research shows that all students achieve higher test T3 All Higher 270 (213) scores when they attend racially integrated schools. Test Scores Research shows that White students are more T4 White 251 (200) accepting of others when they attend racially Accepting integrated schools. Research shows that students of color students are T5 SoC 254 (194) more accepting of others when they attend racially Accepting integrated schools. Research shows that all students are more accepting of T6 All 259 (192) others when they attend racially integrated schools. Accepting 22 This survey experiment was designed so that treatment groups varied along two policy frames: varying the framing of target student group receiving a benefit (White, students of color, or all students) and varying the framing of type of benefit received (academic or nonacademic). These policy frames were informed by the literature that indicated that White participants operate under the assumption that White students do not benefit from school integration and view diverse and academically rigorous schools as mutually exclusive concepts (Roda & Wells, 2013). Two groups of participants received policy frames with White student target groups (T1 and T4). One of these two treatment groups said that White students receive an academic benefit (T1) (increased test scores) and the other said that White students receive a nonacademic benefit (T4) (becoming more accepting of others). Two other groups of participants received policy frames about benefits for students of color (T2 and T5). Again, one of these two treatment groups said that students of color receive an academic benefit (T2) and the other said that students of color receive a nonacademic benefit (T5). Lastly, two other groups of participants received policy frames where “all” students receive a benefit (T3 and T6). Again, one of the two treatment groups received information about “all” students gaining an academic benefit (T3) and the other said “all” students receive a nonacademic benefit (T6). Administering a survey experiment in a controlled environment such as an online survey platform offered several advantages, including the ability to control the information to which the participants were exposed (Haderlein, 2021). While survey experiments do not offer insight into all of the nuances of White individuals’ attitudes regarding racial diversity in schools, they are useful for testing different information frames and associations in the absence of information asymmetries (Haderlein, 2021). Furthermore, experiments randomly sort individuals into 23 treatment groups enabling the results to form causal claims which can provide helpful information for policymakers interested in garnering community support for a similar policy in their community (Druckman et al., 2006). Survey Experiment Item Development To evaluate my survey items, I received feedback in several different environments before the formal survey was conducted to enhance the surveys construct validity. After my survey items were first drafted, I conducted an expert validation check to assess the items ‘clarity and diction to ensure that I avoided using language unfamiliar to the general public (Gehlbach & Brinkworth, 2011). I shared the questions with a group of political science doctoral students and one political science faculty member. I presented my research idea and survey items to this group and received feedback on the item’s wording, scale, and the overall design of the experiment. Their feedback assisted me in changing the wording and scale in my final items and removing some unfamiliar jargon. In addition, I conducted a pilot survey using the online survey program MTurk. MTurk is an online platform that connects employers with workers to complete online surveys for a small compensation (Amazon MTurk, 2018). My survey items were launched as part of a larger survey on October 14, 2020. In this survey, 2,083 participants responded to my survey experiment items and 2,180 participants completed the entire survey during the 24 hours window that it was active for a response rate of 89.4%. By conducting a pilot survey on MTurk, a system that closely resembled the platform of the official survey, I was able to ensure that the survey was readable and clear on an online platform and that the randomization for the experiment was programmed correctly. 24 Empirical Analysis The first research question asked, how does exposing individuals to policy frames emphasizing the student benefits of school desegregation impact their support for increasing the racial diversity in schools in their community? To answer this, I used t-tests to conduct a within subject analysis comparing the mean level of support in between baseline and treatment groups. This within-subject comparison illustrated whether exposing participants to any information about student benefits impacted their support for policies that would increase the racial diversity of schools in their community. Next, I estimated a regression model (Equation 1) to answer the second research question, how does exposing White individuals to policy frames emphasizing White student benefits of desegregated schools’ impact White individuals’ support for increasing the racial diversity of schools in their community? The model predicted the support for increasing racial diversity in schools (Supporti) as a function of (Treatmenti) which indicated whether participant i receives a treatment with the target group being all students, students of color, or White students. I also include the baseline variable capturing participants’ baseline level of support (Baselinei), a vector of control variables (Xi; gender, citizenship status, age, household income, education attainment level, political party, U.S. region of residence, and parental status), the intercept (β0), and the error term (εi) Supporti = β0 + β1Treatmenti + β2Baselinei + γXi + εi In this model, I combined participants in T1 and T4 which both referred to White students, T2 and T5 which both referred to students of color, and T3 and T6 which both referred to all students. The baseline measure is included in the model to further capture the causal impact of the policy frame on support for racial diversity by controlling for their baseline level of support. 25 The vector of control variables included gender, age, educational attainment, income, region of residency, and political ideology—all identified in the literature as factors that have been associated with individual racial attitudes (Fossett & Kiecolt, 1989, Hughes & Tuch, 2003, Krysan, 1998; Taylor & Mateyka, 2011). In the first estimate, I compared treatment groups who received frames referring to White students (T1, T4) versus those groups who received frames referring to students of color (T2, T5). I did this for the aggregate, and then restricted my sample to only White participants, and then restricted my sample again to only participants of color. Next, I compared the level of support for treatment groups who received frames referring to White students (T1, T4) versus treatments groups who received frames referring to all students (T3, T6). Again, I did this for the aggregate, then only the White participants, and then only the participants of color. Lastly, I compared the level of support for groups who received frames referring to students of color (T2, T5) versus those groups who received frames referring to all students (T3, T6). Again, I did this for the aggregate, then only the White participants, and then only the participants of color. My third research question asked, how does exposing White individuals to policy frames emphasizing the nonacademic vs. academic benefits of desegregated schools’ impact White individuals’ support for increasing the racial diversity of schools in their community? To answer this, I relied on the same regression model that was utilized above but a different construction of the treatment groups. I combined T1, T2, and T3 which all referenced an academic benefit and combined T4, T5, and T6 which all referenced a nonacademic benefit and then compared the level of support between these two groups. I also conducted this analysis for the aggregate, and then restricted my sample first to White participants, and then to participants of color. 26 Results The distribution of support for the baseline question and each treatment item are displayed in Figure 1. The neutral position is the most frequently used response in each survey item with about 35% of participants choosing this option. This preference for the middle option or neutral position is common in surveys that ask participants to take a position on a topic (Shaw et al., 2000). The distribution across the baseline item indicates a multimodal distribution with most participants choosing either neutral, strongly oppose, or strongly support. This same pattern is apparent in the distribution for treatment 1. In the other treatment items, there is a bimodal distribution with the most participants choosing either neutral, or strongly support. Figure 1: Average Level of Support between Baseline and Treatment 4 Level of Support (5pt scale) 3.8 3.6 3.4 3.2 3 2.8 2.6 2.4 2.2 2 Figure 1 displays the average level of support from participant’s baseline and treatment responses. This demonstrates that participants are more supportive of increasing diversity in schools in their community after being exposed to any of the treatment conditions compared to their baseline response regardless of the framing. Table 3 shows that these differences are statistically significant at the .01 level. Therefore, when racial diversity in schools is framed as a benefit to students, public support can significantly increase. This supports the first hypothesis 27 and the policy framing literature which argues that exposing participants to new information, especially information framing the policy in a positive light, can influence participants’ support for the policy. Table 3: Difference Table Between Baseline 3: Difference and Treatment Between Baseline Level of SupportLevel of Support and Treatment Treatment Baseline Treatment Treatment Mean Group n Mean Baseline S.D. Mean S.D. Difference T1 508 3.081 1.301 3.205 1.322 0.124*** T2 508 3.12 1.357 3.402 1.329 0.281*** T3 514 3.21 1.311 3.43 1.281 0.219*** T4 518 3.125 1.384 3.313 1.338 0.187*** T5 516 3.145 1.378 3.378 1.326 0.233*** T6 509 3.24 1.321 3.497 1.293 0.257*** Notes: Notes:*** ***p<.01 p<.01**p<.05 **p<.05*p<.1 *p<.1 While all the differences between baseline and treatment support are statistically significant, the largest difference are for participants exposed to treatments that frame all students becoming more accepting (T6) and frame students of color achieving higher test scores (T2) when attending desegregated schools. These treatments differ in their student target groups and types of benefit and indicate that the difference between baseline and treatment support is greater when students of color or “all” students benefit from school desegregation compared to when participants receive a frame referencing White students. Furthermore, the treatment groups with the smallest difference between baseline and treatment support were treatments one and four which both frame White students receiving a benefit from attending a desegregated school with treatment one referring to an academic benefit (increased test scores) and treatment four referring to a nonacademic benefit (becoming more accepting of other). Table 4 compares the means from unpaired t tests to examine if the mean level of support was significantly different between treatment groups. In this table, treatment one, which frames White students receiving higher test scores after attending a desegregated school, acts as the 28 reference group. The results indicate that participants are significantly more supportive of frames where students of color or “all” students receive a benefit compared to White students, regardless of the type of benefit (academic vs. nonacademic) mentioned. However, the differences are greater between treatment one, and the treatments that state that “all” students receive either an academic or nonacademic benefit from attending a desegregated school. Table4:4 Difference Table DifferencesBetween BetweenTreatments’ Treatments' Mean Level of of Support Support Treatment Mean Level n S.D. Mean Difference Group of Support T1 513 3.205 1.317 T2 514 3.389 1.332 .184** T3 517 3.418 1.289 .213** T4 521 3.311 1.334 0.106 T5 516 3.378 1.326 .173** T6 513 3.493 1.295 .288*** Notes: *** p<.001 Notes: p<.001 **p<.05 **p < .05. *p < .01. *p<.01. T1reference T1 is is reference group group. The results in Table 5 reveal that the hypothesis informed my second research question asserting that White participants would be more supportive when White students receive a benefit does not hold. Table 5 includes responses from all participants (columns 4-6), White participants only (columns 1-3), and participants of color only (columns 7-9). I present the results with the baseline level of support included as they are similar to the results when it is omitted although larger in magnitude (see Appendix Table 5b). I estimate the regression three times for the aggregate, three times for the restricted sample with only White participants, and three times for the restricted sample with only participants of color. I first compare treatments framing White students versus students of color, then compare treatments framing benefits to White students versus all students, and lastly comparing treatments framing students of color versus all students. 29 Table 5: Student Target Group Regression Table 5: Student Target Group Regression White Participants Only All Participants Participants of Color Only (1) (2) (3) (4) (5) (6) (7) (8) (9) Variables White v SoC White v All SoC v All White v SoC White v All SoC v All White v SoC White v All SoC v All treat_whitevsoc -0.145** -0.121* -0.0109 -0.0722 (0.0660) (0.162) treat_whitevall -0.172** -0.200*** -0.279* -0.0677 (0.0619) (0.159) treat_socvall -0.021 -0.0785 -0.282** -0.0666 (0.0599) (0.140) baseline 0.556*** 0.558*** 0.593*** 0.557*** 0.562*** 0.594*** 0.557*** 0.561*** 0.585*** -0.0375 -0.0377 -0.0363 (0.0322) (0.0324) (0.0309) (0.0640) (0.0638) (0.0605) Female 0.168** 0.173** 0.0409 0.128* 0.165*** 0.0961 0.0542 0.181 0.314** -0.0726 -0.0687 -0.0652 (0.0666) (0.0627) (0.0604) (0.165) (0.151) (0.141) Person of Color -0.157* -0.0614 -0.0937 (0.0916) (0.0892) (0.0828) Not a US Citizen -0.235 -0.237 -0.216 -0.0811 -0.0822 -0.158 0.0332 -0.104 -0.129 -0.233 -0.211 -0.228 (0.200) (0.166) (0.174) (0.290) (0.227) (0.252) Age 0.0464 0.0199 0.012 0.0980** 0.0450 0.0256 0.309** 0.183 0.199* -0.0491 -0.0466 -0.0457 (0.0447) (0.0427) (0.0412) (0.121) (0.113) (0.114) Income -0.00488 -0.0353 -0.0342 -0.0284 -0.0434 -0.0365 -0.120 -0.0745 -0.0593 -0.0332 -0.0343 -0.0319 (0.0316) (0.0309) (0.0285) (0.0817) (0.0700) (0.0676) Education Level 0.0201 -0.0341 -0.0163 0.0455 -0.0118 -0.00782 0.135 0.0692 0.0176 -0.0491 -0.0463 -0.0475 (0.0464) (0.0440) (0.0427) (0.124) (0.114) (0.0991) Pol Party Ind -0.214** -0.159* -0.103 -0.191** -0.142* -0.103 -0.0933 -0.0189 0.0309 (0.0875) (0.0836) (0.0794) (0.0798) (0.0752) (0.0734) (0.187) (0.164) (0.165) Pol Party Rep -0.345*** -0.421*** -0.296*** -0.284*** -0.353*** -0.252*** 0.110 0.0134 0.0299 (0.0944) (0.0931) (0.0875) (0.0866) (0.0836) (0.0800) (0.240) (0.213) (0.214) U.S. Region -0.0434 -0.0264 -0.0226 -0.0539 -0.0485 -0.00811 -0.0625 -0.0859 0.105 -0.0383 -0.0384 -0.0369 (0.0361) (0.0354) (0.0340) (0.0966) (0.0871) (0.0825) Parent -0.0166 -0.107 -0.200*** -0.0135 -0.0679 -0.154** 0.0488 0.113 0.0659 -0.0761 -0.0713 -0.0685 (0.0710) (0.0660) (0.0633) (0.194) (0.173) (0.170) Constant 2.165*** 2.599*** 2.645*** 2.063*** 2.345*** 2.426*** 1.043 1.462** 0.955 -0.407 -0.417 -0.409 (0.373) (0.362) (0.370) (0.729) (0.641) (0.676) Observations 727 752 745 925 966 965 198 214 220 R-squared 0.418 0.45 0.465 0.405 0.436 0.455 0.395 0.417 0.460 Adj R Squared 0.41 0.442 0.457 0.398 0.429 0.449 0.363 0.388 0.434 Note: Democrats are the political party reference group. Males are the reference group for females. Robust standard errors in parentheses Note: Democrats are the political party reference group. Males are the reference group for *** p<0.01, ** p<0.05, * p<0.1 females. Robust standard errors in parentheses. *** p<.01 **p<.05 *p<.1 I first present the results for the White participants since that is directly related to my research question, and then present the results for the aggregate and the participants of color. Findings from White participants reveal that White participants exposed to frames where White students receive the benefit have significantly lower support (p< 0.05) compared to White participants who are exposed to treatments that reference students of color or “all” students. When White participants receive the student of color benefit framing, their level of support for 30 the policy increases by .145 points on the 5-point scale compared to when they received a treatment that states that White students would receive a benefit. Moreover, when White participants are exposed to the treatment that frames a benefit as something that “all” students receive, their level of support for a policy is .172 points higher compared to when they read that White student receive a benefit from attending a desegregated school. When examining the difference in support by White individuals who vary along partisan lines, the findings further extrapolate which groups are particularly less supportive of the White student framing. I find that White Republicans and White Independents are even significantly less supportive of the White student framings and significantly more supportive of the race neutral framings that reference “all” students compared to their White Democrat peers. This is surprising because it complicates the claims of interest convergence theory. When White participants are told about the benefits to specifically White students, students who identify with their same racial group, they do not support the policy more, and in fact support it less, particularly White Republicans, than when they are told that students of color or all students would receive those same benefits. Furthermore, when the level of support from White participants is compared between those that receive a students of color framing compared an “all” student framing, to there is no significant difference in support, except for White Republicans who show more support when they receive race neutral framing. Overall, when White participants are exposed to the treatments that reference “all” student receiving a benefit, they are the most supportive of a policy that would increase the diversity of schools in their community. In addition, Table 5 shows the regression results when these same comparisons are made for all participants in the sample. The table shows that the same patterns that are present for 31 White participants apply when all participants are included. Lastly, when the sample is restricted to participants of color, there is no significant difference in level of support when participants of color are exposed to White versus students of color framings. However, the participants of color are significantly more supportive (p < .05) of the policy when all students receive a benefit compared to when students of color are the recipients. These findings reveal a similar pattern as those for White participants, when “all” students are referenced to receive a benefit rather than a specific race of students, participants of color are more supportive of the policy. Table 6 presents the results to the third research question which asks, how does exposing White participants to policy frames emphasizing the nonacademic vs. academic benefits of desegregated schools’ impact White individuals’ support for increasing the racial diversity of schools in their community? These regression results use the same regression model with a reconstructed treatment variable. Results in Table 6 include responses from all participants (column 2), responses from White participants only (column 1), and responses from participants of color only (column 3). A more extensive version of the table with baseline support omitted is included in the Appendix (Table 6a). The results from all participants reveal that participants are significantly less supportive (p<.05) of the policy when exposed to academic student benefits frames compared to nonacademic benefits frames. These results remain when the baseline support is excluded. When I restrict the sample to only White participants, the results show that there is no significant difference in support when different types of benefits are referenced. However, when the baseline item is excluded, the difference is significantly different at a .05 level indicating that White participants are less supportive of the policy when exposed to the academic benefits frame when not accounting for their initial level of support. When participants of color are the only 32 participants included, there is no difference in level of support even when the baseline item is excluded. These results defy the third hypothesis in the study and the literature that shows that White parents prioritize academic benefits. These results show that when nonacademic students benefits are referenced, participants are slightly more likely to support a policy that would diversify the student body in schools. 33 Table6:6:Academic Table Academicvs.vs. Nonacademic Nonacademic Student Student Framing Framings (1) (3) (5) Variables White Only All PoC Only Academic Framing -0.0793 -0.0887* -0.0480 (0.0556) (0.0508) (0.129) Baseline 0.570*** 0.571*** 0.562*** (0.0303) -0.0259 (0.0515) Female 0.126** 0.124** 0.188 (0.0566) (0.0517) (0.123) Person of Color - -0.108 - (0.0716) Not a US Citizen -0.260 -0.133 -0.0467 (0.192) (0.146) (0.199) Age 0.0496 0.0237 0.199** (0.0349) (0.0383) (0.0930) Income Level -0.0353 -0.0217 -0.0954 (0.0247) (0.0268) (0.0588) Education Level 0.0132 -0.00396 0.0726 (0.0362) (0.0386) (0.0917) Pol Party Ind -0.162** -0.140** -9.34e-05 (0.0679) (0.0625) (0.142) Pol Party Rep -0.360*** -0.294*** 0.0709 (0.0748) (0.0681) (0.180) U.S. Region -0.0394 -0.0277 -0.0432 (0.0287) (0.0309) (0.0721) Region Central -0.175** -0.189*** -0.146 (0.0741) (0.0680) (0.167) Region East -0.0566 -0.0746 -0.0264 (0.0613) (0.0572) (0.145) Parent -0.104* -0.0755 0.0526 (0.0602) (0.0552) (0.142) Constant 2.004*** 1.881*** 1.082** (0.230) (0.202) (0.442) Observations 1,112 1,428 316 R-squared 0.443 0.431 0.433 Adj R Squared 0.435 0.425 0.403 Notes: Reference Notes: Reference groups are Male, groups are White, male,US citizens, White, Agecitizens, U.S. 18-34, 20K or less annual ncome, HS grad, Democrat, living age 18-34, 20K or less annual income, HS grad, Democrat, in Western US, with school aged children. Robust standard errors in parentheses living in Western U.S., with school aged children. Robust standard *** p<0.01,errors in parentheses. ** p<0.05, * p<0.1 *** p<.01 **p<.05 *p<.1 34 Limitations There are several limitations to note for this survey experiment study. First, while a survey experiment has the benefits of preventing selection bias by randomly sorting participants into treatment groups, previous research shows that opinion changes found in survey experiments may only be short- term attitude changes that may not result in lasting shifts when accounting for real-world social and political contexts (Barabas & Jerit, 2010). Moreover, the participants may have overstated their level of support for the policy to choose the more politically correct answer to avoid appearing racists under the context of social desirability bias. In addition, the survey was conducted online which adds limitations on the generalizability of the sample. Those who do not have a device that can connect to internet are unable to participate in the survey. Internet connectivity and the technological fluency needed to complete an online survey are related to other demographic variables such as rurality which could result in some populations being excluded from the study. For example, individuals living in more remote areas often have less stable internet connections, so participants from these areas may not be represented (Tieken, 2014). In addition, the survey was released during a time of heightened political awareness since it was right before a presidential election which also limits the generalizability of the study because people may be more aware of policy debates and have more opinions on public policies during this time. Bearing this in mind, results should be interpreted with caution when extrapolating to the general U.S. public and different time frames. Additionally, the treatment conditions in my study expose participants to new information about the benefits to students who attend desegregated schools, but these benefits are simply stated in a sentence “Research shows that (target group) students (benefit) when they attend racially integrated schools”. In hindsight, I recognize the potential flaws with this 35 approach given that participants were not given an abundance of new information and the information given continued to refer to these policies in the abstract instead of exposing participants to more concrete examples desegregation policies and students’ benefits. Nevertheless, I did frame the information as coming from researchers which draws on people’s acceptance of information coming from “expert” sources (Page et al., 1987). Page et al. (1987) finds that information coming from different sources has differing degrees of credibility among the public. The public puts more trust when information comes from presumably nonpartisan “experts” when compared to political elites and interest groups. Survey participants could perceive the new information as credible since it is framed as information coming from researchers. Additionally, when policies are relayed to the public, people often do not read all the details in these messages but stop at the headlines to glean the main takeaways (Subramanian, 2017). I only exposed participants to brief but poignant information to increase the likelihood that they read the entire treatment sentence, thus increasing the likelihood that they were indeed exposed to the treatment as intended (Passmore, Dobbie, Parchman, & Tysinger, 2002; Subramanian, 2017). Discussion The results from this analysis provide insight regarding how policy frames and policy primes can influence individuals’ support for a policy that would increase racial diversity in schools. First, I find support for policy framing theory suggesting that providing individuals new information about the benefits of school desegregation for students can significantly increase the public’s level of support for increasing the racial diversity of schools in their community. Participants report greater support for racial diversity after reading about a benefit to students regardless of type of benefit and the student group receiving it. This shows that referring to any 36 student group and either nonacademic or academic student benefits boosts public support for racial diversity in schools compared to when no additional information about student benefits is given. Second, while I hypothesized that White individuals would show the greatest support for treatments that exposed them to information saying that White students would benefit from a policy, this is not realized. Instead, treatments referring to “all” students garner the most support from White participants and participants of color. And, White Republicans and Independents showed significantly more support for the race neutral frames compared to White Democrats. This finding aligns with other research that examines the shift in the framing of U.S. politics and law from a race-conscious frame to one that is race-neutral (McDermott, Frankenberg, & Diem, 2015). Research shows that framing policies like affirmative action with race conscious language can result in less support, particularly by White individuals (Frankenberg & Jacobsen, 2011; Hochschild & Scott, 1998). In the context of school diversity policies, race neutral approaches have become more popular and have become viewed as more politically and legally viable while receiving greater support by White, Black, and Latinx populations (McDermott, Frankenberg, & Diem, 2015; Kahlenberg, 1996). In regards to White individuals specifically, McDermott et al. (2015) explains that when policies are framed to benefit people of color, White individuals know they will never benefit from the policy. But if a policy is framed with race-neutral language such as benefiting a certain socioeconomic group, there is universal appeal because all races and ethnicities can fathom a potential dip in their income. However, race neutral framings fail to address inequities that are tied to racial inequity and the real differences that Americans from different racial groups experience. As King and 37 Smith (2008) assert “the color-blind order is the successor to earlier orders that worked to sustain white advantages” (p. 688.) This finding could also shed light on the influence of framing a race-conscious policy in race-neutral terms and the reaction White individuals have to the term “White” and their own White identity. Even though White individuals and White students share the same race, White individuals may not identify with the term White. As Frankenberg (1993) explains “Whiteness, as a set of normative cultural practices is visible most clearly to those it definitively excludes and those to whom it does violence. Those who are securely housed within its borders usually do no examine it” (pg. 228). White individuals often identify with other aspects of their identity that are marginalized. In Frankenberg’s (1993) book, the White women she spoke with often refer to their identity as a woman or their White ethnic identities, such as having Italian or Irish ancestry, before identifying as a White American. Psychology research on identity finds that individuals internalize their identities that are subject to more external scrutiny (i.e. race for Black women) and do not internalize aspects of their identity that hold a more privileged place in society (i.e. heterosexuality for straight men) (Jones & Abes, 2013; Jones & McEwen, 2000). Because White is the dominant race in the racial hierarchy in the United States, the dominance of the group can make it a less salient part of their identity. Furthermore, White participants may support the race neutral policy framing the most because they do not see race as a particularly important aspect of their lives. Thus, they may operate with a colorblind mindset where they fail to see or minimize the importance of race in society. With colorblindness, not only do White individuals not acknowledge their own racial identity, but they do not acknowledge the racial identity of anyone and the inequities that are tied race (Bonilla-Silva, 2003, 2013). Because many White individuals live racially isolated from 38 others (Fiel, 2013; Lichter et al., 2017; Thompson Dorsey, 2013), they have the privilege of not thinking about racism and are not penalized from abstaining from these thoughts and discussions regarding race (DiAngelo, 2011). Moreover, if White individuals bring up race, they are more likely to be criticized by other White individuals who may view them as racist for directly acknowledging a person’s race instead of subsuming a colorblind stance (DiAngelo, 2011). Supporting the race neutral option could be an extension of White individuals’ colorblind ideologies and illustrate that they are more supportive of frames that align with their dismissal of race altogether under the guise of White comfort and social desirability. In addition, the identity politics surrounding Whiteness in the last few decades have generated considerable angst and White guilt about the oppression carried out by White individuals with the increased attention of White supremacist groups (Kincheloe, 1999). This survey was released the day of the 2020 presidential election with Donald Trump running for reelection after having frequently used racist rhetoric and blatantly supporting White supremacist groups during his campaigning and first term of office (Goldstein & Hall, 2017; Heidt, 2018; Huber, 2016). When Whiteness is explicitly linked to racism or discrimination, as it has been during Donald Trump’s political career, White individuals tend to distance themselves from their own White identities because the social destructive practices of Whiteness are highlighted (Appiah & Gutman, 1996; McDermott & Samson, 2005). The partisan findings from this survey show that White Republicans and Independents may have even further distanced themselves from their White identities compared to White Democrats, even though the Republican candidate for President in 2020 had used language that either overtly or implicitly implied White supremacy. White individuals at the time of this survey may have thought that it is actually in their interest to distance themselves from the term White. If White individuals feel that 39 disassociating with their own race is in their interest, this adds nuance to the interest convergence theory framework where White interests take on a colorblind mindset rather than one that is self- serving to individuals of their own race. By using the word White in my treatments, participants, including those who themselves are White, may have had a negative or uncomfortable reaction to the term and therefore been less supportive of the policy. Future research could further examine this reaction to the term White by comparing support when policies reference White, Caucasian, European American, and other ethnicities that at one time were not considered White but over time have been afforded the benefits of Whiteness (McDermott & Samson, 2005). Lastly, the results indicate that participants are more supportive of the policy when nonacademic outcomes are referenced in the treatment compared to when academic outcomes are mentioned. This finding defies my hypothesis and past literature on parental schooling preferences and their prioritization of academic outcomes, and the tradeoff many White parents note between choosing a diverse or academically “good school. When participants receive treatments that defy their previously conceived notions and pair diverse learning environments with academic benefits, this did not result in increased support for the policy. This could indicate a shifting in preferences where nonacademic traits gleaned from school are valued more or at least as much as academic skills. Policy Implications and Future Research The findings from this study suggest some key implications for policy makers. These initial results are encouraging and show that support for school desegregation can increase among White individuals if they learn that it can have a positive impact on students. This positive impact does not only apply to academic performance in the form of test scores but includes character building outcomes where students learn to become more accepting of others. 40 When White individuals realize that these benefits occur for any race of student, they are more supportive of a policy aimed to increase the diversity in schools compared to when they receive no additional information about student benefits. This is essential information for policy makers trying to attain community buy-in for a new or existing school desegregation policy. This study demonstrates that support among White individuals is malleable. History has shown how powerful White parents and community members are if they do not want a desegregation policy in their school district, so it is imperative that once a policy is created, district leaders take the next step and think carefully about the way it is related to the community so that it can be sustained and implemented with fidelity. This findings from the study also offer future avenues of research examining White attitudes and support for school diversity policies. Further research could explore how White racial identity plays a role in White individuals’ support for policies when different student racial groups are framed. Lastly, this paper represents one of the first attempts to combine policy framing and theories centering race to understand public support for a policy and the findings clearly indicate that policy frames are influential to public support. Further research studying policies frames should center race when determining how individuals react to differing information and factor in the underlying racial landscape that is present in all individuals’ policy viewpoints. 41 REFERENCES Abdulkadiroglu, A., Pathak, P., Schellenberg, J., & Walters, C. (2020). Do parents value school effectiveness? American Economic Review, 110(5), 1502-1539. Adamson, B. L. (2006). The H'aint in the (school) house: The interest convergence paradigm in state legislatures and school finance reform. Cal. WL Rev., 43, 173. Alemán, Jr, E., & Alemán, S. M. (2010). ‘Do Latinx interests always have to “converge” with White interests?’:(Re) claiming racial realism and interest‐convergence in critical race theory praxis. Race Ethnicity and Education, 13(1), 1-21. Amazon Mechanical Turk. (2018). Human intelligence through an API. https:// www.mturk.com/get-started Appiah, K. A. and Gutman, A. (1996) Colour Conscious: The Political Morality of Race. New Jersey: Princeton University Press. Ashenfelter, O., Collins, W. J., & Yoon, A. (2006). Evaluating the role of Brown v. Board of Education in school equalization, desegregation, and the income of African Americans. American Law and Economics Review, 8(2), 213-248. Barabas, J., & Jerit, J. (2010). Are survey experiments externally valid?. American Political Science Review, 104(2), 226-242. Bell Jr, D. A. (1980). Brown v. Board of Education and the interest-convergence dilemma. Harvard law review, 518-533. Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating online labor markets for experimental research: Amazon.com’s Mechanical Turk. Political Analysis, 20(3), 351– 368. Billingham, C. M., & Hunt, M. O. (2016). School racial composition and parental choice: New evidence on the preferences of white parents in the United States. Sociology of education, 89(2), 99-117. Billings, S. B., Deming, D. J., & Rockoff, J. (2014). School segregation, educational attainment, and crime: Evidence from the end of busing in Charlotte-Mecklenburg. The Quarterly Journal of Economics, 129(1), 435-476. Bonilla-Silva, E. (2003). Racial attitudes or racial ideology? An alternative paradigm for examining actors' racial views. Journal of Political Ideologies, 8(1), 63-82. Bonilla-Silva, E. (2013). “New racism,” color-blind racism, and the future of Whiteness in America. In White out (pp. 268-281). Routledge. 42 Boozer, M. A., Krueger, A. B., and Wolken, S. (1993): “Race and School Quality Since Brown v. Board of Education." Brookings Papers on Economic Activity Microeconomics, 269- 326. Byrne, B., & Tona, C. D. (2014). Multicultural desires? Parental negotiation of multiculture and difference in choosing secondary schools for their children. The Sociological Review, 62(3), 475-493. Card, D., & Rothstein, J. (2007). Racial segregation and the black–white test score gap. Journal of Public Economics, 91(11-12), 2158-2184. Carlson, D., & Bell, E. (2021). Socioeconomic Status, Race, and Public Support for School Integration. AERA Open, 7(1), 1-16. Carter, P. L. (2016). Educational Equality Is a Multifaceted Issue: Why We Must Understand the School’s Sociocultural Context for Student Achievement. Russell Sage Foundation. Cashin, S. D. (2005). Shall we overcome-transcending race, class, and ideology through interest convergence. John's L. Rev., 79, 253. Castagno, A. E., & Lee, S. J. (2007). Native mascots and ethnic fraud in higher education: Using tribal critical race theory and the interest convergence principle as an analytic tool. Equity & Excellence in Education, 40(1), 3-13. Chong, D., & Druckman, J. N. (2007). Framing theory. Annual Review of Political Science, 10, 103-126. Coppock, A. (2019). Generalizing from survey experiments conducted on Mechanical Turk: A replication approach. Political Science Research and Methods, 7(3), 613– 628. Crozier, G., Reay, D., James, D., Jamieson, F., Beedell, P., Hollingworth, S., & Williams, K. (2008). White middle‐class parents, identities, educational choice and the urban comprehensive school: dilemmas, ambivalence and moral ambiguity. British Journal of Sociology of Education, 29(3), 261-272. Diamond, J. (2006). Still separate and unequal: Examining race, opportunity, and school achievement in integrated suburbs. Journal of Negro Education, 75(3), 495–505. DiAngelo, R. (2011) White fragility. International Journal of Critical Pedagogy, 3(3), 54-70. Delgado, R. S., & Stefancic, J. J. (2012). Critical race theory: An introduction. 2nd Edition. New York University Press. Denice, P., & Gross, B. (2016). Choice, preferences, and constraints: Evidence from public school applications in Denver. Sociology of Education, 89(4), 300-320. 43 Druckman, J. N. (2001). The implications of framing effects for citizen competence. Political behavior, 23(3), 225-256. Druckman, J. N., Green, D. P., Kuklinski, J. H., Lupia, A. (2006) The growth and development of experimental research in political science. The American Political Science Review, 100(4), 627-635. Edmonds, C., & Killen, M. (2009). Do adolescents' perceptions of parental racial attitudes relate to their intergroup contact and cross-race relationships?. Group Processes & Intergroup Relations, 12(1), 5-21. El-Haj, T. R. A. (2006). Elusive justice: Wrestling with difference and educational equity in everyday practice. Taylor & Francis. Evans, S. A. (2021). “I wanted diversity, but not so much”: Middle-class White parents, school choice, and the persistence of anti-black stereotypes. Urban Education, 0(0). Fiel, J. E. (2013). Decomposing school resegregation: Social closure, racial imbalance, and racial isolation. American Sociological Review, 78(5), 828-848. Follmer, D. J., Sperling, R. A., & Suen, H. K. (2017). The role of MTurk in education research: Advantages, issues, and future directions. Educational Researcher, 46(6), 329– 334. Frankenberg, R. (1993) White women, race matters: The social construction of whiteness. Minneapolis: University of Minnesota Press. Frankenberg, E., & Jacobsen, R. (2011). Trends school integration polls. Public opinion quarterly, 75(4), 788-811. García, E., & Weiss, E. (2016). Making Whole-Child Education the Norm: How Research and Policy Initiatives Can Make Social and Emotional Skills a Focal Point of Children's Education. Economic Policy Institute. Gehlbach, H., & Brinkworth, M. E. (2011). Measure twice, cut down error: A process for enhancing the validity of survey scales. Review of general psychology, 15(4), 380-387. Gillborn, D. (2008). Racism and education: Coincidence or conspiracy?. Routledge. Glazerman, S., & Dotter, D. (2017). Market signals: Evidence on the determinants and consequences of school choice from a citywide lottery. Educational Evaluation and Policy Analysis, 39(4), 593-619. Goldstein, D. M., & Hall, K. (2017). Postelection surrealism and nostalgic racism in the hands of Donald Trump. HAU: Journal of Ethnographic Theory, 7(1), 397-406. 44 Grogger, J. (1996). Does school quality explain the recent black/white wage trend?. Journal of labor economics, 14(2), 231-253. Guryan, J. (2004). Desegregation and black dropout rates. American Economic Review, 94(4), 919-943. Haderlein, S. A. K., (2021) How do parents evaluate and selection schools? Evidence from a survey experiments. American Educational Research Journal, 1-34. Hailey, C.A. (2022) Racial preferences for schools: Evidence from an experiment with White, Black, Latinx, and Asian parents and students. Sociology of Education, 95(2), 110-132. Hannah-Jones, N. (2017, September, 6) The resegregation of Jefferson County. The New York Times. Harbridge, L., & Malhotra, N. (2011). Electoral incentives and partisan conflict in Congress: Evidence from survey experiments. American Journal of Political Science, 55(3), 494- 510. Harris, D. N., & Larsen, M. F. (2019). The identification of schooling preferences: Methods and evidence from post-Katrina New Orleans. Education Research Alliance for New Orleans. Hastings, J., Kane, T., & Staiger, D. (2005). Parental preferences and school competition: Evidence from a public school choice program (No. w11805). National Bureau of Economic Research. Heidt, S. J. (2018). Scapegoater-in-Chief: Racist undertones of Donald Trump’s rhetorical repertoire. In The Trump Presidency, Journalism, and Democracy. Taylor & Francis. Henry, E. E., & Hankins, K. (2012). Halting white flight: Parent activism and the (re) shaping of Atlanta’s “Circuits of Schooling,” 1973-2009. Journal of Urban History, 38(3), 532-552. Hernández, M. (2019). White middle-class families and sociocultural diversity in schools: A literature review. The Urban Review, 51(2), 270-300. Hochschild, J., & Scott, B. (1998). Trends: Governance and reform of public education in the United States. The Public Opinion Quarterly, 62(1), 79-120. Hollingworth, S., & Williams, K. (2010). Multicultural mixing or middle-class reproduction? The white middle classes in London comprehensive schools. Space and Polity, 14(1), 47- 64. Holme, J. J., & Finnigan, K. S. (2018). Striving in Common: A Regional Equity Framework for Urban Schools. Harvard Education Press. Cambridge, MA. 45 Huber, L. P. (2016). Make America great again: Donald Trump, racist nativism and the virulent adherence to white supremecy amid US demographic change. Charleston L. Rev., 10, 215. Johnson, R. C. (2011). Long-run impacts of school desegregation & school quality on adult attainments (No. w16664). National Bureau of Economic Research. Jones, S. R., & Abes, E. S. (2013). Identity development of college students: Advancing frameworks for multiple dimensions of identity. John Wiley & Sons. Jones, S. R., & McEwen, M. K. (2000). A conceptual model of multiple dimensions of identity. Journal of college student development, 41(4), 405-414. Kahlenberg, R. D. (1996). Class-based affirmative action. Cal L. Rev., 84, 1037. Kahlenberg, R. (2016) School integration in practice: Lessons from nine districts. The Century Foundation. Kimelberg, S. M., & Billingham, C. M. (2013). Attitudes toward diversity and the school choice process: Middle-class parents in a segregated urban public school district. Urban Education, 48(2), 198-231. Kincheloe, J. L. (1999). The struggle to define and reinvent whiteness: A pedagogical analysis. College literature, 26(3), 162-194. Kinder, D. R. (1998). Communication and opinion. Annual review of political science, 1(1), 167- 197. King, D. S., & Smith, R. M. (2008). Strange bedfellows? Polarized politics? The quest for racial equity in contemporary America. Political Research Quarterly, 61(4), 686-703. Kurlaender, M., & Yun, J. (2005). Fifty years after Brown: New evidence of the impact of school racial composition on student outcomes. International Journal of Educational Policy, Research and Practice, 6, 51–78. Ladson-Billings, G., & Tate, W. F. (Eds.). (2006). Education research in the public interest: Social justice, action, and policy. Teachers College Press. Leigh, P. R. (2003). Interest convergence and desegregation in the Ohio Valley. Journal of Negro education, 269-296. Leonardo, Z. (2009). Race, whiteness, and education. Routledge. 46 Leonardo, Z. (2013). Race frameworks: A multidimensional theory of racism and education. Teachers College Press. Levine-Rasky, C. (2008). Middle-Classness and Whiteness in Parents' Responses to Multiculturalism: A Study of One School. Canadian Journal of Education, 31(2), 459- 490. Lichter, D. T., Parisi, D., & Taquino, M. C. (2017). Together but apart: Do US Whites live in racially diverse cities and neighborhoods?. Population and Development Review, 229- 255. Lincove, J. A., Cowen, J. M., & Imbrogno, J. P. (2018). What's in your portfolio? How parents rank traditional public, private, and charter schools in post-Katrina New Orleans’ citywide system of school choice. Education Finance and Policy, 13(2), 194-226. Lincove, J. A., Valant, J., & Cowen, J. M. (2018b). You can't always get what you want: Capacity constraints in a choice-based school system. Economics of Education Review, 67, 94-109. Lopez, G. R. (2003). The (racially neutral) politics of education: A critical race theory perspective. Educational Administration Quarterly, 39(1), 68-94. McDermott, K.A., Frankenberg, E., Diem, S. (2015) The “post-racial” politics of change: Changing student assignment policy in three school districts. Education Policy, 29(3) 504-5. McDermott, M., & Samson, F. L. (2005). White racial and ethnic identity in the United States. Annu. Rev. Sociol., 31, 245-261. Mickelson, R. A., Smith, S. S., & Nelson, A. H. (Eds.). (2017). Yesterday, today, and tomorrow: School desegregation and resegregation in Charlotte. Harvard Education Press. Milliken v. Bradley (1974), 418 U.S. 717 (1974). Milner IV, H. R. (2008). Critical race theory and interest convergence as analytic tools in teacher education policies and practices. Journal of teacher education, 59(4), 332-346. Mullinix, K. J., Leeper, T. J., Druckman, J. N., & Freese, J. (2015). The generalizability of survey experiments. Journal of Experimental Political Science, 2(2), 109–138. Orfield, G. (1995). Metropolitan school desegregation: Impacts on metropolitan society. Minn. L. Rev., 80, 825. Orfield, G. (2001). Diversity challenged: Evidence on the impact of affirmative action. Harvard Education Publishing Group. Cambridge, MA 47 Page, B. I., Shapiro, R. Y., & Dempsey, G. R. (1987) What moves public opinion? The American Political Science Review, 81(1), 22-44. Parcel, T. L., & Taylor, A. J. (2015). The end of consensus: Diversity, neighborhoods, and the politics of public school assignments. UNC Press Books. Passmore, C., Dobbie, A. E., Parchman, M., & Tysinger, J. (2002). Guidelines for constructing a survey. Familiy Medicine Kansas City, 34(4), 281-286. Phi Delta Kappan (2017, September). The 49th annual PDK report of the public’s attitudes towards the public schools. Kappan Magazine. Reeves, R. V. & Joo, N. (2018) Do school sessions worsen racial segregation? It’s complicated. Brookings Institute. Retrieved from https://www.brookings.edu/research/do-school- secessions-worsen-racial-segregation-its-complicated/ Rein, M., & Schön, D. (1996). Frame-critical policy analysis and frame-reflective policy practice. Knowledge and policy, 9(1), 85-104. Rhodes, J. H., Shaffner, B. F., Raja, R. J. (2020, September, 3) Research shows just how much more power white voters wield in local elections. The Washington Post Rivas‐Drake, D., Umaña‐Taylor, A. J., Schaefer, D. R., & Medina, M. (2017). Ethnic‐racial identity and friendships in early adolescence. Child Development, 88(3), 710-724. Roda, A., & Wells, A. S. (2013). School choice policies and racial segregation: Where white parents’ good intentions, anxiety, and privilege collide. American Journal of Education, 119(2), 261-293. Rooks, N. (2017) Cutting School: Privatization, Segregation, and the End of Public Education. The New Press. New York. Saatcioglu, A. (2010). Disentangling School-and Student-Level Effects of Desegregation and Resegregation on the Dropout Problem in Urban High Schools: Evidence from the Cleveland Municipal School District, 1977-1998. Teachers College Record, 112(5), 1391-1442. Sandburg, C. (n.d.) The story of Parents Involved in Community Schools. Berkeley Law University of California. Retrieved from https://www.law.berkeley.edu/files/The_Story_of_Parents_Involved_Sandberg.pdf Schofield, J. W. (1981). Unchartered territory: Speculations on some positive effects of desegregation on white students. The Urban Review, 13(4), p. 227-241. Schofield, J. W. (1995). Review of research on school desegregation's impact on elementary and secondary school students. 48 Shaw, J. I., Bergen, J. E., Brown, C. A., & Gallagher, M. E. (2000). Centrality preferences in choices among similar options. The Journal of general psychology, 127(2), 157-164. Siegel-Hawley, G., Diem, S., & Frankenberg, E. (2018). The disintegration of Memphis-Shelby County, Tennessee: School district secession and local control in the 21st century. American Educational Research Journal, 55(4), 651-692. Slavin, R. E. (1979). Effects of biracial learning teams on cross-racial friendships. Journal of Educational Psychology, 71(3), 381. Sniderman, P. M., & Grob, D. B. (1996). Innovations in experimental design in attitude surveys. Annual review of Sociology, 22(1), 377-399. Su, C. (2007). Cracking silent codes: Critical race theory and education organizing. Discourse: studies in the cultural politics of education, 28(4), 531-548. Subramanian, K. R. (2017). Influence of social media in interpersonal communication. International Journal of Scientific Progress and Research, 38(2), 70-75. Svallfors, S. (2012). Contested Welfare States: Welfare Attitudes in Europe and Beyond. Stanford University Press. Taylor, A. J., & Parcel, T. L. (2019). Proximity and the principle-policy gap in White racial attitudes: Insight from views of student assignment policies in Wake County, North Carolina. Social Science Research, 78, 95-103. Thompson Dorsey, D. N. (2013). Segregation 2.0: The new generation of school segregation in the 21st century. Education and Urban Society, 45(5), 533-547. Tieken, M. C. (2014). Why rural schools matter. Chapel Hill: University of North Carolina Press. Torres, E., & Weissbourd, R. (2020). Do parents really want school integration? https://mcc.gse.harvard.edu/ Vaught, S. E. (2011). Racism, public schooling, and the entrenchment of white supremacy: A critical race ethnography. State University of New York Press. Vowden, K. J. (2012). Safety in numbers? Middle-class parents and social mix in London primary schools. Journal of Education Policy, 27(6), 731-745. Weiner, D. A., Lutz, B. F., & Ludwig, J. (2009). The effects of school desegregation on crime (No. w15380). National Bureau of Economic Research. Weinstein, S. M. (2006). A Needed Image Makeover: Interest Convergence and the United States' War on Terror. Roger Williams University Law Review, 11(2), 4. 49 Wells, A. S., & Crain, R. L. (2004). Perpetuation theory and the long-term effects of school desegregation. Review of Educational Research, 64(4), 531-555. Wells, A. S., Holmes, J. J., Revilla, A. T., & Atanda, K. (2009). Both sides now: The story of school desegregation’s graduates. Berkeley: University of California Press. Wells, A. S., Fox, L., & Cordova-Cobo, D. (2016). How racially diverse schools and classrooms can benefit all students. The Education Digest, 82(1), 17. Williams, S. M. (2010). Through the eyes of friends: An investigation of school context and cross-racial friendships in racially mixed schools. Urban Education, 45(4), 480-505. 50 APPENDIX Table 7: Student Target Group Framing with Expanded Control Groups Table 7 Student Target Group Framing with Expanded Control Groups White only All Participants POC only (1) (2) (3) (4) (5) (6) (7) (8) (9) VARIABLES White v SoC White v All SoC v All White v SoC White v All SoC v All White v SoC White v All SoC v All treat_whitevsoc -0.141* -0.117* 0.0117 (0.0728) (0.0662) (0.164) treat_whitevall -0.174** -0.201*** -0.287* (0.0678) (0.0619) (0.157) treat_socvall -0.0278 -0.0855 -0.279** (0.0671) (0.0598) (0.139) Baseline 0.555*** 0.560*** 0.595*** 0.554*** 0.561*** 0.594*** 0.560*** 0.548*** 0.566*** (0.0375) (0.0374) (0.0361) (0.0323) (0.0322) (0.0308) (0.0664) (0.0639) (0.0612) Female 0.168** 0.177** 0.0392 0.125* 0.166*** 0.0936 0.0595 0.195 0.308** (0.0726) (0.0696) (0.0657) (0.0665) (0.0632) (0.0604) (0.166) (0.149) (0.138) Person of Color - - - -0.160* -0.0701 -0.0873 - - - (0.0917) (0.0895) (0.0822) Not a US Citizen -0.242 -0.281 -0.268 -0.0819 -0.0886 -0.186 0.116 -0.0432 -0.108 (0.238) (0.212) (0.231) (0.197) (0.163) (0.175) (0.288) (0.220) (0.239) Age 35-54 0.117 0.0803 0.00688 0.135 0.127 0.0629 0.151 0.237 0.267 (0.113) (0.113) (0.100) (0.0976) (0.0965) (0.0861) (0.204) (0.187) (0.166) Age 55+ 0.102 0.0493 0.0427 0.187** 0.0979 0.0612 0.650*** 0.347 0.285 (0.103) (0.0996) (0.0942) (0.0914) (0.0878) (0.0833) (0.249) (0.231) (0.223) Income 20-40K 0.0288 -0.00631 -0.0826 0.00541 0.00753 -0.114 -0.102 0.105 -0.0907 (0.107) (0.108) (0.108) (0.101) (0.0935) (0.0972) (0.258) (0.186) (0.203) Income 40-80K -0.00107 -0.0939 0.0167 -0.0657 -0.0936 -0.0278 -0.278 0.000883 -0.0501 (0.0983) (0.101) (0.0937) (0.0960) (0.0932) (0.0861) (0.265) (0.227) (0.204) Income 80K+ 0.00239 -0.0885 -0.152 -0.0506 -0.0945 -0.149 -0.300 -0.136 -0.100 (0.106) (0.107) (0.102) (0.101) (0.0960) (0.0927) (0.275) (0.217) (0.229) Education 2 yr dgr 0.0792 -0.0597 -0.123 0.178* 0.0600 0.0216 0.531** 0.548** 0.550*** (0.0946) (0.0857) (0.0896) (0.0919) (0.0832) (0.0846) (0.257) (0.233) (0.208) Education 4 yr dgr 0.0233 -0.0871 -0.0723 0.0926 -0.0358 -0.0347 0.350 0.225 0.164 (0.0999) (0.0938) (0.0956) (0.0966) (0.0899) (0.0861) (0.265) (0.247) (0.208) Pol Party Ind -0.214** -0.159* -0.103 -0.191** -0.142* -0.103 -0.0933 -0.0189 0.0309 (0.0875) (0.0836) (0.0794) (0.0798) (0.0752) (0.0734) (0.187) (0.164) (0.165) Pol Party Rep -0.345*** -0.421*** -0.296*** -0.284*** -0.353*** -0.252*** 0.110 0.0134 0.0299 (0.0944) (0.0931) (0.0875) (0.0866) (0.0836) (0.0800) (0.240) (0.213) (0.214) Region Central -0.122 -0.174** -0.233*** -0.0983 -0.202** -0.253*** 0.114 -0.216 -0.224 (0.100) (0.0867) (0.0861) (0.0903) (0.0821) (0.0796) (0.214) (0.235) (0.198) Region East -0.0871 -0.0542 -0.0485 -0.103 -0.0940 -0.0147 -0.0727 -0.130 0.280* (0.0762) (0.0756) (0.0730) (0.0719) (0.0704) (0.0677) (0.201) (0.175) (0.166) Parent Status -0.0127 -0.109 -0.210*** -0.00830 -0.0683 -0.162** 0.0311 0.106 0.0221 (0.0793) (0.0746) (0.0693) (0.0732) (0.0676) (0.0637) (0.199) (0.173) (0.160) Constant 1.985*** 2.360*** 2.569*** 1.775*** 2.026*** 2.268*** 1.044 1.190** 1.126** (0.369) (0.370) (0.362) (0.310) (0.291) (0.296) (0.662) (0.521) (0.506) Observations 727 752 745 925 966 965 198 214 220 R-squared 0.420 0.453 0.473 0.408 0.440 0.462 0.417 0.443 0.495 Adj R Squared 0.407 0.441 0.461 0.397 0.430 0.453 0.366 0.398 0.455 Notes: Reference groups are Male, White, US citizens, Age 18-34, 20K or less annueal income, HS grad, Democrat, Notes: Reference groups are male, White, U.S. citizens, age 18-34, 20K or living in Western US, with school aged children. Robust standard errors in parentheses less annual income, HS grad, Democrat, living in Western U.S., with school *** p<0.01, ** p<0.05, * p<0.1 aged children. Robust standard errors in parentheses. *** p<.01 **p<.05 *p<.1 51 Table 8 Student Target groups with Baseline Excluded Table 8: Student Target Groups with Baseline Excluded All Participants White Participants Only Participants of Color Only (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) VARIABLES White v SoC White v SoC White v All White v All SoC v All SoC v All White v SoC White v SoC White v All White v All SoC v All SoC v All White v SoC White v SoC White v All White v All SoC v All SoC v All treat_whitevsoc -0.121* -0.190** -0.145** -0.206** -0.0109 -0.137 (0.0660) (0.0816) (0.0722) (0.0888) (0.162) -0.2 treat_whitevall -0.200*** -0.285*** -0.172** -0.239*** -0.279* -0.427** (0.0619) (0.0765) (0.0677) (0.0835) (0.159) (0.193) treat_socvall -0.0785 -0.0847 -0.0210 -0.0226 -0.282** -0.337* (0.0599) (0.0779) -0.0666 (0.0853) (0.140) (0.188) Baseline 0.557*** 0.562*** 0.594*** 0.556*** 0.558*** 0.593*** 0.557*** 0.561*** 0.585*** (0.0322) (0.0324) (0.0309) (0.0375) (0.0377) (0.0363) (0.0640) (0.0638) (0.0605) Female 0.128* 0.180** 0.165*** 0.230*** 0.0961 0.216*** 0.168** 0.190** 0.173** 0.215** 0.0409 0.158* 0.0542 0.210 0.181 0.315 0.314** 0.439** (0.0666) (0.0854) (0.0627) (0.0803) (0.0604) (0.0781) (0.0726) (0.0927) (0.0687) (0.0874) (0.0652) (0.0856) (0.165) (0.208) (0.151) (0.191) (0.141) (0.181) Person of Color -0.157* -0.186 -0.0614 -0.0727 -0.0937 -0.117 (0.0916) (0.113) (0.0892) (0.110) (0.0828) (0.107) Not a US Citizen -0.0811 -0.242 -0.0822 -0.166 -0.158 -0.286 -0.235 -0.134 -0.237 -0.176 -0.216 -0.0541 0.0332 -0.291 -0.104 -0.259 -0.129 -0.425 (0.200) (0.257) (0.166) (0.269) (0.174) (0.247) (0.233) (0.368) (0.211) (0.400) (0.228) (0.357) (0.290) (0.350) (0.227) (0.345) (0.252) (0.336) Age 0.0980** 0.0792 0.0450 0.0402 0.0256 -0.0299 0.0464 -0.00164 0.0199 0.0219 0.0120 -0.0652 0.309** 0.407*** 0.183 0.175 0.199* 0.258* (0.0447) (0.0539) (0.0427) (0.0525) (0.0412) (0.0542) (0.0491) (0.0594) (0.0466) (0.0572) (0.0457) (0.0586) (0.121) (0.133) (0.113) (0.140) (0.114) (0.145) Income Level -0.0284 -0.0346 -0.0434 -0.0629 -0.0365 -0.0116 -0.00488 -0.0481 -0.0353 -0.0651 -0.0342 -0.0240 -0.120 -8.06e-05 -0.0745 -0.0547 -0.0593 -0.0135 (0.0316) (0.0403) (0.0309) (0.0392) (0.0285) (0.0406) (0.0332) (0.0435) (0.0343) (0.0428) (0.0319) (0.0440) (0.0817) (0.0984) (0.0700) (0.0912) (0.0676) (0.0950) Ed Level 0.0455 0.139** -0.0118 0.0612 -0.00782 0.0563 0.0201 0.109* -0.0341 0.0240 -0.0163 0.0559 0.135 0.261* 0.0692 0.191 0.0176 0.0662 (0.0464) (0.0575) (0.0440) (0.0539) (0.0427) (0.0562) (0.0491) (0.0620) (0.0463) (0.0572) (0.0475) (0.0601) (0.124) (0.141) (0.114) (0.133) (0.0991) (0.140) Pol Party -0.143*** -0.328*** -0.170*** -0.412*** -0.121*** -0.364*** -0.172*** -0.366*** -0.209*** -0.469*** -0.150*** -0.405*** 0.0103 -0.139 -0.00409 -0.159 0.0364 -0.139 (0.0429) (0.0491) (0.0415) (0.0462) (0.0395) (0.0466) (0.0471) (0.0524) (0.0465) (0.0492) (0.0436) (0.0511) (0.110) (0.138) (0.101) (0.130) (0.0975) (0.118) U.S. Region -0.0539 -0.0326 -0.0485 -5.53e-06 -0.00811 -0.00443 -0.0434 -0.000829 -0.0264 0.0245 -0.0226 -0.0113 -0.0625 -0.0831 -0.0859 -0.0296 0.105 0.0987 (0.0361) (0.0450) (0.0354) (0.0430) (0.0340) (0.0434) (0.0383) (0.0487) (0.0384) (0.0453) (0.0369) (0.0470) (0.0966) (0.111) (0.0871) (0.111) (0.0825) (0.110) Parent -0.0135 0.0281 -0.0679 -0.0324 -0.154** -0.127 -0.0166 -0.0112 -0.107 -0.0771 -0.200*** -0.201** 0.0488 0.250 0.113 0.185 0.0659 0.233 (0.0710) (0.0872) (0.0660) (0.0812) (0.0633) (0.0839) (0.0761) (0.0947) (0.0713) (0.0882) (0.0685) (0.0912) (0.194) (0.228) (0.173) (0.203) (0.170) (0.216) Constant 2.063*** 4.137*** 2.345*** 4.445*** 2.426*** 4.713*** 2.165*** 4.185*** 2.599*** 4.639*** 2.645*** 4.730*** 1.043 2.314** 1.462** 3.126*** 0.955 2.873*** (0.373) (0.451) (0.362) (0.456) (0.370) (0.424) (0.407) (0.507) (0.417) (0.541) (0.409) (0.483) (0.729) (0.969) (0.641) (0.853) (0.676) (0.830) Observations 925 927 966 967 965 966 727 729 752 753 745 746 198 198 214 214 220 220 R-squared 0.405 0.072 0.436 0.107 0.455 0.077 0.418 0.088 0.450 0.135 0.465 0.099 0.395 0.081 0.417 0.071 0.460 0.078 Adj R Squared 0.398 0.0621 0.429 0.0981 0.449 0.0676 0.410 0.0770 0.442 0.124 0.457 0.0879 0.363 0.0367 0.388 0.0303 0.434 0.0390 Robust standard errors in parentheses Note: Robust standard errors in parentheses. *** p<.01, **p<.05, *p<.1 *** p<0.01, ** p<0.05, * p<0.1 52 Table 9: Academic Table 9: Academicvs. vs.Nonacademic Nonacademic StudentStudent BenefitBenefits Referenced Referenced with Baseline Excluded with Baseline Excluded (1) (2) (3) (4) (5) (6) Variables All All White Only White Only PoC Only PoC Only Academic Framing -0.0923* -0.157** -0.0837 -0.166** -0.0879 -0.0686 (0.0505) (0.0637) (0.0553) (0.0694) (0.124) (0.157) Baseline 0.573*** 0.569*** 0.575*** (0.0260) (0.0304) (0.0507) Female 0.125** 0.202*** 0.124** 0.183** 0.180 0.321** (0.0516) (0.0664) (0.0561) (0.0722) (0.125) (0.159) Citizenship -0.125 -0.258 -0.233 -0.118 -0.0971 -0.351 (0.148) (0.210) (0.190) (0.306) (0.209) (0.280) Age 0.0496 0.0192 0.0237 -0.0188 0.199** 0.239** (0.0349) (0.0435) (0.0383) (0.0474) (0.0930) (0.111) Income Level -0.0353 -0.0363 -0.0217 -0.0406 -0.0954 -0.0479 (0.0247) (0.0327) (0.0268) (0.0353) (0.0588) (0.0774) Education Level 0.0132 0.0935** -0.00396 0.0711 0.0726 0.183 (0.0362) (0.0457) (0.0386) (0.0487) (0.0917) (0.113) Political Party -0.144*** -0.369*** -0.179*** -0.418*** 0.0285 -0.125 (0.0338) (0.0386) (0.0374) (0.0415) (0.0835) (0.105) U.S. Region -0.0394 -0.0160 -0.0277 0.00736 -0.0432 -0.0431 (0.0287) (0.0358) (0.0309) (0.0384) (0.0721) (0.0907) Parent -0.0713 -0.0319 -0.0982* -0.0765 0.0686 0.200 (0.0542) (0.0685) (0.0583) (0.0744) (0.143) (0.171) Constant 2.275*** 4.457*** 2.432*** 4.479*** 1.241** 2.869*** (0.301) (0.360) (0.336) (0.412) (0.580) (0.741) Observations 1,428 1,430 1,112 1,114 316 316 R-squared 0.428 0.080 0.441 0.103 0.411 0.052 Adj R Squared 0.424 0.0731 0.435 0.0958 0.392 0.0240 Robust standard errors in Note: Robust standard errors parentheses in parentheses. *** p<.01, **p<.05, *p<.1 *** p<0.01, ** p<0.05, * p<0.1 53 PAPER 2: INTENTIONALLY DIVERSE CHARTER SCHOOLS: UNDERSTANDING WHITE PARENTS’ MOTIVATIONS FOR ENROLLING THEIR CHILDREN IN DIVERSE LEARNING ENVIRONMENTS Introduction Neighborhood schools are racially segregated due to the lasting consequences of restrictive housing policies. and individual residential decisions with White individuals continually choosing to live in majority White areas and some Black individuals prefering to live in White minority neighborhoods (Holme & Finnigan, 2018; McWhorter 2000; Rothstein, 2017; Thernstrom & Thernstrom 1997). Districts throughout the United States have attempted to desegregate schools in the past but have often failed to successfully implement policies because of a lack of support from White parents (Kahlenberg, 2016; McDermott et al., 2015; Holme & Finnigan, 2018). Public opinion research finds that White individuals have confounding opinions concerning policies that attempt to diversify schools. One national survey finds that 66% of White parents report that it is “very” or “somewhat” important for their child to attend a diverse school (Wells et al., 2009). However, most White parents ultimately choose to enroll their children in majority White schools rationalizing their choices by explaining that these schools hold students to higher education standards, have more positive classroom climates, and have higher quality teachers (Crozier et al., 2008; Cucchiara, 2013; Vowden, 2012). Research shows that charter schools contribute to this segregation resulting in more racially isolated schools than their traditional public-school counterparts (Bifulco & Ladd, 2007; Rich et al., 2021). However, some argue that school choice policies can provide a potential solution to school segregation by giving students choices that no longer confine them to their segregated neighborhood school catchment zones (Diem et al., 2019; Orefield, 2013; Wohlstetter et al., 2021). Over the past several years, more than one hundred charter schools have opened 54 across the country with commitments to diversity and plans for intentional integration (Kahlenberg & Potter 2012; Jabbar & Wilson, 2018). These charter schools could offer a means to desegregate schools by sidestepping the historically segregated traditional school zones, but it is unclear why White parents are motivated to choose these specific schools for their children. In the context of White parents claiming to value diversity while ultimately enrolling their children in majority White schools, I aim to understand White parents’ perceptions of these uniquely diverse charter schools and their motivations to enroll their children in these diverse schooling environments. Two opposing theories could explain White parents’ perceptions and motivations. First, colorblind racism asserts that White individuals make choices that deny the influence of race and its role in stratifying society (Bonilla-Silva, 2003). White parents and guardians may express colorblind ideologies when explaining their enrollment in an intentionally diverse charter school by being unaware of the school’s diversity mission and claim that the racial makeup of the student body was unimportant or not considered when making their decision. In contrast to colorblind motivations, the diversity ideology framework could explain White parents’ perceptions and enrollment decisions. This framework asserts that White individuals are race conscious and actively seek out multiracial and multiethnic spaces because they see the diversity in these spaces as benefitting themselves (Smith & Mayorga-Gallo, 2017) and a way to distinguish themselves as nonracist, “good White” individuals (Underhill, 2019). Nevertheless, under diversity ideology, White individuals’ view and use of these spaces and the diversity within them maintains White racial dominance in racial and ethnically diverse settings (Smith & Mayorga-Gallo, 2017). Several studies have analyzed how school leaders have conceptualized diversity in intentionally diverse schools (Jabbar & Wilson, 2018), and evaluated the academic 55 outcomes of students attending these schools (Wohlstetter et al., 2021), but there is little research examining how White parents perceive these schools and why they choose these schools for their children. Given the lack of theoretical consensus regarding White parents’ motivations for exposing their children to diverse environments, and the limited research on White parents enrolling in intentionally diverse charter schools, I interviewed White parents who enrolled their children in intentionally diverse charter schools in the Denver Public School district. My research seeks to answer two main research questions: RQ1: Why do White parents choose intentionally diverse charter schools? RQ2: How do White parents interpret and perceive their chosen intentionally diverse charter school? My study expands upon the current research on intentionally diverse schools and unpacks the perspectives and motivations of a critical group of stakeholders, White parents, who have been shown to have mixed feelings about diversity in schools and have been powerful inhibitors in previous school desegregation efforts (Kahlenberg, 2016; McDermott et al., 2015; Holme & Finnigan, 2018). Additionally, I extend the application of the diversity ideology framework to a new multiracial and multiethnic space, intentionally diverse charter schools. In the following sections, I first summarize literature on intentionally diverse charter schools and White parents’ preferences when selecting K-12 schools. Next, I further explain my theoretical frameworks of colorblindness and diversity ideology. I then describe the data collection strategy and analysis. This is followed by my results section, where I connect parents’ explanations to colorblindness and the guiding tenets of diversity ideology. Lastly, I end with a final section on concluding thoughts from the analysis. 56 Literature Review Intentionally Diverse Charter Schools There are specific criteria that separate intentionally diverse charter schools from other charter schools. Potter and Quick’s synthesis of intentionally diverse charter schools in the United States (2018) defines these schools as charter schools that have mission statements emphasizing a commitment to diversity, employ enrollment practices that intentionally attract a diverse student body, describe the benefits of diversity, and show an appreciation for diversity on their school website. In addition to demonstrating an intent to attract a diverse student body through their websites, schools must also achieve a specific level of diversity based on the actual study body enrollment. Schools are considered racially diverse if their largest racial or ethnic groups comprise 70 percent or less of the student body and 30-70% of the students qualify for free or reduced-price lunch (Potter & Quick, 2018). Research finds that school choice, particularly the presence of charter schools, exacerbates segregation in urban areas (Bifulco & Ladd, 2007; Frankenberg et al, 2010) because of selective enrollment practices, charter schools being built in higher income areas, or school staff discouraging students labelled “hard to serve” from applying (Gulosino & d’Entremont, 2011; Henig & MacDonald, 2002; Jabbar, 2015; Koller & Welsch, 2017; Lubienski, Gulosino, & Weitzel, 2009; Weiler & Vogel, 2015). Intentionally diverse charter schools work to combat these negative outcomes with their enrollment strategies that target specific student groups to attain a diverse student body. Nevertheless, there are still relatively few charter schools that meet all the required qualifications of intentionally diverse. Potter and Quick’s (2018) analysis find only 125 charter schools across the United States that meet all the criteria. While there is limited research exploring intentionally diverse charter schools, the majority of the research to date 57 focuses on how school administrators open and operate intentionally diverse charter schools (Jabbar & Wilson, 2018; Wohlstetter et al., 2016). Jabbar & Wilson (2018) conduct interviews with school leaders in New Orleans and Minneapolis analyzing the marketing practices used in 13 intentionally diverse charter schools. They find that schools put in considerable work to specifically recruit students of color and students from lower income households. To the school leaders’ surprise, they do not have any problem attracting students from affluent White families. The school leaders note that the schools’ missions may have more directly appealed to these individuals, but they could not say what specifically about the missions would have appealed to this group. Wohlstetter et al. (2016) also interview school leaders at 21 intentionally diverse charter schools across the United States to learn about their recruitment strategies and administer a survey to parents asking about their school decision making and levels of engagement. Their interview findings mirror Jabber & Wilson’s (2018) conclusions discovering that school leaders engage in targeted recruitment efforts to encourage students of color and lower income students to apply, while easily attracting higher-income parents. Their parent survey data indicate that some affluent parents may be attracted to schools because of the messages about diversity, but others indicate that they were attracted to these schools because of the pedagogical offerings. In this study I add to this existing body of literature to provide clarity to this previous research by interviewing White parents about their decision-making process and perceptions of the diversity in these schools. White Parents’ School Preferences Parents and/or guardians continue to be some the most important individuals in a child’s racial socialization (Aboud, 2008; Loyd & Gaither, 2018; Sullivan et al., 2021; Waxman, 2021). 58 The term racial socialization encapsulates an individual’s lifelong process of understanding race and racism (Abaied & Perry, 2021; Castelli et al., 2009; Hazelbaker et al., 2022; Hughes, 2003; Hughes et al., 2006; Huguley et al., 2019). Parents are key agents when this process begins for children by establishing a child’s social context which includes their neighborhood, school, and accessible information (Pugh, 2009; Mose, 2016; Underhill, 2017). One implicit racial socialization practice that parents and guardians can engage in when determining their child’s social world is “exposure to diversity” (Underhill, 2019). Exposure to diversity consists of a parent or guardian’s active efforts to foster interracial contact for their children (Underhill, 2019). These efforts can include travel to diverse areas, consumption of food or music from other cultures, or enrolling their child in a racially diverse school (Hagerman, 2014; Posey- Maddox, 2014). Parents and guardians of all races and ethnicities can make decisions with their child’s exposure to diversity in mind, but the reasons that parents seek diverse environments for their children vary by racial group. Research on White individuals’ opinions around school desegregation and school diversity find that White individuals have conflicting attitudes regarding the racial diversity of a student body. Research using data from public opinion polls report that White respondents are overwhelmingly in favor of the idea of increasing diversity in schools. For example, a 2017 Phi Delta Kappan survey finds that 48% of White participants indicate that a racially and ethnically diverse student body is an extremely/very important school factor, and 70% of White respondents indicate a preference for racially diverse schools. Additionally, results from a 2007 NORC poll indicate that 95% of White respondents support Black and White students attending the same school (Frankenberg & Jacobsen, 2011). However, White attitudes towards increasing racial diversity in schools are less supportive when desegregation policies are actually 59 implemented in their own communities. This division is known as the principle policy gap (Carlson & Bell, 2021, Dixon et al., 2017; Smith & Mayorga-Gallo, 2017) where White individuals claim to support ideas that promote diversity, but then attempt to block actual policies that turn the idea into a reality. This disconnect becomes evident when examining White reactions to local school desegregation policies. For example, almost every district that was forced to implement mandatory busing policies has experienced public pushback from local White residents (Carlson & Bell,) and the vast majority of districts released from a court mandated desegregation plans have refrained from enacting policies that voluntarily pursue school desegregation (Carlson & Bell, 2021). While other districts have voluntarily attempted to desegregate public schools by redrawing attendance zones or offering voluntary school choice programs, they have also experienced significant resistance from White residents (Parcel & Taylor, 2015). Thus, even though public opinion polls indicate that White individuals conceptually support the idea of increasing racial diversity in schools, their actions tell a different story when they are faced with a policy that will directly change the makeup of the student body of their community. Furthermore, the principle policy gap may be explained by research revealing White parents’ multiple priorities and considerations when choosing a school for their child (Roda & Wells, 2013). One national public opinion survey finds that 66% of White parents want their child to attend a diverse school because they believe the school’s diversity will help their child develop nonacademic skills that will prepare their children for an increasingly diverse twenty- first century (Hernandez, 2019; Kimelberg & Billingham, 2012; Wells et al., 2009) and expose their children to the “real world” (Byrne & De Tona, 2014; Hollingworth & Williams, 2010). Research indicates that racially diverse schools can benefit students who identify with all races 60 and ethnicities (Edmonds & Killen, 2009; García & Weiss, 2016; Johnson, 2011; Rivas-Drake, Umaña-Taylor, Schaefer, & Medina, 2017; Williams, 2010). For example, during the school desegregation efforts of the 1960s and 1970s, Black and White students reported more positive racial attitudes, less fear, and more tolerance of students from different races (Kurlander & Yun, 2005; Schofield, 1995). Studies routinely find that attending desegregated schools has a positive impact on Black students’ academic achievement (Billings et al., 2014; Card & Rothstein, 2007; Johnson, 2011; Mickelson et al., 2017), high school and college graduation rates (Guryan, 2004; Saatcioglu, 2010; Orfield, 2001), college attendance (Billings et al., 2014; Wells & Crain, 2004); prevalence of cross racial friendships (Slavin, 1979; Williams, 2010), income and job aspirations (Ashenfelter et al., 2006; Boozer, et al., 1993; Grogger, 1996), and decreases the probability of incarceration (Johnson, 2001; Weiner et al., 2009). Less research has focused on the outcomes for White students and graduates of desegregated schools (Schofield, 1981, 1995; Orfield, 2001; Wells et al. 2016), but scholars do conclude that there are benefits to White students such as improved critical thinking (Wells et al., 2016) and a decreased probability of dropping out of high school (Billings et al., 2014). Despite the student benefits and parents’ stated desire for racially diverse schools, research concludes that White parents ultimately choose schools based on academic performance measures. Findings from a survey of 1,898 White parents, reveal that while White parents say they desire racially and socio-economically diverse schools for their children, they ultimately choose schools based on the school’s academic profile, safety record, and location (Hernandez, 2019; Torres & Weissbourd, 2020). Qualitative studies that have asked parents about their school choices find that White parents living in urban areas report feeling hesitant to send their children to their neighborhood schools citing fears of unsafe classroom environments leading to 61 higher rates of bullying (Crozier et al., 2008), and lower academic standards with teachers having to spend the majority of their time meeting the needs of lower achieving students, or non- native English speaking students (Levine-Rasky, 2008; Vowden, 2012). Another source of data on school preferences comes from enrollment applications in portfolio-style choice cities. These studies indicate that family’s first choice schools are schools with higher academic performance (Glazerman & Dotter, 2017; Lincove et al. 2018; 2018b), more experienced and educated teachers (Abdulkadiroglu et al., 2020; Harris & Larson, 2019) and more diverse extracurricular offerings (Harris & Larson, 2019). These studies indicate that a school’s academic performance, rather than the diversity of the student body, is consistently one of the most valued aspects for families who are selecting schools in choice-rich environments. Nevertheless, school safety and academic achievement are often proxies for a school’s racial composition (Billingham & Hunt, 2016). In Billingham & Hunt’s (2016) survey experiment analyzing the influence of racial composition on parent’s school choices in New York City, they find that even when factors that proxy for race are controlled for, White parents prefer schools with a higher percentage of White students. Billingham & Hunt (2016) claim that the social desirability bias may underestimate a parent’s true priorities in regard to the racial composition of a school. While parents claim that safety, academics, quality teachers, and extracurricular activities are their top considerations when examining a potential school, the race of the student body is also an important consideration for parents, leading most White families to opt for majority White schools. 62 Theoretical Frameworks Colorblind Racism Bearing in mind this research on parental preferences and school choice, it begs the question, why are some White parents going against these patterns and choosing to enroll in schools that are intentionally diverse? One theory that could explain White parents’ choices is colorblind racism. Individuals demonstrate colorblind racism when they deny the influence of race and its role in stratifying society (Bonilla-Silva, 2003). Pett’s (2020) survey experiment finds that when parents chose a Whiter school as a school that modeled diversity, they justify their choices using colorblind rhetoric and explain that they do not “see race” and so their choice is not specifically about the racial makeup of the school. Whereas those that choose a school with a higher percentage of Black and Latinx students as the model diversity school explain their rationale by directly addressing the race the ethnicity of the student body. Other research draws on colorblind ideologies as a way of explaining White parents’ neighborhood choices (Hagerman, 2014; Underhill, 2019; Vittrup, 2018). For example, Hagerman’s (2014) ethnographic work studying White parents living in adjacent communities to understand how the racial demographics of each neighborhood play a role in parents’ residential finds that parents who live in the predominantly White neighborhood justify their choice of neighborhood using colorblind rationales. They explain that the lack of diversity in the community is a “‘non-issue’” (pp. 2604), something that is rarely discussed, or something that is not a “‘big deal’”(pp. 2605) and that they try to teach their children “‘it doesn’t matter what color you are, . . .it’s how hard you work’” (pp. 2604). This study builds on this literature that utilizes colorblindess by analyzing if these same colorblind ideologies surface when White parents are asked about their schooling choices after 63 they have ultimately chosen a school whose mission is to intentionally enroll a diverse student body. Even though this is a part of the school’s mission statement, White parents and guardians may express similar colorblind sentiments when explaining their rationale for choosing a school by claiming that the racial makeup of the school was unimportant when making their decision. These schools also tout their college preparation focus and STEM offerings on their websites, and some White parents may have focused on these aspects and not considered the school’s diversity mission when making their choice. Applying a color-blind ideology in this context, one in which it would appear that White parents are making a race-conscious decision, can further reveal if the appearance of a race-conscious school choice is laced with a colorblind mindsets. Diversity Ideology An opposing framework to explain White parents’ and guardians’ selecting into multiracial schools is diversity ideology. Exposure to diversity has become an indicator for White individuals to distinguish themselves as “good white” people (Underhill, 2019). Bourdieu (1984) contends that dominant group members distinguish themselves from subordinate groups by developing specific priorities or preferences. Peterson and colleagues (1996) argue that a shift in priorities and preferences among high-status White individuals has occurred where their status is no longer indicated by exclusion but instead by celebrating diversity. This shift distinguishes the “good White” person as someone who is antiracist and engages in multiracial and multiethnic spaces from other White individuals who do not make these same diversity- conscious decisions (Underhill, 2019). Research examining how White individuals use diversity to achieve a form of model Whiteness draws upon the diversity ideology framework. Diversity ideology centers race and an appreciation for racial differences to ultimately maintain White racial dominance in 64 multiracial/multiethnic spaces (Smith & Mayorga-Gallo, 2017). Since colorblind ideology and diversity ideology both sidestep issues of racial equality and omit White individuals’ role in past and present racial inequalities, these seemingly contradictory frameworks achieve similar ends. They both allow White individuals to believe that racial equality has been achieved by either concluding that race does not matter in society (colorblindness), or through the good intentions of White individuals through their promotion of racial inclusion (diversity ideology) while still avoiding reconciling past actions that have resulted in current racial inequalities. Diversity ideology originated as an organizational theory framework critiquing the diversity initiatives in corporations (Embrick, 2006, 2011). Embrick (2011) describes diversity ideology as "a set of beliefs held by many individuals in US society that women and minorities are not only treated equally in comparison to their white male counterparts, but that institutions such as major US businesses are sincerely invested in creating a racially and gender diverse workplace.” (542). While businesses may intend to prioritize diversity, Embrick (2006, 2011) finds that there is a disconnect between their stated intents and their outcomes. Embrick (2006, 2001) rationalizes this disconnect using the diversity ideology framework and demonstrates how racial and ethnic diversity is prioritized when it is viewed as an asset for corporations. Mayorga (2014) extends Embrick’s work by applying diversity ideology to understand White families’ neighborhood residential choices in a racially and ethnically diverse neighborhood. Later, Mayorga-Gallo (2019) further refines the framework by defining four tenets of diversity ideology that highlight different ways White individuals practice race consciousness: diversity as acceptance, diversity as intent, diversity as commodity, and diversity as liability. Diversity as acceptance calls for the inclusion of people from different racial backgrounds to remedy racial inequality (Mayorga-Gallo, 2019). Diversity as intent focuses on 65 the good intentions of diversity initiatives and actions, particularly in the institutional context (Mayorga-Gallo, 2019). Diversity as commodity is the treatment of diverse spaces or people of color as objects for the benefit of White individuals, dictated by White racial comfort (Mayorga- Gallo, 2019). Lastly, diversity as a liability is showcased when White individuals express appreciation and support for racially diverse spaces while also seeing them as a threat to White status (Mayorga-Gallo, 2019). Diversity as a liability demonstrates the principle policy gap where Whites support diversity as an abstract concept, but then show a lack of support for specific instances or policies that promote diversity. In all four tenants of diversity ideology, White individuals view diversity as a way of recognizing differences between races and ethnicities devoid of recognition of past or present forms of racial inequality and systemic racism more broadly, thus perpetuating the racial hierarchy that privileges Whiteness (Mayorga-Gallo, 2019). Since the creation of the diversity ideology framework by Embrick in 2006, scholars have applied diversity ideology to understand White’s perceptions of race and diversity in a variety of contexts such as higher education, urban studies of gentrification, and education. In higher education, diversity ideology is used to understand millennial college students’ perception of race and policies such as affirmative action (Smith & Mayorga-Gallo, 2017), and the practices and language choices in predominantly White institutions that maintain White dominance (Lang & Yandel, 2019). In urban studies, researchers apply diversity ideology to explain White parenting practices such as taking their children to play in multiracial parks or living in multiracial neighborhoods aiming to expose their children to diverse environments (Mayorga, 2014; Underhill, 2019). Additionally, research in urban studies utilizes diversity ideology to show urban planning’s centering of Whiteness and White priorities (Goetz, Willaims, & 66 Damaino, 2021) and neighborhood gentrification patterns in Boston (Walton, 2021) and Portland, Oregon (Woody, 2021). Some education research also employs diversity ideology such as Emerick’s (2021) analysis studying school leadership in a career and technical education institution and how a leader’s conceptions of diversity results in a lack of support for emergent bilingual students. Nevertheless, diversity ideology has never been applied to the context of White enrollment in intentionally diverse charter schools. This paper aims to expand upon the current diversity ideology canon by applying diversity ideology in a new multiracial/multiethnic space, intentionally diverse charter schools in Denver Public Schools. The Denver Context The National Community Reinvestment Coalition (NCRC) finds that Denver is the second most gentrified city in the United States (Rubino, 2020). Gentrification is not new to Denver, but this recent study by NCRC shows that Denver is becoming more gentrified. NCRC finds that 27% of the eligible neighborhoods in Denver gentrified between 2013-2017, compared to 15% between 2000 and 2013. The gentrifying neighborhoods in Denver are located in areas which have historically been home to people of color, indicating that these populations are disproportionately being forced to relocate due to increased rent prices and property values (Rubino, 2020). The decades of gentrification in Denver coincide with expanding school choice policies in Denver Public Schools (DPS) to meet the needs of the new, typically affluent and White, residents and parents (Diem et al, 2019). In 2008, DPS expanded their school choice policies when they adopted a portfolio management model and passed the Innovation Schools Act (ISA). This ISA encourages the opening of nontraditional public schools, such as charter and magnet schools, and gives schools more autonomy over school budgets, curriculums, schedules, staffing, 67 and resource allocation. Under the ISA, the local board of education operates as the oversight committee for choice schools, holding reviews every three years to determine whether adequate progress had been made. With the introduction of these expanded choice options, parents in DPS find themselves with an array of learning options to choose from outside of their zoned traditional public school. In DPS, all parents are required to fill out an application and list their top five schools for middle school or high school. If a family wants to attend their zoned neighborhood school, they still fill out this application and simply list their zoned school as their first choice. Those zoned for a school are given preference. This centralized and mandated application system ensures that every family engages in the school choice process further illustrating the scope of school choice in DPS. There is a strong link between gentrification and the proliferation of school choice. Pearman and Swain’s (2017) national study found that when school choice options are substantial, the chance of a traditionally non-White neighborhood experiencing increases housing prices and an influx of White families more than doubles. Findings from a 2017 KIDS COUNT report indicate that during the time DPS expanded their school choice policies, schools within the DPS district lines became more segregated (Schimke, 2017). Furthermore, the Colorado Children’s Campaign finds that the level of segregation is highest in DPS compared to any other district based on the dissimilarity index (Schimke, 2017). In light of the segregated nature of schools in DPS after years of gentrification and after the expansion the district’s school choice policy, the district set out to actively pursue new desegregation efforts (Diem et al., 2019). In March 2016, school officials in Denver met to discuss the declining enrollment numbers as a result of gentrification (Diem et al., 2019) and concluded that the decline was the result of two groups exiting the public school system; families 68 forced to relocate due to increasing housing prices, and largely White families taking advantage of Denver’s portfolio management model and choicing out of the traditional public schools in their neighborhood. In 2017, the Denver School Board passed the Resolution to Strengthen Schools because of these enrollment patterns and instituted the Strengthening Neighborhoods Committee (SNC). This committee is made up of multiple stakeholders including parents, students, and teachers who regularly assess the segregation patterns in the district and submit recommendations to the school board (Diem et al., 2019). Since the committee’s inception, the district has invoked several enrollment policies pursuing integration through choice-based enrollment. These include redrawing school boundary lines around select schools of choice to intentionally include families of different races, ethnicities, and income levels, and creating several intentionally diverse charter schools. This policy context in DPS, where intentionally diverse charter schools were implemented to specifically address the racial segregation in the district after years of gentrification and policy expansions to school choice, makes Denver an appropriate location to study White parents’ perceptions and enrollment decisions concerning these intentionally diverse charter schools. Data and Methodology Qualitative research methods were particularly well suited for explanatory research where the goal was to unpack the reasoning behind an individual’s attitude and decision. The aim of this work was to understand White parents’ attitudes and perceptions of intentionally integrated charter schools and the process by which they came to make their decision to send their child to such a school. As Ritchie (2003) asserts, “qualitative research provides a unique tool for studying what lies behind, or underpins, a decision, attitude, behavior or other phenomena. It also allows associations that occur in people's thinking or acting - and the meaning these have for 69 people - to be identified.” (p. 28). For this study, a qualitative interview approach using semi- structured interviews allowed me to identify White parents’ attitudes and perceptions of intentionally integrated charter schools and why they ultimately chose this type of school for their child. Research Methodology and Analysis Interviews were advantageous as they gave me the opportunity to understand the participant’s contextual environment and motivations. I purposefully went into each interview aiming to create a friendly and agreeable environment where I asked each parent to educate me about their experiences in attempt to position the parent as the expert. Conducting semi- structured interviews in this manner allowed me to probe White parents to provide additional information to inform the research (Creswell, 2003; Morris, 2015; Welsh & Williams, 2018). Several methodologists have defined interviewing in the context of qualitative research (deMarrais, 2004, Morris, 2015; Seidman 2013). For example, deMarrais (2004) defined interviewing as “a process in which a researcher and participant engage in a conversation focused on questions related to a research study” (p. 54). Seidman (2013) further extrapolated asserting that the value of interviewing comes from the ability to gain an understanding of human action in context by providing an entry point to understand the meaning of individuals’ behaviors (Seidman, 1998; 2013). The essence of interviewing is a genuine interest in people’s stories as seeing the value in these stories (Seidman, 2013). I conducted interviews with White identifying parents to gather detailed information about their initial decisions when choosing intentionally diverse charter schools and their perceptions of the schools after enrollment. The heart of an interview is “understanding the lived experience of other people and the meaning 70 they make of that experience” (Seidman, 2013, p. 9). Thus, the use of this method of data collection was well suited for the goal of this research. To ensure that I conducted quality interviews, I followed the recommendations of Kvale (1996) and Roulston (2010a, 2010b) where: 1) Shorter interview questions and longer interview answers were the ideal, 2) I followed up with participants to clarify the meaning of answers if relevant, 3) I verified my interpretation of participant’s responses throughout the interview to probe deeper into responses, and 4) The interviews were “self-communication” and relayed a contained story that required little explanation. I used a unique semi-structured interview protocol with open-ended questions to elicit participants’ in-depth responses about their intentionally diverse charter schools. Other interview protocols from research focused on understanding White parents’ decisions to expose their children to diverse communal spaces (Underhill, 2018) and parental motivations for enrolling in schools in diverse neighborhoods (Gillen-O’Neel, 2021) were used to inform my interview protocol. Participants were asked phenomenological questions regarding their initial enrollment decisions, their experiences and satisfaction with the school post enrollment, and how diversity and race played a role in their decision-making process directly (see Appendix for complete interview protocol). Following the guidelines of Maxwell (2013), the components of this research design and interview protocol were modified as new patterns and developments unfolded. For example, after the first interview, the interview protocol was revised to add clarifying questions around the school’s discipline procedures since that appeared to be a relevant factor for parents’ perceptions of diversity and their experiences with the school in general. After each interview, I wrote a reflection memo about topics and quotes that stood out from the interview to help recognize reoccurring themes. After all the data from the interviews 71 was collected, the transcripts were cleaned and de-identified, and I began my analysis of the interview transcripts. First, I coded the transcripts with a set of deductive codes based on the original research questions and the colorblindness and diversity ideology theoretical frameworks. I used the qualitative analysis software, Dedoose, for all transcript coding. After the initial round of coding, I developed a series of memos to describe emergent trends. From this, I conducted a second round of coding with inductive codes exploring the role of the parents’ own school experiences in their decision making for their children. The final analysis involved interpreting the meaning of these related inductive and deductive themes while providing descriptions of the study’s context. Data Collection and Sample I relied on purposeful sampling to collect data from a specific group of White parents who have chosen to send at least one child to an intentionally diverse charter school in the Denver Public School (DPS) system. Patton (2002) asserts that purposeful sampling is an effective tool because “information-rich cases” are selected to learn about a specific issue of central importance. I specifically sought White parents who send one or more of their children to either Byers Middle School, Byers High School, Conservatory Green High School, Green Valley Ranch High School, or Montview High School. These five schools are all charter schools that are a part of the Denver Schools of Science and Technology (DSST) charter network, and the only charter schools in the district that met all the requirements of Potter and Quick’s (2018) definition of intentionally diverse. On each of their websites and student handbooks, they each emphasize a strong commitment and appreciation to diversity on their websites and describe the benefits of diversity. In addition, they each employ enrollment practices that result in an intentionally diverse student body where their largest racial or ethnic groups comprised 70 72 percent or less of the student body and 30-70% of the students qualified for free or reduced-price lunch (Potter & Quick, 2018). While the student body at the 13 schools across the DSST network as a whole met Potter & Quick’s (2018) enrollment criteria for intentionally diverse, some individual campus student demographics did not meet these conditions. Therefore, I limit my sample to these five campuses. Table 10 displays the students demographics at all of the DSST campuses and ensures that the five schools selected met all the requirements based on their student demographic data for the 2021-2022 school year. Table 10 Student Demographics of DSST SchoolsTable 10: Demographics of DSST Students School Female (%) White (%) Black (%) Latinx (%) Asian (%) Other Race (%) FRPL (%) ELL (%) SPED (%) Byers Middle School 49 40 20 31 3 6 48 32 10 Byers High School 45 41 16 33 4 6 42 38 11 Cole Middle School 46 3 22 70 1 5 91 55 18 Cole High School 50 4 11 83 2 1 87 69 17 College View Middle School 49 3 3 87 5 1 84 71 10 College View High School 45 3 1 87 8 2 81 81 9 Conservatory Green Middle School 47 13 20 59 3 5 70 50 10 Conservatory Green High School 47 10 25 56 5 5 62 56 11 Green Valley Ranch Middle School 48 5 25 58 10 4 74 60 10 Green Valley Ranch High School 48 5 23 59 10 6 68 66 10 Montview Middle School 39 18 35 36 5 7 74 59 9 Montview High School 42 17 31 41 4 7 65 47 10 Source http://dps.schoolmint.net/school-finder/home Furthermore, I also specifically selected these schools to allow for geographic comparisons. Byers Middle and High School, Conservatory Green High School, Green Valley Ranch High School, or Montview High School are located in different neighborhoods within the larger Denver metro area. While these charter schools are not limited to their geographic boundaries, they are located in neighborhoods with different racial makeups (see Table 12 in appendix). Research shows that parents consider the transit time and school location relative to their home when making school choice decisions and often rank schools closer to home higher on enrollment applications (Denice & Gross, 2016; Glazerman & Dotter, 2017; Harris & Larsen, 2019; Hastings et al., 2005; Lenhoff et al., 2021; Lincove et al., 2018). Interviewing White parents who chose schools located in four different neighborhoods throughout Denver sheds light 73 on the nuances between Denver neighborhoods and how Whites parents in each community evaluate their nearest schooling options while considering diversity exposure. To select and recruit participants for this study, I first emailed the director of DSST charter schools in February 2022 to explain my research and ask for their assistance recruiting parents. The director connected me with the organization’s Diversity, Equity, and Inclusion (DEI) officer. I met with DEI officer over zoom in February 2022 and explained the purpose of my project to see if DSST would be willing to connect me with parents. The DEI officer gave me the go-ahead for my research and then connected me with the DSST Senior Manager of Donor Stewardship and Events in March 2022 to assist with parent recruitment. After multiple email exchanges explaining my project and discussing recruitment methods, they gave me emails for each school director and each campus’s Community and Engagement Manager (CEM) and suggested I contact them in June 2022 to coordinate recruitment at the individual campus level. Each school director and CEM oversaw the middle school and high school. On June 1, 2022 I emailed each campus’s school director and CEM explaining my study, the individuals at DSST who had connected us, and asked for their assistance and guidance to recruit White parents at their schools. A few days later, the CEM from Montview responded with a list of five White parents at their high school to contact for interviews. I immediately emailed the five families and heard back from two who said that they were willing to participate. In the recruitment email, I told parents that I wished to interview them to understand their school decision-making process but refrained from centering race or diversity in the study description. I did this for two reasons: (1) I feared that mentioning race may hinder participant recruitment (Frankenberg, 1993) and (2) I wanted to give the interviewees the opportunity to comment on the school’s diversity and their own race unprompted. One week later, I sent a follow up email to 74 the other three Montview High School families who did not initially respond. This follow up did not result in any more participants. I emailed a second follow up email one week after that, but to no avail. I tried emailing these individuals one more time in September, after the school year had commenced thinking that they may check their emails more during the school year, but unfortunately this did not result in any more participants. I then emailed the Montview CEM in September asking if they could send out my information to all White parents at their High School since the school year was back in session. The CEM agreed to send out my information at the end of September, but I did not hear from any additional interested Montview parents. I had more success with parents at the DSST Byers campus. The CEM responded to my initial email offering to send out my recruitment email to all of their White parents on June 13, 2022. After they sent my information out, 16 parents emailed me willing to participate in my study. Out of these 16 initial responses, I was able to interview 12 Byers parents. At the Conservatory Green Campus, the CEM responded to my initial email and agreed to email my recruitment information to all White High School parents. While the CEM at Conservatory Green took a similar approach to Byers’ CEM, I only received one response from an interested parent. Because of the low response rate, I emailed the school director and CEM at Conservatory Green at the beginning of July and again in September after the beginning of the school year asking if they could send my information out again, but these additional efforts did not result in any more interested parent participants. Lastly, I never heard back from the school director or the CEM at the Green Valley Ranch campus after I emailed them information about my study. After the initial email was sent on June 1, 2022, I sent three additional follow up emails the second week of June and the last week of June. I did not receive a response from either of these emails. I emailed the Senior 75 Manager of Donor Stewardship and Events who had originally supplied their contact information asking for assistance and verified that the school director and CEM had not experienced a personnel change during the summer. The Senior Manager confirmed that they were still in their positions and gave me the contact information for the assistant school director. I emailed the assistant school director three times throughout July, but never got a response. I emailed all three contacts at Green Valley Ranch once more in September, but never received a response. I conducted 15 semi-structured interviews from June 2022 through October 2022. Parents were given the option to be interviewed over zoom or in person. One parent chose to do the interview in person and 14 were conducted over zoom. Each interview lasted between one and one and a half hours and all were audio recorded for transcription. At the end of each interview, I asked a series of demographic questions to collect data on participants’ gender identity, their child’s gender identity, approximate household income, highest education level, and political ideology. This data is displayed in Table 11 below. 76 Table 11 Sample Demographics Table 11: Sample Demographics Participant Montview Conservatory Characteristics Byers Campus Campus Green Campus Male: 3 Male: 0 Male: 0 Gender Female: 9 Female: 2 Female: 1 Male: 9 Male: 1 Male: 0 Female: 2 Female: 1 Female: 1 Child's Gender Nonbinary: 1 Nonbinary: 0 Nonbinary: 0 50-100K: 3 50-100K: 0 50-100K: 0 100-200K: 5 100-200K: 1 100-200K: 0 200K +: 3 200K +: 1 200K +: 1 Household Income Not given: 1 Not given: 0 Not given: 0 Bachelors: 7 Bachelors: 2 Bachelors: 0 Education Level Postgrad: 5 Postgrad: 0 Postgrad: 1 Very Liberal: 4 Very Liberal: 0 Very Liberal: 0 Liberal: 5 Liberal: 2 Liberal: 0 Center: 1 Center: 0 Center: 1 Political Ideology Conservative: 2 Conservative: 0 Conservative: 0 The demographic data reveal that most of the parents were very well-educated upper middle class women who identified as either liberal or very liberal. These parent demographics align with past research that finds that democrat, middle class, highly educated White women report lower level of racial resentment and are more likely to exhibit race conscious attitudes (Fossett & Kiecolt, 1989, Hughes & Tuch, 2003, Krysan, 1998; Taylor & Mateyka, 2011). Thus, the fact that the majority of the White parents who send their children to an intentionally diverse school associate with these demographics was not a surprise. In addition, most recounted their experiences enrolling their male child in one of the DSST schools. One parent noted the lack of White female students at her son’s school saying that “there are very few Caucasian females at that school. Very, very few, if any.” Most of the parents interviewed enrolled their child at one of the Byers campuses. This overrepresentation from one campus may limit the interpretations of my results because these findings may pertain more to the Byers campus than other DSST locations. 77 Positionality Statement My racial and cultural background greatly influenced how I experience the world and how I approach research. I am a White woman who grew up in a predominantly White suburb of Denver. My path to understanding my own Whiteness informs the research the I pursue and my interest in understanding how other White individuals make sense of their own race and privilege. I entered these interviews with a recognition of assumptions that may be made since I am a White person interviewing White people about race in a space absent of people of color. It has been argued that the quality of qualitative research suffers when there is a mismatch between the racial identity of the researcher and the research subject (May, 2014). I realized that our shared race could create a space in which some level of racial bonding may occur (Underhill, 2018). However, I was also aware that some ideas may be implied rather than explicitly named because of this perceived bonding. I was prepared to probe further and directly name race in the conversation at the risk of fracturing this bond and disrupting the “culture of niceness” in the interview. White individuals often dance around directly naming race in conversations (Gordon, 2005) but directly naming race is important for this research so that I am able to understand how my participants think about race in their children’s schools. Additionally, because of our shared Whiteness and the absence of a person of color, I made an active effort in the interviews to acknowledge the race of other students at the school and the White race of the participants themselves so that they see themselves as a racial being having a racial presence at the school. Throughout the interviews I restrained from divulging my own opinions but instead asked participants probing questions to clearly understand their own opinions and give them ample opportunity for explanations. This research centers on the thoughts of the participants rather than my own as I did not want them to be influenced by my own interests and research agenda. 78 Limitations The aim of the study is to provide insight about a specific population (White parents who send children to one of the five intentionally diverse schools in Denver) which inherently makes the sample non-random. The findings from this study cannot be generalized to other contexts but can provide insight into this specific context which can motivate future research inquires. Furthermore, the overrepresentation of White parents from the Byers DSST campus may limit the generalizability of these findings even further. Additionally, the nature of this qualitative study requires parents to recall their decision-making process, which could have occurred years ago. Parents could find it difficult recalling their reasoning behind their initial school decisions and may have different rationales for keeping their child in this school in light of the COVID-19 pandemic. Findings In this section, I will present the findings and relate them back to the colorblindness and diversity ideology frameworks. First, I will explain how White parents explained their priorities when making their initial schooling decisions. Next, I will describe how parents’ decisions to enroll and experiences in an intentionally diverse charter school align with the diversity as commodity tenet of the diversity ideology framework. Lastly, I will explore ways that parents supported the school’s diversity while also referring to it as a liability. White Parents’ School Priorities When White parents were asked what appealed to them about their DSST school, 12 mentioned the diversity at the school unprompted. Most of the parents in the sample said that the diversity was a positive aspect about the school, but they never mentioned diversity as the first 79 factor that was considered when they recounted their decision-making process and priorities. For example, one parent who sends both of their daughters to DSST Byers stated: We liked DSST Byers because it went through high school. DSST Byers. It's, it's very nearby, it's only a few miles away, but it's, you know, we leapfrogged a couple of schools to get there geographically. . . Looking at some of the stats, very, very strong academic scores. And that was a plus [that] I very much wanted. My kids did really well, in this case, they did really well in elementary school, but they're very strong academically.. . . And then diversity was very, very important. . . being able to look up some of those statistics and you know, it's, it's 45 or 55%, free reduced lunch. And so very diverse socio-economic backgrounds at the school, combined with strong academic testing, I thought, okay, that's, that's the perfect combo. This parent first explained that they chose the school because of the social aspect and its structure that they hoped would help their daughters make lasting friendships by offering grades 6 through 12. Then they listed the distance, followed by the academic scores. The diversity of the school was mentioned after all these other factors. Explanations like these were common amongst White parents where other factors such as school size, distance, academics/STEM focus were listed before the school’s commitment to diversity. The ordering of these factors is important because it confirms other research on parental preferences which finds that while White parents want a diverse school for their children, they prioritize other things before the school’s diversity. Even these White parents who did ultimately choose a school that markets itself as intentionally diverse also considered other school factors when selecting the school, and the school’s diversity was not the driving factor behind their decisions. 80 Diversity as a Commodity When White parents in my sample explained why they valued the diversity in the school, they used language that aligned with the tenet of diversity as a commodity. Diversity as a commodity argues that White individuals view diversity as one more available good that they can consume (Mayorga-Gallo, 2019; Centeno & Cohen, 2012). In a school setting, this consumption often involved preparing students for the future. For example, one White parent said that that school’s diversity would benefit his child in college, whereas another White parent claimed that “this school will prepare [child’s name] for the real world.” One White parent further extrapolated how the diverse environment would benefit their child and said that they wanted a diverse school for their son because “as a middle-class White male. He has every privilege in the world but the world he'll go into is not that world, and it's not going to be that world in the future.” Another parent said that they liked the diversity because they wanted “[child’s name] to see how privileged they are and understand that not everyone comes from a home like they do”. These examples reveal that parents saw diversity at the school as something that they hoped would yield benefits to their children in the future. The school’s diversity became a commodity where the presence of students of color would prepare their White children for the future in some way. In some instances, this was a more direct benefit, such as preparing them for college, but for others it was to offer a comparison so that the White children of these parents could contextualize their privilege. As diversity ideology asserts, while White parents were actively seeking diverse environments, they often reduced students of color or students from lower socioeconomic backgrounds and saw them as mere objects for their own benefit. 81 White parents’ own educational experiences and childhood community often played a part in their enrollment decision and doubled down on their views of diversity as a commodity. Parents who had attended a racially homogeneous school typically wanted their children to have a different education experience. For example, one parent of a Byers high schooler explained: So like, little background of me, I grew up in Boulder, which is one of the least diverse places on the planet, particularly in the 1980s, it was even worse than it is now. And I moved to Chicago after college, and it was just, you know, a wonderful experience of just so many things that, you know, different socioeconomic, different race, different ethnicity, different like the food, my God, you know, it’s incredible. And it also I learned a lot about everything. I don't think that the slice of East Denver in which we live is representative of what the entire country looks like, and so I think it is good for [child] to be exposed to people who live differently and have been raised differently and experience life differently. And that will serve her as an adult. This parent had perceived that attending a majority White high school had not prepared her for other communities that differed from Boulder circa 1980. This parent also talked about consuming diversity during their time in Chicago and spoke about the diversity in a positive light and as an eye-opening experience from which they were able to grow. This time spent outside of their hometown showed them “what they didn’t know” and these lessons are things that they hope their child will learn in an intentionally diverse charter school. By enrolling their child in a more diverse school, they believed that their child would have the opportunity to consume this diversity earlier. 82 Colorblind Ideology Three parents demonstrated colorblind ideologies when describing the diversity of the school. Interestingly, these three parents were the parents who did not identify as liberal or very liberal. All three of these parents did not mention the school’s diversity when recounting their decision-making process and priorities, but only reflected on it when asked explicitly about the school’s diversity commitment. One of these parents explained that they were drawn to DSST because of the school’s STEM focus. This parent of a Montview high schooler explained that “the primary thing. . . you know the name of the school is the Denver School of Science and Technology. And so [we] thought that it what it was going to be . . . all students who want to have a science focused high school experience, they're all choosing to go there.” The other two parents chose the school because of curriculum options that they felt would best fit their individual child’s learning needs. One parent of a Conservatory Green high schooler explained that they liked the school because they did not mandate a foreign language class like most schools. They explained “My kid has a learning disorder, [learning disability]. So they have a learning disorder where it's really hard for them to do language.”, making the lack of language requirement very appealing. Another parent had a child with an intellectual disability and chose this school because of its inclusive education model where her child would stay in the regular education classroom throughout the day instead of being pulled out for special services. When these three parents were directly asked about the school’s diversity, one parent described the diversity as being a “happy accident” or as something that was unimportant. The parent of the child with a learning disability explained that “It's [the diversity] a happy accident honestly, I’m not I mean people can say this, so they're right. Yeah, I don't care about the diversity, everyone's like I want more diversity. I'm like I don't care, I want my kids to be in 83 good places.” One of the other parents explained that while they did notice the diversity in the student body after their child was enrolled they said “it's not something where I would say like, I want to go to school because it looks like this [diverse]. I was looking more for what's the best education that is going to work for my son”. All the parents that exhibited colorblind ideologies explained that their primary concern was sending their children to a school that met the curriculum needs of their children, whether that was an emphasis on STEM, a lack of foreign language requirement, or an inclusive special education curriculum. For these parents, the academics of the school took precedent while the racial makeup of the student body was given little thought or overlooked entirely. Diversity as a Liability Post Enrollment For many parents, even those that valued the schools’ diversity when they enrolled, the diversity at the school was seen as a liability when they reflected on their child’s learning experiences and their own satisfaction with the school. Diversity ideology’s tenet of diversity as a liability is unique because it is illustrated when people praise diversity while also identifying negative aspects. In this sense, people can appear to be supportive of diversity and exhibit things a “good White” person does while also explaining the boundaries of diversity that make it unacceptable. Diversity as a liability emerged in some of the White parents’ initial enrollment decisions, but it was more prevalent when parents described their experiences and perceptions of the school’s diversity post enrollment. This sentiment was something shared between those parents that had mentioned the diversity at DSST as being a positive factor during their enrollment decision and between parents who did not mention it at all. Parents referenced the diversity of the school as a liability in three ways. First, while White parents were supportive of the diverse educational environment at DSST schools, they did 84 not elect to enroll in other schools specifically because of the racial or socioeconomic makeup of their student body. One parent explained why they chose not to send their son to a neighborhood school: when [son] would have gone to [neighborhood public school] for kindergarten. . .it was a like 97% free and reduced lunch. I mean well, that much low income is a problem when you can't raise money for the PE teacher and the library, and whatever, but it's not necessarily a total deal breaker. What's a total deal breaker is the percent of the kids there that are in foster care, English language learners, the percent who are homeless, it's the percent who are living with a relative other than their parents. It is just, they concentrate every kind of disadvantage in that school. DSST Byers is about 50% free and reduced lunch and it is racially diverse. . . Not that it's extremely disadvantaged, it's not, we didn't. . . we didn't really want to send [son] to school that was 90% people who qualify for free and reduced lunch . . . So yeah, just because Denver doesn't fund public schools very well. So like you need parents donating money. You need parents volunteering you, just need that. Any public school is like that, and if you don't have that, which is what happens if 90% are very disadvantaged, those schools just don't, you know, they get special funding but it just doesn't, they just don't work very well. As this parent explained, a school with 50% free and reduced lunch with racial diversity is a good schooling option with a good level of diversity, but a school with 90% free and reduced lunch with a large proportion of homeless children and English Language Learners makes for an unacceptable schooling option. In essence, there are some diverse environments that are acceptable while others are not. Diversity broadly may be seen as a desirable thing, but when White parents unpacked the specific makeup of the student body, it was clear that only certain 85 diverse contexts were seen in a positive light. This parent justified this unacceptable diversity by explaining that the school would be chronically underfunded because schools in Denver rely on financial support from parents for what many White parents perceived as basic needs such as a full-time school librarian. This differentiation in support for diversity is a form of diversity as a liability. As Mayora-Gallo (2019) explains, White people engage in diversity as a liability when “rather than avoiding multiracial spaces, as color-blind logic prescribes. . . [they] aim to control it”. In the context of White parents at an intentionally diverse charter school, that control translates into sending their children to a diverse school where they feel comfortable with the diversity and not sending their children to other schools where the diversity is seen as too much of a risk to their child’s education, and therefore unacceptable. Another reason parents noted that the diversity was acceptable at DSST but not at other schools was because of the school’s additional resources. One parent of a DSST Byers Middle Schooler said the “[the school’s] ratio of student support, like student counselors, to students is the highest in DPS, and it's a smaller school. There's probably like at least 15 like full time staff and that is all they do. So like we know that they're equipped for working with this demographic.” They valued the diversity at the school, but the other schools in the area that had similar demographics but that did not have these personnel were not viewed as positively. This further displays White parents’ perceptions of acceptable diversity. Instead of categorizing acceptable diversity by the makeup of the student body, this parent categorized it by the presence of support staff at the school who they implied were needed to work with a diverse student population. Next, parents touted their support for the school’s diversity while also explaining that the diversity required the school to adopt stricter discipline procedures. As one parent of a Byers 86 Middle Schooler explained “The school has 50% like requirement for admission, they were giving 50% of the spots to children from various districts in the city that were on free lunches. And we were like, this is amazing, you know, so and that's why they have these disciplinary things. That's why it's so strict because they need that.” Some of the strict discipline parents referenced included silent hallways during passing period, mandatory uniforms, and receiving demerits (i.e. a negative consequence) for being tardy to school. Some parents saw the strict discipline at DSST as more of a problem than others. As another parent explained: So we love the idea of a uniform we love the idea of special effort being put into trying to kind of help [students] coexist having come from very different backgrounds and kind of blend. I know that a lot of people felt like this was too much. And I know some people who actually left Byers in the middle of their middle school experience or never stayed for their high school experience, because kids identified this as being too rigid. . . We knew why this discipline was in place, because kids were coming from a variety of backgrounds. There's a lot of like first generation in learning English as a foreign language type of situation. . .We knew they needed this structure to kind of amalgamate. When this parent references amalgamating children from different parts of Denver, diversity as liability reinforces Whiteness as property. They imply that those that are assimilating are students who are first generation immigrants and English Language Learners who are assimilating to the White ideal of schooling. This perception of the strict discipline shows that White racial comfort is prioritized in the school and that while White parents want diversity, they want it on their terms and in an environment where they feel comfortable. Other parents praised the diversity at the school but then explained their dissatisfaction with what they felt was lower 87 academic rigor because of the school’s diverse student body. One parent of a high schooler at Montview explained: So it's a really interesting, and great diverse atmosphere, because you have a lot of students whose families are opting for this because they want their. . . They know that they want their kids to be accepted into college and that's the environment we want him to go into. So we're White, [son] is White he's in the minority, which is a really great experience, I think, for a high school student. But it's not a science focused school and there's not, like he can't even take, there's no AP classes available. He can't even like, they just discontinued AP chemistry for tenth graders, because they found that they weren't having much success with tenth graders taking that class. . . I think they are sort of using ninth and tenth grade, I think, to kind of bring them [other students] up to that level if that makes sense. This parent never directly says that the diversity at the school in resulting in lower rigor, but they explain that they are unsatisfied that their tenth grader cannot take AP classes and note that the reason the school gave for not offering tenth graders AP courses is to avoid within school segregation. They said this while praising the diversity at the school and saying that they think that is it a good thing that their child is in the racial minority. This same parent later explained that they considered changing schools in the middle of the year because of the lack of AP courses but ultimately decided against it because of their child’s insistence on staying. This further demonstrates that while parents want a diverse school, diversity is not their top priority and they do not want to compromise on academic rigor to be a part of a diverse school. 88 Discussion and Conclusion In this research, I study how White parents in Denver chose intentionally diverse charter schools and how they perceived the diversity of the student body in their initial decisions and during their time at the school thus far. I interviewed White parents at these schools and utilized the theoretical frameworks of colorblindness and diversity ideology to understand how they perceived the diversity at the schools. The data reveal that most White parents did consider the schools’ diversity when initially choosing their schools, although it was usually listed after other factors such as the schools’ academic reputation. Furthermore, the participants often said they valued the schools’ diversity, but saw this diversity as a commodity that gave them the opportunity to expose their White children to peers from different demographics with hopes of preparing their White children for the “real-world.” The White parents also spoke of the schools’ diversity as a liability where the diversity necessitated the school to hire more student support personnel and adopt strict discipline procedures. Many White parents noted that this was also how the schools framed their rationale for their discipline protocol. In these interviews, parents often spoke about the diversity at these schools as a commodity, something that would benefit their children in both concrete and abstract ways. This echoes arguments related to interest convergence theory which asserts that White people will only support a policy if it benefits themselves in some way (Bell, 1980). This illustrates that White parents are supportive of diversity when they see that they can gain something from the diversity, but it becomes a liability if the diversity means decreased funding, fewer AP courses, and stricter discipline policies. Based on these findings, White parents’ participation in intentionally diverse schools may appear to be a step in the right direction by using race-conscious language that appears to value 89 diversity, but their good intentions still prioritize Whiteness ideals and comfort. White people’s good intentions often perpetuate racial hierarchies that maintain White privilege, White superiority, and White normalcy (Applebaum, 2021). As Underhill (2019) describes, the presence of children of color is seen as an “add on” to the school experiences of children of White parents. The White parents in my sample operate in a context where they have the privilege to choose to remove their children from the school if they feel the level of diversity has resulted in an unacceptable schooling option. As one parent clearly articulated, there are contexts with acceptable and unacceptable diversity. If these schools were to approach the threshold where the diversity became unacceptable by being seen as more of a liability than a commodity, these parents have the resources to engage with the choice process again and choose a different school for their children. Exposure to diversity by enrolling in an intentionally diverse school also helps White parents and students distinguish themselves as “good Whites” and separate themselves from other White individuals. As one mother explained when describing her choice of neighborhood, they explicitly did not want to live in the neighboring suburban, and mostly White, community because it was “unacceptable and full of awful and mean people”. So while they liked the academic reputations of the suburban schools, they did not want to live around “those people.” White parents saw intentionally diverse schools as a commodity for their children to consume to get ahead in life, but also as a reflection of their beliefs and a symbol of their antiracist views. This study adds to the current research on intentionally diverse schools by examining the perspective of White parents. While these parents may have had good intentions when choosing an intentionally diverse school for their child, these good intentions still perpetuate the current racial hierarchy and as James Baldwin (2013) asserts “it is the innocence that constitutes the 90 crime.” Additional research is needed to assess the reliability of the parental perceptions and decision-making processes around school diversity that I observed. Future research could ask similar questions to White parents at intentionally diverse schools in other communities to understand how these parents view diversity in different contexts. Additionally, it is imperative that future research investigates how parents of color perceive the diversity in intentionally diverse schools. If White parents in these schools are selecting them with good, albeit problematic intentions, how do students and families of color operate in these environments and how do they view the diversity that the White students bring to the school? 91 REFERENCES Abaied, J. L., & Perry, S. P. (2021). Socialization of racial ideology by White parents. Cultural diversity and ethnic minority psychology, 27(3), 431. Abdulkadiroglu, A., Pathak, P., Schellenberg, J., & Walters, C. (2020). Do parents value school effectiveness? American Economic Review, 110(5), 1502-1539. Aboud, F. E. (2008). A social-cognitive developmental theory of prejudice. Handbook of race, racism, and the developing child, John Wiley & Sons. Applebaum, B. (2021) The good intentions of White people. Syracuse University Diversity and Inclusion. https://diversity.syr.edu/the-good-intentions-of-white-people/#_edn3 Ashenfelter, O., Collins, W. J., & Yoon, A. (2006). Evaluating the role of Brown v. Board of Education in school equalization, desegregation, and the income of African Americans. American Law and Economics Review, 8(2), 213-248. Baldwin, J. (2013). The fire next time. Vintage. Bell Jr, D. A. (1980). Brown v. Board of Education and the interest-convergence dilemma. Harvard law review, 518-533. Berrey, E. (2015). The enigma of diversity. In The Enigma of Diversity. University of Chicago Press. Bifulco, R., & Ladd, H. F. (2007). School choice, racial segregation, and test‐score gaps: Evidence from North Carolinas charter school program. Journal of Policy Analysis and Management, 26(1), 31–56. Billingham, C. M., & Hunt, M. O. (2016). School racial composition and parental choice: New evidence on the preferences of white parents in the United States. Sociology of education, 89(2), 99-117. Billings, S. B., Deming, D. J., & Rockoff, J. (2014). School segregation, educational attainment, and crime: Evidence from the end of busing in Charlotte-Mecklenburg. The Quarterly Journal of Economics, 129(1), 435-476. Bonilla-Silva, E. (2003). Racial attitudes or racial ideology? An alternative paradigm for examining actors' racial views. Journal of Political Ideologies, 8(1), 63-82. Boozer, M. A., Krueger, A. B., and Wolken, S. (1993): “Race and School Quality Since Brown v. Board of Education." Brookings Papers on Economic Activity Microeconomics, 269- 326. Bourdieu, P. (1984) Distinction. Harvard University Press. 92 Byrne, B., & Tona, C. D. (2014). Multicultural desires? Parental negotiation of multiculture and difference in choosing secondary schools for their children. The Sociological Review, 62(3), 475-493. Card, D., & Rothstein, J. (2007). Racial segregation and the black–white test score gap. Journal of Public Economics, 91(11-12), 2158-2184. Carlson, D., & Bell, E. (2021). Socioeconomic Status, Race, and Public Support for School Integration. AERA Open, 7(1), 1-16. Castelli, L., Zogmaister, C., & Tomelleri, S. (2009). The transmission of racial attitudes within the family. Developmental psychology, 45(2), 586. Centeno, M. A., & Cohen, J. N. (2012). The arc of neoliberalism. Annual review of sociology, 38, 317-340. Creswell, J. W. (2003). A framework for design. Research design: Qualitative, quantitative, and mixed methods approaches, 2003, 9-11. Crozier, G., Reay, D., James, D., Jamieson, F., Beedell, P., Hollingworth, S., & Williams, K. (2008). White middle‐class parents, identities, educational choice and the urban comprehensive school: dilemmas, ambivalence and moral ambiguity. British Journal of Sociology of Education, 29(3), 261-272. Cucchiara, M. B. (2013). Marketing schools, marketing cities. University of Chicago Press. Darrah–Okike, J., Harvey, H., & Fong, K. (2020). “Because the world consists of everybody”: Understanding parents’ preferences for neighborhood diversity. City & Community, 19(2), 374-397. deMarrais, K. (2004). Elegant communications: Sharing qualitative research with communities, colleagues, and critics. Qualitative inquiry, 10(2), 281-297. Denice, P., & Gross, B. (2016). Choice, preferences, and constraints: Evidence from public school applications in Denver. Sociology of Education, 89(4), 300-320. Diem, S., Holme, J. J., Edwards, W., Haynes, M., & Epstein, E. (2019). Diversity for whom? Gentrification, demographic change, and the politics of school integration. Educational policy, 33(1), 16-43. Dixon, J., Durrheim, K., & Thomae, M. (2017). The principle‐implementation gap in attitudes towards racial equality (and how to close it). Political Psychology, 38, 91-126. 93 Edmonds, C., & Killen, M. (2009). Do adolescents' perceptions of parental racial attitudes relate to their intergroup contact and cross-race relationships?. Group Processes & Intergroup Relations, 12(1), 5-21. Embrick D. G. 2006. “The Making and Selling of an Illusion: An Examination of Racial and Gender Diversity in Post–Civil Rights U.S. Corporations.” PhD dissertation, Department of Sociology, Texas A&M University. Embrick, D. G. (2011) The diversity ideology in the business world: A new Oppression for a new age. Critical Sociology, 37(5), 541–56. Emerick, M. R. (2021) Diversity ideology and school leadership: Obscuring inequities for emergent bilingual students in career and technical education. Education Administration Quarterly, September, 1-17. Fossett, M. A., & Kiecolt, K. J. (1989). The relative size of minority populations and white racial attitudes. Social Science Quarterly, 70(4), 820. Frankenberg, R. (1993) White women, race matters: The social construction of whiteness. Minneapolis: University of Minnesota Press. Frankenberg, E., & Jacobsen, R. (2011). Trends school integration polls. Public opinion quarterly, 75(4), 788-811. Frankenberg, E., Siegel-Hawley, G., & Wang, J. (2010). Choice without equity: Charter school segregation and the need for civil rights standards. Civil Rights Project/Proyecto Derechos Civiles. Retrieved from https://files.eric.ed.gov/fulltext/ED509773.pdf García, E., & Weiss, E. (2016). Making Whole-Child Education the Norm: How Research and Policy Initiatives Can Make Social and Emotional Skills a Focal Point of Children's Education. Economic Policy Institute. Glazerman, S., & Dotter, D. (2017). Market signals: Evidence on the determinants and consequences of school choice from a citywide lottery. Educational Evaluation and Policy Analysis, 39(4), 593-619. Gillen-O’Neel, C. (2021). Sense of belonging and student engagement: A daily study of first-and continuing-generation college students. Research in Higher Education, 62(1), 45-71. Goetz, E. G., Williams, R. A., & Damiano, A. (2020). Whiteness and urban planning. Journal of the American Planning Association, 86(2), 142-156. Gordon, J. (2005). White on white: Researcher reflexivity and the logics of privilege in white schools undertaking reform. The Urban Review, 37, 279-302. 94 Grogger, J. (1996). Does school quality explain the recent black/white wage trend?. Journal of labor economics, 14(2), 231-253. Gulosino, C., & d'Entremont, C. (2011). Circles of influence: An analysis of charter school location and racial patterns at varying geographic scales. Education Policy Analysis Archives/Archivos Analíticos de Políticas Educativas, 19, 1-29. Guryan, J. (2004). Desegregation and black dropout rates. American Economic Review, 94(4), 919-943. Hagerman, M. A. (2014) White families and race: Colour-blind and colour-conscious approaches to White racial socialization. Ethnic and Racial Studies, 37(14), 2598–2614. Hagerman, M. A. (2016) Reproducing and reworking colorblind racial ideology: Acknowledging children’s agency in the White habitus. Sociology of Race and Ethnicity, 2(1), 58–71. Harris, D. N., & Larsen, M. F. (2019 July). The identification of schooling preferences: Methods and evidence from post-Katrina New Orleans. Education Research Alliance for New Orleans. Retrieved from https://educationresearchalliancenola.org/files/publications/Harris-Larsen-How-Parents- Choose-2019-07-05.pdf Hastings, J., Kane, T., & Staiger, D. (2005). Parental preferences and school competition: Evidence from a public school choice program (No. w11805). National Bureau of Economic Research. Hazelbaker, T., Brown, C. S., Nenadal, L., & Mistry, R. S. (2022). Fostering anti-racism in white children and youth: Development within contexts. American Psychologist, 77(4), 497. Henig, J. R., & MacDonald, J. A. (2002). Locational decisions of charter schools: Probing the market metaphor. Social Science Quarterly, 83(4), 962-980. Hernández, M. (2019). White middle-class families and sociocultural diversity in schools: A literature review. The Urban Review, 51(2), 270-300. Hollingworth, S., & Williams, K. (2010). Multicultural mixing or middle-class reproduction? The white middle classes in London comprehensive schools. Space and Polity, 14(1), 47- 64. Holme, J. J., & Finnigan, K. S. (2018). Striving in Common: A Regional Equity Framework for Urban Schools. Harvard Education Press. Cambridge, MA. Hughes, D. (2003) Correlates of African American and Latino parents’ messages to children about ethnicity and race: A comparative study of racial socialization.” American Journal of Community Psychology, 31(1–2),15–33. 95 Hughes, D., Smith, E., Stevenson, H., Rodriguez, J., Johnson, D. J., and Spicer, P. (2006) Parents’ ethnic-racial socialization practices: A review of research and directions for future study.” Developmental Psychology, 42(5),747–70. Hughes, M., & Tuch, S. A. (2003). Gender differences in whites' racial attitudes: Are women's attitudes really more favorable?. Social psychology quarterly, 384-401. Huguley, J. P., Wang, M. T., Vasquez, A. C., & Guo, J. (2019). Parental ethnic–racial socialization practices and the construction of children of color’s ethnic–racial identity: A research synthesis and meta-analysis. Psychological bulletin, 145(5), 437. Jabbar, H. (2015). “Every kid is money” market-like competition and school leader strategies in New Orleans. Educational Evaluation and Policy Analysis, 37(4), 638-659. Jabbar, H., & Wilson, T. S. (2018). What is diverse enough? How" intentionally diverse" charter schools recruit and retain students. education policy analysis archives, 26(165), n165. Johnson, R. C. (2011). Long-run impacts of school desegregation & school quality on adult attainments (No. w16664). National Bureau of Economic Research. Kahlenberg, R. (2016) School integration in practice: Lessons from nine districts. The Century Foundation. Kahlenberg, R., & Potter, H. (2012). Diverse charter schools: Can racial and socioeconomic integration promote better outcomes for students? The Century Foundation. Kimelberg, S. M., & Billingham, C. M. (2013). Attitudes toward diversity and the school choice process: Middle-class parents in a segregated urban public school district. Urban Education, 48(2), 198-231. Koller, K., & Welsch, D. M. (2017). Location decisions of charter schools: an examination of Michigan. Education Economics, 25(2), 158-182. Krysan, M. (1998). Privacy and the expression of white racial attitudes: A comparison across three contexts. Public Opinion Quarterly, 506-544. Kurlaender, M., & Yun, J. (2005). Fifty years after Brown: New evidence of the impact of school racial composition on student outcomes. International Journal of Educational Policy, Research and Practice, 6, 51–78. Kvale, S. (1996). Interviews. An introduction to qualitative research interviewing. Thousand Oaks, CA: Sage. Lang, J.A. & Yandell, L. (2019) Diversity language as system maintenance: Toward alternative frameworks for addressing racism at predominantly White institutions, Christian Higher Education, (18)5, 343-355 96 Lenhoff, S. W., Singer, J., Pogodzinski, B., & Cook, W. (2020). Exiting Detroit for school: Inequitable choice sets and school quality. Journal of Education Policy, 1-23. Levine-Rasky, C. (2008). Middle-Classness and Whiteness in Parents' Responses to Multiculturalism: A Study of One School. Canadian Journal of Education, 31(2), 459- 490. Lewis, A. E., Hagerman, M. A., & Forman, T. A. (2019). The sociology of race & racism: Key concepts, contributions & debates. Equity & Excellence in Education, 52(1), 29-46. Lincove, J. A., Cowen, J. M., & Imbrogno, J. P. (2018). What's in your portfolio? How parents rank traditional public, private, and charter schools in post-Katrina New Orleans’ citywide system of school choice. Education Finance and Policy, 13(2), 194-226. Lincove, J. A., Valant, J., & Cowen, J. M. (2018b). You can't always get what you want: Capacity constraints in a choice-based school system. Economics of Education Review, 67, 94-109. Loyd, A. B., & Gaither, S. E. (2018). Racial/ethnic socialization for White youth: What we know and future directions. Journal of Applied Developmental Psychology, 59, 54-64. Lubienski, C., Gulosino, C., & Weitzel, P. (2009). School choice and competitive incentives: Mapping the distribution of educational opportunities across local education markets. American Journal of Education, 115(4), 601-647. Maxwell, J. A. (2021). Why qualitative methods are necessary for generalization. Qualitative Psychology, 8(1), 111. May, R. (2014). When the methodological shoe is on the other foot: African American interviewer and white interviewees. Qualitative Sociology, 37(1), 117–136. Mayorga, S. (2014). Behind the white picket fence: Power and privilege in a multiethnic neighborhood. UNC Press Books. Mayorga-Gallo, S. (2019). The white-centering logic of diversity ideology. American Behavioral Scientist, 63(13), 1789-1809. McDermott, K.A., Frankenberg, E., Diem, S. (2015) The “post-racial” politics of change: Changing student assignment policy in three school districts. Education Policy, 29(3) 504-5. McWhorter, J. H.(2000) Losing the Race: Self-Sabotage in Black America. Free Press. Mickelson, R. A., Smith, S. S., & Nelson, A. H. (Eds.). (2017). Yesterday, today, and tomorrow: School desegregation and resegregation in Charlotte. Harvard Education Press. 97 Morris, A. (2015). A practical introduction to in-depth interviewing. Sage. Mose, T. R (2016) The playdate: Parents, children, and the new expectations of play. New York: New York University Press. Orfield, G. (2001). Diversity challenged: Evidence on the impact of affirmative action. Harvard Education Publishing Group. Cambridge, MA Orfield, G. (2013). Housing segregation produces unequal schools. Closing the opportunity gap: What America must do to give every child an even chance, 40-60. Parcel, T. L., & Taylor, A. J. (2015). The end of consensus: Diversity, neighborhoods, and the politics of public school assignments. UNC Press Books. Patton, M. Q. (2002). Two decades of developments in qualitative inquiry: A personal, experiential perspective. Qualitative social work, 1(3), 261-283. Pearman, F. A., & Swain, W. A. (2017). School choice, gentrification, and the variable significance of racial stratification in urban neighborhoods. Sociology of Education, 90(3), 213-235. Peterson, R. A., & Kern, R. M. (1996). Changing highbrow taste: From snob to omnivore. American sociological review, 900-907. Petts, A. L. (2020) It’s all in the definition: Color-blind interpretations of school diversity. Eastern Sociological Society, 35(2), 465-487. Phi Delta Kappan (2017, September). The 49th annual PDK report of the public’s attitudes towards the public schools. Kappan Magazine. Posey-Maddox, L. (2014) When middle-class Parents choose urban schools: Class, race & the challenge of equity in public education. Chicago: University of Chicago Press. Potter, H. & Quick, K. (2018, May) Diverse-by-design charter schools. The Century Foundation. Retrieved from https://tcf.org/content/report/diverse-design-charter-schools/?agreed=1 Pugh, A. (2009) Longing and belonging: Parents, children, and consumer culture. Berkeley, CA: University of California Press. Rich, P., Candipan, J., & Owens, A. (2021). Segregated neighborhoods, segregated schools: Do charters break a stubborn link?. Demography, 58(2), 471-498. Ritchie, J. (2003). The Applications of Qualitative. Qualitative research practice: A guide for social science students and researchers, 34, 24. 98 Rivas‐Drake, D., Umaña‐Taylor, A. J., Schaefer, D. R., & Medina, M. (2017). Ethnic‐racial identity and friendships in early adolescence. Child Development, 88(3), 710-724. Roda, A., & Wells, A. S. (2013). School choice policies and racial segregation: Where white parents’ good intentions, anxiety, and privilege collide. American Journal of Education, 119(2), 261-293. Rothstein, R. (2017). The color of law: A forgotten history of how our government segregated America. Liveright Publishing. Roulston, K. (2010a). Reflective interviewing: A guide to theory and practice. Thousand Oaks, CA: Sage. Roulston, K. (2010b). Considering quality in qualitative interviewing. Qualitative Research, 10(2), 199- 228. Rubino, J. (2020, July 8) Denver No. 2 for gentrification in recent years, national study finds. The Denver Post. Retrieved from https://www.denverpost.com/2020/07/08/denver- gentrification-san-francisco/ Saatcioglu, A. (2010). Disentangling School-and Student-Level Effects of Desegregation and Resegregation on the Dropout Problem in Urban High Schools: Evidence from the Cleveland Municipal School District, 1977-1998. Teachers College Record, 112(5), 1391-1442. Schimke, A. (2017, April 27) The thorny problem of segregated schools and Denver’s newest plan to address it. Chalkbeat Colorado. Retrieved from https://co.chalkbeat.org/2017/4/27/21099759/the-thorny-problem-of-segregated-schools- and-denver-s-newest-plan-to-address-it Schofield, J. W. (1981). Unchartered territory: Speculations on some positive effects of desegregation on white students. The Urban Review, 13(4), p. 227-241. Schofield, J. W. (1995). Review of research on school desegregation's impact on elementary and secondary school students. Seidman, I. E. (1998). Interviewing as qualitative research: A guide for researchers in education and the social sciences (2nd ed.). New York, NY: Teachers College Press. Seidman, I. E. (2013). Interviewing as qualitative research: A guide for researchers in education and the social sciences (4th ed.). New York, NY: Teachers College Press. Slavin, R. E. (1979). Effects of biracial learning teams on cross-racial friendships. Journal of Educational Psychology, 71(3), 381. 99 Smith, C. W., & Mayorga-Gallo, S. (2017). The new principle-policy gap: How diversity ideology subverts diversity initiatives. Sociological Perspectives, 60(5), 889-911. Sullivan, J., Wilton, L., & Apfelbaum, E. P. (2021). Adults delay conversations about race because they underestimate children’s processing of race. Journal of Experimental Psychology: General, 150(2), 395. Taylor, M. C., & Mateyka, P. J. (2011). Community influences on white racial attitudes: what matters and why?. Sociological Quarterly, 52(2), 220-243. Thernstrom, A., & Thernstrom, S. (1997). The prescience of Myrdal. Public Interest, (128), 36. Torres, E., & Weissbourd, R. (2020). Do parents really want school integration? https://mcc.gse.harvard.edu/ Underhill Megan R. 2017. “Parenting during Ferguson: Making Sense of White Parents’ Silence.” Ethnic and Racial Studies 41(11):1934–51. Underhill, M. R. (2018). Parenting during Ferguson: Making sense of white parents’ silence. Ethnic and racial studies, 41(11), 1934-1951. Underhill, M. R. (2019) “Diversity is important to me”: White parents and exposure-to-diversity parenting practices. Sociology of Race and Ethnicity, 5(4), 486–499. Vittrup, B. (2018). Color blind or color conscious? White American mothers’ approaches to racial socialization. Journal of Family Issues, 39(3), 668-692. Vowden, K. J. (2012). Safety in numbers? Middle-class parents and social mix in London primary schools. Journal of Education Policy, 27(6), 731-745. Walton, E. (2021) Habits of Whiteness: How racial domination persists in multiethnic neighborhoods. Racial Ideologies in Context, 7(1), 71-85. Waxman, S. R. (2021). Racial awareness and bias begin early: Developmental entry points, challenges, and a call to action. Perspectives on Psychological Science, 16(5), 893-902. Weiler, S. C., & Vogel, L. R. (2015). Charter school barriers: Do enrollment requirements limit student access to charter schools?. Equity & Excellence in Education, 48(1), 36-48. Wells, A. S., & Crain, R. L. (2004). Perpetuation theory and the long-term effects of school desegregation. Review of Educational Research, 64(4), 531-555. Wells, A. S., Holmes, J. J., Revilla, A. T., & Atanda, K. (2009). Both sides now: The story of school desegregation’s graduates. Berkeley: University of California Press. 100 Wells, A. S., Fox, L., & Cordova-Cobo, D. (2016). How racially diverse schools and classrooms can benefit all students. The Education Digest, 82(1), 17. Welsh, R. O., & Williams, S. M. (2018). Incentivizing improvement or imposition? An examination of the response to gubernatorial school takeover and statewide turnaround districts. Education Policy Analysis Archives, 26(124), 1-36. Williams, S. M. (2010) Through the eyes of friends: Investigation of school context and cross- racial friendships in racially mixed schools. Urban Education, 45(4), 480-505. Wohlstetter, P., & Buck, B. (2016). Intentionally diverse charter schools: Why and how. AERA Online Paper Repository. Paper presented at the Annual Meeting of the American Educational Research Association (Washington, DC, Apr 8-12, 2016). Wohlstetter, P., Cordes, D., Smith, J., (2021) Moving the needle on desegregation: Performance outcomes and implementation lessons from diverse-by-design charter schools. Teachers College Columbia University Woody, A. (2021). Emotions and ambient racism in America’s whitest big city. Social Problems, 00, 1-18. 101 APPENDIX Interview Protocol 1. Can you tell me a little bit about your child’s school? a. PROBE What is the name of your child’s school? b. PROBE What grade is he/she in? c. PROBE How many years have they attended this school? 2. When you were growing up, what was your school like? a. PROBE Where did you go to school? b. PROBE Was it a public school, private school, ect.? 3. Why did you choose this school? a. PROBE What features were you looking for when you were deciding which school to send your child to? b. PROBE What about this school appealed to you? 4. Were their other schools that you considered sending your child to? a. PROBE What are their names? 5. What made you decide against sending your child to these other schools and their local school? 6. Are you happy with your child’s school experience? Why or why not? 7. When you hear the word diversity, what comes to mind? 8. How would you describe the demographics of the students in your child’s class? 9. ## is a school that is a part of the diverse charter school coalition which supports schools whose mission is to intentionally integrate students across race, economic status, and language. Were you aware of this when you chose this school? 102 a. If you were aware, how did that factor into your decision for choosing this school? 10. When you were growing up, what was your school like? What was the racial makeup of your school? 11. How does your child identify in terms of race? 12. (If they send child to school in charter network) According to the charter school website the school’s racial demographic breakdown is 60% Latinx, 20% Black, abut 15% White and about 5% Asian. How do you feel about your child being in the minority racially at their school? 13. Have you ever had a conversation with your child about race? Why or why not? a. What prompted the conversation? b. What were your initial thoughts when your child instigated this conversation? c. What did you or your child say during the conversation? 13. Have you noticed any changes in how your child thinks or talks about race since attending ## school? 14. Over the past year, there’s been a lot of news coverage instances of racial violence after the death of George Floyd by a police officer. Is this something you’ve spoken to your children about? Why or why not? 103 Table 12: Table 12 DSST Campus Neighborhood DSST Campus Demographics Neighborhood Demographics American Indian/Alas Two or kan Native Native More Races Campus White (%) Black (%) Latinx (%) Asian (%) (%) Hawaiian (%) (%) Byers 80 > <5 5-20 3-5 <3 1 5-10 Conservatory Green 80 > 5-10 5-20 5-10 <3 <3 10-20 Green Valley Ranch 20-40 25-50 20-50 5-10 <3 1-5 10-20 Montview 60-80 5-10 5-20 2-5 <3 0 20-30 Source https://mtgis-portal.geo.census.gov/arcgis/apps/MapSeries/index.html?appid=2566121a73de463995ed2b2fd7ff6eb7 104 PAPER 3: DUAL LANGUAGE IMMERSION IN NORTH CAROLINA: A MIXED METHODS APPROACH EXAMINING CHANGES IN ACCESS TO DUAL LANGUAGE IMMERSION Introduction Dual Language Immersion (DLI) education, providing education content instruction in English and another language, has experienced rapid growth over the last three decades (Howard & Sugarman, 2007; Uzzel & Ayescue, 2021; Valdez et al., 2016). Research shows that DLI programs are the most effective way for designated English language learners (ELLs) to acquire English fluency (Howard, et al., 2003; Lindholm-Leary & Genesee, 2014; Lindholm-Leary & Howard, 2008; Thomas & Collier, 2011). Additionally, monolingual English-speaking students and parents may be attracted to DLI programs because they offer the most successful curriculum option for students attempting to become bi or multilingual (Thomas & Collier, 2011). Furthermore, research shows that many native English-speaking parents, and particularly White parents, believe that bilingualism makes one more competitive in the ever-globalizing economy (Delavan et al., 2017; Parkes, 2008; Petrovic, 2005; Ricento, 2005; Valdés, 1997). During this time of national program expansion, North Carolina’s Latinx population more than doubled and the state has greatly expanded DLI in schools (Cervantes-Soon, 2014). In January 2013, a task force report was released that was commissioned by the state board of education which recommended expanding these programs throughout the state. North Carolina is currently ranked fifth in the nation for the number of DLI programs offered and ranked first among states in the Southeast region (Cervantes-Soon-2014; NCDPI, 2022). Bearing in mind discourses such as globalized human capital, Whiteness as property, and English hegemony that emerge in studies of the equity of DLI in other states (Delavan et al 2021), I began to wonder how equitable the expansion of DLI was in North Carolina. Other 105 research on dual language programs finds that these programs are often concentrated in wealthier and majority White schools, but does this same concentration apply in a state that is expanding DLI while simultaneously experiencing a population change resulting in a decrease of White residents and an increase in Latinx, Asian, and non-native English-speaking population? Because of the research that shows how DLI programs can benefit all students, are these schools being accessed by designated ELL students, like they were originally designed, or are they oversubscribed by other students who hope to become bilingual or multilingual? In this paper I apply the processes from Valdez et al. (2016) analysis of Utah’s dual language policy to the DLI context in North Carolina. The analysis is guided by the following research questions: 1) How do the most recent policy documents that recommend an expansion and change in North Carolina’s DLI program advance or defy the global human capital, English hegemony, Whiteness as property frameworks? 2) How did access to DLI programs change before and after the task force report was released in 2013? I utilize a mixed methods approach by first critically analyzing central policy documents for North Carolina’s DLI program. Then, I conduct a quantitative analysis comparing students demographics in DLI schools that instituted a DLI program before the release of the 2013 task force recommendation to the demographics in schools that adopted the program after the release of the task force report. In general, I find that while policy documents critique English monolingualism for students in North Carolina, the primary rationale of North Carolina’s DLI policy is to develop a competitive human capital that can interact and compete on a global scale. In this pursuit, folk bilinguals and multilinguals are often silenced or written about as an 106 afterthought in policy documents. The demographic findings also demonstrate that schools that adopted DLI programs before the task force report were significantly less White than those who adopted the program after the report, even though they are comparable on the percent of free and reduced prince lunch students, and percent of English language learning students served. Background of Dual Language Immersion (DLI) programs Bilingual education has been linked to immigrant and social movements since its inception. Historically, Mexican American students were segregated from White students in the southwestern United States for the majority of the early 1900s and were punished for speaking Spanish in school (Snyder, 2020). Bilingual education became a priority of the Chicanx social movements of the 1960s to demonstrate the value of Spanish culture and language fluency as a means of countering past language repression policies that were seen as a means of cultural assimilation (Flores, 2016). The Bilingual Education Act (BEA), passed in 1968, designated Title III federal funds to states through competitive grants to specifically fund education programs for linguistic minorities (Trujillo, 2005). However, bilingual education priorities began to shift to emphasize English acquisition rather than bilingualism in response to increased immigration and the United States’ population becoming more racially and ethnically diverse (Uzzel & Ayescue, 2021). While no federal language policy exists in the United States, some states went so far as to implement English-only language policies in an attempt to stress English acquisition (Uzzel & Ayescue, 2021). Other federal education programs such as No Child Left Behind (NCLB) and the Every Student Succeeds Act (ESSA) mandated that all students take high stakes tests in English (Uzzel & Ayescue, 2021). While Title III does not prohibit teaching in non-English languages, it does not contain any specific mandates around bilingual education. Even though some states employ language 107 policies that limit instruction in non-English languages, others utilize an array of policies that emphasize bilingual instruction models including transitional bilingual education, one-way immersion programs, and two-way immersion programs (Uzzel & Ayescue, 2021). Transitional bilingual education consists of designated ELLs receiving special support in their native language while learning English with the ultimate goal of transitioning to 100% English instruction by mid-elementary school (Kelly, 2018). Two other forms of bilingual education all fall under the label dual language or dual language immersion. In one-way immersion (OWI) programs, students study content through a non-English language throughout elementary school, often extending the secondary grades (Kelly, 2018). Lastly, in two-way immersion (TWI) programs students study content material in both English and another language. Schools with TWI programs aim to create a balance of learners with different language backgrounds to provide an immersive language experience and develop bilingualism in English-first students and in students whose first language is not English (Giacchino-Baker & Piller 2006; Uzzel & Ayescue, 2021; Whiting & Feinaur, 2011). TWI programs can be implemented in a variety of formats. Some utilize a 90:10 model where students start their bilingual education journey with 90% of the instruction in one language and then ultimately transition to a point where the languages are taught 50:50 (de Jong, 2016). Others may start in a 50:50 model and choose to break up this time with mornings devoted to one language and afternoons to another, or alternate languages each day or week. Even though DLI programs were limited in the 1960s and 1970s with English acquisition being the dominant trend in bilingual education policy, DLI programs are now quickly growing throughout many states (de Jong, 2016). There is not an up-to-date list of all DLI programs nationwide, but the Center for Applied Linguistics compiled a list in 2016 and 108 reported over 300 programs across the United States. Others estimate that there may be more than 1,000 DLI programs currently in existence (Uzzel & Ayescue, 2021). Part of the growth of DLI programs likely stems from research applauding their high achievement outcomes for native English students and designated ELL students. Scholars examining DLI immersion in several states find that both English and non native-English speaking students reach their academic goals and language proficiency in these programs (Howard, et al., 2003; Lindholm-Leary & Genesee, 2014; Lindholm-Leary & Howard, 2008; Thomas & Collier, 2011). In addition, research indicates that DLI programs are the “best educational program model available for language minoritized students” (Freire et al, 2022, pp. 29) when achievement levels are compared between designated ELLs in DLI to designated ELLs in other language programs (Howard & Sugarman 2007; Lindholm-Leary 2001; Lindholm-Leary & Block 2010; Shannon & Milian 2002; Thomas & Collier 2002). Statewide data evaluating DLI programs in North Carolina find that students from both language backgrounds score 1-2 years ahead of their peers in non-DLI programs on state math assessments (Thomas & Collier, 2011). Furthermore, research has also found that students in these programs show more positive attitudes towards bilingual education and peers with language backgrounds different than their own (Feinauer & Howard, 2014; Ramos, 2007). Access to DLI Programs Research examining parents’ motivations for enrolling in DLI programs using surveys or interviews find that the most common rationale parents give for enrolling in DLI is that they want their child to be fluent in a language other than English (Giacchino-Baker & Piller 2006; Parkes, 2008; Shannon & Milian, 2002). This is consistent across parents of different races and whether they transfer into the school or are attending their designated neighborhood school that 109 happens to house a DLI program. The second most common reason parents choose DLI programs is because they see some future career advantages from attending the schools and learning a second language (Parkes, 2008; Shannon & Milian, 2002). Research examining DLI programs finds that while dual language programs originated as a policy to help designated ELLs students become fluent in English, these programs have become the target of White, native English-speaking parents and students, pushing out designated ELL students (Flores, 2016; Flores & Garcia, 2017; Friere et al., 2017; Freire et al., 2022; Palmer et al., 2020; Valdez et al., 2016) Freire and Aléman Jr.’s (2021) ethnographic research analyzing the perceptions of students in the TWI strand within a school in Utah find that the TWI program is viewed as elitist by teachers and administrators because students have to test into the program in Kindergarten. This gatekeeping mechanism marginalizes the Spanish speaking students, which the school labels as underperforming, and thus prohibits them from the TWI program. Other research examining policy documents and marketing materials of TWI programs in Utah, Georgia, Delaware, and Wyoming finds that these documents use language that prioritizes the perspectives of native English-speaking families and White families (Freire et al., 2017; Freire et al., 2022). ELLs students are often dismissed from these policy documents or referred to as a tool to provide a “diverse perspective” for the other students in the TWI classroom. These patterns have led scholars to deem DLI as a new frontier for gentrification where historically privileged groups are able to capitalize on the policy for their own benefit while pushing others out (Valdez et al., 2016). Theoretical Frameworks To understand how DLI is framed and accessed in North Carolina, I draw on several frameworks including globalized human capital theory, Whiteness as property, and English 110 hegemony. In this section I will explain each of these frameworks in more detail and describe how they can each play a part in DLI policy. Globalized Human Capital Scholars have investigated the connection between the explosive growth of DLI programs and bilingualism drawing on Ruiz’s (1984) theory of language-as-resource (Delavan et al., 2017; Flores, 2016; Petrovic, 2005; Ricento, 2005; Valdés, 1997). Ruiz (1984) proposes three orientations of language policies as advocacy tools for minorities: language-as-problem, language-as-right, or language-as-resource. Ruiz (1984) argues that the policies prioritizing English acquisition view a lack of English fluency as a problem and gaining English fluency as a personal right. However, policies such as DLI view bilingualism as a resource or a competitive asset in an increasingly globalized world (Delavan et al., 2017; Flores, 2016; Ruiz, 1984). Viewing bilingualism with this framework adds status to bilingual and multilingual individuals and views bilingualism as a sought-after resource for one to possess. Scholars have offered cautionary warnings regarding the language-as-resource ideology asserting that this promotion of bilingual status could result in elites commodifying bilingualism for their own competitive advantage (Delavan et al., 2017; Petrovic, 2005; Ricento, 2005; Valdéz, 1997). The commodification of bilingualism draws on neoliberal ideals and the globalization of human capital (Fairclough, 2006). Globalized human capital in this context relates to bilingualism operating as a sought-after skill that makes one more competitive in a global market (Valdez et al., 2016). The material effects of this often translate into an unequal distribution of opportunities to attain such a skill where hegemonic discourses dominate the distribution of resources (Valdez et al., 2016). Some have referred to this shift in bilingual status as a form of gentrification that enables the dominant group to define and exploit a valued resource for their 111 own gain while silencing or dismissing the skills and resources of less dominant groups and limiting their access (Flores & Beardsmore, 2015; Heimanan & Yanes, 2018; Valdez et al., 2016). Furthermore, this framework undermines the original equity minded nature of DLI by limiting access to DLI by less dominant groups, thus creating further inequities. Whiteness as Property and English Hegemony The neoliberal perspective on bilingualism intersects with theories of Whiteness as property (Chavez-Moreno, 2021; Harris, 1993) and English hegemony. Harris (1993) argues that there are advantages for those who identify as White, and these privileges have kept White individuals on top of the racial hierarchy for generations. The orientation of Whiteness as property is one mechanism that perpetuates White privilege. According to Harris (1993), Whiteness as property extends the notion of property, encompassing the allocation of public and private rights as well as the ability to exclude others from such rights based on race. Subsequently, Whiteness acts simultaneously as ‘‘something that can both be experienced and deployed as a resource [. . .] at the social, political, and institutional level to maintain control’’ (Harris, 1993, p. 1734). Therefore, White individuals may see the high status of bilingualism and exploit their privilege and resources to attain this status to have a competitive advantage (Delavan et al., 2017; Giacchino-Baker & Piller 2006; Heimann & Yanes, 2018; Peck et al., 2009; Petrovic, 2005; Ricento, 2005). Scholars have applied Whiteness as property to bilingualism to study the verbiage used in official bilingual policy that privileges White individuals (Freire et al., 2017; Snyder, 2020) and as a way to explain inequities in policy implementation in case studies by noting the dominance of White individual’s perspective DLI policy implementation (Burns, 2017; Palmer, 2010). 112 Whiteness as property intersects with English hegemony which highlights the power and privilege afforded to English speakers in the United States (Macedo et al., 2003; Shannon, 1995, Valdez et al., 2016). English hegemony is closely linked to monanglicization which describes the lack of value the United States places on immigrants’ fluency in their heritage language and the pressures immigrants face to abandon their first language for the sake of mastering English (Friere et al., 2016). Monanglicization is highlighted when comparing the value placed on elite multilingualism to folk multilingualism which represent different paths to multilingualism. Elite multilingualism occurs when an individual, typically from a higher societal class, becomes multilingual through formal education. Folk multilingualism occurs when and individual, usually from a lower societal class, becomes multilingual through the transmission of language between family and community members. While these both achieve the same end, DLI instruction caters to the former and because English is usually one of the languages in DLI programs, it demotes the knowledge of some languages that are less often taught through formal instruction. The North Carolina Policy Context North Carolina offers an interesting context to examine changes in access to DLI because it has experienced a rapid increase of immigrants and designated ELL students over the last two decades (Park et al., 2017). Between 2000 and 2010, the Latinx population in North Carolina more than doubled in size (Cervantes-Soon, 2014; Ennis, Rios-Vargas, & Albert, 2011). The Latinx population in the southeastern region of the United States mostly consists of Mexican immigrants, but about a third of the population comes from Central America or from other cities in the United States (Beck & Allexsaht-Snider, 2002; Cuadros, 2011; Ennis et al., 2011). Economic and social changes in the southeast during the latter half of the 20th century paved the 113 way for lower income Black and White residents to move on from lower paying, labor intensive jobs, which left a gap in the economy that migrant and immigrant workers have filled (Cervantes-Soon, 2014). Additionally, this most recent migration and immigration movement has involved Latinx family units moving together, in contrast to the past when it was typically only males who relocated (Cervantes-Soon, 2014). Because of this, this population shift coincided with an increase of designated ELL students in schools. Between 2007 and 2013, North Carolina experienced a 10% increase in designated ELLs, the largest increase of any state in the southeast region (USDE, 2015). North Carolina has also witnessed a rapid increase in DLI programs during this time. The state first implemented a program in 1997 and now has over 76 TWI and 115 OWI programs covering eight languages including Spanish, Chinese, Urdu, French, German, Greek, Cherokee, and Japanese (Cervantes-Soon, 2014; NCDPI, 2022). Part of the increase in programs occurred after Thomas and Collier’s (2011) study on the effectiveness of two-way language immersion programs in North Carolina, and, most recently, after the recommendations from the State Board of Education’s task force on Globalized Education. Between 2008-2010, North Carolina’s DLI programs were the subject of Thomas and Colliers study assessing the academic outcomes for designated ELL and non- ELL students in DLI programs throughout the state. They concluded that these programs could be an opportunity to close the achievement gap between designated ELL students and English native students. The test scores in Math and Reading were significantly higher for students in DLI programs, regardless of their language status, compared to their designated ELL and English native peers in English monolingual classrooms. However, while they primarily focused on the positive outcomes of DLI programs, they failed to highlight 114 that designated ELL students and native English speaking Black students performed lower than their native-English speaking White peers (Cervantes-Soon, 2014). After these findings were released, the State Board of Education commissioned a task force for the state on “globalized education” in 2011. In January 2013 the task force released its report with five policy recommendations for the state. These included: (1) developing robust and cutting-edge teacher support and materials; (2) implementing new language instruction programs; (3) cultivating new school models; (4) creating a process for district networking and recognition; and (5) developing strategic international relationships. (NCDPI, 2013) Under the second recommendation for new language studies, the task force specifically recommended the state “implement a plan for statewide access to dual language/immersion opportunities beginning in elementary school and continuing through high school” (NCDPI, 2013, p.6) Currently, North Carolina is ranked fifth in the nation for the number of DLI programs offered and ranked first among states in the southeastern region (NCDPI, 2022). The task force recommendations act as the primary policy document guiding DLI immersion in the state (Cervantes-Soon, 2014; USDE, 2015). North Carolina’s DLI policy efforts focused on developing DLI teaching standards after the task force report was released. North Carolina is one of three states, along with Ohio and Utah, that have adopted language proficiency standards from the American Council on the Teaching of Foreign Languages (ACTFL). These standards were adopted in 2014, after the release of the task force report, and outline specific education standards for DLI programs spanning Kindergarten through twelfth grade. Because of a lack of other DLI implementation guidelines from the state, much of the DLI policy is left to local discretion (USDE, 2015). For example, while some states such as Texas, New Jersey, and New York, require districts to provide bilingual education if they have a 115 certain number of designated ELL students in the same school and in the same grade level, North Carolina does not have such requirements (USDE, 2015). Additionally, DLI programs are expected to run from kindergarten through 12th grade, but this is not mandated. The state also never outlines a clear ratio between native and non-native English-speaking students in TWI classrooms in North Carolina, however the task force’s report does stipulate that the “ideal ratio” in TWI classrooms is one to one (USDE, 2015). Overall, North Carolina’s task force report and department of education website clearly indicate a desire for a statewide expansion of DLI programs, but most of the decisions regarding this expansion are left to the local districts and schools. Data and Methodology My research relied on multiple sources of data to conduct a mixed methods analysis. The research questions that I aim to answer, and my interpretation of the findings are formed by my unique positionality as a researcher. I identify as White women who was raised in an upper middle-class environment. I am a monolingual English speaker who attended schools with few designated ELLs students and no dual language program offerings. However, I did take five years of Spanish world language courses throughout my middle school and high school careers which showed me first-hand the limited nature of language acquisition that comes from this form of foreign language education. As a critical scholar undertaking critical research, I believe it is important to be transparent about my positionality and how it informs my research. I am committed to understanding changes in DLI programs to better understand how equitably these programs are accessed and if they have become dominated by groups who may share characteristics of my own identity. 116 Critical Policy Analysis of North Carolina’s DLI Policy Documents I collected three policy documents to conduct a critical discourse analysis of North Carolina’s DLI program. These included the 2013 task force report titled, Preparing Students for the World: Final Report of the State Board of Education’s Task Force on Global Education, and two documents outlining state educational standards specifically pertaining to DLI programs. The task force report is the primary document guiding DLI for the state and clearly outlines the state’s goal. I included the first standard document titled “World Language Essential Standards Crosswalk: A Document to Assist with the Transition from the 2004 Standard Course of Study to the 2010 Essential Standards” because it includes teaching standards that specifically guide the K-12 DLI programs across the state. The second standards document is titled “Instructional Support Tools for Achieving New Standards.” I included this document as part of the policy analysis since it was released by the State Department of Education for the 2012-2013 school year to act as a guide and teaching tool for schools and teachers with DLI programs as they transitioned to the new standards. After collecting these documents, I conducted a critical policy analysis (CPA) of the written policy language. CPA is an extension of critical discourse analysis (CDA). CDA asserts that language is a vehicle that transmits and reproduces forms of social order and power (Fairclough, 2013; Taylor, 1997). CPA specifically examines the language employed in policy and contends that the existence of a policy, or the absence of a policy, reveals the existence of power structures (Prunty, 1985; Williams et al., 2022; Wright et al., 2020). CPA assumes that no policy is inherently neutral, but instead is written with winners and losers in mind (Williams et al., 2022). CPA is a tool to reveal the members of these groups and identify explicit and covert language that may marginalize specific groups while privileging others. 117 When conducting CPA on DLI policy materials, I followed the process of Valdez et al. (2016) who had completed a similar analysis of the DLI policies in Utah. I utilized the qualitative coding software, Dedoose, for all my coding to identify reoccurring patterns. I completed a first round of analysis to determine the source, intent, and structure of each of the policy documents included in the study to understand which sections pertained to DLI specifically. I then completed the second round of CPA coding using the three hegemonic frameworks globalized human capital, Whiteness as property, and English hegemony as deductive codes. During this round, I read all three policy documents and coded instances in the text that showcased or implied these frameworks, as well as instances where these frameworks silenced a specific group (Thiesmeyer, 2003). Next, I conducted a third round of inductive coding by analyzing excerpts that had been coded as examples of each theoretical frameworks to identify subcodes within these larger categories. This third round of coding revealed the mechanisms the policy documents employed that advanced or challenged these three frameworks. The third round of coding also helped to reveal the intended beneficiaries and the marginalized groups of the TWI policy that were often subtly referenced. For example, under the broader code of globalized human capital, I created the subcodes of globally competitive graduates, globally competitive state, influence of business community, international influence, and most efficient path to bilingualism to further categorize how the globalized human capital framework was actualized in the policy language. Quantitative Analysis of Access to DLI To determine to what extent access to DLI programs in North Carolina changed after the release of the task force’s recommendations, I quantitatively compared student demographics in early adopter schools, schools that launched a DLI program before January 2013, to 118 demographics in late adopter schools, schools that launched a DLI program after January 2013. I conducted two rounds analyses, first comparing the 76 early and late adopting TWI programs, and then following the same steps to compare the 124 early and late adopting OWI programs. I compared the early and late adopters’ demographics across 24 years from the 1997-1998 school year to the 2020-2021 school year to capture the full trajectory of DLI in North Carolina. This analysis allows me to compare these two groups of schools throughout time while controlling for year level variation. North Carolina experienced changes in population with a rise of designated ELL students and an overall decrease in White identifying individuals (USA Facts, 2022). Comparing these groups with school level demographic data for each year allows me to account for these wider population shifts. I drew on North Carolina’s Dual Language Department’s current list of OWI and TWI programs to determine which schools housed these programs. This list also included data on the year the program launched which allowed me to identify early and late adopters. I compared the late and early adopting schools across three demographics that are related to the theoretical frameworks of globalized human capital, Whiteness as property, and English hegemony. These demographics included student income, race/ethnicity, and English language proficiency status. To compare student income and race/ethnicity, I utilized NCES’s school level Common Core Data (CCD) from the 1997-1998 school year to the 2020-2021 school year. I used two available measures as proxies for student income, percent of students eligible for free lunch and percent of students in the school eligible for reduced price lunch. When comparing English proficiency status, I drew on the school level data from the Civil Rights Data Collection (CRDC) for the school years 2010-2011, 2012-2013, 2013-2015, and 2015-2017 since the CCD 119 data did not include data on language proficiency. I use the percent of students labeled “limited English proficiency” as a proxy for the number of designated ELLs in the school. Findings from Critical Policy Analysis Critiques of English Only All three documents consistently highlighted flaws to English monolingualism instruction and ability. English monolingualism was referred to as being “uncompetitive compared to international education leaders” and “inefficient to prepare students for the world today” in the task force recommendation document. These critiques of English monolingualism offered a critique to the English hegemony framework more generally. While they still privileged a person’s ability to speak English, they argued that that was insufficient and that fluency in other languages, in addition to English, was preferred. This revealed a new set of values attached to multilingualism. While historically the United States placed little value in bilingualism or multilingualism and even viewed it as a threat to English fluency, the criticism of English monolingualism in these documents, particularly the task force recommendation, showed that education policy makers in North Carolina share sentiments with DLI policy makers in other states and now see bilingualism or multilingualism as a valued skill. Nonetheless, as the following sections will outline, this critique of English monolingualism was paired with language that reinforced English hegemony, Whiteness, and globalized human capital and neglected to situate multilingualism as a solution to the inequities that exist in education between non-native English speakers and their peers. Whiteness and English-First Students Prioritized There was a consistent pattern in the policy texts, especially those that outlined the new learning standards for dual language programs, that demonstrated the English hegemony and 120 Whiteness as property, particularly normative Whiteness. Normative Whiteness stems from Critical Whiteness and argues that one of the privileges of Whiteness comes from discourses that position those that are White as the default and normal group (Harris, 1993) English speaking community members acted as the reference group in many of the standards demonstrating a bias for the native English-speaking individuals in the DLI classroom. For example, one standard read “the student will compare the students’ culture to the target culture”. The standards make clear that the non-English language is the “target language” and so we can assume the “target culture” is the culture of individuals from the countries who speak the non-English language. This standard only accounts for the English-speaking students’ culture and perspective, whereas those whose first language and cultural identity match the target language are not considered and are left to compare their culture to itself. In addition, the dual language standards utilized the term “familiar” in a way that disregarded students from other backgrounds that may be acquainted with different topics. For example, one standard read “[students will] understand the subtleties and stylistic features of texts on unfamiliar topics,” whereas another says “ [students will] understand the subtleties and stylistic features of texts on familiar topics.” In TWI classroom that the task force said should be comprised of 50% native English-speaking students and 50% non-native English speaking students, these standards position assumed that all students would be familiar with the same topics even though students familiarity with a topic may differ based on differences in lived experiences. While this less overtly marginalizes nonnative English-speaking students compared to the previous examples, this showcased normative Whiteness by situating a certain group of students who have the common definition of familiar and unfamiliar as “normal” while disregarding other students in the classroom who might have different world views. 121 This showcased how ideas of English hegemony and Whiteness are often interconnected. The language used in the learning standards centered the world view of White, English speakers, who were often considered the reference or default group in the United States, by assuming that the students’ own culture must differ from that of the target language. Here this on full display by the assumptions that were made in the word choice of these standards. These standards were not an isolated incidents but three of 43 that use similar language to disregard certain groups of students. Multilingualism to Compete in the Global Market All three policy documents referred to DLI as a means of developing essential skills to prepare students for a global economy. As the North Carolina standards Crosswalk document asserted “in comparison to the rest of the world where multilingualism is the norm for all students, our students are not as globally-ready as they should be.” In response to this problem, the task force report supported expanding DLI throughout the state. Dual language/immersion students master subject content from other disciplines, using the target language or both languages, and become bilingual and biliterate as a result, which supports the North Carolina State Board of Education’s goals for Future-Ready Students for the 21st Century to be competitive in an international economy by being multilingual, knowledgeable global citizens. (Task Force Report) In addition, the task force report specifically highlighted the role of North Carolina business leaders in the push to expand DLI education. Globalization presents significant challenges and enormous opportunities for education in North Carolina, and demands approaches that ensure students are prepared. The Task Force was emboldened particularly by the voices of business leaders who suggested that 122 this agenda is core to their future success and the success of graduates of our public schools who seek to enter the workforce in today’s global economy. The task force also reported the states then-current engagement in the global market as further evidence for the need to expand DLI programs in the state. The task force recommendations for DLI concluded by situating DLI in the larger context of the state’s economic development. “Consistent communication about the competitive advantage and opportunity that an education which includes global knowledge will offer to individuals, communities, and our state is critical.” These references highlighted that the globalized human capital framework was a key motivator in the expansion of DLI programs. The language in the documents clearly linked DLI to bilingualism and expounded on the benefits of this skill on an individual and statewide level. However, this focus on bilingualism for the sake of global competitiveness forsook the potential equity impact of DLI and the historical origin of the policy. It is also worth noting which groups were absent in the task force report as well and other rationales that could have been used but were silenced. Here, there is no mention of a specific target student group who may benefit from this program, such as non-native English speakers who were the original intended beneficiaries of dual language when it was originally implemented in the 1960s. The policy assumed that boosting the global competitiveness of the North Carolina student body as a whole, or the state’s economy broadly, will help all students and North Carolinians. This disregards Valdés (1997) warning that White individuals will reap more benefits when DLI is viewed through a neoliberal lens and neglects to consider DLI as a targeted intervention that could help specific students to promote equity in language education. 123 Silencing Folk bilinguals All three policy sources referred to specific groups as partners or advisors for dual language education. In the task force report, the task force recommended naming “partner countries to serve as the primary source of information about skill requirements and projections, inform development of K-12 curriculum and teacher preparation and professional development, and serve as a high priority source and destination for administrator, principal, and teacher exchanges and visits.” The report also recommended identifying “priority languages” by working with the North Carolina businesses. These excerpts reveal how the international non-English speaking world was valued and seen as a resource for information while they neglected to see the local immigrant and folk bilingual or multilingual communities in North Carolina as a potential source of knowledge. This exemplified globalized human capital by positioning the business community as the entity to determine “priority languages” based on their business interests and connections rather than looking to the local communities to determine which dual language options were most needed to serve their language needs. DLI for All Lastly, the policy documents continuously touted the benefits of DLI for all students. For example, the task force report noted “students in dual language/immersion programs perform better in all subjects, and acquire enhanced and critical ‘non-cognitive’ skills such as creativity, perseverance and lateral thinking.” The unpacked standards document asserted “research shows that students in dual language/immersion programs: Develop high levels of proficiency in the target language and English [and] demonstrate academic performance at or above grade level on statewide standardized tests when compared to monolingual peers.” These examples illustrate 124 how a discourse of all silences concerns for equity and is a stark departure from the DLI models of the 1960s when a primary goal was to maintain ELL students’ heritage languages. This emphasis of DLI being a program suitable and beneficial for all students corresponded to the globalized human capital mindset which disregarded concerns for equity under the assumption that “a rising tide will float all boats” (Gould & Robert, 2013). No specific group was targeted or seen as benefitting more from this program than others and thus DLI was painted as something that all parents and students should try to pursue. However, this general pursuit could lead to inequitable access when society privileges White, wealthy, and English-first individuals who may have the means to monopolize these programs and their benefits. Growth and Changes in Access to DLI programs The North Carolina education system followed the task force recommendations to expand DLI programs and experienced a growth in DLI programs after January 2013. Figures 2 illustrates this growth for two way immersion (TWI) programs and one way immersion (OWI) programs denoting the time when the task force recommendations were released with the solid blue line between 2012 and 2013. 125 Figure 2: North Carolina DLI Program Enrollment Increases Over Time Twelve TWI programs had been established before the task force report with a gradual increase in programs until the fall 2012 school year. In the years after the task force report, the number of TWI programs throughout the state increased nearly six-fold with 68 additional programs launching between fall 2013 and fall 2021, with eight of these programs launching in the year immediately after the task force’s recommendations. The majority of the programs introduced in the years before the task force report taught Spanish as the partner language, with 20 teaching Spanish and one teaching Chinese. This pattern continued after the task force report. Of the 68 programs launched in between fall 2014 and fall 2021, 57 taught Spanish, one taught Chinese, and one taught Urdu. Thirty-nine OWI programs had opened before the release of the task force report. Similar to the TWI, the number of OWI also increased after the policy change with 81 programs opening between the 2013-2014 and 2020-2021 school year, with 11 beginning in the 2013-2014 school year immediately after the report was released. Eleven of these late adopter programs 126 taught Chinese, 1 taught Japanese, 1 taught German, and 1 taught French, and the rest taught Spanish as the target language. The majority of the early adopter OWI programs also taught Spanish, but 1 taught Cherokee, and 1 taught Greek. To answer the second research question (how did access to DLI programs change after the Task Force report was released in 2013?) I compared student demographics in the early adopter schools, those that housed DLI before the task force recommendations were released, to student demographics in late adopter schools, schools that housed DLI after January 2013 when the task force released their recommendations. Similar to Valdez et al’s (2016) Utah analysis, I compared early and late adopter TWI programs first, and then performed the same analysis comparing early and late adopter OWI programs. I compared demographics that were proxies for a student’s income or wealth, White racial privilege, and English language status. I extended upon their Valde et al.’s (2016) work by also comparing the percent of Black, Latinx, and Asian students in late adopter versus early adopter schools. I first compared the percent of students who qualified for free lunch (FL) and the percent of students who qualified for reduced price lunch (RPL) in the early adopter and late adopter TWI schools each year between the 1997-1998 school year until the 2021-2022 school year. I first performed a F test to determine the type of t test to execute to determine if the groups had equal or unequal variances. These results indicated that the null hypothesis asserting that the two groups had equal variances could be rejected at the .1 level. I then proceeded to perform two sample independent t tests assuming unequal variances comparing the percent of FL students and RPL students between the early and late adopter TWI schools for each year. I then performed another two-sample independent t test assuming equal variances and compared early and late adopter schools housing OWI programs. Table 13 summarized the results for TWI schools and 127 Table 14 summarized the results for OWI schools with both including the effect size or the practical significance of the difference. Table Table 13: Percent 13 Percent FreeEligible Free Lunch LunchDifference EligibleinDifference in TWI Early vs. Late Early vs. Late Adopting Adopting TWI Schools Schools Early Adopter TWI schools Late Adopter TWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 - no observations 1998-1999 10 0.43 0.20 49 0.41 0.19 0.03 .713 -0.13 1999-2000 10 0.46 0.19 51 0.41 0.18 0.05 .503 -0.25 2000-2001 9 0.48 0.22 51 0.41 0.18 0.07 .421 -0.35 2001-2002 10 0.48 0.19 52 0.43 0.19 0.05 .437 -0.28 2002-2003 10 0.27 0.36 52 0.42 0.21 -0.15 .235 0.62 2003-2004 11 0.47 0.27 52 0.48 0.18 -0.01 .882 0.07 2004-2005 12 0.51 0.29 52 0.49 0.20 0.03 .77 -0.12 2005-2006 12 0.51 0.30 54 0.43 0.26 0.08 .402 -0.30 2006-2007 12 0.45 0.26 55 0.50 0.16 -0.05 .52 0.28 2007-2008 12 0.34 0.26 55 0.35 0.25 -0.02 .852 0.06 2008-2009 12 0.36 0.28 55 0.43 0.23 -0.07 .434 0.29 2009-2010 12 0.54 0.24 55 0.58 0.17 -0.04 .621 0.20 2010-2011 12 0.54 0.26 58 0.62 0.18 -0.08 .345 0.39 2011-2012 13 0.60 0.26 59 0.63 0.17 -0.03 .717 0.15 2012-2013 13 0.61 0.24 60 0.65 0.17 -0.04 .54 0.24 2013-2014 13 0.61 0.24 60 0.64 0.16 -0.03 .664 0.17 2014-2015 13 0.63 0.29 60 0.68 0.21 -0.05 .541 0.24 2015-2016 13 0.66 0.31 60 0.69 0.21 -0.03 .735 0.13 2016-2017 13 0.67 0.33 60 0.71 0.22 -0.04 .689 0.16 2017-2018 13 0.67 0.35 60 0.72 0.26 -0.04 .682 0.16 2018-2019 13 0.66 0.34 62 0.73 0.25 -0.07 .468 0.28 2019-2020 13 0.61 0.34 63 0.72 0.26 -0.11 .273 0.41 2020-2021 12 0.60 0.32 62 0.68 0.28 -0.08 0.41 0.29 The findings in Table 13 revealed that there was little difference between the percent of students eligible for free lunch services between the 1998-1999 school year and the 2020-2021 school year for the late and early adopter TWI schools. In six of the school years, all before the release of the task force report, the early adopter TWI schools enrolled slightly more free lunch students than the late adopter school, but this difference was never significant. In the other 17 years included in this study, the late adopter TWI schools enrolled a greater percentage of free lunch students, and in all of the years after the release of the task force report (2013-2014 to 2020-2021) the late adopter TWI schools enrolled a greater percentage of free lunch students 128 compared to the early adopter TWI schools. Nevertheless, these differences were never statistically significant. The results comparing the percent of students eligible for reduced lunch in TWI schools services indicate a similar conclusion illustrating little difference between the late and early adopter schools. Table Table 14: Percent 14 Percent FreeEligible Free Lunch LunchDifference Eligible in Difference in Early Early vs. Late vs. LateSchools OWI Adopting Adopting TWI Schools Early Adopter OWI schools Late Adopter OWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 - no observations 1998-1999 23 0.38 0.22 67 0.28 0.17 0.10 .03 -0.53 1999-2000 23 0.37 0.23 70 0.29 0.18 0.08 .107 -0.39 2000-2001 21 0.40 0.24 61 0.29 0.19 0.12 .028 -0.57 2001-2002 24 0.36 0.23 76 0.27 0.18 0.08 .058 -0.45 2002-2003 25 0.40 0.30 79 0.23 0.20 0.17 .002 -0.73 2003-2004 25 0.42 0.27 79 0.33 0.21 0.09 .072 -0.42 2004-2005 25 0.43 0.27 82 0.33 0.19 0.10 .049 -0.45 2005-2006 25 0.42 0.27 85 0.30 0.19 0.12 .016 -0.56 2006-2007 25 0.40 0.27 89 0.30 0.22 0.10 .071 -0.41 2007-2008 26 0.30 0.35 90 0.19 0.23 0.11 .062 -0.42 2008-2009 26 0.39 0.29 91 0.23 0.25 0.16 .005 -0.64 2009-2010 26 0.46 0.27 93 0.35 0.22 0.11 .04 -0.46 2010-2011 26 0.46 0.26 94 0.38 0.22 0.08 .097 -0.37 2011-2012 26 0.49 0.26 94 0.39 0.22 0.10 .059 -0.42 2012-2013 26 0.50 0.26 94 0.41 0.23 0.08 .107 -0.36 2013-2014 26 0.50 0.27 95 0.42 0.24 0.09 .108 -0.36 2014-2015 26 0.55 0.33 96 0.47 0.30 0.09 .189 -0.29 2015-2016 26 0.56 0.33 96 0.48 0.32 0.08 .257 -0.25 2016-2017 26 0.57 0.34 95 0.47 0.31 0.10 .156 -0.32 2017-2018 26 0.57 0.35 97 0.46 0.32 0.11 .119 -0.35 2018-2019 26 0.57 0.34 98 0.47 0.33 0.10 .154 -0.32 2019-2020 26 0.58 0.36 98 0.46 0.34 0.12 .129 -0.34 2020-2021 26 0.56 0.37 95 0.46 0.35 0.11 0.17 -0.31 When comparing the percent of students eligible for free lunch in early and late adopting OWI schools, there was no significant difference between late and early adopter schools in the years after the policy change. However, in seven of the fifteen years pre-policy, late adopter schools had a significantly smaller percent of free lunch eligible students compared to early adopting schools. These late adopter schools attracted students and families from a higher socioeconomic group after they became OWI schools. Nevertheless, the results comparing the 129 percent of students eligible for reduced price lunch in early and later adopting OWI schools revealed no significant different between early and late adopting schools throughout the 22 of the 23 years included in this study. TableELL Table 15 Percent 15: Difference Percent ELL Difference in Early in Early vs. Late TWI vs. Schools Adopting Late TWI Adopting Schools Early Adopter TWI schools Late Adopter TWI schools T test Year (fall semester) obs mean sd obs mean sd difference p value effect size 2011 12 0.26 0.19 58 0.20 0.11 0.05 .349 -0.43 2013 13 0.20 0.12 60 0.18 0.11 0.02 .554 -0.19 2015 13 0.17 0.08 60 0.17 0.11 0.01 .847 -0.05 2017 13 0.18 0.07 60 0.17 0.11 0.01 0.70 -0.09 TableELL Table 16 Percent 16:Difference Percent ELL Difference in Early in Early vs. Late OWI vs.Schools Adopting Late OWI Adopting Schools Pre Policy OWI schools Post Policy OWI schools T test Year (fall semester) obs mean sd obs mean sd difference p value effect size 2011 25 0.11 0.14 94 0.08 0.08 0.04 .195 -0.40 2013 26 0.11 0.14 94 0.06 0.07 0.04 .144 -0.47 2015 26 0.11 0.13 96 0.06 0.06 0.05 .103 -0.54 2017 26 0.11 0.12 97 0.07 0.07 0.04 0.08 -0.54 Second, I ran two analyses and first compared the percent of designated ELL students in early adopter TWI and late adopter TWI schools and then compared the early and later adopter OWI schools. Data were not available on this measure in the Common Core Data (CCD) files utilized to compare student income and racial/ethnicity, so I used data from the Civil Rights Data Collection. This data was only available for years 2011, 2013, 2015, and 2017 with 2011 representing the pre-task force period and 2013, 2015, and 2017 representing the post-task force period. I conducted a two-sample independent t test assuming unequal. The results from the t test for TWI schools are presented in Table 15 and for OWI schools are presented in Table 16. The findings indicated that there was no significant change in the percent of designated ELL students in early adopter and later adopter schools TWI or OWI schools before and after the task force report was released. 130 Lastly, I compared the percent of White, Black, Asian, and Latinx students in the early adopter and later adopter TWI schools from the 1997-1998 school year to the 2020-2021 school year. Again, I conducted four F tests to test the null hypothesis of equal variances to determine the type of t test that was suitable for these two groups. The F tests for each racial/ethnic group resulted in F statistics that were less than .05 so I rejected the null hypothesis of equal variances. I then ran two sample independent t tests with unequal variances for comparing the percent of students identifying with each of the racial/ethnic groups in the late adopter and early adopter schools. I followed this same process for to compare the percent of White, Black, Asian, and Latinx students in early and late adopter OWI schools. TableWhite Table 17 Percent 17: Percent White Difference Difference in Early vs. Late TWIin Adopting Early vs.Schools Late TWI Adopting Schools Early Adopter TWI schools Late Adopter TWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 10 0.46 0.16 49 0.59 0.23 -0.12 .115 0.56 1998-1999 10 0.44 0.16 49 0.58 0.23 -0.14 .07 0.64 1999-2000 10 0.40 0.17 51 0.55 0.24 -0.15 .059 0.67 2000-2001 10 0.36 0.17 52 0.53 0.24 -0.17 .035 0.75 2001-2002 10 0.34 0.17 52 0.52 0.24 -0.17 .032 0.76 2002-2003 10 0.33 0.18 52 0.50 0.23 -0.16 .039 0.73 2003-2004 11 0.36 0.21 52 0.48 0.23 -0.13 .088 0.57 2004-2005 12 0.32 0.20 52 0.47 0.22 -0.16 .032 0.70 2005-2006 12 0.31 0.20 54 0.46 0.23 -0.15 .044 0.66 2006-2007 12 0.31 0.20 55 0.46 0.23 -0.15 .046 0.65 2007-2008 12 0.30 0.19 55 0.43 0.22 -0.13 .056 0.62 2008-2009 12 0.28 0.17 55 0.42 0.22 -0.14 .035 0.69 2009-2010 12 0.27 0.17 55 0.41 0.22 -0.14 .043 0.66 2010-2011 12 0.27 0.17 58 0.43 0.24 -0.16 .03 0.70 2011-2012 13 0.32 0.25 59 0.42 0.22 -0.10 .153 0.44 2012-2013 13 0.27 0.16 60 0.41 0.22 -0.14 .038 0.65 2013-2014 13 0.26 0.16 60 0.39 0.20 -0.12 .046 0.62 2014-2015 13 0.26 0.16 60 0.38 0.20 -0.12 .054 0.60 2015-2016 13 0.25 0.15 60 0.37 0.20 -0.12 .047 0.62 2016-2017 13 0.25 0.15 60 0.37 0.20 -0.12 .047 0.62 2017-2018 13 0.25 0.15 60 0.35 0.20 -0.11 .07 0.56 2018-2019 13 0.24 0.15 62 0.35 0.19 -0.11 .061 0.58 2019-2020 13 0.23 0.14 63 0.34 0.20 -0.11 .063 0.58 2020-2021 13 0.23 0.14 63 0.33 0.19 -0.09 0.11 0.50 131 Table 17 illustrates the difference in the percent White/Non Latinx identifying students in the early adopter and late adopter TWI schools. In all but one school year between 1997-1998 and 2020-2021, late adopter schools had a significantly greater percentage of White students than the early adopter schools at a .05 significance level with medium effect sizes. In the years before the task force report, late adopter schools had an average of 14 percentage points more White students than the early adopter schools. In the years after the task force report was released, late adopter schools had an average of 11 percentage points more White students than the early adopter schools. Late adopter TWI school were consistently Whiter than early adopter TWI schools throughout this time period. I also examined the difference in the percentage of Black, Latinx, and Asian students in early adopter TWI schools versus late adopter TWI schools in Tables 21, 22, and 23 respectively (see appendix). Early adopter schools had a greater percentage of Black, Latinx, and Asian students than late adopter schools throughout the 24-year period included in the study, but the differences for these groups were never significant. 132 Table Table 18 18:White Percent Percent WhiteinDifference Difference Early vs. LateinOWI Early vs. Late Adopting OWI Schools Adopting Schools Early Adopter OWI schools Late Adopter OWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 23 0.54 0.23 67 0.71 0.21 -0.17 .001 0.80 1998-1999 23 0.53 0.24 69 0.70 0.21 -0.17 .002 0.78 1999-2000 23 0.51 0.25 72 0.69 0.22 -0.18 .002 0.78 2000-2001 23 0.49 0.26 73 0.68 0.22 -0.19 .001 0.82 2001-2002 24 0.49 0.28 76 0.67 0.22 -0.18 .002 0.74 2002-2003 25 0.50 0.28 79 0.66 0.22 -0.16 .005 0.66 2003-2004 25 0.49 0.28 79 0.65 0.23 -0.15 .006 0.64 2004-2005 25 0.48 0.28 82 0.64 0.22 -0.17 .003 0.70 2005-2006 25 0.47 0.28 85 0.64 0.22 -0.18 .001 0.75 2006-2007 25 0.45 0.28 89 0.65 0.23 -0.20 .000 0.82 2007-2008 26 0.46 0.29 90 0.64 0.23 -0.18 .001 0.72 2008-2009 26 0.44 0.27 91 0.62 0.23 -0.19 .001 0.76 2009-2010 26 0.43 0.27 93 0.62 0.23 -0.19 .001 0.79 2010-2011 26 0.43 0.26 94 0.61 0.23 -0.18 .001 0.75 2011-2012 26 0.42 0.26 94 0.59 0.23 -0.18 .001 0.75 2012-2013 26 0.42 0.26 94 0.59 0.23 -0.17 .001 0.72 2013-2014 26 0.41 0.26 95 0.58 0.24 -0.17 .002 0.70 2014-2015 26 0.41 0.26 96 0.56 0.24 -0.15 .005 0.63 2015-2016 26 0.40 0.25 96 0.55 0.24 -0.16 .003 0.66 2016-2017 26 0.40 0.25 96 0.55 0.24 -0.16 .003 0.66 2017-2018 26 0.38 0.24 97 0.54 0.23 -0.15 .004 0.65 2018-2019 26 0.38 0.24 98 0.52 0.23 -0.14 .006 0.61 2019-2020 26 0.37 0.23 98 0.50 0.23 -0.14 .008 0.60 2020-2021 26 0.36 0.23 98 0.50 0.23 -0.13 0.01 0.58 When comparing the demographics for OWI schools, the difference between early and late adopter schools were more striking than in TWI schools. First, similar to findings for TWI schools, late adopter OWI had a significantly greater percentage of White students (p < .05) compared to early adopter OWI schools with medium effect sizes. Whereas late adopter TWI schools had an average of 11 percentage points more White students in the years after the task force and an average of 14 percentage points more White students in the years before the task force, late adopter OWI schools had an average of 15 percentage points more White students in the years after the task force compared to early adopter OWI schools and an average of 18 percentage points more White students in the years before the task force. The late adopter OWI schools have consistently been Whiter than the early adopter OWI schools. 133 Table Table 19: Percent 19 Percent Black Difference Black Difference in Early vs.in Early Late OWIvs.Adopting Late OWI Adopting Schools Schools Early Adopter OWI schools Late Adopter OWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 23 0.39 0.21 67 0.24 0.19 0.14 .003 -0.73 1998-1999 23 0.39 0.22 69 0.25 0.20 0.15 .004 -0.72 1999-2000 23 0.40 0.23 72 0.25 0.20 0.15 .003 -0.74 2000-2001 23 0.41 0.24 73 0.25 0.20 0.16 .001 -0.79 2001-2002 24 0.39 0.25 76 0.25 0.19 0.14 .004 -0.69 2002-2003 25 0.37 0.25 79 0.25 0.19 0.12 .013 -0.58 2003-2004 25 0.37 0.24 79 0.25 0.19 0.11 .018 -0.55 2004-2005 25 0.36 0.23 82 0.25 0.18 0.12 .011 -0.59 2005-2006 25 0.36 0.23 85 0.24 0.18 0.11 .01 -0.59 2006-2007 25 0.31 0.22 89 0.24 0.22 0.07 .147 -0.33 2007-2008 26 0.34 0.25 90 0.21 0.19 0.12 .01 -0.58 2008-2009 26 0.36 0.23 91 0.25 0.20 0.11 .021 -0.52 2009-2010 26 0.35 0.22 93 0.25 0.19 0.10 .026 -0.50 2010-2011 26 0.29 0.21 94 0.20 0.17 0.09 .021 -0.52 2011-2012 26 0.29 0.21 94 0.20 0.18 0.09 .025 -0.50 2012-2013 26 0.29 0.21 94 0.20 0.17 0.09 .025 -0.50 2013-2014 26 0.28 0.20 95 0.21 0.20 0.07 .105 -0.36 2014-2015 26 0.27 0.19 96 0.21 0.19 0.06 .162 -0.31 2015-2016 26 0.28 0.19 96 0.21 0.19 0.06 .146 -0.32 2016-2017 26 0.28 0.19 96 0.21 0.19 0.06 .146 -0.32 2017-2018 26 0.28 0.19 97 0.21 0.18 0.07 .106 -0.36 2018-2019 26 0.28 0.19 98 0.22 0.19 0.06 .134 -0.33 2019-2020 26 0.28 0.19 98 0.21 0.19 0.06 .129 -0.34 2020-2021 26 0.27 0.18 98 0.22 0.19 0.05 0.20 -0.29 Next, while early and late adopter TWI schools did not serve significantly different percentages of Black and Latinx students, there were significant differences in the percentage of Black and Latinx students in early and late adopter OWI schools. Table 19 shows that late adopter schools had a significantly (p < .05) smaller proportion of Black students compared to early adopter schools in the years before the task force report was released. However, in the years since the policy change, there was no significant difference between the percentage of Blacks students in early and late adopter OWI schools with the early OWI schools percentage of Black students steadily decreasing. For Latinx students, the opposite pattern emerges. In the years before the task force report, the difference in the percentage of Latinx students served in early and late adopter schools was insignificant (p < .05). Nevertheless, in the years after the 134 task force report was released, early adopter OWI schools served a higher percentage of Latinx students compared to the late adopter schools. When comparing the percent of Asian students in early and late adopter OWI schools, I found no significant difference between early and later adopter OWI schools (see appendix). TableTable 20: Percent 20 Percent LatinxinDifference Latinx Difference inOWI Early vs. Late EarlyAdopting vs. Late OWI Adopting Schools Schools Early Adopter OWI schools Late Adopter OWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 23 0.05 0.04 67 0.03 0.04 0.02 .133 -0.37 1998-1999 23 0.05 0.04 69 0.03 0.05 0.02 .099 -0.40 1999-2000 23 0.06 0.05 72 0.04 0.05 0.02 .152 -0.35 2000-2001 23 0.07 0.06 73 0.05 0.07 0.02 .184 -0.32 2001-2002 24 0.09 0.09 76 0.06 0.07 0.03 .123 -0.36 2002-2003 25 0.10 0.10 79 0.07 0.07 0.03 .099 -0.38 2003-2004 25 0.11 0.11 79 0.07 0.08 0.04 .079 -0.41 2004-2005 25 0.13 0.14 82 0.08 0.09 0.05 .048 -0.46 2005-2006 25 0.15 0.16 85 0.09 0.09 0.06 .028 -0.51 2006-2007 25 0.16 0.18 89 0.13 0.19 0.03 .454 -0.17 2007-2008 26 0.20 0.25 90 0.11 0.14 0.09 .021 -0.52 2008-2009 26 0.17 0.19 91 0.12 0.14 0.05 .14 -0.33 2009-2010 26 0.18 0.20 93 0.12 0.14 0.06 .089 -0.38 2010-2011 26 0.19 0.18 94 0.13 0.11 0.06 .032 -0.48 2011-2012 26 0.20 0.18 94 0.14 0.11 0.07 .022 -0.52 2012-2013 26 0.21 0.18 94 0.14 0.12 0.07 .024 -0.51 2013-2014 26 0.22 0.18 95 0.16 0.15 0.06 .084 -0.39 2014-2015 26 0.23 0.18 96 0.15 0.12 0.08 .013 -0.56 2015-2016 26 0.24 0.18 96 0.16 0.12 0.08 .006 -0.61 2016-2017 26 0.24 0.18 96 0.16 0.12 0.08 .006 -0.61 2017-2018 26 0.25 0.16 97 0.17 0.13 0.07 .014 -0.55 2018-2019 26 0.25 0.16 98 0.18 0.12 0.07 .015 -0.54 2019-2020 26 0.26 0.16 98 0.18 0.12 0.07 .013 -0.56 2020-2021 26 0.26 0.16 98 0.19 0.12 0.07 0.01 -0.57 These results may appear initially deceiving because there is a steady downward trend in the percent of White and Black students in early and late TWI adopter schools and a steady increase in Latinx and Asian students during this time. Nevertheless, the increase in Latinx and Asian students and the decrease in Black students was comparable between the late adopter and early adopter TWI schools and followed the general trajectories of the state population. However, as Figures 3 and 4 illustrate, the downward trend of the percentage of White students was steeper for the early adopter TWI and OWI schools than the later adopter TWI and OWI 135 schools indicating that the late adopter schools were retaining their White populations at a higher rate between the 1997-1998 and 2020-2021 school years. While both early and later adopter schools were teaching a lower percentage of White students in 1997-1998 compared to 2020- 2021, this population shift was slower for later adopter schools. For the late adopter OWI schools, Figure 5 illustrates the trajectories of the Black students indicate that there was little change in the percent of Black students served at late adopter OWI schools throughout the 24 years. The early adopter OWI schools experienced a steady decline in their percentage of Black students and thus began to resemble the lower percentage of Black students enrolled in late adopting OWI schools. Regarding the percent of Latinx students, late adopter OWI have been more resistant to increasing the Latinx student population over these 24 years compared to early adopter OWI schools. The resistance of the late adopter’s White student populations resulted in fewer slots available to Latinx students, even though the Latinx population in the state has more than doubled between 2000 and 2021 (USA Facts, 2022) Comparing the trajectories of these student populations demonstrates that it was imperative that the greater North Carolina context and population change was taken into account when interpreting these findings. During the course of this 24-year time span, North Carolina experienced an influx of Latinx and Asian individuals and this was reflected in the schools. However, those schools most resistant to the changes in the states’ decline of White residents were the later adopter TWI and OWI schools. Discussion Findings from the CPA analysis suggest that the expansion of TWI policies in North Carolina in 2013 use language that privileges White, wealthy, and native English-speaking students while marginalizing others under the pursuit of global capital competitiveness. 136 Bilingualism or multilingualism accessed through DLI is viewed as an elite resource that is targeted towards certain groups, while folk bilingual individuals in North Carolina are omitted from the conversation, and thus demoted. The demographic findings also reveal that DLI programs in North Carolina are not accessible to all students equitably. When comparing student demographics in schools that adopted TWI or OWI programs before the task force report (i.e. early adopters) to schools that adopted TWI or OWI programs after the task force report was released (i.e. late adopters), inequities in access are revealed. Early and later adopter schools are similar to one another in terms of the percent of students eligible for free lunch and reduced-price lunch services and the percent of students with limited English proficiency status. Therefore, even though language used in the policy documents silences folk bilingual and multilingual students, the schools who adopted the programs after the policy change did not reduce access for designated ELL and low- income students to DLI programs. However, early and late adopting TWI and OWI schools do significantly differ when looking at the race/ethnicity of students. Late adopter TWI and OWI are significantly Whiter than early adopter schools throughout the 24-year time period of this study. This shift, in addition to the language employed in the policy documents that reimagines DLI standards and recommends expanding DLI programs in the state, shows that DLI programs in general are appealing to White individuals, and that they may be capitalizing on the rationale used in the policy language to secure their place in DLI schools. These findings may speak to ideals of interest convergence theory. Interest convergence theory was first introduced by Bell in 1980 when he claimed that “the interest of blacks in achieving racial equality will be accommodated only when it converges with the interests of whites.” (Bell, 1980, p. 523). For 137 interests to converge, it is often assumed that the dominant group must give something up (Lopez, 2003; Milner, 2008). Interest convergence theory acknowledges that there is a fear that this change will threaten the dominant group’s status, specifically the power, privilege, and value of Whiteness as property, and have lasting effects for future generations (Bell, 1980; Ladson- Billings & Tate, 1995). In the context of DLI, however, White individuals may not have this fear because they are not giving something up when they enroll in or advocate for DLI policies, and in fact gain a skill which increases their status. While research clearly indicates that TWI is the best method for designated ELLs student to become proficient in English, research also touts the academic benefit for all students and states that it is the best way for English monolingual students to become bilingual, and thus globally competitive. This advocacy for DLI in North Carolina is an example of a converging of interests where what is best for designated ELLs is also sought after by White individuals. However, the power of White individuals when they use their Whiteness as property suggests that White individuals may be able to take advantage of this aligning of interests at a greater rate than the designated ELL population. Harris’ (1993) conception of Whiteness as property defines property as the right of “possession, use, and disposition. . . the rights to enjoyment, and the right to exclude others” (p. 1731) White individuals hold a privileged and competitive position that allows them to exploit policies that support bilingualism and multilingualism now that policy makers and businesses have deemed these as competitive and valued skills. This study is not without its limitations. First, the use of school level data restricts the ability to know who exactly is taking part in the OWI and TWI programs within the school. Research shows that internal segregation can occur in DLI schools (Friere & Aléman Jr., 2021), 138 but using school level data limits my ability to capture this in North Carolina. However, student level data capturing whether a student is enrolled in a DLI program within a school was not gathered and thus not available for a more fine-grained analysis. Similar analyses of Utah’s dual language programs have also had to rely on school level data because of a lack of classroom and student level data collection (Valdez et al., 2016). Conclusion The findings from this study indicate that DLI in North Carolina has the potential to become a vehicle for education inequity based on the language used in documents driving the DLI policy discussions in the state, but that inequity has not materialized across some demographics. Furthermore, when comparing these results to a similar study on Utah’s Dual Language policy and access, North Carolina appears to have a more equitable system based on enrollment despite both systems relying on policies that use language that suggests English hegemony, Whiteness as property, and globalized human capital. The findings in Utah reveal similar patterns from a critical discourse analysis of policy documents highlighting themes of normative Whiteness, English hegemony, and globalized human capital. However, all three of these discourses then materialized when comparing schools who housed a dual language program before and after the state’s policy change (Valdez et al., 2016). In Utah, researchers find that schools housing dual language programs established after the policy change serve a higher proportion of White, English-first, and higher income students. North Carolina can learn from other states like Utah to avoid falling into their inequitable practices where the inequities in the policy language related to English hegemony and globalized human capital develop into inequitable access to DLI based on student income and English language status. 139 Nevertheless, North Carolina can improve on their own policy to promote equity in DLI. First, North Carolina policy makers can consult the folk bilingual and multilingual populations in the state for guidance on ways to further expand DLI programs to meet their needs. These conversations could help advocate for DLI programs in languages beyond the current varieties that are offered to help groups honor other heritage languages. Additionally, North Carolina could look to states such as Texas that have guidelines in place requiring bilingual education when a certain number or percentage of designated ELL students is present in the school to ensure that these students do not become inequitably served if DLI programs continue to expand in the future. 140 REFERENCES Beck, S A. & Allexsaht-Snider, M. (2002) Recent language minority education policy in Georgia: Appropriation, assimilation, and Americanization." Department of Middle Grades and Secondary Education Faculty Presentations. Presentation 310. https://digitalcommons.georgiasouthern.edu/teach-secondary-facpres/310 Bell Jr, D. A. (1980). Brown v. Board of Education and the interest-convergence dilemma. Harvard law review, 518-533. Burns, M. (2017). ‘‘Compromises that we make’’: Whiteness in the dual-language context. Bilingual Research Journal, 40(4), 339–352. https://doi.org/10.1080/152358 82.2017.1388303 Cervantes-Soon, C. G. (2014). A critical look at dual language immersion in the new Latin@ diaspora. Bilingual Research Journal, 37(1), 64-82. Cervantes-Soon, C. G., Degollado, E. D., & Nuñez, I. (2020). The Black and brown search for agency: African American and Latinx children’s plight to bilingualism in a two-way dual language program. In Bilingualism for All?: Raciolinguistic Perspectives on Dual Language Education in the United States (pp. 199-219). Channel View Publications. Chavez-Moreno, L. C. (2021) Dual language as White property: Examining a secondary bilingual-education program and Latinx equity. American Educational Research Journal, 58(6), 1107-1141. Cuadros, P. (2011). We play too: Latina integration through soccer in the “New South.” Southeastern Geographer, 51(2), 227–241. de Jong, E. J. (2016). Two-way immersion for the next generation: Models, policies, and principles. International Multilingual Research Journal, 10(1), 6-16. Delavan, M. G., Freire, J. A., & Valdez, V. E. (2021). The intersectionality of neoliberal classing with raciolinguistic marginalization in state dual language policy: A call for locally crafted programs. Bilingualism for All? Raciolinguistic Perspectives on Dual Language Education, 19-39. Delavan, M. G., Valdez, V. E., & Freire, J. A. (2017). The marketing of dual language education policy in Utah print media. Educational Policy, 30(6), 849–883. Ennis, S. R., Rios-Vargas, M., & Albert, N. G. (2011). The Hispanic population. US Census Bureau, US Department of Commerce Economics and Statistics Administration. Fairclough, N. (2006). Language and globalization. New York, NY: Routledge. Fairclough, N. (2013). Critical discourse analysis: The critical study of language. Routledge. 141 Flores, N. (2016). A tale of two visions: Hegemonic whiteness and bilingual education. Educational Policy, 30(1) 13–38. https://doi.org/10.1177/0895904815616482 Flores, N., & Beardsmore, H. B. (2015). Programs and structures in bilingual and multilingual education. The handbook of bilingual and multilingual education, 203-222. Flores, N., & García, O. (2017). A critical review of bilingual education in the United States: From basements and pride to boutiques and profit. Annual Review of Applied Linguistics, 37, 14-29. Freire, J. A., & Alemán Jr, E. (2021). “Two schools within a school”: Elitism, divisiveness, and intra-racial gentrification in a dual language strand. Bilingual Research Journal, 44(2), 249-269. Freire, J. A., Gambrell, J., Kasun, G. S., Dorner, L. M., & Cervantes-Soon, C. (2022). The expropriation of dual language bilingual education: Deconstructing neoliberalism, whitestreaming, and English-hegemony. International Multilingual Research Journal, 16(1), 27-46. Freire, J. A., Valdez, V. E., & Delavan, M. G. (2017). The (dis)inclusion of Latina/o interests from Utah’s dual-language education boom. Journal of Latinos and Education, 16(4), 276–289. https://doi.org/10.1080/15348431.2016.1229617 Giacchino-Baker, R., & Piller, B. (2006). Parental motivation, attitudes, support, and commitment in a Southern Californian two-way immersion program. Journal of Latinos and Education, 5(1), 5-28. Gould, A. M., & Robert, M. (2013). The neoliberal pea and thimble trick: Changing rhetoric of neoliberal champions across two periods of economic history and two hypotheses about why the message is less sanguine. Advances in Applied Sociology, 3(01),79-84. Harris, C. (1993). Whiteness as property. Harvard Law Review, 106, 1707–1791. Heimann, D., & Yanes, M. (2018). Centering the fourth pillar in times of TWBE gentrification: ‘‘Spanish, love, content, not in that order.’’ International Multilingual Research Journal, 12, 173–187. Howard, E.R., and J. Sugarman. 2007. Realizing the vision of two-way immersion: Fostering effective programs and classrooms. Washington, DC: Center for Applied Linguistics. Howard, E. R., Sugarman, J., & Christian, D. (2003). Trends in two-way immersion education. A review of the research. Center for Research on the Education of Students Placed At Risk. Kelly, L. B. (2018). Interest convergence and hegemony in dual language: Bilingual education, but for whom and why?. Language Policy, 17(1), 1-21. 142 Ladson-Billings, G. (2004). New directions in multicultural education. Handbook of Research on Multicultural Education, 2, 50-65. Lindholm-Leary, K.J. 2001. Dual language education. Toronto, ON: Multilingual Matters. Lindholm-Leary, K., & Block, N. (2010). Achievement in predominately low SES/Hispanic dual language schools. International Journal of Bilingual Education and Bilingualism, 13(1), 43–60. https://doi.org/10.1080/13670050902777546 Lindholm-Leary, K., & Genesee, F. (2014). Student outcomes in one-way, two-way, and indigenous language immersion education. Journal of Immersion and Content-Based Language Education, 2(2), 165-180. Lindholm-Leary, K. J., & Howard, E. R. (2008). Language development and academic achievement in two-way immersion programs. In T. W. Fortune & D. J. Tedick (Eds.), Pathways to multilingualism: Evolving perspectives on immersion education (pp. 177– 200). Oxford: Blackwell. Lopez, G. R. (2003). The (racially neutral) politics of education: A critical race theory perspective. Educational Administration Quarterly, 39(1), 68-94. Macedo, D., Dendrinos, B., & Gounari, P. (2003). The hegemony of English. Boulder, CO: Paradigm Publishers. Milner IV, H. R. (2008). Critical race theory and interest convergence as analytic tools in teacher education policies and practices. Journal of teacher education, 59(4), 332-346. North Carolina Department of Public Instruction (2013) Preparing students for the world: Final report of the state board of education’s task force on global education. North Carolina Department of Public Instruction Dual Language Immersion. Retrieved from https://www.dpi.nc.gov/documents/globaled/preparing-students-world-final-report-state- board-educations-task-force-global/download North Carolina Department of Public Instruction (2016a) Instructional support tools for achieving new standards. North Carolina Department of Public Instruction Dual Language Immersion, Standards. Retrieved from https://www.dpi.nc.gov/documents/cte/curriculum/languagearts/scos/unpacking/world- lang/unpacking-document-world-language-0 North Carolina Department of Public Instruction (2016b) World language sssential Standards crosswalk: A document to assist with the transition from the 2004 standard course of study. North Carolina Department of Public Instruction Dual Language Immersion, Standards. 143 North Carolina Department of Public Instruction (2022) Dual Language Immersion. North Carolina Department of Public Instruction, Programs and Initiatives. Retrieved from https://www.dpi.nc.gov/districts-schools/classroom-resources/academic- standards/programs-and-initiatives/dual-language-immersion Palmer, D. (2010). Race, power and equity in a multiethnic urban elementary school with a dual- language “strand” program. Anthropology and Education Quarterly, 41(1), 94-114. https://doi.org/10.1111/j.1548-1492.2010.01069.x Palmer, D. K., Cervantes-Soon, C., & Dornier-Heiman, D. (2020). Bilingualism, biliteracy, biculturalism, and critical consciousness for all: Proposing a fourth fundamental goal for two-way dual language education. In Dual Language Education in the US (pp. 34-50). Routledge. Park, M., O’Toole, A., & Katsiaficas, C. (2017) Dual language learners: A national demographic and policy profile. Migration Policy Institute. Retrieve from https://www.migrationpolicy.org/research/dual-language-learners-national-demographic- and-policy-profile Parkes, J. (2008) Who chooses dual language education for their children and why. International Journal of Bilingual Education and Bilingualism, 11(6), 635-660. Peck, J., Theodore, N., & Brenner, N. (2009). Neoliberal Urbanism. The SAIS Review of International Affairs, 29(1), 49-66. Petrovic, J. E. (2005). The conservative restoration and neoliberal defenses of bilingual education. Language Policy, 4, 395–416. doi:10.1007/s10993-00502880y. Prunty, J. (1985). Critical signposts for a critical educational policy analysis. Australian Journal of Education, 29, 133-140 Ramos, F. (2007). Opinions of Students Enrolled in an Andalusian Bilingual Program on Bilingualism and the Program Itself. Revista Electrónica de Investigación Educativa, 9(2), 1-15. Ricento, T. (2005). Problems with the ‘language‐as‐resource’ discourse in the promotion of heritage languages in the USA. Journal of Sociolinguistics, 9(3), 348-368. Ruiz, R. (1984). Orientations in language planning. The Journal for the Association for Bilingual Education, 8(2), 15–34. Shannon, S. M. (1995). The hegemony of English: A case study of one bilingual classroom as a site of resistance. Linguistics and education, 7(3), 175-200. Shannon, S.M., and M. Milian. 2002. Parents choose dual language programs in Colorado: A survey. Bilingual Research Journal, 26 (3), 681-696. 144 Snyder, R. (2020). The right to define: analyzing whiteness as a form of property in Washington state bilingual education law. Language policy, 19(1), 31-60. Taylor, S. (1997). Critical policy analysis: Exploring contexts, texts and consequences. Discourse: Studies in the cultural politics of education, 18(1), 23-35. Thiesmeyer, L. (2003). Discourse and silencing. John Benjamins. Thomas, W. P., & Collier, V. P. (2002). A national study of school effectiveness for language minority students’ long-term academic achievement. Santa Cruz, CA: Center for Research on Education Diversity and Excellence, University of California-Santa Cruz. Retrieved at http://repositories.cdlib.org/crede/finalrpts/1_1_final. Thomas, W., Collier, V., & Collier, K. 2011. English learners in North Carolina, 2010. Fairfax, VA: George Mason University. Retrieved from http://esl.ncwiseowl.org/resources/dual_language/ Trujillo, A. (2005). Politics, school philosophy, and language policy: The case of Crystal City schools. Educational Policy, 19(4), 621-654. USA Facts (2022). Our changing population: North Carolina. U.S. Census. Retrieved from https://usafacts.org/data/topics/people-society/population-and-demographics/our- changing-population/state/north-carolina?endDate=2021-01-01&startDate=2020-01-01 United States Department of Education (2015) Dual language education programs: Current state policies and practices. U.S. Department of Education Office of English Language Acquisition. Retrieved from http://www2.ed.gov/about/offices/list/oela/resources.html Uzzel, E. M., & Ayscue, J. B. (2021) Racial integration through two-way dual language immersion: A case study. Education Policy Analysis Archives, 29(48), 1-35. Valdés, G. (1997). Dual-language immersion programs: a cautionary note concerning the education of language minority students. Harvard Educational Review, 67(3), 391-429. https://doi.org/10.17763/haer.67.3.n5q175qp86120948 Valdez, V. E., Freire, J. A., & Delavan, M. G. (2016). The gentrification of dual-language education. Urban Review, 48, 601-627. https://doi.org/10.1007/s11256-016-0370-0 Williams III, J. A., Mallant, C., & Svajda-Hardy, M. (2022). A Gap in Culturally Responsive Classroom Management Coverage? A Critical Policy Analysis of States’ School Discipline Policies. Educational Policy, 00(0), 1-26. Whiting, E. F. & Feinauer, E. (2011) Reasons for enrollment at a Spanish–English two-way immersion charter school among highly motivated parents from a diverse] community, International Journal of Bilingual Education and Bilingualism, 14(6), 631-651. 145 Wright, J., Whitaker, R. W., Khalifa, M., & Briscoe, F. (2020). The color of neoliberal reform: A critical race policy analysis of school district takeovers in Michigan. Urban Education, 55(3), 424-447. 146 APPENDIX TableBlack Table 21 Percent 21: Percent Black Difference Difference in Early in Adopting vs. Late TWI Early vs.Schools Late TWI Adopting Schools Early AdopterTWI schools Late Adopter TWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 10 0.39 0.18 49 0.33 0.21 0.06 .376 -0.31 1998-1999 10 0.40 0.19 49 0.33 0.21 0.07 .367 -0.32 1999-2000 10 0.40 0.20 51 0.34 0.22 0.07 .356 -0.32 2000-2001 10 0.42 0.21 52 0.33 0.21 0.09 .236 -0.41 2001-2002 10 0.41 0.22 52 0.33 0.20 0.08 .26 -0.39 2002-2003 10 0.40 0.22 52 0.33 0.20 0.07 .29 -0.37 2003-2004 11 0.36 0.24 52 0.32 0.19 0.04 .538 -0.21 2004-2005 12 0.40 0.25 52 0.32 0.18 0.08 .186 -0.43 2005-2006 12 0.38 0.24 54 0.31 0.18 0.07 .252 -0.37 2006-2007 12 0.31 0.23 55 0.27 0.20 0.04 .528 -0.20 2007-2008 12 0.31 0.21 55 0.26 0.17 0.05 .344 -0.30 2008-2009 12 0.37 0.22 55 0.31 0.17 0.07 .251 -0.37 2009-2010 12 0.37 0.23 55 0.31 0.17 0.07 .249 -0.37 2010-2011 12 0.29 0.22 58 0.27 0.21 0.02 .821 -0.07 2011-2012 13 0.34 0.29 59 0.25 0.19 0.09 .15 -0.45 2012-2013 13 0.29 0.21 60 0.25 0.19 0.04 .482 -0.22 2013-2014 13 0.28 0.21 60 0.24 0.16 0.04 .47 -0.22 2014-2015 13 0.28 0.21 60 0.24 0.16 0.04 .477 -0.22 2015-2016 13 0.27 0.20 60 0.24 0.16 0.03 .526 -0.19 2016-2017 13 0.27 0.20 60 0.24 0.16 0.03 .526 -0.19 2017-2018 13 0.26 0.18 60 0.24 0.15 0.02 .622 -0.15 2018-2019 13 0.26 0.17 62 0.24 0.15 0.03 .568 -0.17 2019-2020 13 0.25 0.16 63 0.24 0.18 0.01 .858 -0.05 2020-2021 13 0.25 0.16 63 0.24 0.15 0.01 0.81 -0.07 147 TableLatinx Table 22 Percent 22: Percent Latins Difference Difference in Early in Early vs. Late TWI vs.Schools Adopting Late TWI Adopting Schools Early Adopter TWI schools Late Adopter TWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 10 0.11 0.09 49 0.06 0.06 0.05 .03 -0.77 1998-1999 10 0.13 0.11 49 0.06 0.06 0.07 .012 -0.90 1999-2000 10 0.16 0.13 51 0.08 0.07 0.07 .012 -0.90 2000-2001 10 0.18 0.14 52 0.10 0.09 0.08 .026 -0.79 2001-2002 10 0.21 0.17 52 0.12 0.10 0.08 .031 -0.76 2002-2003 10 0.22 0.18 52 0.15 0.10 0.07 .081 -0.61 2003-2004 11 0.24 0.19 52 0.16 0.11 0.07 .097 -0.56 2004-2005 12 0.24 0.21 52 0.19 0.12 0.05 .231 -0.39 2005-2006 12 0.26 0.21 54 0.21 0.13 0.06 .225 -0.39 2006-2007 12 0.27 0.21 55 0.24 0.17 0.04 .528 -0.20 2007-2008 12 0.28 0.21 55 0.24 0.13 0.04 .421 -0.26 2008-2009 12 0.29 0.21 55 0.25 0.13 0.04 .36 -0.29 2009-2010 12 0.29 0.21 55 0.26 0.14 0.04 .466 -0.23 2010-2011 12 0.33 0.21 58 0.30 0.19 0.03 .63 -0.15 2011-2012 13 0.38 0.26 59 0.30 0.16 0.08 .149 -0.45 2012-2013 13 0.35 0.19 60 0.31 0.17 0.04 .497 -0.21 2013-2014 13 0.37 0.18 60 0.30 0.14 0.07 .12 -0.48 2014-2015 13 0.38 0.19 60 0.32 0.13 0.06 .155 -0.44 2015-2016 13 0.40 0.18 60 0.32 0.14 0.07 .099 -0.51 2016-2017 13 0.40 0.18 60 0.32 0.14 0.07 .099 -0.51 2017-2018 13 0.40 0.19 60 0.33 0.14 0.07 .137 -0.46 2018-2019 13 0.41 0.19 62 0.34 0.14 0.06 .155 -0.44 2019-2020 13 0.42 0.19 63 0.36 0.16 0.06 .264 -0.34 2020-2021 13.00 0.42 0.18 63.00 0.36 0.14 0.06 0.16 -0.43 148 TableAsian Table 23 Percent 23: Percent Asian Difference Difference in Early in Early vs. Late TWI vs. Late Adopting TWI Adopting Schools Schools Early Adopter TWI schools Late Adopter TWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 10 0.03 0.05 49 0.02 0.03 0.01 .339 -0.33 1998-1999 10 0.03 0.04 49 0.02 0.04 0.01 .304 -0.36 1999-2000 10 0.04 0.05 51 0.03 0.04 0.01 .51 -0.23 2000-2001 10 0.04 0.06 52 0.03 0.04 0.01 .562 -0.20 2001-2002 10 0.04 0.05 52 0.03 0.05 0.01 .55 -0.21 2002-2003 10 0.04 0.07 52 0.03 0.04 0.02 .337 -0.33 2003-2004 11 0.04 0.07 52 0.02 0.03 0.02 .198 -0.43 2004-2005 12 0.04 0.08 52 0.02 0.03 0.02 .21 -0.41 2005-2006 12 0.04 0.08 54 0.02 0.03 0.02 .127 -0.49 2006-2007 12 0.04 0.08 55 0.04 0.14 0.01 .881 -0.05 2007-2008 12 0.05 0.09 55 0.02 0.03 0.04 .018 -0.78 2008-2009 12 0.06 0.09 55 0.02 0.03 0.04 .017 -0.78 2009-2010 12 0.06 0.09 55 0.02 0.03 0.04 .014 -0.80 2010-2011 12 0.05 0.09 58 0.05 0.18 0.00 .951 -0.02 2011-2012 13 0.12 0.27 59 0.03 0.13 0.09 .087 -0.53 2012-2013 13 0.05 0.08 60 0.03 0.13 0.01 .73 -0.11 2013-2014 13 0.04 0.10 60 0.02 0.03 0.02 .11 -0.49 2014-2015 13 0.05 0.11 60 0.02 0.03 0.03 .115 -0.49 2015-2016 13 0.05 0.11 60 0.02 0.03 0.03 .127 -0.47 2016-2017 13 0.05 0.11 60 0.02 0.03 0.03 .127 -0.47 2017-2018 13 0.05 0.10 60 0.02 0.04 0.03 .108 -0.50 2018-2019 13 0.05 0.09 62 0.02 0.04 0.02 .148 -0.45 2019-2020 13 0.04 0.09 63 0.04 0.13 0.00 .916 -0.03 2020-2021 13 0.04 0.10 63 0.02 0.05 0.02 0.25 -0.35 149 TableAsian Table 24 Percent 24: Percent Asian Difference Difference in Early in Early vs. Late OWI vs. Schools Adopting Late OWI Adopting Schools Early Adopter OWI schools Late Adopter OWI schools T test Year obs mean sd obs mean sd difference p value effect size 1997-1998 23 0.03 0.02 67 0.01 0.02 0.01 .019 -0.58 1998-1999 23 0.02 0.02 69 0.01 0.02 0.01 .091 -0.41 1999-2000 23 0.03 0.03 72 0.02 0.02 0.01 .072 -0.44 2000-2001 23 0.03 0.03 73 0.02 0.02 0.01 .111 -0.38 2001-2002 24 0.03 0.03 76 0.02 0.02 0.01 .153 -0.34 2002-2003 25 0.03 0.03 79 0.02 0.02 0.01 .139 -0.34 2003-2004 25 0.03 0.03 79 0.02 0.02 0.01 .173 -0.31 2004-2005 25 0.03 0.03 82 0.02 0.02 0.01 .215 -0.29 2005-2006 25 0.03 0.03 85 0.02 0.02 0.01 .129 -0.35 2006-2007 25 0.03 0.03 89 0.05 0.18 -0.02 .5 0.15 2007-2008 26 0.07 0.19 90 0.03 0.11 0.04 .215 -0.28 2008-2009 26 0.03 0.03 91 0.03 0.11 0.00 .886 0.03 2009-2010 26 0.03 0.04 93 0.03 0.10 0.00 .946 0.02 2010-2011 26 0.03 0.04 94 0.02 0.02 0.01 .299 -0.23 2011-2012 26 0.03 0.03 94 0.02 0.02 0.01 .243 -0.26 2012-2013 26 0.03 0.04 94 0.02 0.03 0.01 .229 -0.27 2013-2014 26 0.03 0.04 95 0.03 0.10 0.00 .924 0.02 2014-2015 26 0.03 0.04 96 0.03 0.03 0.01 .234 -0.26 2015-2016 26 0.03 0.05 96 0.03 0.03 0.01 .348 -0.21 2016-2017 26 0.03 0.05 96 0.03 0.03 0.01 .348 -0.21 2017-2018 26 0.04 0.05 97 0.03 0.04 0.01 .442 -0.17 2018-2019 26 0.04 0.05 98 0.03 0.05 0.01 .589 -0.12 2019-2020 26 0.04 0.06 98 0.03 0.05 0.00 .731 -0.08 2020-2021 26 0.04 0.06 98 0.03 0.06 0.00 0.80 -0.06 150 Figure 3: Change in the Percentage of White Students in TWI Schools Figure 4: Change in the Percentage of White Students in OWI Schools 151 Figure 5: Change in the Percentage of Black Students in OWI Schools Figure 6: Change in Percentage of Latinx Students in OWI Schools 152 CONCLUSION The three papers that comprise this dissertation provide insight on a variety of policies and mechanisms that have the potential to increase racial and ethnic diversity in the schools that choose to adopt these programs. Nevertheless, each of the policies is uniquely designed to potentially achieve a diverse student body. School desegregation policies such as intentionally diverse charter schools incorporate diversity as part of their school mission and enrollment process and espouse the students benefits of learning amongst racially and ethnically diverse peers. Dual language immersion programs may also rely on controlled enrollment policies to aide in creating a diverse classroom environment, but they also advocate for a linguistically diverse classroom with a mix of students with varying native languages to help all students become bilingual learners. While diversity is a potential for each of these policies, these three papers explore whether this potential is realized by specially examining White individuals’ perspectives and actions when these policies are proposed and then implemented. By focusing on White individuals, I extend the application of Critical Whiteness theories to new policy contexts and add nuance to the understanding of interest convergence theory in education spaces. I use mixed methods and examine White individuals in a variety of contexts using data from two different states in the United States. White is not a monolith, and by including data on White individuals from a national survey, administrative data encompassing the state of North Carolina, and interview data on specific group of White parents in Denver, I add nuance to our understanding of how Whiteness plays out in multiple policy and geographic contexts. In my first paper which utilizes national survey data in a survey experiment that studies how policy frames regarding student benefits influence White individuals’ support for a policy that would increase the racial diversity of schools in their community, I find that White individuals are the 153 most supportive of the policy when it is framed in race neutral language and the least supportive of the policy when White students are specifically mentioned. In my second paper that draws on the interview data from White parents who enroll in an intentionally diverse charter school in Denver, I find that the majority of the parents value school diversity, but largely view it as something to for their consumption that will ultimately help their White children be even more prepared for college and the “real world.” And lastly, my third paper analyses policy document language and enrollment data in North Carolina examining how access to DLI programs in the state has changed over time, specifically after the policy change to expand DLI programs in the state. I find that policy documents primarily base their support for expanding the programs by citing goals of developing a globally competitive human capital while silencing folk bi and multilingual communities. Additionally, I find that the early and late adopter DLI are comparable in student demographics except that late adopter DLI schools serve a significantly higher percentage of White students than early adopter schools. Together, these papers provide insight into White individuals’ policy perspective and provide information about how to structure policies that have the potential to diversify schools so that they are not exploited by individuals with greater social capital. For many of these policies, the White interest and support for the polies can help expand the policy and give it greater reach, but through this expansion, it is especially important that equity in terms of access and voice in policy decisions remain front-of-mind so that all groups continue to have access and input during the policy making and implementation processes to prevent the interests of White individuals to become prioritized over others. 154