Racial Integration Through Two-Way Dual Language Immersion: A Case Study

Despite increasingly diverse public school enrollment, students across the U.S. are still segregated by race and poverty, and English learners (ELs) often experience triple segregation by race, poverty, and language. Two-way immersion (TWI) programs may create racially integrated learning environments, by offering a dual language model that balances native English speakers and speakers of the partner language. Through semi-structured interviews, observation, and document analysis, this qualitative case study examined how a Spanish TWI program facilitates integration in a rural elementary school. Findings show that students from different backgrounds may have equal status in mutually beneficial environments, can become bilingual and bicultural, and may experience lifelong benefits. Implications include the need for increased federal, state, and local funding to support districts using TWI to achieve integration as well as a federal language policy that promotes TWI.


Introduction
Although the United States has become increasingly racially and linguistically diverse in the past century, many schools around the country do not reflect that diversity. More than 65 years ago, the U.S. Supreme Court acknowledged that segregated schools are inherently unequal. After a slow start, schools began to desegregate, and the peak of school desegregation occurred in 1988 . Since that time, public schools across the United States have been resegregating, with many students of color experiencing isolation from students of noncolor . Research consistently shows that segregation is systematically linked to unequal educational opportunities and outcomes for students of color and students living in poverty, yet the federal impetus to intervene in segregated districts has diminished (Orfield & Eaton, 1996). Following court cases (e.g., Milliken v. Bradley, 1974;Missouri v. Jenkins, 1995;Parents Involved in Community Schools, 2007) and significant changes to Civil Rights Era legislation, the tools through which school districts can achieve voluntary desegregation have become severely limited, resulting in a need for research to identify tools for facilitating integration in the 21 st Century.
While the landmark Brown v. Board of Education (1954) decision reversed the lawful segregation of Black students, and Brown II called for schools to desegregate with "all deliberate speed," there were still court decisions to be made in the following decades concerning students who had immigrated to the United States with varying levels of English fluency. Each of these factors, immigration and language proficiency, has presented as a contentious political issue over the years, resulting in immigration reforms and language policies influenced by xenophobia, white supremacy, and English hegemony. The precedent set by pre-Brown cases like Maestas v. Shone (1914) and Mendez v. Westminster (1947), in which the rights of Mexican American students to equal educational opportunities were brought forward and affirmed, has been challenged and undermined such that many still suffered from within-school segregation (Gándara & Orfield, 2010). The right to supplemental language instruction was established in Lau v. Nichols (1974), but the guidance following that case was insufficient when it came to implementing effective programs (Stewner-Manzanares, 1988). The unique needs of immigrant students were recognized, but many school districts often did not know or could not afford programs that best served them, ultimately resulting in the racial and linguistic isolation that legislation intended to avoid (Gándara & Orfield, 2010). Although there is no official language in the United States, the preference for English language instruction is longstanding, and nonwhite English learners (EL) have had to fight for educational access mostly through English acquisition rather than being offered the curriculum in their home language.
Early proponents of bilingual education sought reforms that could work toward dismantling white supremacy and reversing the damaging psychological effects of linguistic segregation, though ultimately the first wave of legislation took an individual rather than systemic route (Flores & García, 2017). The Bilingual Education Act of 1968 (Title VII of the Elementary and Secondary Education Act [ESEA]), and subsequent amendments, offered funding through competitive grants to schools serving ELs, often in their home language and with the support of parents (Stewner-Manzanares, 1988). However, the focus of federal and state guidelines soon shifted to English acquisition, and little regard for the home language, with some states going so far as to enforce English-only policies (Combs et al., 2005). The preference for English proficiency over bilingualism was confirmed in the reauthorization of ESEA as No Child Left Behind (2002), mandating that all students, including ELs, perform on high-stakes assessments. What began as a celebration of culture and language became, through multiple revisions, a doctrine of assimilation, which has not recovered to this day (Every Students Succeeds Act [ESSA], 2015). Title III of ESSA offers rigorous guidelines for the instruction of ELs, intent on bringing these students up to proficiency and increasing academic achievement. Unlike now-defunct English-only policies, Title III does not prohibit instruction in a language other than English-but it does not emphasize bilingual education either. The benefits of learning in two languages include increased cognitive performance and executive control (Bialystok, 2011), greater probability of college enrollment compared to monolingual peers (Santibañez & Zárate, 2014), lower dropout rates and higher earning potential (Rumbaut, 2014), and even preference in hiring (Porras et al., 2014). On the whole, these advantages, along with programs that support becoming bilingual, bicultural and biliterate (Rolstad et al., 2005;Thomas & Collier, 2014;Valentino & Reardon, 2015), have been understated in favor of programs that promote English acquisition. However, despite the federal support of transitional bilingual programs, designed to get ELs into mainstream classrooms as quickly as possible, there are a number of developmental bilingual programs growing in popularity in states across the country. These types of programs, that promote acquisition of another language and maintenance of the home language, have potential to positively impact historically marginalized students through integration and immersion.
Two-way dual language immersion (TWI) programs offer one possibility for states and school districts to voluntarily create integrated learning environments. In the United States, TWI programs offer a dual language model in which instruction is provided in two languages: English and a partner language. The nature of TWI programs presents an opportunity for education policy, not only for bilingual education, but also for integration. Because TWI programs are ideally designed to balance native English speakers and speakers of the partner language, such programs have the potential to create integrated classrooms in which students from different racial backgrounds can have equal status in mutually beneficial environments. The goals of bilingualism and biculturalism relate directly to integration in that they embrace diversity and advocate knowledge of and interaction with other cultures. TWI can pull students out of the linguistic, racial, and socioeconomic segregation that ELs and students of color experience far too often in our schools (Gándara & Orfield, 2010). Thus, the purpose of this qualitative case study is to examine whether and how a TWI program facilitates integration. Through interviews and observations of the TWI program in one rural elementary school, we investigate the question: How do TWI programs facilitate integration?

Literature Review
This study brings together research on school integration and TWI programs to examine how TWI programs facilitate integration, an area that has yet to be explored in the literature. Despite our nation's increasingly diverse student enrollment, students of color often experience double segregation by both race and poverty , while ELs experience triple segregation by race, socioeconomic status, and language (Gándara, 2010). While there is no federal language policy in the United States (Gándara & Hopkins, 2010), many states have adopted policies regarding how to teach ELs, some that focus on providing adequate bilingual instruction to ELs and English speakers, and others that focus on English-only instruction. Just as segregation is related to lower academic achievement for students of color (Mickelson et al., 2013;Mickelson et al., 2020), segregating ELs in separate classrooms does not improve academic outcomes (Gándara & Aldana, 2014). However, research suggests that both integration and immersion carry a number of benefits for all students, as dual language immersion (DLI 1 ) programs continue to grow in popularity. TWI programs in particular are expanding in several states, including North Carolina, the focus of this study.

Double Segregation
Prior to 1954, public schools in the United States were organized along racial lines. De jure segregation, after several decades of challenges (Green, 2004), was finally overturned following the landmark Brown v. Board of Education (1954) acknowledgement that separate schools were not providing an equal education. In addition to Brown, immigration reform of the Civil Rights Era contributed to the need for integration as the racial makeup of the country changed. From 1968 to 2011, the White student enrollment decreased by 28%, Black enrollment increased 19%, and Latinx enrollment increased 495% (Orfield & Frankenberg, 2014). Students of color now account for over 50% of public school enrollment nationwide.
However, students of color still experience double segregation. Using measures of exposure and isolation, Orfield and Frankenberg (2014) found that the majority of White students attended schools that are mostly White, while students of color became increasingly isolated in majorityminority and sometimes hyper-segregated schools. Higher concentrations of students of color often result in higher concentrations of poverty (Orfield & Frankenberg, 2014). In 2016, the typical Black and Latinx students attended schools that were 74-75% nonwhite . Many disadvantages are associated with extreme levels of segregation: segregated schools have fewer experienced and qualified teachers (Clotfelter et al., 2005;Jackson, 2009), higher rates of teacher turnover (Clotfelter et al., 2010), higher rates of student mobility (Rumberger, 2003), and less advanced curricular options (Yun & Moreno, 2006). Consequently, minority segregated schools have lower academic performance (Mickelson et al., 2013;Mickelson et al., 2020), higher dropout rates, and lower graduation rates (Balfanz & Legters, 2004;Swanson, 2004).
Segregation is often framed in terms of students of color being separated from White students, but many White students are attending segregated schools as well-that is, with other White students. Because the initial process of desegregation was framed as a movement to allow Black students into White schools-rather than to provide more funding to Black schools-the racist idea that Black students reap academic benefits from exposure to White students has remained intact (Kendi, 2016;Ladson-Billings, 2004). The decision to create integrated learning environments was in part fueled by a desire to undo the damaging effects of segregation on Black children's psychology (Green, 2004), and provide a semblance of racial equality on the world stage, not to eradicate the notion of white supremacy from the education system (Ladson-Billings, 2004). While students of color may benefit from attending schools with middle class White students because of the resources that are more readily available, there are also benefits to White students attending desegregated schools, including the important, if often overlooked, ability to function in diverse workplaces and interact with people from different racial backgrounds (Siegel-Hawley, 2012). Segregation can negatively impact all aspects of society, while desegregation has documented benefits.
However, inequality, racism, white supremacy, and xenophobia can and do persist within desegregated learning environments (Lewis & Diamond, 2015). Through homogeneous groupings and academic tracking, students of color are often provided with access to inferior academic opportunities (Mickelson, 2001;Oakes, 2005), while White students have disproportionately greater access to more advanced curriculum as well as gifted and talented programs (Ford et al., 2008;Roda, 2015). Racial disproportionality in discipline (Losen & Martinez, 2020) and special education (Losen & Orfield, 2002;Sullivan & Bal, 2013) also persist within schools that may appear to be desegregated. Thus, ensuring authentic integration within desegregated learning environments is essential.

Triple Segregation
The rapidly expanding Latinx student population in the United States is even more likely to experience segregation in three ways: race/ethnicity, poverty, and language. Although the Maestas v. Shone (1914) decision in the early part of the twentieth century challenged the segregation of Mexican American students on racial and linguistic grounds, and landmark cases like Mendez v. Westminster (1947) andGonzales v. Sheely (1951) confirmed the rights of Mexican American students to attend White schools, these rights continue to be violated. In the fall of 2017, 60% of Latinx students attended majority-minority schools (National Center for Education Statistics, 2020), while accounting for 25% of the total school-aged population (Shapiro & Partelow, 2017). Not only do Latinx ELs experience racial segregation, but also in some cases they experience linguistic segregation, spending a part of if not the whole day isolated from English-speaking peers, in pursuit of English proficiency, suggesting a return to so-called "Mexican schools" (Gándara & Orfield, 2010). This strategy has not been found to improve student achievement or language proficiency (Gándara & Aldana, 2014), and may be fueled by discrimination and the desire to assimilate Latinx students, whose language skills are inevitably racialized (Flores & Rosa, 2015).
Linguistic segregation also hinders achievement through isolation from resources: lack of positive peer role models, lack of support staff, and lack of communication in academic English (Gándara, 2010;Garver, 2020). Without meaningful interactions with English-speaking classmates, ELs can struggle in language development and academic achievement. Schools with high levels of poverty are marked by high rates of absenteeism and turnover for both students and teachers, so stability and effective instruction also impact proficiency (Gándara, 2010). The quality of EL instruction often comes down to program implementation and teacher support (Garver, 2020). Gándara (2010) notes that ELs attending segregated schools and living in linguistically isolated communities lack regular access to mainstream English, making it more difficult for them to achieve in academic English. Unfortunately, tests of English language proficiency cannot capture the dexterity in languages that bilinguals may possess (Flores & Rosa, 2015). If language is a hierarchy, academic English is often viewed as the top, so although emergent bilinguals are gaining access to the advantages of knowing two or more languages, their skills are not always recognized as such. The desire to parse out languages for Latinx ELs, replacing Spanish with English through isolated classroom experiences, is rooted in racist ideas about the inferiority of languages other than English, or linguistic practices that blend two or more languages in ways that monolinguals do not understand (Flores & Rosa, 2015). The linguistic segregation of Latinx ELs cannot be separated out from the racial segregation of these same students, and its effects are compounding.
The history of discrimination which began in segregated "Mexican schools" continues to replicate itself through racialized linguistic instructional practices like sheltered immersion (Combs et al., 2005) and inadequate learning opportunities, the cumulative effects of which are still seen today. On average, Latinx students are one to three grade levels behind White peers (Educational Opportunity Project, 2019; Reardon et al., 2019), a gap which may be attributable to reduced opportunities for learning from qualified teachers and higher-achieving peers (Gándara, 2010). Inattention to disparities in the early grades can lead to higher levels of high school dropout and lower instances of postsecondary attainment (Gándara & Mordechay, 2017), while Latinxs who do receive support have higher GPAs and are more likely to attend college (Bordes-Edgar et al., 2011). Discrimination against Latinx students has been a constant over the last century, but has increased substantially since the 2016 election: school principals report an increase in anxiety over immigration policies targeting Latinx families, creating deleterious effects on students' health and well-being (Ee & Gándara, 2020). While not the only racial/ethnic group affected by poverty and linguistic isolation, Latinx students are the largest minority group, and their treatment has implications for other groups experiencing marginalization.

Resegregation
Efforts to integrate students from different racial backgrounds that ostensibly began with Brown v. Board of Education (1954) have shifted to resegregation in the last several decades. Since the mid-1970s, the Supreme Court has limited the ways in which school districts can attain desegregation (Board of Education of Oklahoma v. Dowell, 1991;Freeman v. Pitts, 1992;Milliken v. Bradley, 1974;Missouri v. Jenkins, 1995). Most recently, the Parents Involved in Community Schools v. Seattle (2007) decision ruled that an individual student's race could not be the deciding factor in determining where the student would be assigned to school. Between 1991 and 2009, nearly 200 districts were released from court-ordered desegregation and subsequently experienced resegregation after having been declared unitary (Reardon et al., 2012). The focus has become on attending neighborhood schools, as parents would rather see students attend schools close to home, even as suburban schools become more diverse (Parcel & Taylor, 2015). This can be problematic when neighborhood segregation mirrors school segregation (Frankenberg, 2013), as inequitable housing opportunities create homogenous district zones and coincide with a push for local control. Exacerbating resegregation trends are the handful of schools that have created majority White schools by seceding from the larger, more racially and socioeconomically diverse district (Rojas, 2019;Siegel-Hawley et al., 2018). These factors combine to show-using segregation measures of isolation and exposure (Orfield et al., 2014) rather than unevenness (Reardon & Owens, 2014)-that many students are as segregated now as they were in 1954.
There are, however, a number of schools pushing back against resegregation. Recognizing the relationship between race and poverty, some districts are using socioeconomic status (SES) to create more racially diverse schools. 91 districts and charters across the country use free and reduced lunch eligibility, sometimes combined with census data and other indicators, to create integration policies along socioeconomic lines (Potter et al., 2016). Additionally, Taylor, Anderson, and Frankenberg (2019) estimated that 59 public school districts in the United States are currently pursuing voluntary desegregation efforts, many of which rely on desegregation based on SES. While this approach is effective in some places, using SES as a proxy for race is not effective everywhere and its success depends on multiple factors, including how SES is defined and how many students and schools the SES-based plan encompasses (Reardon & Rhodes, 2011); this approach is likely not as effective as using race in permissible ways would be (Siegel-Hawley, Frankenberg, & Ayscue, 2017). Districts also use magnet schools and controlled choice to facilitate integration (Taylor, Anderson, & Frankenberg, 2019). However, in some urban neighborhoods of color, the expansion of schools of choice facilitates gentrification as White families view these choices as a way to opt out of sending their children to the racially segregated neighborhood school (Pearman & Swain, 2017). Although unregulated choice options, such as charter schools, often contribute to inequities (Ayscue et al., 2018;Clotfelter et al., 2019;Scott, 2011), regulated choice programs, like TWI, may disrupt resegregation trends and provide an integrated learning environment.

Bilingual Education Policy
The argument for adequate language instruction for ELs in the United States came to the national stage in the 1974 Supreme Court case Lau v. Nichols, in which the plaintiffs argued that their access to equal education was dependent on the ability to have command of the English language. The Lau Remedies inspired implementation of various bilingual education programs, including developmental bilingual education as well as one-way immersion and the increasingly popular TWI programs (de Jong & Howard, 2009). While both types of immersion offer a dual language model supporting instruction in two languages (e.g., English and Spanish), one-way programs typically enroll speakers of the same language background, whereas TWI intentionally creates a balance of learners from each of the language backgrounds. Indeed, the ideal design of a TWI program includes 50% partner language speakers and 50% native English speakers.
The trend of bilingual education policy in the United States has historically focused on the lack of ability to speak English as a deficit rather than the ability to speak another language as an asset. As Gándara and Aldana (2014) note, "Despite the increase of multilingual individuals, and Spanish speakers in particular, U.S. schools have failed to capitalize on the linguistic assets these students and their families bring with them" (p. 736). In some instances, the focus was solely on English acquisition, with states like Massachusetts, California, and Arizona implementing Englishonly policies (Combs et al., 2005). California and Massachusetts have since repealed these policies, and Arizona was expected to follow suit in 2020 (Mitchell, 2019). Other states have chosen to approach bilingual education from an asset perspective, though not necessarily with the intent of lifting students out of linguistic isolation; in fact, some scholars argue that DLI will primarily benefit historically advantaged students (Cervantes-Soon, 2014;Valdez et al., 2016). Privileged students with access to DLI will be able to take more advanced courses because they have received enrichment, while these same courses may not be available to historically disadvantaged students (Valdez et al., 2016).
The intention behind the implementation of TWI programs has led to some concern. While they have the ability to desegregate and pull ELs out of linguistic isolation, they could also primarily benefit English speakers. Valdés (1997) was one of the first to caution against DLI programs, particularly with regards to language-minority students. This scholar sees some bilingual and desegregation programs as "narrow solutions to far broader problems," especially when these programs approach students from a deficit perspective. She acknowledges that DLI does take an asset, rather than deficit, position, and offers the added benefits of educating language-minority students with mainstream students. In order to be effective, these programs must: ensure quality instruction, in which academic Spanish is the target and Spanish-speaking students are not slowed down by their English-majority classmates; pay close attention to intergroup relations, so that students are not just working together in class, but also socializing outside of class, thus forming strong relationships across lines of race, ethnicity and language; and attempt to actually empower language-minority students (Valdés, 1997). Valdés is cited and supported by later scholars (Cervantes-Soon, 2014;Dorner, 2011;Knight et al., 2016;Valdez et al., 2016) who find that DLI programs need to place equity for all students at the forefront.
Without a federal language policy, states and schools are left to decide how to address the needs of ELs, with guidelines from ESSA still mainly centered around English acquisition. Furthermore, states can vary in their definitions of DLI and requirements for programs, leaving important equity decisions up to local education agencies (U.S. Department of Education, 2015). Those who implement are not always aware of how equity issues can manifest, so careful attention must be paid to the design and implementation of TWI programs to ensure that their potential for integration is fully realized. For example, Palmer (2010) investigated the nature of a TWI program in California, particularly regarding its racial makeup; while the school in which the program resides is integrated, the TWI strand contains mostly White students and very few Black students. Conducting her research in a school that includes a 1:1 ratio of Spanish-speaking Latinx students to Englishspeaking students in the TWI program, Palmer notes the predicament of "integrated on paper." From the outside, the school seems diverse, but most of the White students are funneled into the immersion program, meaning that the population of Black students in non-TWI classes is inflated. Furthermore, the resources that accompany affluent, often White middle-class families also get funneled into that program, and these resources leave when White middle-class students exit the program. Ultimately, Palmer concludes that the program would be most effective if used throughout the whole school, but that it should make a stronger effort to include students of all backgrounds, not just White and Latinx students.
Historically, desegregation and language policy have been treated as separate and sometimes competing initiatives rather than as interrelated means for achieving equitable diversity and serving a diverse population (Gándara, 2020;Gándara & Aldana, 2014). TWI programs can promote integration and will likely be most effective if implemented with the goal of serving marginalized students and intentionally facilitating authentic integration. Although it would seem that the requisite balance of speakers from each partner language would necessarily lead to a racially integrated program, some scholars are skeptical (Cervantes-Soon, 2014;de Jong & Howard, 2009;Palmer, 2010;Valdés, 1997). An emerging body of research would suggest that TWI programs can reinforce the very problems they seek to solve, including resource hoarding by White families and cultural appropriation rather than appreciation (Valdés, 1997). Along with potential gatekeeping of who is admitted into the program, immigrant EL students may experience inequality within the TWI program (de Jong & Howard, 2009;Flores & García, 2017). Some scholars (e.g., Jaffe-Walter & Miranda, 2020) go so far as to suggest that immigrant ELs benefit from attending segregated schools specifically designed around their needs, as they are provided a kind of safe haven wherein their teachers are adequately prepared and they receive the support they need. However, as the research below demonstrates, integrated environments provide many benefits to students; therefore, if designed with equity as a foundation, TWI programs have the potential to offer the benefits of both immersion and integration.

Benefits of Immersion and Integration
Immersion DLI programs not only improve abilities in English and the partner language, but increase academic achievement for all students (Lindholm-Leary & Hernández, 2011). Many scholars (Knight et al., 2016;Steele et al., 2017;Thomas & Collier, 2014) have studied and reported the academic benefits of DLI programs, and part of their popularity can be attributed to the recognition of bilingualism as a global resource. A three-year study of 12 DLI programs in North Carolina found that TWI programs do indeed raise test scores for Black, Latinx, and White students (Thomas & Collier, 2014). This study also suggested that Black students have the most to gain from TWI programs and recommended that greater effort be made to include them in such programs. Additional research confirms that DLI programs improve academic achievement for ELs and native English speakers (Rolstad et al., 2005;Steele et al., 2017;Umansky & Reardon, 2014;Valentino & Reardon, 2015). Along with higher test scores while in school, DLI programs help students become bilingual, which increases career and earning potential (Porras et al., 2014;Rumbaut, 2014). Students acting as language models for each other promote language acquisition as well as cooperation and cultural exchange (Lindholm-Leary et al., 2007), which provides an academic as well as a social benefit. The social benefits of TWI cannot be understated: the balance of native speakers from both languages exposes students to diverse cultural and linguistic backgrounds which may foster friendships and promote positive interactions across cultures (Block, 2011;Feinauer & Howard, 2014). TWI in particular offers these benefits for all students, and has been found to be especially effective for ELs (Lindholm-Leary & Block, 2010;Valentino & Reardon, 2015). Students in these programs were found to have increased proficiency in both languages and some even outperform peers on standardized tests (Lindholm-Leary et al., 2007). Furthermore, TWI programs are one possibility for counteracting triple segregation and promoting racial and linguistic desegregation (Gándara & Aldana, 2014).

Integration
Turning specifically to integration, numerous studies have shown that its benefits include higher achievement in mathematics, science, language arts, and reading (Borman et al., 2004;Kain & O'Brien, 2002), and extend to all students within a diverse school regardless of race or SES (Mickelson, 2016;Siegel-Hawley, 2012). Middle and high school students experience compounding benefits of integration over time and are more likely to graduate from high school and go to college than their peers in segregated schools. Additionally, integrated schools are associated with a reduction in prejudice, negative attitudes, and stereotypes as well as increased friendships among members of different groups (Allport, 1954;Pettigrew & Tropp, 2006;Tropp & Prenovost, 2008). In the long term, students who have attended integrated schools are more likely to live and work in desegregated environments later in life (Braddock & McPartland, 1989). They experience increased educational and occupational attainment, greater economic returns, health benefits, and less adult poverty (Johnson, 2011(Johnson, , 2019. In other words, the effects of attending a diverse school are lifelong and even intergenerational (Johnson, 2019;Mickelson, 2016;Mickelson & Nkomo, 2012). Diverse schools produce global-ready students who enter the workplace with requisite social skills and typically receive a higher income.
Given the unintended historic and contemporary consequences of school desegregationjob loss for Black teachers and leaders, a lack of culturally relevant teaching, the persistence of the Black-White achievement gap, and the failure to address white supremacy in education-some scholars suggest that the promises of integration have not been fully realized (Walker, 2009;Stuart Wells et al., 2004). Therefore we contend that these benefits depend on true integration, that is, the fair and equitable treatment of students of different racial and ethnic groups within a desegregated environment (Ayscue & Frankenberg, 2016).

Integration Theory of Choice
This study is grounded in the integration theory of choice, which emerged from the Civil Rights Era and focuses on regulated choice as a tool for pursuing integration, equity, and equality of opportunity (Cobb & Glass, 2009;Orfield, 2013). The premise of the integration theory of choice is that social inequality underlies educational inequality, and choice can be used as a voluntary approach to facilitate integration. This theory relies on providing attractive educational options so that a diverse set of students will choose to enroll in the school or program. Although perhaps not the primary goal for students and families, this process can result in students learning in environments with more diverse classmates.
The integration theory of choice suggests that being able to voluntarily enroll in a program like TWI, that encourages diversity in enrollment and equal status contact among racial groups, will benefit students of all backgrounds, and can be used as a tool to achieve integration (Orfield, 2013). Consistent with this theory, Gándara (2010) suggests that a bilingual immersion program can promote gains in achievement by providing ELs access to English speakers as well as promoting friendships and respect for their native language. Furthermore, TWI programs are one possibility for counteracting triple segregation and promoting racial and linguistic desegregation (Gándara & Aldana, 2014).
The broader concept of choice is often critiqued for its association with privatization and neoliberalism, and within this broad concept, there are important distinctions among different forms of choice (Cobb & Glass, 2009). In contrast to the integration theory of choice, the market theory of choice is based on the premise that a competitive educational marketplace will improve the quality of education for students (Chubb & Moe, 1990). Charter schools grew out of the market theory, which prioritizes competition, deregulation, and privatization. Unlike the integration theory of choice, the market theory of choice is not guided by integration or equity. An important distinction exists between unregulated market-based choice (e.g., charters and vouchers) that promote privatization versus regulated integration-based choice (e.g., magnets and specifically in this case, TWI programs) that function within the traditional public schools governance structure and are used as mechanisms to promote integration voluntarily. Therefore, guided by the integration theory of choice, this study examines whether and how TWI programs can be used as a tool for integration.

Two-Way Immersion in North Carolina
Although there is not an up-to-date database of all existing DLI programs in the United States, the Center for Applied Linguistics has compiled an impressive directory of over three hundred DLI programs, including type, languages offered, school level, and instructional model (Center for Applied Linguistics, 2016). The implementation of Spanish DLI programs in the Southern states has coincided with a 200% increase in the Latinx population (Gándara & Mordechay, 2017). As states continue to implement DLI, there may now be over one thousand programs nationwide. With the popularity of bilingual education on the rise, North Carolina is one of a handful of states that has begun to promote DLI programs specifically, and provided guidelines and funding for schools implementing programs (U.S. Department of Education, 2015). The state began adding DLI programs in the 1990s, and the first TWI program began in 1997 (Thomas & Collier, 2012). Professors Wayne Thomas and Virginia Collier have been conducting longitudinal studies of extant programs in the state, making North Carolina a case study of the lasting academic, cognitive and social benefits of DLI (Thomas & Collier, 2012. North Carolina has been a leader in dual language education, although the implementation of DLI programs may have been based on a desire to provide foreign language education for English-speaking students rather than a need to promote equity (Cervantes-Soon, 2014). In January of 2013, the State Board of Education Task Force on Global Education proposed that schools should attempt to "graduate students with advanced cultural and language skills" and recommended a renewed effort to implement DLI programs (North Carolina State Board of Education, 2013). So, the popularity of TWI solidified and grew. Concurrently, over the past few decades, North Carolina experienced an increase in the enrollment of ELs, due, in part, to an 84% increase in immigration between 2000 and 2016, so that ELs now account for 8% of the school-aged population (Sugarman & Geary, 2018). This spike, alongside an increase in the Latinx population, presents a unique opportunity for schools to provide a high-quality, integrated education, and for policymakers to implement inclusive language policy. The expansion of equitable TWI programs might be one way to achieve both goals.
The North Carolina Department of Public Instruction (NCDPI) has dedicated a number of guidelines for DLI program implementation, including standards, definitions, assessment, and funding. The focus is on developing proficiency in English and the partner language, as well as mastering subjects in the curriculum using both languages (NCDPI, 2015). The state's 60 TWI programs, 96% of which are Spanish immersion, offer opportunities for English speakers and Spanish speakers to learn both languages by grouping them equally in the same environment. Most schools begin the program in kindergarten with a 90/10 model: 90% of the day is taught in Spanish, and 10% in English. As students progress through the program, instruction moves to a 50/50 model. In North Carolina, most schools prefer a team teaching model, with students switching between a Spanish class and an English class (Thomas & Collier, 2012). Students receive content in both Spanish and English throughout the day.
Once a leader of desegregation efforts in the South, North Carolina has been changing student assignment policy in its largest school districts so that race is no longer a priority, and effectively resegregating schools in the largest counties Williams & Houck, 2013). The problem with "race-neutral" assignment policy is that it enforces the same colorblind mentality that allows the dominant culture to remain dominant while continuing to marginalize people of color. With support for school choice on the rise, North Carolina may have an opportunity to renew integration efforts with its expansion of TWI programs, though not without explicit attention to equity.
In an increasingly diverse society, acceptance of differences among students needs to start early. Given that most TWI programs begin in kindergarten, this exposure at a young age is an inherent strength of the model. The nature of TWI programs presents an opportunity for education policy, not just for bilingual education, but also for integration. In addition to the direct benefits to the students, integration through TWI could promote the foundation that is needed for a diverse, democratic society (Mickelson & Nkomo, 2012). Despite this potential of TWI programs, researchers have yet to explore the ways in which TWI programs facilitate integration. It is this gap in the literature that the current study seeks to address.

Methods
This qualitative research project followed a case study methodology (Yin, 2018). Data collection took place in one elementary school and included interviews with teachers and other school personnel, observations of TWI classrooms, and analysis of relevant documents such as school board meeting minutes. Although we believe this case has implications for education policy and practice, findings should not be perceived as generalizable.

Data Collection
Our research was conducted on-site at an elementary school over the course of two months. The second author recruited participants at a staff meeting. We interviewed all eight current Spanish TWI teachers, four of the six current TWI English teachers, the school guidance counselor, and the principal. The guidance counselor, along with knowing the students in the program, also had a son in the TWI program.
Our interview protocol questioned what the participants perceived as the benefits and challenges of TWI, how they saw students interacting in the classroom, how students were grouped and provided opportunities to collaborate, and how students' backgrounds were incorporated into the classroom environment. Given our interest in integration, we specifically asked about race, ethnicity, and language. Interviews were semi-structured to allow for probing and follow-up questions. Interviews mostly occurred in the participant's classroom, although in a few instances we met in private spaces in the media center due to space restrictions. The interviews were audio recorded and lasted between 20 and 90 minutes. Recordings were transcribed by an outside vendor, and the first author read each transcript to confirm accuracy.
We also conducted eight observations in TWI classrooms, six of which were being taught in Spanish and two in English. Observations included lessons in math and reading, centers, health, and indoor recess. We focused on how students interacted with each other in both structured and unstructured situations. Observations in some classrooms lasted only 20 minutes, while others were over an hour and a half.

Data Analysis
We began open coding by reading through two transcripts together to establish a general consensus on codes. We then uploaded transcripts to an online qualitative coding program. Our codebook included 24 initial codes and sub-codes. The codes relevant to this article are: students working together; close knit group; and advantages associated with TWI, a parent code which included bilingual, interact with people not like you, broaden perspectives, exposure to other cultures, rigorous, and open doors in the future. These 8 codes mapped evenly onto four themes that captured the essence of integration in the TWI classroom. Two remaining codes, English default and changes in enrollment, helped us generate an additional theme describing the challenges associated with TWI. We triangulated interview data with observations and documents from the school website and school board meeting minutes.

Silverthorne Elementary
The site where our research took place is a rural school district in North Carolina. While populous counties like Wake and Mecklenburg have seen substantial growth in the Latinx population (Gándara & Mordechay, 2017), many rural counties in the state have experienced a population boom as well. The share of Latinx school-aged children in our site's school district is comparable to the state average of 16% (Gándara & Mordechay, 2017). In response to changing demographics, the school district began offering TWI strand programs in the 2013-2014 school year. While it is not the only county in the state to recognize both the need for English acquisition and the benefits of instruction in another language such as Spanish, this district showed a commitment to equity among racial, ethnic, and linguistic groups with its preference for TWI over other options like one-way immersion.
The town in which our school resides has a population of approximately 2,000. The school, Silverthorne Elementary (pseudonym), enrolled 700 students in the 2018-2019 school year. It is a Title I school, and at 32%, the Latinx population is higher than the county average, likely due to the existence of the TWI program; the school also enrolls 14% Black students, 49% White students and 4% two or more races (National Center for Education Statistics, 2019). Observational data indicates that enrollment in the TWI program was approximately 59% Latinx, 32% White, and 9% Black. The school provides instruction from kindergarten through fifth grade, and teachers from each grade level participated in our study. Each grade has two TWI classrooms, one taught by a Spanish teacher and the other taught by an English teacher. The kindergarten and first grade classrooms each have a teacher's assistant. In Silverthorne's model of 50% English and 50% Spanish instruction, students alternate between classrooms daily, Monday through Thursday; Fridays are split between the two teachers.
While we did not specifically collect demographic information from teachers, most of them volunteered their country of origin. The English teachers and the guidance counselor are from the state where we conducted this study. One White English teacher, Dawn Cortez, is married to a man of Mexican descent. The Spanish teachers had moved from Puerto Rico, Central America, and South America (Table 1). Among the Spanish teachers, some had come through a program that places international teachers in dual language classrooms, while others had moved their families to the state to teach in the school. All participating teachers had more than five years of teaching experience, with an average of 13 years. Spanish teachers, many of whom had taught for 10 or more years in their home countries, began at Silverthorne as co-teachers or teacher assistants, with the opportunity to have their own classroom after one year. We created pseudonyms for all our participants.

Limitations
Although we sought participation from all TWI teachers, two of the TWI English teachers declined to participate. Because we did not interview students in TWI, the conclusions we draw about their interactions are based solely on observation and participant insight. We were able to read some lessons and student work while on-site, but we did not collect additional artifacts. Although we had planned on member checking, we were not able to share our findings with participants due to the interruption caused by COVID-19. The context of this county offers a unique perspective of TWI in a rural setting, but may be different socioeconomically, demographically, and culturally than a more urban setting would be, so our recommendations should not be considered generalizable. Despite these limitations, the combination of our interviews and observations allowed us to collect rich data that provides for an in-depth description of how TWI promotes integration at Silverthorne.

Positionality
Both authors of this study are White, monolingual, cisgender women from the South, which might have affected the responses of our participants. Furthermore, we recognize that our backgrounds inevitably influenced our approach to this project. We are also native English speakers, and the majority of our participants were native Spanish speakers. Although many participants began the interviews by claiming their English was "not so good" or even "bad," this language difference did not present as a barrier for us. We did, however, have several opportunities to experience what students experience in the TWI classrooms during our observations. Neither of us is fluent in Spanish, and on several occasions, we were "in the dark" about what was happening in the lesson. While we do not believe that this situation seriously hindered our data collection, we are aware that we inevitably missed some of the classroom interaction.

Findings
True integration promotes equity among historically marginalized and historically advantaged groups through a program that acknowledges the value of different abilities and embraces diversity in collaborative, mutually beneficial environments. Based on our findings at Silverthorne Elementary, we propose that TWI can facilitate integration in four central ways: The program attempts to create equal status among linguistically diverse students, uplifts languages other than English by helping students become bilingual, explores global cultures by helping students become bicultural, and provides an environment where all students may be successful. Along with the benefits of this program, there are some challenges as well, due in part to the way TWI was implemented at this school.

TWI Attempts to Create Equal Status
A key facet of an integrated classroom is equal status among students from different cultural backgrounds (Allport, 1954), which TWI enables by creating a language balance between English and the partner language. By having a classroom where all students are emergent bilingual, not just the ELs, language acquisition acts as the leveler for the effects of segregation and isolation. The classroom environment is one in which students are encouraged to collaborate and learn from each other (Ayscue & Uzzell, in preparation). This equal status has the potential to facilitate interaction across racial and ethnic lines, strengthen student relationships, and promote positive collaboration through shared challenges.
Being in a classroom that encourages acquisition of English and Spanish is particularly beneficial for students who would otherwise be classified as ELs and might find themselves lost in a mainstream English classroom. In our case, this approach meant that Latinx students brought similar value to the classroom as non-Latinx students. Having Latinx and non-Latinx peers who speak their language helps pull Spanish-dominant students out of linguistic isolation. This point was made by Mrs. Cortez: "The dual language is good for helping those students not feel so isolated and struggle so much." When all students in the classroom are emergent bilingual, it is likely that fewer will feel excluded and confused because of the language barrier. Most participants said their students interacted well with one another despite racial differences, which was confirmed by our classroom observations. For example, in a second-grade mathematics lesson, students chose partners in a speed dating style activity. When given the option, students seemed comfortable finding a partner and working with someone of another race or gender. With each mixing, all six White students worked with a student of color; it was only in the third pairing that the two White boys chose each other.
Also, it was notable that while students were seated at their desks, their assigned seats were usually next to someone of a different race, a choice made by Ms. Speller, the second-grade English teacher.
Along with exposure to students from different backgrounds, TWI can facilitate building relationships among students, as well as between students and teachers. The TWI environment helps students form friendships across racial lines, as Principal Gutierrez noted, "It's not just the intellectual benefits, it's the interpersonal relationships, et cetera, from students and everyone." In the fourth grade classrooms, we observed students getting along and working well in racially diverse, assigned partners during structured class time, but it was their informal interactions that we found most telling. During indoor recess time, students chose games to play and some formed groups: a Latino boy and a White girl played a board game, while two Latina girls and two White girls sorted through blocks to construct houses. Our participants shared that student friendships are formed both inside and outside of the classroom. Mrs. Richards, the guidance counselor whose son participated in TWI, shared, "I've loved that my son has friends that he can sit down at the [lunch] table with and they speak in Spanish." Students in the TWI program are able to form friendships because of their shared status as emergent bilinguals. They can also capitalize on interactions with their bilingual teachers. As Mrs. Cortez observed about her colleague, Mrs. Diaz: "They'll talk to her in Spanish, and they can have full conversations and can relate to their teacher. Versus if they were just with an English teacher, we can't do those things." Along with the students communicating with and learning from each other, the TWI program helps create positive relationships between students and teachers.
Instances of students working together and helping each other were common throughout the data collection. Sometimes this cooperation happened naturally, and sometimes it was facilitated by the teacher. Ms. Veracruz noted, "Well, in my class I make the Hispanic students who speak Spanish work with the Americans who don't know [Spanish], so they help each other. … I tell them, 'Okay, if you don't know any word, your friends know Spanish, ask them.' That's how they interact and work." This peer teaching is an essential feature of the TWI program, as Mr. Rodrigo expressed: They're going to learn how to speak in English from the English speakers, as well as the English speakers to learn from them. So they respect each other. They know that they are facing the same difficulties even if they are from different backgrounds. So they have the same objective and that is great to see them working together in different languages.
The shared challenge of learning in two languages helps students develop mutual respect from a young age, and students are able to act as language models for each other in a natural, collaborative way. Mrs. Bautista remarked, "There's always cooperative learning, which helps a lot for a dual language where there are some students that have less vocab than others… That's for me, very important for them to be sharing and helping each other." Although the emphasis on collaboration is not unique to the TWI classroom, building that skill may be facilitated by TWI, as Principal Gutierrez stated: "When you learn to negotiate meaning, then you learn to negotiate other things and work with other people. And that's part of how we're preparing the students for the future, not just about Spanish or English or two languages. It's learning to interact with other people who are different from you." Thus, the environment in which students learn from and help each other is facilitated by the TWI program, and equal status, which is a critical aspect of true integration, may be provided in that environment through the language balance.

TWI Can Help Students Become Bilingual
TWI gives students the opportunity to become bilingual, which can include many academic, cognitive, and social benefits. Participants in this study, more than half of whom were bilingual themselves, expressed that learning in two languages helps students with understanding difficult concepts and often provides an opportunity for deeper learning. Additionally, bilingual education challenges the English hegemony, or what we referred to in coding as "English default," since participants expressed that some students slipped into English during Spanish class. This challenge presents an important alternative to English-only policies, and it might help provide teachers with empathy for what their ELs experience, both within and outside of TWI settings. Bilingualism offers a lifelong benefit that enables students to interact and communicate with other cultures, ensuring their participation in a diverse society.
Several participants expressed their belief that knowing two languages could help illuminate concepts for students. Mrs. Diaz, the third-grade Spanish teacher, mentioned, "I think they can learn better and maybe if one of them can get it in English, can get it in Spanish and the learning will be there no matter what, in Spanish, in English." This kind of learning is aided by the teachers often giving the same lesson in both languages, so that difficult concepts may become more accessible. In the third-grade classrooms we observed, the students were learning about bones (huesos) and creating skeletons labeled with Spanish terms. Similarly, each of us observed a second-grade math lesson on even and odd numbers in which the concepts seemed complex to the students, but the learning was reinforced with multiple opportunities in each language. Mrs. Alvarez, the fourth-grade Spanish teacher, attributed this process of understanding to the neural pathways created by studying in two languages: If they do not understand something in English and they come to the dual language with me, in Spanish, they can understand, and they can develop the outcome in a different form, that is, a different way. I believe that something happens in the brain for them just with that exercise, switching on and off, that is amazing for me.
These teachers underscored their belief that learning the same concept in two languages provides students with a greater depth of understanding. An appreciation and focus on bilingualism through TWI may also help address the English hegemony. The preference for English above all other languages, sometimes expressed in Englishonly policies, is a symptom of segregation, xenophobia, and white supremacy. We noticed during observation of the fourth-grade Spanish class that students were English dominant regardless of race, and several of the Spanish teachers commented that their students would sometimes slip into English. The teachers got their students back on track without being punitive; if students working in groups spoke English during Spanish class, Mrs. Bautista would joke, "'oh, that's, wow. I didn't know that was in Spanish,' because they were speaking English." Indeed, there were a few students in her fifth-grade classroom who worked together in a vocabulary center on Spanish words, but helped each other understand the objective by speaking English. Mrs. Bautista also shared that even students who had begun learning Spanish from their parents at home may be English dominant "because maybe parents were raised here in the States and it's the grandparents that belonged to South America or to Mexico." Still, enrollment in the TWI program foregoes the American preference for replacing the home language with English, as Mrs. Gutierrez observed, "It used to be that Hispanic parents would [say], 'No, I want him to learn English and then delete that native language,' and now they see the value in, 'Let's continue to learn Spanish, continue to improve the skills in your native language, and learn English as well.'" Despite the tendency to slip into speaking English, students, teachers and parents recognizing the value of bilingualism helps everyone move away from the norm of using English as a default.
One of the benefits of TWI is the challenge to English hegemony, but this can create difficulties as well. Often this sentiment was expressed by the monolingual English teachers, who seemed to get a sense of how their Spanish teacher counterparts and even their ELs might feel. When asked about the challenges of TWI, Mrs. Cortez said, "A challenge that we've had this year, and it's a good challenge to have, we have two students that just came from Honduras. So they speak absolutely no English, which you will have in a typical classroom too." She recognized that the same EL students would face even more difficulty in an all-English classroom. Mrs. Palmer sympathized with American students in the program because an "English-speaking child, of course, is going to struggle at first because they simply just don't understand what the [Spanish] teacher is saying. But that's how a Hispanic kid feels when they're here for the first time too." Just as increased cultural awareness may help students see that the world is bigger than the town where they live, dual language can help them see that there are other languages besides English and those languages also have value. Ms. Garcia, the second-grade Spanish teacher, summed it up simply: "I love my language. So I want kids to love the language, too." Promoting bilingualism rather than English acquisition alone suggests that TWI supports the cross-cultural interactions needed for true integration.

TWI Can Help Students Become Bicultural
Along with the goal of becoming bilingual, TWI exposes students to other cultures and can help them become bicultural. Students learn about other cultures from their teachers and classmates, who are often from other countries. This learning helps broaden students' worldviews and promotes global awareness. Students are able to and encouraged to interact with classmates who are different from them. Becoming bicultural aids in integration as students can understand and appreciate people who are unlike themselves.
Learning another language is often accompanied by gaining understanding of the customs and individualities of the countries where that language is spoken. As Mr. Rodrigo, the first-grade Spanish teacher assistant, commented, "It is one way to bridge the gap between cultures. And when students, when people learn a different language, they also learn about a different culture. So basically what we are trying to do is to build bonds between different countries, different cultures, too." Ms. Garcia also commented, "Being bilingual, it's not just the language. They get to know the culture of listening to Spanish for the Hispanic countries." Many teachers mentioned sharing about their country's traditions with students and fellow teachers, and most classrooms had a display about that country. Kindergarten teacher Ms. Blanca said, "The students that are not Hispanic can embrace our culture and can be exposed to our culture, our language. And we as teachers, we share that part of us with them." Spanish teachers who celebrate their country of origin help students understand that Latinx people are not a monolithic group. As students learn about many other countries, they develop interest in those countries. Ms. Veracruz proudly shared, "When I teach my country, they were like, 'Oh my God, I want to go to Ecuador. Your country's awesome.'" Beyond exposing students to other countries, the TWI program encourages students to appreciate them.
Learning about other cultures can also expand students' worldviews, as Mrs. Bautista noted, "Students will be able to explore another world, [which] gives them a big perspective that the world is not [just] where they live. It's more than what they see." For students in a rural community, the opportunity to learn intimately about other countries can combat xenophobia that they may experience outside of school. Mrs. Bautista was also able to video conference with a classroom in Bogotá, in her home country of Colombia, offering students access to that world beyond their town. Mrs. Cortez pointed out that the push to incorporate other cultures was exciting because it was different from how students' cultures have been treated in the past: They learn more about Hispanic culture. … And the kids get real excited, and the parents too, because it's being celebrated in a positive light instead of a negative connotation. … It's really awesome because the kids learn so much, and they get to be excited about their own heritage instead of it being a taboo kind of thing.
In essence, TWI encourages cultural inclusion rather than exclusion or assimilation. Mrs. Alvarez saw cultural inclusion as a "natural" extension of the TWI curriculum, while Mr. Montez, one of the kindergarten Spanish teachers, shared that his reference to other countries was often spontaneous: For example, I would say to them that in Honduras, we say for kids we say cipotes [Salvadoran or Honduran slang for kid], and they don't even know what's that, because it's a word that we use in my country. … We would make a little discussion about that, that in other countries the kids are called in different ways, like in Mexico chamacos [Mexican slang for kid], like in Honduras cipotes.
Thus, with the TWI teachers pointing out and celebrating cultural differences, any stigma is removed from the topic and students learn more about the world in the process. TWI also gives students an opportunity to interact with people who are not like themselves. In today's climate, when cultural differences are exaggerated and keenly felt, a program that acknowledges and affirms those differences is especially important. The students also benefit from "having teachers from different countries," working alongside American teachers, according to Mr. Montez, because that "is a good role model for kids in a way to teach them that we work as a team." Mrs. Richards acknowledged, "I don't know that [my son] would have been best friends with a Hispanic child or learned anything about their culture, if he hadn't been involved in this program." The TWI classroom can provide a positive forum for all the students to experience intercultural interaction.
As beneficial as many participants found cultural inclusion to be, others found it challenging. Both first-grade Spanish teachers expressed that their students did not notice race or ethnicity, and they did not want to emphasize cultural differences because they wanted to avoid conflict (Ayscue & Uzzell, in preparation). However, both teachers were willing to share with students about their native countries. Spanish teachers seemed to think of cultural inclusion as applying to Spanishspeaking countries, and English teachers often did not see cultural inclusion as their responsibility at all. We found it difficult to see how non-Latinx cultures were overtly represented in the TWI classrooms. For example, during a fifth-grade observation, students in one reading group read a book about the history of Native Americans and tribal differences, but there was no mention that indigenous people still exist today. Another group read about the Civil War, and although the book recognized slavery as the cause, the voices and actions of White abolitionists (abolicionistas) were uplifted more than those of the enslaved people. In these cases, Native and Black cultures are relegated to the history books, with no sense that these cultures continue to exist and contribute to this day. This experience gave the sense that even in a Spanish classroom, these topics were whitewashed. In this instance, the emphasis on becoming bicultural may actually have inhibited students from being multicultural.

TWI Can Provide an Environment Where ELs and Non-ELs Can Be Successful
With its ideal language balance (50/50), the Silverthorne TWI program provides a potential "win-win situation" in which all students can be successful. TWI can be beneficial for historically advantaged and disadvantaged students. The curriculum is challenging and offers a depth of understanding for students. TWI offers opportunities, such as learning in another language, that are not available in a non-TWI classroom. The benefits reaped from the program can have lasting effects beyond the classroom and may serve to open doors for students in the future, laying the foundation for integrated workplaces and a diverse, democratic society.
The TWI program, according to our participants, is more rigorous than the general education curriculum. This rigor might be due to the setup of the TWI program at this school, in which students alternate between Spanish and English classes every day. According to Mrs. Richards, the workload is "tough" in part because students have "double homework." For example, they have two sets of spelling words instead of one. However, for her son and for other more advanced students, the TWI program offers a needed challenge. Fourth-grade English teacher Mrs. Palmer noted that the challenge has paid off: "Our data has shown from the past three years that the kids that are in the dual language programs are performing better [than non-TWI students at Silverthorne] on both reading and math EOGs." While this outcome is consistent with data showing the academic benefits of TWI (e.g. Steele et al., 2017;Thomas & Collier, 2014), Mrs. Palmer attributed her students' success to their struggle with the material.
Not only do the academic benefits show when teachers compare data on end-of-grade (EOG) examinations, but they also are noticeable in the classroom in the way TWI students approach difficulties and persevere to address them. Fourth-grade Spanish teacher Mrs. Alvarez sees that the program is challenging for students, both ELs and non-ELs, who may need to work harder to understand something that is not in their home language: The challenge that I found on them is that sometimes they don't understand a word, for instance, students that are American, that their families speak English, sometimes it's very easy for them to be in the English class and just all of the reading passages that they have, and here [in the Spanish class] it looks like they have to pay more attention. They have to make a big effort and that happens the same to the students that their first language is Spanish.
The TWI classroom can make students more conscious and aware during the learning process, and instill the necessary skill of perseverance. The difficulty of the TWI program at Silverthorne was used during initial implementation to attract academically gifted, mostly White students, and still has that appeal after five years of operation. During observation of the third grade Spanish class, we encountered three White students who had just entered the program because, according to the teacher, they needed the extra challenge. Another opportunity that TWI may provide is the lifelong skill of knowing another language. For Mrs. Cortez, the timing of this program is important as well: For students, it's a huge benefit because it is so much easier to learn a second language when you're young. It's so much easier. And if they start off in kindergarten being completely fluent, especially when they leave to go to middle school, reading, writing and speaking Spanish, I mean, that's a lifelong skill that's going to benefit them in work and everything else.
Although students in elementary school may not be considering their futures yet, the teachers are excited about their students' potential after being in this program. The benefits of TWI can extend beyond the classroom, as students carry the skills they acquire with them. Not only that, TWI teachers are communicating the potential to the students, as Mrs. Bautista stated, "One of the things that I am always telling them is that they have, they always have advantages. They have one step ahead of other students that's around them because that will open a lot of doors for them." Beyond advantages over other students, Mrs. Alvarez expressed specific goals for her students: "It gives them lots of opportunities that they have another language. What I tell them, when they grow up and they finish college, they're going to have the possibility to talk with people that are Hispanic." Mrs. Richards, as the guidance counselor, directs students toward specific college and career advantages: "We talk about careers and their future. … I always say, 'You guys have an advantage over most of the world. You have English, you have Spanish, and there are careers out there that are going to be begging for you to come work for them.'" As TWI students become young adults, their earning and job potential increase due to their ability to speak another language (Rumbaut, 2014), as many employers do in fact prefer bilingual employees (Porras et al., 2014). Thus the benefits of TWI can be lifelong. The principal equates cultural proficiency with the ability to be a productive member of society: "It's just a life skill, it's the future. You know, they need to be able to function in a society where not everybody looks the same. Not everybody speaks the same language. Not everybody has the same accent." TWI can help develop academic, social and even economic benefits of being bilingual and bicultural for all students who participate in the program, so that they may experience the lifelong benefits of integration.

Challenges Due to Program Implementation
Some of the aforementioned challenges that students and teachers at Silverthorne Elementary face-such as managing the rigorous workload, incorporating cultural inclusion, and teaching monolingual students-might be due more to the way TWI is implemented rather than to the program itself. These challenges raise potentially important considerations for districts that want to use TWI as a tool for integration. The most distinct difference between Silverthorne and other schools with TWI strands is that the former has separate classrooms for English and Spanish, while some of the latter practice immersion in one classroom where both languages are spoken. This class structure means that Silverthorne can enroll more students in TWI, which results in a burdensome workload for everyone and reduced time with students. The number of students and the changes in who enrolls in the TWI program presented a challenge for participants as well.
The study participants often cited the workload, for both students and teachers, as a challenge. Mrs. Cortez emphasized the burden on teachers: "We have twice as many grades, twice as many sheets to grade, twice as many report cards, twice as many parents to deal with, you know, twice as many behaviors and just all the things that a normal classroom has, we have twice as much." Mrs. Alvarez also struggled because "it's double everything, the planning, the centers, because I have different levels. It's always a challenge that, the amount of students." Our participants who had their own children in the program also noted the workload for students. Mrs. Bruce said, "I mean, it is a lot of extra work for the kids. It's a lot on them. My son typically has to learn it in both languages...It's a lot more for students to have to understand." The heavier workload seems to account for the perception that only higher-performing students can handle TWI. However, it also increases the potential academic benefits and holds students to high expectations.
In addition to a heavy workload for everyone, the TWI class structure at Silverthorne also results in teachers having less time with their students. Because TWI teachers at Silverthorne switch off English and Spanish days, they see their students for half the time compared to non-TWI teachers. According to Mr. Montez, this reduced time results in a necessity to cut non-academic instruction: "We struggle with time...because in my country, I take an hour or 30 minutes to talk about topics that are not academic, that are things that I think that they might know, to be humble, about responsibility, about teaching values." Nevertheless, while the class structure means that the teachers face the challenge of having less time with their students, it also creates the opportunity for students to learn from two different teachers with different backgrounds, experiences, and approaches to learning. Thus, in this way the class structure may contribute positively to integration.
Alongside the benefits of expanded access to the program, the changes in enrollment in TWI also have created some challenges at Silverthorne. Although this program started at Silverthorne with a group of high-performing, mostly White and Latinx students who seemed able to "handle" the challenge, it has since evolved into a program that encourages anyone to enroll. We inferred from the teachers' comments that most Latinx students were referred for enrollment in the program, while it was becoming increasingly harder to enroll non-Latinx students. Mrs. Cortez said, "Last year and this year have been very, very heavy with Hispanic." One possible explanation for this change in enrollment, offered by Mrs. Richards, is the turnover of teachers in the lower grades, which makes some parents, particularly White, native English-speaking parents hesitant to enroll their children. Another explanation, which was not mentioned by participants but can be inferred, is the influence of racism and xenophobia. Regardless of the explanation, heavier Latinx enrollment disturbs the requisite linguistic balance, which poses a challenge to linguistic, racial, and socioeconomic integration.
We note these challenges with the intent of informing anyone interested in implementing this program to be aware of the potential drawbacks of this model. It is possible that a different model, such as one in which students learn English and Spanish in the same classroom without the need to transition, would not result in the same heavy workload. Despite these challenges, students enrolled in the TWI program at Silverthorne experience a learning environment in which teachers seek to create equal status among the students, and where students may be able to become bilingual and bicultural. Successful implementation requires significant and sustained effort, which may not always be realized, such that all students are not able to reap these program benefits. Under ideal circumstances, the many benefits of the TWI program can promote integration among all the students who participate in the TWI classes.

Discussion
TWI can be an effective tool for integration, creating environments of equal status and opportunity for all students, especially those who have been historically marginalized (Gándara & Aldana, 2014;Kotok & DeMatthews, 2017;Orfield, 2013). The case of Silverthorne Elementary provides an example for other schools wishing to pursue integration through TWI, as it offers a mutually beneficial environment in which Latinx and non-Latinx students can become bilingual and bicultural. While there are some challenges for students and teachers, many of which might be due more to implementation than to program theory, there is still important knowledge to be gained from this particular TWI program. TWI is likely to be most effective when implemented with a goal of integrating diverse student populations.
Although our observational findings on the racial balance in the classroom are consistent with other research (deJong & Howard, 2009;Palmer, 2010;Valdez et al., 2016) showing that mostly Latinx and White students enroll in TWI, our experiences at Silverthorne Elementary were encouraging. There were Black students enrolled in the program, especially in the lower grades, and on average made up 9% of the TWI classrooms. Teachers in Silverthorne's TWI program strive to create equal status among the students, and the voices and experiences of Latinx, Black, and White students are heard and elevated. It has been proposed that TWI has the potential to close achievement gaps (Groom & Hanson, 2013;Steele et al., 2017;Thomas & Collier, 2014) for students of color in general and Latinx students in particular. Latinx students often suffer from triple segregation, experiencing gaps in opportunity along with the inadequate support and linguistic isolation that can come with being labeled as ELs (Gándara, 2010). The TWI program at Silverthorne Elementary addresses the language barrier by providing ELs with access to rigorous curriculum in their home language and adequate resources such as qualified bilingual teachers and English-speaking peer role models. Additionally, non-Latinx students receive the same curriculum with the additive benefit of learning another language alongside Spanish language role models.
Students at Silverthorne have the opportunity to become bilingual and bicultural, which provides academic as well as social benefits. Although there are challenges to doing so, our observations at Silverthorne revealed that a TWI program can embrace the objectives proposed by Valdés (1997) effectively. TWI teachers at Silverthorne demonstrated a commitment to academic Spanish, which was given a level of importance equal to that of academic English. The challenge to the English hegemony in turn can act as a challenge to the dominant culture and the norms of whiteness (Flores, 2016). Regarding the development of relationships among the students, the teachers encouraged interactions between racial groups in class activities, while socializing among students outside of class seemed more natural than deliberate. It has been proposed that these interactions aid in cross-cultural awareness as students are encouraged to understand and embrace differences (Lambert & Cazabon, 1994;Lindholm, 1994;Lindholm-Leary, 2003). Efforts to empower language-minority students in Silverthorne's TWI program included having both English and Spanish students help each other, and a commitment to learning about and respecting various cultures. Gándara (2020) describes potential equity pitfalls that must be addressed if TWI programs are to facilitate integration. To ensure equitable access, no more than 50% native English speakers should be enrolled in the TWI program; without this guideline, it is possible that the demand among middle-class White families could push out low-income Latinx families. Additional equity issues include developing TWI programs in locations that are accessible to low-income families, choosing a partner language that is beneficial for ELs in the local context, assessing student learning in both languages, and addressing the shortage of bilingual teachers. In our study, Silverthorne attempts to enroll a balance of ELs and native English speakers, but that has become a concern in recent years. Unlike the equity issue of opportunity hoarding among White families, however, Silverthorne has experienced a decline in White participation, resulting in more Latinx ELs enrolled in the program than is ideal. While this may upset the balance of partner language speakers, it may also present an opportunity to address discrimination in education and bring the focus back to historically marginalized students (Flores & García, 2017). We did not get a clear answer for what caused the decline in enrollment, so it is plausible that racism and white supremacy played a role. Additionally, the location of the school, situated intentionally in the rural community where Latinx students live, and the choice of Spanish as the partner language are helpful for promoting equity in the TWI program at Silverthorne. Finally, partnering with an organization that recruits bilingual teachers from abroad has helped Silverthorne address the shortage of bilingual teachers in the local community. Similarly, Kotok and DeMatthews (2017) provide recommendations for districts implementing dual language programs: Promote programs with the intention of increasing diversity, make efforts to find and develop dual language teachers, pursue the program that fits community needs and values, and create a system that supports the long-term implementation of a district-wide DLI program. The school district in this case study follows many of these guidelines, and could prioritize diversity by paying attention to enrollment in TWI programs and by providing resources and materials that facilitate cultural inclusion.
The success of TWI at this school may provide an example for other schools and districts as well. The principal, Ms. Gutierrez, saw how her school was helping not only her students, but also the district to be more open-minded: "It's been a great thing for the school and for the county. I think the county, they're now realizing that this is the way we want to go." The school district in which our study took place had intentionally implemented TWI in a number of its schools in order to meet the needs of the growing Latinx population, and could add more programs. Although facilitating integration was not the goal of implementing TWI in this district, it is an unintentional benefit of the TWI program. At Silverthorne, the TWI program demonstrates an avenue through which districts can pursue voluntary desegregation and halt resegregation, in spite of Supreme Court rulings and in the absence of a statewide language policy.
Although one may be used to facilitate the other, TWI and integration have a lot in common, including the attempt to create equal status among racially and linguistically diverse groups, the need for community support and a program rooted in the local context, and a promise of long-term academic, social, and economic benefits. Both are often called on to respond or conform to norms of the dominant culture. Both have to be "sold" to attract middle-class White families by providing advantages, and are thus prone to resource hoarding. The success of both programs may vary depending on implementation, as we saw with the challenges at Silverthorne Elementary, though the benefits seem to outweigh the challenges. Although not central to our study, many participants mentioned how beneficial they felt this program was, and how "it should be everywhere." The interactions across racial and linguistic lines in TWI classrooms at Silverthorne were reminiscent of a truly integrated learning environment (Orfield, 2013). We agree with Kotok and DeMatthews (2017) that implementation should be rooted in the local context, and we believe that many communities across the United States would be able to implement TWI to further integration efforts.

Implications for Policy and Future Research
While its implications for local education agencies are encouraging, federal support for integration through TWI would be helpful. For example, a federal language policy that supports and encourages bilingual education, specifically TWI, would go a long way in promoting implementation of TWI programs and integration of students from different racial and linguistic backgrounds. Also, the federal government can develop and support grant programs that would allow local districts to implement TWI programs for integration purposes. Even amending Title III of ESSA to specifically mention TWI as a viable bilingual option for educating ELs, and providing adequate grants to support implementation, may help accomplish the goals of integration and bilingualism. Funding should be accompanied by technical assistance for districts and schools, and guidance for how to implement equity-centered programs that foster true integration and challenge the status quo of white supremacy and English hegemony.
Similarly, state governments should provide financial and technical support for districts and schools that are implementing TWI for integration. At the local level, school districts should consider TWI programs as a tool for integration and should do so with consideration of the larger context of the district and how such programs would impact desegregation across the district. They must also ensure that programs have access to adequate resources and materials in the partner language. One of the most valuable resources in a TWI program is the teachers. Therefore, support and possibly additional compensation that aligns with the increased workload described in our study are necessary to recruit, develop, and retain TWI teachers. Because integration benefits students of all races, schools and districts should monitor TWI programs to ensure that they are serving all students and therefore allow the benefits of integration to accrue, rather than allowing for opportunity hoarding by White families or stigmatization resulting in only Latinx families enrolling in TWI.
This study brought up several interesting avenues to pursue in future research regarding race and racism, xenophobia, white supremacy, and English hegemony. As previously stated, neither of us speak Spanish, though we recognize the variety of Spanish dialects that our participants brought to their classrooms; future research might investigate how these differences manifest, and whether schools pursue more formal versions. We are also interested in further research into raciolinguistics (Flores & Rosa, 2015), and how race and racism influence integration in programs like TWI. While we are interested in learning how race and racism affects students in TWI, the current study did not collect data directly from students, so we are interested in a follow-up study to understand the student perspective.
Additional research is needed to gain a deeper understanding of how other districts in North Carolina and around the country are implementing TWI programs as a tool for integration. Heretofore, most TWI programs have been available only in elementary school (Howard, Sugarman & Christian, 2003). Therefore, the opportunities for integration presented in earlier grades may not be sustained in secondary school. Research about TWI programs in secondary schools would be useful and could perhaps encourage the growth of such programs if found to be beneficial. A longitudinal analysis of students who have participated in TWI programs could explore the ways in which the benefits of integration through TWI persist over time. As resegregation continues to intensify across the country, the potential of TWI programs as a tool for integration must not be lost.