Exploring Quality Programs for English Language Learners in Charter Schools : A Framework to Guide Future Research

Although there has been a great deal of debate about the effectiveness of charter schools in the research literature, there has been surprisingly little attention paid to English language learners (ELLs) in charter schools. Moreover, the charter school research has predominantly focused on whether or not charter schools are effective rather than how or why high-performing charter schools work, particularly for ELLs. We contend that researchers must expand their focus beyond access and achievement and begin to grapple with questions related to the quality of programs for ELLs in charter schools. To meet an emerging need in the field, we synthesize several strands of existing research—related to charter schools, school improvement, and ELLs in traditional public schools— to propose a five-component framework that describes essential elements of quality programs for ELLs in charter schools. We conclude with a discussion of implications of our framework for research, policy, and practice. Education Policy Analysis Archives Vol. 24 No. 53 2


Introduction
Although there has been a great deal of debate about the effectiveness of charter schools in the research and policy literature, there has been surprisingly little attention paid to the quality of programs for English language learners (ELLs) in charter schools.Moreover, the charter school research has predominantly focused on whether or not charter schools are effective rather than how or why high-performing charter schools work, particularly for ELLs.After setting the context with relevant background information, we describe our review of the literature related to charter schools and ELLs.Based on our findings, we propose a framework outlining key components of charter with tax dollars, charter schools are not permitted to discriminate when enrolling students (Henig, 2008).While charter school advocates argue that ELLs are served equitably in charter schools (National Alliance for Public Charter Schools, 2014), some researchers and critics have claimed that ELLs are underrepresented in charters (Buckley & Sattin-Bajaj, 2011;Frankenberg, Siegel-Hawley, & Wang, 2011).In 2013, Congress charged the Government Accountability Office (GAO) with examining this claim, but the GAO researchers were unable to complete the study because enrollment data from the U.S. Department of Education were unreliable or insufficient.
However, some researchers have been able to use state or local data to answer this question.For example, in their 2011 study, Buckley and Sattin-Bajaj examined the enrollment of ELLs in charter schools and traditional public schools in New York City.They found that most charter schools enrolled a smaller percentage of ELLs than did the district, but noted that there was also variation in the sample, with some charter schools enrolling a similar percentage of ELLs or a greater percentage than the district average (Buckley & Sattin-Bajaj, 2011).A follow-up study conducted by Winters (2014) came to a similar conclusion.He found that a key reason to explain the underrepresentation of ELLs in New York City charter schools was due to the fact that ELLs were less likely to enroll in charter schools during gateway grades, such as kindergarten for elementary school and ninth grade for high school (Winters, 2014).
Further, in her study of ELLs' access to charter schools in 10 urban districts with high populations of ELLs that were located in 10 different states, P. Garcia (2015) found that ELLs in charter schools were significantly overrepresented in three districts, significantly underrepresented in four districts, equitably represented in one district, and that the quality of the data was not sufficient to make a determination in two districts.Consistent with Buckley and Sattin-Bajaj's (2011) finding, P. Garcia also found that proportions of ELLs ranged widely across both charter and traditional public school portfolios, with some schools serving small proportions of ELLs while others served much larger proportions.Certainly, improving the quality of the available data is a critical first step to help researchers determine if ELLs are equitably served in charter schools.Moreover, as concentrations of ELLs vary across the country, it is likely that the proportion of ELLs served in charter schools also ranges widely by state and community.Consequently, further research is necessary-at national, state, and local levels-to examine if ELLs are being equitably served in charter schools across a range of communities.
ELLs' access to charter schools has become the subject of much debate in the policy world (Stewart, 2010), but more research is needed to better understand if ELLs are equitably represented in charter schools, and if not, why that is the case.In her study, P. Garcia (2015) found that one potential explanation for the overrepresentation of ELLs in the districts she studied was that charter school design teams might have strategically established their attendance zones to be inclusive of communities with large populations of ELLs.In some states, such as Oklahoma and Missouri, design teams can propose attendance boundaries specifically for their school (Missouri Revised Statues,§ 160.410.1;Oklahoma Statues,.These policies present a unique opportunity for design teams who are interested in developing ELL-specific models because they can select a particular location, within a district or a state, and then strategically draw an attendance boundary that maximizes the school's ability to serve communities with large concentrations of multilingual residents. In districts in which ELLs are underrepresented, researchers have proposed three sets of potential explanations-related to outreach, mission, and structural elements-to explain the enrollment gap.In terms of outreach, some critics have suggested that charter schools might deliberately exclude some students, specifically ELLs and students with disabilities, in order to increase the school's test scores (Buckley & Sattin-Bajaj, 2011).In addition, families of ELLs might not have accurate and consistent information about charter schools, may not be aware that charter schools are a tuition-free option for their children, or they might not know how to navigate the application processes (Buckley & Sattin-Bajaj, 2011;Frankenberg et al., 2011;P. Garcia, 2015;Villavicencio, 2013).In fact, P. Garcia (2015) found that the charter schools that were most successful in attracting and retaining ELLs often utilized strategic "boots on the ground campaigns" in which school leaders intentionally reached out to the families of ELLs, connected with community organizations, and provided information about their school in multiple languages.Conversely, charter schools tended to be less successful if ELLs were not intentionally recruited, if the cultural backgrounds and linguistic skills of the charter school's staff did not reflect those of communities in which ELLs were concentrated, or if the charter school staff's lack of ELL-specific expertise and capacity discouraged them from actively recruiting ELLs (P.Garcia, 2015).
Another category of potential explanations to explain enrollment gaps is related to the mission-driven nature of charter schools.As one example, some charter schools have developed models that are designed for secondary students.However, ELLs generally, and particularly those at lower English proficiency levels, tend to be concentrated in elementary schools because ELLs develop their proficiency in English over time and exit the ELL subgroup.As a result, ELLs may be underrepresented in charter school portfolios that offer a large number of secondary options (P.Garcia, 2015).Furthermore, charter missions that are specific to a particular subgroup might contribute to both over-and underrepresentation of ELLs.In working to fulfill their mission to narrow the achievement gap, some mission-driven charter schools may strategically decide to locate their school in predominantly African American neighborhoods, for example, in which there might not be large concentrations of ELLs, but where there is a need for high-quality educational options.As a result of their location, these charter schools will likely reflect the racial and ethnic makeup of the neighborhoods they serve (Henig, 2008;Lazarin & Ortiz-Licon, 2010;National Alliance for Public Charter Schools, 2012).If an authorizer's portfolio includes a large number of charter schools that have a mission related to serving the African American community, ELLs might be underrepresented in that portfolio.Conversely, if charter schools have a mission that is explicitly related to ELLs or language acquisition, they might be more likely to attract large numbers of ELLs.For example, P. Garcia (2015) found that several charter schools in one district in her sample had an explicit focus in their missions on multilingualism, multiculturalism, or on serving culturally and linguistically diverse young people.These models might have been particularly popular because they were meeting a need in the community, which had a large population of ELLs (P.Garcia, 2015).
The final category of potential explanations to explain enrollment gaps is related to structural elements that tend to be associated with charter schools.One such element is the lottery, which is typically used if the charter school is oversubscribed.Unlike traditional public schools that serve all students in a neighborhood, charter schools often draw students from across a district or from several different districts.As students cannot automatically enroll in charter schools solely because of their residential address, they must formally apply to these schools.If there are more applicants than seats, charter schools generally conduct a lottery in which they randomly select some students to be part of the new cohort, but everyone who applied may not get a seat.Thus, even if the charter school made good faith efforts to recruit ELLs into their pool of potential students, those ELLs may not have been randomly selected during the lottery (P.Garcia, 2015).Another structural element that is common to charter schools is a lack of backfill.In their charters, design teams often commit to serve a particular number of students per grade.They open the school with one cohort, and then expand grade by grade, adding a new cohort each year.For example, a sixth to eighth grade school that is designed to serve 150 students will add 50 new sixth graders each year, growing to the full capacity of 150 over three years.In communities in which ELLs might be recent immigrants or in which there is a great deal of mobility, newly arrived ELLs would only be able to enroll in a charter school if they applied for a seat in the initial grade that was offered (P.Garcia, 2015).While researchers have proposed a range of potential reasons to explain why ELLs may not be equitably represented in charter schools, more empirical evidence is needed to better understand the types of policies and practices that can be implemented by statewide policymakers, charter school authorizers, and school leaders to ensure that ELLs have equitable access to charter schools.

ELLs' Achievement in Charter Schools
It is critical to examine whether or not ELLs have equitable access to charter schools, but perhaps more importantly, researchers must also explore the quality of the educational experiences that ELLs receive in charter schools.The majority of achievement studies that have focused on the effectiveness of charter schools have not analyzed outcomes for ELLs, although more recent studies are disaggregating the results by subgroup more frequently.As one example, the most recent national study conducted by the Center for Research on Education Outcomes (CREDO, 2013) found that ELLs in charter schools gained approximately 43 additional days of learning in reading and 36 additional days in math as compared with their peers in traditional public schools (pp. 37-38).Although this is an encouraging finding, it is important to note that the CREDO researchers did not disaggregate their data by language proficiency level.Consequently, it was not clear that the ELLs in the charter schools were equivalent to the ELLs in traditional public schools.As researchers have noted in other studies, it is possible that ELLs with higher proficiency levels in English were enrolling in charter schools, while ELLs with lower proficiency levels remained in traditional public schools (Buckley & Sattin-Bajaj, 2011).
Even if the achievement data were disaggregated by language proficiency levels, standardized statewide assessments are generally not valid and meaningful measures of the achievement of ELLs, who, by definition, are in the process of developing their English language proficiency skills.When ELLs are assessed with standardized assessments in English, it is often the case that the assessment was normed for native English speakers (Abedi, 2004).Furthermore, if a school uses a dual language or transitional bilingual program, instruction in the content areas may be delivered in the student's home language.In this case, an assessment in English would not be valid because it does not reflect the language of instruction.In short, standardized assessments are often not valid for ELLs because they often underestimate ELLs' content knowledge by measuring students' ability to read in English more than their content knowledge or skills (Abedi, 2004).
In addition, statewide assessments that are used for accountability purposes do not take into consideration ELLs' level of language proficiency or the amount of time that it takes to attain proficiency.Not surprisingly, proficiency rates for ELLs on state assessments increase as their language proficiency increases (Cook, Boals, & Lundberg, 2011).Several researchers have found that it takes ELLs at least five years to acquire proficiency in academic English and that the rate of acquisition is influenced by the level of proficiency at which students begin, with students at higher levels of proficiency acquiring English more quickly than students who begin at lower levels of proficiency (Cook et al., 2011;Genesee, Lindholm-Leary, Saunders, & Christian, 2006;Hakuta, Butler, & Witt, 2000).To address these concerns about validity, researchers have recommended that educators use multiple sources of data to assess the academic progress of ELLs, to make decisions based on assessments that are valid and reliable for ELLs, and to take into account, as much as possible, assessments of skills in the home language (Echevarria & Hasbrouck, 2009).This is a specific area of opportunity for charter schools because they are often able to propose missionspecific measures on which they will be evaluated.Thus, when charter school design teams develop their models, they could outline how they will use multiple measures-that are valid and meaningful for ELLs-to evaluate the success of ELLs in their schools.By testing out a range of metrics, charter schools can lead the field in the development of accountability measures that provide more accurate assessments of ELLs' success than do statewide achievement tests.

Expanding Definitions of Quality for ELLs in Charter Schools
In building on the research related to access and achievement for ELLs in charter schools, we argue that more research is necessary in a third area that has not yet been explored in the peerreviewed literature: high-quality programs for ELLs in charter schools.By drawing on the literature related to charter schools generally and to ELLs in traditional public schools, we propose a framework in the section that follows that could be used to guide future research.This framework could also inform charter school design teams that are interested in responding to former Secretary Duncan's challenge by developing models that support the success of ELLs.Finally, while we encourage the charter sector to be more proactive about serving ELLs, we also urge them to develop innovative models that reconceptualize quality programs for ELLs and go well beyond measures of success based solely on standardized tests.
As a result of the one-size-fits-all accountability system under No Child Left Behind (NCLB), charter schools, like all public schools, were held accountable for ensuring that all students were proficient on statewide achievement tests.In the last section, we described how these one-sizefits-all accountability systems are particularly problematic for ELLs.However, in a post-NCLB policy environment, there may be more space for innovation in the charter sector to the extent that charter schools can propose the use of multiple measures and mission-specific metrics to evaluate their success.
For charter schools that explicitly include ELLs in their designs, these mission-specific goals are a potential affordance because the schools could propose measures that are valid for ELLs, are consistent with their mission and programs, and that provide more accurate measures of ELLs' performance than do statewide assessments.For example, dual language schools might include a metric related to the proportion of students who have demonstrated academic proficiency in two or more languages or who have earned the seal of biliteracy (Seal of biliteracy, 2015).Secondary schools might use on-track measures that are highly predictive of high school graduation rates and that have been empirically validated for ELLs (Gwynne, Pareja, Ehrlich, & Allensworth, 2012).By expanding definitions of quality beyond proficiency on statewide assessments, the charter sector can reignite Albert Shanker 's (1988) original vision of charter schools as decentralized laboratories of innovation that empower educators to develop a wide variety of models that are designed to maximize learning for many different types of students.This focus on quality might also help researchers on both sides of the charter school debate to move away from studies that focus only on whether charter schools outperform traditional public schools on statewide achievement tests and toward a more in-depth analysis of how or why high-performing charter schools work, particularly for ELLs.

Conceptual Framework.
To structure our discussion of expanding definitions of quality for ELLs in charter schools, we begin with the framework of essential supports for school quality that was developed by the Consortium on Chicago School Research, but we consider it through an ELL lens.Explicitly considering ELLs in educational policies is a practice that researchers and advocates for ELLs have recommended and one that we have tried to reflect throughout this paper (Hakuta, 2011;O'Day, 2009).Based on years of research in Chicago Public Schools, the Consortium on Chicago School Research (Consortium) developed a framework of five essential elements of effective schools: school leadership, parent-community school ties, professional capacity, instructional guidance, and student-centered learning climate (Bryk, Sebring, Allensworth, Luppescu, & Easton, 2010).The researchers emphasized that there is no silver bullet; instead all of these elements must be present and aligned in a coherent manner.The school leadership component included program coherence, implementation of the school improvement plan, and instructional leadership.Within the parentcommunity school ties element, the researchers evaluated teachers' knowledge of the community, their use of community resources, and parental involvement.Professional capacity included the quality of teachers' undergraduate preparation, the frequency and quality of professional development at the school, and the degree to which teachers collaborated within a professional community.The student-centered learning climate measured school safety, the level of academic press at the school, and peer support for academic work.Finally, instructional guidance referred to clear mapping and alignment of the curriculum across all grades as well as the instructional resources available to teachers to support their implementation of the curriculum (Bryk et al., 2010).
We will examine each of these components in turn through an ELL lens and then discuss research findings related to the component from the charter schools literature generally or from studies focused on ELLs in traditional public schools.It is important to note that the research base related to effective programs for ELLs in traditional public schools is quite limited.Only one of the studies that we cite was peer-reviewed and three of the four used statewide tests of academic achievement to identify schools in which ELLs were successful, a measure that we critiqued as overly narrow and not valid for ELLs earlier in this paper (Horowitz et al., 2009;Lucas, Henze & Donato, 1990;Parrish et al., 2006;Williams et al., 2007).While these studies provide us with a starting point, it is clear that more research is needed-based on a broader range of valid and meaningful measures-to identify key components of successful programs for ELLs in both the charter sector and in traditional public schools.
The components of our adapted framework include: a coherent schoolwide model; a focus on language development; building the capacity of all educators to serve ELLs; a data-driven focus on continuous improvement; and family and community engagement.After reviewing the research related to each component, we illustrate each of the five components with an example from a charter school that supports the success of ELLs.We selected these examples from webinars that were hosted by the National Charter School Resource Center (Resource Center) in 2012. 1 The Resource Center is a national technical assistance center funded by the U.S. Department of Education and all of these webinars are publicly available on their website.Table 1 illustrates the diversity of the four charter schools that we highlight in this article.
It is important to point out that ELLs are an incredibly diverse population in terms of the languages they speak, their cultural resources, their proficiency in English and their home language, and their prior schooling.The schools from the webinars capture this diversity because they varied widely in terms of location, student population, and program model, but all shared a commitment to providing ELLs with high-quality programs.However, these webinars share the limitation of the research related to ELLs in traditional public schools because these charter schools were selected to present largely on the basis of their ELLs' performance on statewide tests of academic achievement.Coherent schoolwide model.The first theme, a coherent schoolwide model, is related to the Consortium's school leadership and instructional guidance elements (Bryk et al., 2010).When viewed through an ELL lens, this component describes how charter schools might consider ELLs at the inception, when they design their models, and then integrate research-based practices for ELLs throughout every aspect of their design.This component of our model is consistent with previous studies that have found positive correlations between student achievement and instructional programs that thoughtfully integrate curriculum, instruction, and assessment (e.g., Newmann, Smith, Allensworth, & Bryk, 2001).Furthermore, existing research related to ELLs in traditional public schools has suggested that effective schoolwide or district-level models are comprehensive, coherent, and explicitly consider the needs of ELLs (Goldenberg & Coleman, 2010).In these effective schools and districts, leaders valued the linguistic and cultural diversity of ELLs, set high expectations for language development and academic performance, and then aligned programs, resources, and staff to help ELLs meet those standards (Horowitz et al., 2009;Lucas, Henze & Donato, 1990;Parrish et al., 2006;Williams et al., 2007).
In fact, this ability to design a model that considers ELLs from the inception is an important affordance of charter schools.In related research from the charter schools literature, Zimmer and Buddin (2007) found that charter schools were more likely than traditional public schools to develop an educational model that was targeted for a particular student population, in part because they can often draw students from a wider attendance area than can a neighborhood school.If more charter schools took advantage of this affordance, they could potentially design a wide variety of models that thoughtfully consider how to support the success of ELLs.
Another important affordance that charter schools can capitalize upon when developing their models is the flexibility they often have to expand learning time by offering an extended school year, longer school days, and summer school (Lazarin & Ortiz-Licon, 2010).This is particularly important for ELLs because, as discussed earlier, research has indicated that it takes ELLs at least five years to acquire proficiency in academic English (Cook et al., 2011;Genesee, Lindholm-Leary, Saunders, & Christian, 2006;Hakuta, Butler, & Witt, 2000).By taking advantage of this affordance, charter schools can facilitate the language acquisition process for ELLs by increasing the amount of opportunities they have to produce academic language and receive feedback through interactions with their peers (Gass & Mackey, 2007;Gass & Selinker, 2008;Peregoy & Boyle, 2001).
In their study that focused on charter schools generally, Zimmer and Buddin (2007) found that charter schools often extend the school day or school year to increase the amount of instructional time for students.In their charter school study, Therriault, Gandhi, Casasanto, and Carney (2010) also found that charter schools extended the school day, but they stressed that the critical difference was not just more time, but how that time was used and organized.In the most successful schools, expanded school days were used to increase instructional time, to develop structures that embedded a variety of academic supports into the regular school day for struggling students, and to provide teachers with time to review student data and plan collaboratively to meet the needs of students (Therriault et al., 2010).
An illustrative example of a coherent schoolwide model that supports the success of ELLs is provided by Chula Vista Learning Community Charter School, a school in southern California.To advance their mission of preparing all students for success as multilingual, global citizens, the school leaders developed a 50/50 dual language model.This model equally values English and Spanish and is designed to help all language learners develop strong skills in both languages.Approximately half of their students are ELLs, but they consider all of their students to be second language learners because they are learning either English or Spanish.In addition to fulfilling their mission to prepare multilingual citizens, Chula Vista has realized a number of benefits by implementing their dual language program.First, although the majority of their students are Latino, many of these students were not fluent in academic Spanish.Thus, the dual language model helps students to develop their academic language skills in both English and Spanish at the same time that it affirms their culture and identity.Second, students have improved their skills in both languages because of the schoolwide emphasis on reading, writing, and vocabulary development across the curriculum and the infusion of language learning strategies throughout the entire school day.Finally, as one example of how the expanded learning time affordance might be used in charter schools, during Chula Vista's afterschool program, students design and run small businesses in a vibrant marketplace, which provides them with multiple opportunities to develop their skills in both languages within an informal, real-world context (National Charter School Resource Center, 2012d).
Focus on language development.The second theme in our framework focuses on how charter school educators deliver high-quality instruction for ELLs, but with a "language lens" (University of Colorado, 2011).This theme is closely related to the Consortium's student-centered learning climate element (Bryk et al., 2010).Combining academic rigor with a focus on language development is critical because ELLs must learn grade-level content while developing language proficiency in English (Short & Fitzsimmons, 2007;Understanding Language Initiative, 2013).As implementation of the Common Core State Standards and Next Generation Science Standards ramps up, ELLs will be required to navigate complex grade-level texts, to identify and make evidence-based arguments, and to write for a variety of purposes-all tasks that require well-developed skills with academic language in English that many ELLs are still developing (Bunch, Kibler, & Pimentel, 2012).Researchers point out that, because language and content are inextricably intertwined, effective content instruction for ELLs must pay explicit attention to language development to help ELLs learn content through language at the same time that they learn about language (Fang & Schleppegrell, 2010;Halliday, 1993).To effectively teach language and content concurrently, teachers will need specific training about language development and research-based practices for ELLs.For example, effective teachers might build on ELLs' existing knowledge, strategically use the home language, provide explicit vocabulary instruction, or utilize scaffolding strategies to reduce the language load for ELLs to ensure that they understand the content that is being presented (August & Shanahan, 2006, 2010;Coleman & Goldenberg, 2012;Himmel, Short, Richards, & Echevarria, 2009;Understanding Language Initiative, 2013;VanPatten & Williams, 2012).In addition, to build their oral language proficiency skills-which are correlated with strong skills in reading comprehension-ELLs will need frequent opportunities to interact with their peers in collaborative discussions involving academic language throughout the school day (August et al., 2009;Himmel et al., 2009).
The most rigorous studies about the language of instruction for ELLs have demonstrated positive impacts for bilingual education programs, particularly those that are long term and high quality (August & Shanahan, 2006, 2010;Goldenberg & Coleman, 2010).Consequently, more charter schools might consider offering bilingual or dual language programs to enable ELLs to build on the linguistic and cultural knowledge they bring with them and to help them further develop academic language and literacy skills in two languages concurrently.Even in states that have restrictive language policies, such as California, charter schools might have more flexibility to offer bilingual instruction than do traditional public schools because they might not have to adhere to all state laws.In fact, several charter schools across the country have implemented dual language models, which are in increasing demand as parents recognize the value of these models in preparing students to be global, multilingual citizens (A.Garcia, 2015).As in traditional public schools, researchers have found that charter schools have also provided ELLs with home language support in other ways, including hiring bilingual paraprofessionals, filling school libraries with books in multiple languages, pairing bilingual students with less proficient ELLs, and communicating with parents in the home language (Cervantes, 2006;Lazarin & Ortiz-Licon, 2010).
Depending on the population of ELLs they serve, it may not be feasible for charter schools to offer dual language programs for all ELLs.However, charter schools can focus intently on language development whether their model is related to the arts, the sciences, or offers a more traditional college preparatory curriculum.Within a STEM-focused model, for example, charter school educators might help ELLs to build their academic language with explicit vocabulary instruction related to both technical scientific terms and vocabulary associated with disciplinary practices in the sciences (e.g., hypotheses, prediction) (August et al., 2009;Lee & Buxton, 2013).To ensure that science content is comprehensible for ELLs, teachers could scaffold instruction (August et al., 2009;August, Artzi, & Mazrum, 2010) and immerse ELLs in rich and supportive inquirybased learning experiences (Lee, Quinn, & Valdes, 2013).An inquiry-based approach in science, which includes hands-on activities and experiments, can be particularly helpful for ELLs because this approach engages ELLs in rigorous academic content and the practices of the discipline, while it also reduces the language load through the use of experiential learning techniques (Lee, 2005;Lee & Buxton, 2013).In addition, these types of collaborative inquiry-based explorations provide ELLs with authentic opportunities to practice academic language, through interactions with their peers, as they make sense of the scientific principles involved (Lee & Buxton, 2013).
One charter school network that is exemplary in terms of its focus on language development is Community Day Charter Public Schools in Lawrence, Massachusetts.Community Day is a network of schools that combine a language rich school environment with a data-driven approach to achieve their mission of eliminating the achievement gap for ELLs.Education at Community Day begins with a full-day program for four-year-olds that is designed to give students a head start with language development and with building oral proficiency.This is one of the ways that Community Day capitalizes upon the affordance that charter schools have to increase instructional time for ELLs.At all grade levels, teachers encourage students to speak in complete sentences, focus on developing academic language, provide students with multiple opportunities to develop their oral proficiency (for example through turn and talks, partner discussions, etc.), explicitly teach vocabulary, and continuously celebrate words and nurture a love of reading through schoolwide events like the annual Vocabulary Parade and Dr. Seuss night (National Charter School Resource Center, 2012c).One of the school leaders observed that, ". . .all teachers are language teachers all day long, so whether you teach math or you teach social studies-even our chefs in our buildingeverybody is aware of the fact that our environment needs to be language rich all of the time" (National Charter School Resource Center, 2012c, p. 8).Community Day schools have also implemented family literacy initiatives, through which parents partner with educators to nurture the development of students' language and literacy in ways that leverage students' home language and culture.For example, the school sends students home with books to read with their families every night, in both Spanish and English.In addition, parents are regularly invited to the school to be "mystery readers," who surprise their child's class and read a story to them in either Spanish or English (National Charter School Resource Center, 2012c).
Building the capacity of all educators to serve ELLs.The third theme in our framework, building the capacity of educators to serve ELLs, is closely aligned with the Consortium's professional capacity element (Bryk et al., 2010).As the number of ELLs grows across the country, more and more mainstream classrooms include students from diverse linguistic and cultural backgrounds.For example, a recent survey conducted by the Council of Great City Schools revealed that 91.6% of general education teachers in St. Paul, Minnesota have ELLs in their classrooms, as do 87.9% of teachers in Wichita, Kansas (Uro & Barrio, 2013, p. 55).Thus, across the nation, an increasing number of teachers will need to increase their knowledge about language acquisition and effective instructional practices for ELLs, but many of them have not yet received the training they need to serve ELLs well.In fact, in the U.S. Department of Education's (2012) recent evaluation of Title III districts, 73% of the districts surveyed indicated that the lack of ELL-specific expertise among general education teachers was a "moderate or major challenge" (p.117).
To meet this need, researchers and advocates for ELLs have recommended that school and district leaders consider how to build the capacity of teachers to serve ELLs and to ensure that all educators are accountable for the success of ELLs (Horowitz et al., 2009;Working Group on ELL Policy, 2010).This recommendation is supported by evidence from traditional public schools in which ELLs were successful.In these effective schools and districts, teachers had strong content knowledge, were skilled in aligning curriculum and instruction, recognized the assets that ELLs brought with them by emphasizing their skills in their home language as a rich resource, and connected their practice to research related to second language acquisition and effective instructional practices for ELLs (Parrish et al., 2006;Short & Fitzsimmons, 2007;University of Colorado, 2011).These schools and districts also built the capacity of their staff by providing all teachers with high-quality, job-embedded professional development that enabled them to implement researchbased instructional practices that are effective with ELLs (Horowitz et al., 2009;Lucas, Henze & Donato, 1990;O'Day, 2009).For example, in St. Paul, when the district implemented their collaborative teaching model, they offered joint professional development, for both ELL experts and general education teachers, which was focused on instructional strategies the teachers could use to support ELLs when they taught together in one classroom (Horowitz et al., 2009).
Strengthening the capacity of charter school educators to serve ELLs could also be connected to another affordance of charter schools.Charter school leaders have the flexibility to hire skilled teachers who have specialized knowledge related to ELLs and language acquisition.In their study of different types of public schools in Boston, Therriault et al. (2010) found that the flexibility charter school leaders had to hire and remove teachers resulted in schools that were staffed by teachers who were committed to the mission of the school and intently focused on student learning.We do not intend to be cavalier about the ability to fire teachers, a practice which can be easily abused.However, we note that the flexibility with hiring is an affordance of charter schools because it can be challenging to hire highly-qualified teachers who are trained to serve ELLs and who have strong bilingual skills.If schools are unable to hire highly-skilled ELL experts, many high-performing charter schools build the capacity of existing staff by investing heavily in induction models, offering professional development related to teaching ELLs, and providing teachers with time to work collaboratively to review data, reflect on their practice, and modify their instruction to meet the needs of ELLs (Buckley & Sattin-Bajaj, 2011;Cervantes, 2006;Lazarin & Ortiz-Licon, 2010;University of Colorado, 2011).This is something that other public schools can learn from, about what it takes to train and support teachers of ELLs effectively.
The exemplar for this component, building the capacity of all educators to serve ELLs, is also Community Day Charter Public Schools.They have succeeded in narrowing the achievement gap for ELLs in large part because of their team-based approach in which all educators in the building are accountable for the success of ELLs.There are several notable components of their model aimed at building teacher capacity.First, the schools are led by instructional leaders.The heads of school are responsible for setting coherent schoolwide policies, monitoring student progress, developing teachers, and aligning resources with student needs.Second, Community Day uses a co-teaching model in which an experienced teacher is paired with a novice teacher.Coteaching serves as an effective induction and mentoring model for novice teachers at the same time that it allows the two teachers to easily differentiate instruction in small groups.Third, teachers work collaboratively, not only within their classrooms, but also in team meetings that are held every other week.They identify gaps in student learning through their analysis of student data and then develop individualized action plans for each student that will drive instruction for the next cycle.Fourth, instructional leaders strategically assign staff to classrooms in order to align existing resources with the student needs teachers identified during their collaborative planning sessions.Thus, depending on the action plan developed by teachers, several educators might be working in any given classroom at the same time to differentiate instruction and deliver personalized supports and interventions for each child.For example, in addition to the two co-teachers in the classroom, additional support might be provided by the reading specialist, a generalist, the head of school, or a teacher with specialized expertise in serving students with disabilities.One of the school leaders succinctly summed up this component of the model when she remarked that, "all hands are on deck at all times" (National Charter School Resource Center, 2012c, p. 14).Finally, instructional leaders have high expectations for teachers, but also provide them with the personalized support they need to meet these expectations.Classrooms at Community Day are open and transparent, and teachers are constantly observing, mentoring, and supporting one another.Heads of school and experienced teachers serve as coaches to help teachers hone their craft through a continuous improvement process.In fact, much of the professional development at the school is driven by teachers based on the needs they have identified to improve their own practice (National Charter School Resource Center, 2012c).
Data-driven focus on continuous improvement.The fourth theme in our framework, a focus on data-driven continuous improvement, is related to the Consortium's professional capacity, school leadership, and instructional guidance elements (Bryk et al., 2010).One common finding in the studies focused on ELLs in traditional public schools was that educators in effective districts had frequent access to data from regular assessments of student progress.They used this data to diagnose student needs and then developed personalized learning plans that aligned student needs with appropriate supports and interventions (Cervantes, 2006;Dessof, 2010;Horowitz et al., 2009;Lazarin & Ortiz-Licon, 2010;Parrish et al., 2006;University of Colorado, 2011;Williams et al., 2007).Gathering a variety of data about each student is particularly important with this subgroup because ELLs vary widely in terms of their prior schooling, age, motivation, socioeconomic status, proficiency in English, literacy skills in their native language, and mastery of content knowledge.Thus, educators need access to data from multiple measures to appropriately individualize instruction and provide ELLs with opportunities to build on their existing strengths and assets (Understanding Language, 2013).
One example of a charter school that has effectively used a data-driven approach is Twin Cities International Elementary School (TIES), located in Minneapolis, Minnesota.This charter school was founded by educational leaders in the East African community who wanted to provide their children with a rigorous educational option in a culturally sensitive environment.Educators at the school use a variety of assessments to gather information about student progress and then collaboratively analyze that data.As part of their commitment to use data to drive their curriculum, the school invested in a data coordinator.The coordinator meets weekly with grade-level teams to discuss student data and plans for how that data can be used to inform instruction.Perhaps more important than the available tools and time to collaborate is how school leaders have nurtured the development of a school culture in which teachers share data and discuss it with their colleagues.One of the school leaders noted that, "it has taken us a few years to get to the point where our teachers are comfortable looking at each other's data and being able to share that data.But they are able to do that now in a learning environment" (National Charter School Resource Center, 2012a, p. 18).After analyzing their data, teachers discuss how to adjust their instruction to meet student needs, how to flexibly group students to provide them with interventions or additional support, and how to feed the information collected into their Response to Intervention model.Similar to Community Day, TIES also develops individual learning plans for each student to identify goals, track progress, and assess accomplishments.These plans are linked to school report cards to ensure that parents, students, and teachers are monitoring progress toward the goals and updating them as necessary (National Charter School Resource Center, 2012a).
Family and community engagement.The final theme in our framework, family and community engagement, mirrors the Consortium's parent-community school ties element (Bryk et al., 2010).Viewing this component through an ELL lens encourages charter schools to communicate with families in multiple languages, conduct outreach to involve organizations in the community, invite parents to volunteer in the school, hire bilingual staff who can communicate with parents, or provide services or learning opportunities for parents at the school (Buckley & Sattin-Bajaj, 2011;Lazarin & Ortiz-Licon, 2010;Orlick, 2013;University of Colorado, 2011).In their study of charter schools more generally, Zimmer and Buddin (2007) found that parental involvement at charter schools was higher than at similar traditional public schools, and further, that this increased parental involvement was positively and significantly correlated with student outcomes.Research conducted in traditional public schools also found that educators in schools and districts in which ELLs were successful aimed to ensure that their schools were culturally and linguistically responsive to ELLs and their families (Buckley & Sattin-Bajaj, 2011;Lazarin & Ortiz-Licon, 2010;Orlick, 2013;University of Colorado, 2011).Charter school design teams could potentially engage communities with large proportions of ELLs by developing culturally responsive models that meet a need in the community.For example, Buchanan and Fox (2003) described three Hawai'ian charter schools that were designed to preserve Hawai'ian culture and values, incorporated Hawai'ian language into their curricula, and involved elders and community members in learning at the school.
The family and community engagement component of the framework is exemplified by Folk Arts-Cultural Treasures Charter School (FACTS) in Philadelphia's Chinatown neighborhood, which was founded by parents and two community-based organizations.After filing numerous civil rights complaints, the founders concluded that while they could legally require traditional public schools to serve ELLs, ultimately they wanted high-quality programs for their students that went beyond compliance with the law (Folk Arts-Cultural Treasures Charter School, 2012).Consequently, they decided to design a school that celebrated the culture of their community and was responsive to the unique needs of their students.FACTS uses several strategies to engage families and the community.First, the school model was designed to celebrate the arts and global cultures.Teachers work closely with folk artists from the community to connect their curriculum with a particular genre of the arts.For example, through a residency program for artists, the school invited a local spoken word artist to conduct workshops with students while they were studying poetry in their language arts class.As a culminating activity, the students publicly shared their work in a poetry slam.Second, teachers and school leaders explicitly target immigrant neighborhoods for recruitment, work hard to ensure that parents in the community are aware that their charter school is an option for their children, and conduct outreach in multiple languages.Third, the school continues their multilingual communication by translating documents that are sent home with students and using in-house translators and telephone-based interpretation services to communicate with parents.Finally, drawing on the research related to funds of knowledge (Gonzalez, Moll, & Amanti, 2005), FACTS teachers engage students in inquiry projects in their communities and conduct home visits for multiple purposes, such as building relationships with families, developing a deeper understanding of students and their strengths, and learning more about the cultural practices and knowledge that ELLs bring with them (National Charter School Resource Center, 2012b).
Many of the practices we have summarized here are not unique to charter schools.Traditional public schools also engage families, develop coherent schoolwide models, provide ELLs with high-quality instruction, and use data to inform their decisions.However, charter schools have several unique affordances, including the ability to explicitly design models that serve ELLs, to hire skilled teachers, to expand learning time, and to use ELLs' home languages, even if bilingual instruction is forbidden by state law.As the number of charter schools grows and the population of ELLs increases, charter schools can capitalize upon their unique affordances to develop a variety of high-quality educational options that are explicitly designed to build on ELLs' strengths.

Discussion
This paper takes an initial step toward filling a gap in the literature.We have argued that researchers must expand their focus beyond access and achievement and begin to grapple with questions related to the quality of programs for ELLs in charter schools.To provide a guide for future research, we have offered a five-component framework that is grounded in the literature related to school improvement, charter schools, and ELLs in traditional public schools.More research is necessary to empirically test this framework and to disentangle the key elements of effective programs for ELLs in charter schools.The five-component framework we outlined in this paper might be used by charter school leaders, researchers, policymakers, and authorizers in a number of different ways.School leaders might reflect upon the components of the framework and consider the extent to which each component has been implemented at their school.Qualitative researchers might use a case study approach to determine if this framework holds when they examine a small number of schools and then generate rich and contextualized descriptions of practices in those schools.Quantitative researchers might analyze outcomes for ELLs more broadly-using valid and meaningful measures-to determine if the elements of our framework are, in fact, associated with stronger outcomes for ELLs.Policymakers might consider how to amplify the opportunities charter schools have to develop a range of innovative models for ELLs and allow these schools to use mission-specific measures to evaluate their success.Charter school authorizers might incorporate this framework into their authorizing practices to ensure that there are a range of high-quality educational options for ELLs in their portfolios that are responsive to the local context, but that are also held to a high standard of quality.
In the charter sector, authorizers have the potential to become an important policy lever in scaling up success for ELLs in charter schools because they can implement authorizing practices that are designed to provide ELLs with equitable access to the charter schools in their portfolios and to ensure that programs for ELLs in those schools are research-based, high-quality, and build on the assets and strengths of ELLs.To further investigate the role of authorizers, we offer below a number of recommendations for lines of future research related to charter school authorizers, particularly those in communities with large concentrations of ELLs.
There is currently a small body of research about charter school authorizers, but we are not aware of any peer-reviewed studies that examine the intersection between ELLs and authorizing practices.Future studies could explore how the framework we developed is reflected in the practices of charter school authorizers.Charter school authorizers serve three essential functions: (1) they review applications from design teams and grant charters to those who meet the standards; (2) they monitor the performance of charter schools under the terms of the contract with the authorizer; and (3) they evaluate the charter school at the end of the charter term and decide to renew or revoke the charter (Bulkley, 2001;Vergari, 2000).It is possible that authorizers could dramatically increase the number of high-quality charter school options for ELLs if they incorporate the framework we outlined into their authorizing practices.
We briefly outline below three avenues for future research related to authorizers, associated with each of the key responsibilities of authorizers.First, in response to concerns expressed about underrepresentation of ELLs in charter schools, researchers might determine the extent to which ELLs have access to charter schools in specific communities, and if not, why that is the case.In addition, future studies might investigate authorizing policies or practices that are deliberately designed to ensure that ELLs have equitable access to high-quality charter schools in particular communities.For example, authorizers' requests for proposals for new charter schools might prioritize the development of charter schools in communities with large concentrations of ELLs or incentivize the development of innovative models that are explicitly designed to serve ELLs.
Second, researchers might compare authorizers, particularly those in communities with large concentrations of ELLs, to determine the extent to which these authorizers focus on ELLs in their authorizing practices.As one component of this investigation, researchers might explore whether authorizers use their applications process to encourage charter schools to develop comprehensive models that incorporate the five components of the framework we outlined in this paper.As we mentioned earlier, a unique affordance of charter schools is their ability to develop models, from the inception, that are explicitly designed to consider ELLs.Most authorizers currently include a standard question on their application that asks design teams to describe how they will comply with federal requirements for ELLs, but more research is needed to determine how authorizers design application processes that go beyond compliance to foster the development of high-quality, comprehensive models for ELLs in their portfolios.
Finally, authorizers are responsible for monitoring the progress of charter schools and for closing them if they do not meet the authorizer's standards for quality.Researchers might study the extent to which authorizers rigorously review the quality of instructional programs for ELLs as part of their monitoring process and incorporate valid ELL-specific metrics into their performance frameworks.Researchers might also examine if charter schools have been closed as a result of ELLs' performance or explore the types of technical assistance that authorizers offer to charter schools who struggle to serve their ELLs well.
We have argued that the growth in charter schools and the increasing number of ELLs, together with the unique affordances of charter schools, provides the charter school sector with an opportunity to develop a variety of innovative models that shed light on promising practices for ELLs and expand definitions of quality beyond performance on statewide assessments.If charter schools capitalize upon their unique affordances and deliberately design schools that include the five components of the framework we outlined, they could potentially develop a range of high-quality educational options that could scale up the success of ELLs across the charter sector.

Table 1
Selected characteristics of the four schools featured in the Resource Center webinars Note: Information in the table was compiled from the webinars that were presented by the National Charter School Resource Center in 2012 and may not reflect the most recent demographic statistics for each school.Webinars were retrieved from: http://www.charterschoolcenter.org/webinars?keyword=&topic=142