Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Separate and Not Equal: The Implementation of Structured English Immersion in Arizona’s Classrooms

by Karen E. Lillie, Amy Markos, M. Beatriz Arias & Terrence G. Wiley - 2012

Background/Context: Over the last ten years, a convergence of laws and decrees has impacted the development and implementation of Arizona's current program for English language learners (ELLs): the four-hour Structured English Immersion (SEI) model. Arizona’s new model, while being touted by some as the most effective program for ELLs (Clark, 2009), is raising concerns with researchers and educators involved with the education of ELLs (Combs et al., 2005; Faltis & Arias, 2007; Krashen, Rolstad, & MacSwan, 2007, Lillie et al., 2010; Wright, 2005). A recent Auditor General’s report (Davenport, 2011) indicates that instructional programs for ELLs have lacked uniformity and that the impact of the SEI model on ELLs’ language development and academic achievement is unknown. Therefore, although it has been over a decade since the Flores Consent Order, we must continue to examine the evidence regarding implementation of the SEI model in order to determine if the instruction offered to ELLs is comparable in amount, scope, and quality to that provided to English-proficient students.

Purpose/Objective/Research Question/Focus of Study: This paper utilizes the trajectory of these language policies as a lens for examining how ideas stemming from the Flores Consent Order and the English-Only voter initiative (Proposition 203), and continuing through to the legislative response of the court order (House Bill 2064) are evidenced in Arizona's SEI classrooms.

Research Design: The purpose of this qualitative study is to document and describe the characteristics of the four-hour SEI policy in practice. Over a seven-week period during the spring semester of the 2009/2010 school year, researchers utilized ethnographic observation methods, interviews, and a collection of artifacts to document the implementation of the SEI model and instruction of ELLs. Using a constant comparison analysis, findings were re-examined to show how this restrictive language policy has impacted and shaped the SEI model in practice today.

Conclusions/Recommendations: Arizona’s current four-hour model, as implemented in K-12 SEI classrooms, hinders policy intentions: ELLs do not have access to quality instruction, or a curriculum that is comparable in amount, scope, and quality to that of non-ELLs. Furthermore, deficiencies in funding abound, limiting ELLs' opportunities for the compensatory education needed to help them catch up academically. The authors contend that a re-examination of the SEI policy is warranted if the state is to fully realize the intentions of the Flores Consent Order and equal educational opportunity for ELLs.


In order to fully appreciate the parameters of Structured English Immersion (SEI) that exist today, it is necessary to understand the confluence of language and education policies of Arizona in the last decade. A brief review of the Flores Consent Order and the major state laws that shaped this instructional policy will shed light on the depth and scope of the consequences of this policy. The SEI model that is currently in place results from the intersection of the Flores Consent Order (2000) issued by Federal District Judge Marquez as part of the Flores v. Arizona (1992) case; Proposition 203, the voter initiative mandating English-only instruction in Arizona; and legislation emanating from a response to the Flores Consent Order, House Bill (H.B.) 2064, in 2006. Both the Consent Order and the English-Only instructional initiative were mandated in 2000. Between 2000 and 2006, programs that had been in place to serve ELLs were dismantled and replaced with loosely defined alternatives (Davenport, 2008). In 2006, the legislature finally responded to the mandates of the Court Order with H.B. 2064. The Consent Order provided the impetus for ELL program definition, Proposition 203 mandated English-only instruction, and H.B. 2064 set parameters for the program that is referred to as the four-hour SEI model.


In this study we have applied a policy trajectory approach (Ball, 2000) to the analysis of current ELL policy in Arizona. We assert that the SEI model is a result of the struggle between the Court, the legislature, and the State Board of Education, and we examine evidence of model implementation to explore its impact on students.

Figure 1 displays the convergence of the decrees and laws as they impacted the definition of the four-hour SEI model. This figure shows the intersection of the Consent Order, the English-Only voter initiative (Proposition 203) and the legislative response to the court order, H.B. 2064. The figure depicts that Proposition 203 arose from a context of English-Only ideologies and that the Flores Consent Order emerged as a result of the Flores v. Arizona lawsuit. H.B. 2064 was the legislature’s response to the mandate of the Court Order that the state would establish policies regarding comparable instruction, curriculum, and assessment for ELLs.

Figure 1. Policy trajectory of Arizona’s SEI model


Flores Consent Order

Over a decade ago, in a Consent Order ruling in the Flores case, Judge Marquez ruled on behalf of the plaintiffs. He ruled that state funding of programs for ELLs was “arbitrary and capricious” and that it contributed to the overall lack of qualified teachers, not enough classrooms for ELL students, poor tutoring/after-school programs, and not enough materials to teach the students. The Consent Order required that the State define the ELL program that would be funded but it did not address the type of instruction to be given to ELLs (i.e., bilingual, SEI, or ESL). The goal was twofold: define the ELL program in order to adequately determine the cost to fund it.

The Flores Consent Order addressed several programmatic issues regarding the delivery of ELL programs in Arizona including: instruction in both subject matter and English Language Development (ELD); comparability with curricular programs for English-proficient students; and assessment including proficiency tests and ongoing evaluation. With regard to instruction, the Consent Order required:   

daily instruction in English Language Development appropriate to the level of English proficiency of the student including listening and speaking, reading and writing skills, and cognitive and academic development in English;  

daily instruction in basic subject areas that is understandable and appropriate to the level of academic achievement of the LEP student, and is in conformity with accepted strategies for teaching LEP students; and

the curriculum of all bilingual education and ESL programs shall incorporate the Board’s Academic Standards and shall be comparable in amount, scope, and quality to that provided to English proficient students.  (Flores v. Arizona, 172 F. Supp.2d 1225. D. Ariz. 2000)


With regard to assessment, the Order required the selection of English proficiency tests by the Superintendent and the reassessment of ELL students after two years of exiting the program. The challenge of Flores is to adequately fund instructional programs for ELLs that include these components of daily instruction in ELD, and subject area instruction comparable to English-proficient students. Almost twenty years later, it is debatable whether ELLs have yet to receive adequate schooling comparable to English-proficient students in Arizona.

Proposition 203

Much has been written regarding Proposition 203, one of three current state laws that mandate English-only instruction (see e.g., Combs, Evans, Fletcher, Parra, & Jimenez, 2005; Crawford, 2000; Wiley & Wright, 2004; Wright, 2005; Wright & Pu, 2005). It effectively barred and dismantled bilingual programs in Arizona and replaced them with a loosely designed program defined as Structured English Immersion. A.R.S. •15-751 revised statutes state:

SEI means an English language acquisition process for young children in which nearly all classroom instruction is in English but with the curriculum and presentation designed for children who are learning the language. Books and instructional materials are in English and all reading, writing, and subject matter are taught in English. Although teachers may use a minimal amount of the child’s native language when necessary, no subject matter shall be taught in any language other than English, and children in this program learn to read and write solely in English. This educational methodology represents the standard definition of “sheltered English” or “structured English” found in educational literature.

This explicitly creates a model meant for K-12 students on a faulty understanding of “language acquisition…for young children”, where minimal native language can be used but yet is ordered not to be implemented that way in the classroom, and where no content will be taught so as to learn English quickly and effectively.

After the passage of Proposition 203 there was wide variation in ELL program consistency due to confusion over what Proposition 203 meant (Davenport, 2008). Wright (2005) noted that there had been no operational definition of SEI provided by the Arizona Department of Education (ADE). Others commented that the definition of Structured English Immersion as an instructional model was confusing (Mora, 2000); that it merged two very different methods, Structured English Immersion and Sheltered English Immersion instruction (Arias & Faltis, 2007); that it did not differentiate a SEI classroom from a mainstream classroom (Combs et al., 2005); and that it prescribed a “one size fits all” preparation for a SEI endorsement for K-12 teachers (Wright & Choi, 2005). Teachers reported that they and their administrators were confused about what Proposition 203 allowed with regard to primary language support in SEI classrooms (Wright & Choi, 2005). Practices varied widely from school to school.  

H.B. 2064

It took the passage of H.B. 2064 in 2006 to finally set the policy framework for SEI instruction. H.B. 2064 was a response from the legislature to the Consent Order mandate. This House Bill decreed that a statewide sheltered/structured English immersion model be defined for all ELLs. A Task Force was charged with selecting the appropriate model for SEI instruction, that adhered to the core principles of H.B. 2064, including: (1) four hours minimum daily instruction, (2) English-only instruction, (3) that the model be research-based, and (4) cost-effectiveness. This statewide model has evolved to be designated as the “four-hour block” (or “four-hour SEI model”).

The State Auditor General (Davenport, 2008) reported that the effectiveness of the SEI model in place between 2006 and 2008 was difficult to determine due to the fact that it was not until 2006 that the instructional components of the SEI program were developed by ADE and that the first year of actual SEI implementation was 2008. The 2008 report was a baseline study on the condition of ELL programs that were in place in 2006 and 2007, prior to the implementation of the statewide four-hour SEI model. Davenport examined ELL programs in 18 sample district and charter schools and documented wide variation in program implementation:

In 2007…more than half of all ELL students in the sample districts and charter schools attended programs that mainstreamed all ELL students, providing no hours of ELD instruction in a SEI setting.  Forty-two percent were in programs that provided up to 2 hours of daily ELD instruction.  The remaining 6 percent provided more than 2 and up to 4 hours of daily ELD. (p. ii)

Although programs at the sample districts and charter schools were aligned with the new models’ requirements regarding assessment and English-only classroom materials, their instructional approaches were significantly different from the models’ future SEI requirement regarding English language development (ELD). (p. ii)

For the sampled districts and charter schools, about 7 percent of the approximately 8,700 ELL students became fully proficient in fiscal year 2007, and most of them had been in the program for at least 2 years… (p. 11)

Implementation Issues

Statewide implementation of the SEI model began in 2008. A more recent Auditor General’s report (Davenport, 2011) indicates that many schools in Arizona are not implementing the model adequately. The Auditor made two findings: (1) the SEI models are not fully implemented, and (2) the impact of the SEI model is unknown. The Auditor’s report comments that “in fiscal year 2010, almost two-thirds, or 46 of the 73 districts and charter schools (districts) reviewed either by the Auditor General’s Office or the Arizona Department of Education (ADE) had not fully implemented the SEI models” (Davenport, 2011, p. 17). For example, nearly one-half of the reviewed districts failed to provide the four hours of daily ELD, a key requirement of the SEI model. Other deficiencies included, but were not limited in, failing to (1) group students properly, (2) provide qualified teachers, and (3) provide required instructional components of ELD. The 2011 report also underscores the fact that the SEI model has not been universally adopted. The report concludes that “because the SEI programs are relatively new and not fully implemented at many districts, data must be gathered over a longer period of time to identify the impact of those programs on ELL students” (Davenport, 2011, p. 6).

It has been ten years since ADE required the SEI approach, five years since the SEI model was articulated and over this period of time, the evidence is clear that instructional programs for ELLs have lacked uniformity. Over a decade since the Consent Order, we continue to examine the evidence regarding implementation of the SEI curriculum in order to determine if the instruction offered to ELLs is comparable in amount, scope, and quality to that provided English-proficient students.


In order to understand the trajectory of language policies in practice, specifically, how ideas stemming from the Flores Consent Order, H.B. 2064, and Proposition 203 are evidenced in Arizona’s SEI classrooms, findings from a previous study were revisited and re-analyzed (see Lillie et al., 2010). Since the Flores Consent Order was intended to be applicable for any state-adopted language programs, the authors assert the Consent Order, and emanating H.B. 2064, is a valid angle for evaluating educational practices in Arizona’s currently mandated SEI model. After an overview of the initial research project, the authors describe the way findings of the Policy in Practice study were re-analyzed from a policy trajectory viewpoint. Through this lens, the authors present an interpretation of the trajectory of language policies in Arizona, starting with the Flores Consent Order and Proposition 203 and continuing through the current practices in SEI classrooms, as dictated by H.B. 2064.


The purpose of the initial qualitative study was to document and describe the characteristics of the four-hour SEI policy in practice. Over a seven-week period during the spring semester of the 2009/2010 school year, researchers utilized ethnographic observation methods to document the four-hour SEI model1 implementation and instruction of English to students classified as ELLs in Arizona.  


With the intent of seeking participation from a variety of districts in order to make claims that would be representative of the different school types within Arizona, researchers anticipated using purposeful sampling of school participants. Unfortunately, the topic of the study was a sensitive one for many schools and districts. Many schools were under scrutiny from the ADE, accused of being out of compliance with the mandated four-hour SEI model. Even with researchers promising confidentiality for all involved, gathering participants for a study intended to examine the practices within SEI classrooms proved difficult. Overall, six school districts were contacted and five agreed to participate.  

Once district officials granted permission, researchers contacted the ELL coordinators, who assisted the research team with setting up contact with school level participants. This process resulted in eighteen classroom (ten at the high school level and eight at the K-8 level), across nine schools within the five districts participating in the study. Table 1 presents the characteristics of the participating districts.  

Table 1. Description of Population



% Free/Reduced Meals



Grade Level

# Participating Classrooms

District A




High School


District B




High School


District C






District D




High School


District E






Note. ELL percentages taken from the 2007-2008 year Census estimates per district, retrieved from www.greatschools.org. All other information from NCES, Common Core of Data, 2009, which represents the 2008-2009 school year, which can be accessed from http://nces.ed.gov/ccd/districtsearch/index.asp.

Researchers gained access to the five participating districts because of their previous working relations with each. The five districts represent various types of Arizona schools including those with both high and low percentages of ELLs, and schools in both rural and urban areas. Participating schools offered a diverse picture regarding percentage of ELL population, locale, and number of economically-disadvantaged students. Therefore, the researchers assert that practices within participating schools afford a depiction of salient themes across SEI classrooms, which may be present in schools statewide.  

In order to preserve anonymity, all districts were coded alphabetically (A through E). To maintain clarity across data sets another alphabet letter was added to the district code to specify school sites. This was helpful when looking at multiple schools within one district. Teachers were then coded by number within the schools at which they taught, adding this number to the alphabetical codes. For example, one district had three schools and four teachers. Therefore, if researchers used a specific quote from a teacher, this was coded as District C, school A, teacher 1 (CA1). A further illustration of coding methods is purposefully withheld in order to maintain anonymity of all participants.


Seven researchers utilized the following data collection methods informed by ethnographic techniques: interviews, observations, and the collection of artifacts and archival data. The researchers collecting data were advanced doctoral students who had past experience with ethnographic note-taking and interviews. Additionally, researchers’ familiarity with schools and prior experience as classroom teachers and ELL Coordinators enabled them to establish rapport and credibility with teachers and district staff. Four researchers had experience teaching at the high school level and three at the elementary level. Researchers were assigned to school sites based on their prior teaching experience.  


At each campus, at least one staff member (e.g., ELL coaches/coordinators, teachers) assigned to work with the four-hour block met with the researchers and participated in a semi-structured initial visit interview prior to the start of observations. This interview served as a means for gathering baseline data about each school, their ELL population, and the implementation process of the four-hour model at each site. Informal interviews were also conducted with additional school staff (e.g., teachers, instructional assistants, and office personnel) in order to provide clarification as needed throughout the data collection and analysis process. Overall, researchers spent a minimum of twenty hours conducting interviews across all sites.

Classroom Observations

In addition to interviews, researchers conducted a total of 264 observation hours, across the 18 classrooms participating in the study. Researchers worked with teachers to schedule observations that would reflect “typical” classroom practices, meaning dates for testing or other school events (e.g., field trips, assemblies) were avoided. Researchers observed each SEI classroom on three separate occasions between late February and early April of 2010. Since the four-hour SEI classes were scheduled differently at high schools than they were at elementary schools, high school classroom visits lasted for four hours each and elementary classroom visits lasted six hours. Before researchers began observations in classrooms, they reviewed information from the ADE, including the state’s SEI Observation Protocol and the PowerPoints used by the state to train four-hour model teachers. This information served as a starting point for considering what might be seen in SEI classrooms. Researchers moved beyond aspects of the four-hour model as deemed important and observable by the ADE to record ethnographic observation notes on all actions in the classrooms. Observation notes included, but were not limited to, examples of instructional practices, teacher/student interactions, and the social and physical environment.

Background Information and Artifacts

Researchers collected background information and artifacts from each site. Some background information, such as teacher certification, was available to the public and easily accessible. Additional artifact collection varied across sites in relation to the availability and willingness of personnel to share such artifacts. Examples of various items collected included lesson plans, course materials, curriculum maps for SEI, district curricular overviews for the ELD levels, and classroom rosters showing class size and proficiency levels of students. Researchers also accessed archival data from information provided online by the ADE to the public (see e.g., http://www.ade.state.az.us/oelas/). This included specific policies, laws, instructional suggestions per the SEI training to teachers and administrators, and other SEI model implementation PowerPoints created by the ADE.  

Data Analysis

Data were analyzed using Erickson’s (1986) method of modified analytic induction. Data were organized chronologically, as well as by data collection method (i.e., interviews, observation, artifacts). Seven researchers worked on the study, and each simultaneously collected and analyzed data on specific observation sites. While data were collected, researchers independently read over their data, taking notes along the way to make sense of the data as a whole. Researchers then used weekly meetings to discuss emerging themes within each independent data set. Afterwards, researchers came to a consensus on a list of four themes that described what the researchers had observed: (1) SEI classroom organization and environment, (2) materials and resources used in SEI classrooms, (3) English language development practices, and (4) promotion and graduation of ELL students. Researchers used these themes as codes for the final stages of collaborative data analysis.  

After all data had been collected, researchers convened for four days, during which time more than 30 hours were spent doing a final joint analysis of the data. Once all data were coded, researchers read the evidence under each code and created a list of beginning assertions. Researchers took each assertion one at a time and looked for supporting and disconfirming evidence across classrooms, time, and data collection methods. Researchers ended up with a set of thirteen major assertions that were used to describe the characteristics of the four-hour policy as practice in K-12 classrooms.


For the purposes of this paper, the authors revisited findings from the Policy in Practice study (Lillie et al., 2010)2. What follows is a discussion of the findings as relevant to the trajectory of language policies in the state of Arizona, beginning with Flores and continuing through H.B. 2064 (2006), as evidenced in K-12 classrooms. Using a constant comparison analysis (Miles & Huberman, 1994), the authors coded the findings descriptively, based on the following three themes interpreted from the Consent Order and the ensuing H.B. 2064: (1) instruction, (2) curriculum, and (3) assessment/reclassification.


Marquez’s 2000 ruling emphasized deficiencies in ELL programs resulting from inadequate funding. Some of these deficiencies directly influenced the instruction ELLs received in their language program, including too many students, not enough classrooms for the students, and a lack of qualified teachers. Additionally, the Consent Order that followed required that the State Board establish rules for daily instruction appropriate to the level of the ELL students. Findings are applicable in understanding themes from the Consent Order because the current four-hour SEI model, as defined by H.B. 2064, delineates student grouping and class-size requirements, mandates teacher training, and describes rules for daily instruction (ADE, 2009). What follows is a condensed version of the findings that relate to ideas surrounding instruction for ELLs as addressed by the Marquez ruling (and thus eventually the Flores Consent Order and H.B. 2064). Specifically, findings related to (1) teacher quality, (2) rules for instruction, and (3) grouping of students in the four-hour SEI model.

Teacher Quality  

The background, qualifications, and skills of the educators who were teaching in the four-hour SEI model were key factors for understanding instruction for ELLs. Teachers are required to be highly qualified and, as a result of post-Proposition 203 mandates, also hold a SEI endorsement. The only exception to the SEI endorsement mandate is for those teachers who already hold a bilingual (BLE) or English as a Second Language (ESL) endorsement. The SEI endorsement can be Provisional (15 hours) or Full (45 hours)3. Table 2 illustrates the endorsements held by the eighteen teachers observed. Although the SEI endorsement is not as comprehensive or in-depth as a BLE or ESL endorsement (de Jong, Arias, & Sanchez, 2010), this was the only endorsement for almost half of the observed teachers (8:18).

Table 2. Teacher Certification at School Districts

Certification Type


Total %

SEI Provisional



SEI Full Endorsement



ESL Endorsement



BLE Endorsement



Note. N = 18, also represents highest completed certification.

Whether they held a BLE, ESL, or SEI endorsement in accordance with state policy, all SEI classroom teachers were observed to focus their instruction on English. However, researchers identified differences in instructional practices between teachers who held a BLE or ESL endorsement and those who held a SEI endorsement. Some teachers did not feel adequately supported to “really know what we’re doing” (AA1, personal communication4, 2/25/2010); this lack of confidence was noted by many teachers, regardless of how experienced they were or thoroughly trained they had been in implementing the four-hour SEI model. However, many secondary teachers who held only the SEI endorsements indicated that they felt “confused” about how the SEI model was supposed to work within the discrete hour-by-hour ADE timeframe (AA2 & BA1, 3/2010).

Teachers who held a BLE or ESL endorsement reported they drew on their past experiences teaching in ESL or bilingual classes, as well as their previous preparation for said endorsements, as means for negotiating what the state was asking them to do in the four-hour model and what they felt was best for their students. As one teacher noted while referencing lessons plans and the posted objectives in her classroom, “I do what I have to do on paper and then I teach the way I need to for my students” (CA, p.c., 3/19/10). This quote illustrates that some teachers were aware of the tensions between what was expected of their instruction within their SEI classrooms and what they understood to be best practices for ELLs.  

In regards to teachers being “highly qualified”, a discussion about credentials and background knowledge in content areas becomes important regarding instruction for ELLs at the secondary level. Although the intent of the four-hour SEI model is to teach ELLs English through direct English instruction related to components of reading, writing, speaking, and listening (ADE, 2009), a few teachers at the secondary level were observed to incorporate content into their English instruction. While it is admirable that teachers attempted to integrate content, the participating SEI teachers did not identify themselves as being highly qualified in any content area outside of English. When SEI teachers do not hold the credentials required of them to be considered “highly qualified” in the content areas, attempts to incorporate science, social studies, or math concepts cannot be considered comparable to the content instruction other students receive from mainstream, “highly qualified” content area teachers.  

Along with specifying regulations for teacher endorsements and qualifications for working with ELLs, post-Proposition 203 directives outline rules for instructional practices within the four-hour SEI model. Findings illustrate how teachers enacted the rules for instructional practices in different ways. A look at the findings specific to instructional practices affords an understanding of how the rules for the current SEI model relate to the intent of the Flores Consent Order.

Rules for Instruction  

Rules for instruction within the current SEI model stipulate that all materials and subject matter instruction be in English (A.R. S. •15-751; A.R.S. •15-752) and that error correction should be used as a means for accelerating English acquisition (OELAS, 2009-2010). In the classrooms observed, researchers noted all materials were in English and all teachers used English for instruction. However, the ways in which teachers used English instructional practices and promoted language correctness varied across classrooms. Some teachers were prepared to value students’ language acquisition process and use of their native language (L1), but others effectively silenced their students by remaining steadfast to the ADE’s focus on using error correction and discrete English instruction.  

English-only instruction. Although the rule, “Arizona law requires materials and subject matter instruction to be in English” (A.R.S. •15-751. Definitions, 5 and A.R.S. •15-752), does not explicitly exclude student use of their L1, the overwhelming interpretation and implementation of the policy is one that strictly enforces the use of English-only in classrooms (Mahoney, MacSwan, Haladyna, & García, 2010). This “English-only” interpretation was evidenced in 11 of the 18 classrooms observed in the study. In these rooms, teachers posted “Speak in English” posters on their classrooms walls, made it taboo for students to use their L1, and rejected students’ prior knowledge in their native language. For example, one teacher encouraged a type of “language police” in her classroom, asking students to report out when they heard the use of a language other than English, to then receive school “money” for doing so (BA1, 3/30/2010). Another teacher scolded a student for referring to a word (collar) in class as a Spanish word, telling the student, “You may think that is a Spanish word, but it is not, we only have English words in this room” (CA4, p.c., 3/26/2010). In the majority of the classrooms observed, the interpretation of instruction in English to mean “English-only” prohibited students from communicating in their L1 during and outside of direct instruction, but not all teachers observed promoted these English-only practices.  

Seven of the eighteen teachers allowed ELLs to use their L1 in the classroom. Examples of such instructional practices included a teacher who encouraged students to communicate in their shared L1 as they worked collaboratively toward a final product in English (CB2, 3/5/2010) and one who allowed students to seek clarification or guidance from their peers in their shared L1 (DB2, 3/26/2010). Additionally, teachers permitted students to access information in their native language (on the computer, in native language/English dictionaries, or in L1 texts) to further their understanding of classroom lessons (CB3 & DB3, 3/2010). Finally, some teachers used students’ L1 to promote a comfortable environment in the classroom (CA1, 3/19/2010). In these classrooms, teachers maintained a focus on instruction in English, while recognizing the value of students’ linguistic knowledge as a foundation for English language development (Wong-Fillmore & Snow, 2002).    

Out of the seven teachers observed promoting English acquisition while simultaneously valuing students’ native language and culture, six held their BLE or ESL endorsement. Of the teachers who interpreted instruction in English to mean “English-only” in the classroom, the majority held only an SEI endorsement. Concepts related to the value of a students’ L1 in developing English are underemphasized in the curriculum for the SEI endorsement training (see e.g., ADE’s SEI Curricular Framework, ADE 2007). In this way, a teacher’s preparation was seen to influence the way they interpreted the instruction in English rule.  

Error Correction  

While Arizona law mandates ELLs receive four hours of English instruction and the Task Force materialized this law into policy when they created the SEI model, it was the ADE who emphasized that in order to accelerate English acquisition, SEI teachers should focus on error correction (OELAS, 2009-2010). Error correction was emphasized as a “SEI Super Strategy” at ADE trainings and meetings with ELL Coordinators. Error correction was also included in the Observation Protocol created and used by ADE to monitor compliance with the four-hour SEI model. While observing in SEI classrooms, researchers observed that teachers corrected student errors in different ways.  

Although some teachers focused on language as a communicative act and modeled the correct language without interrupting regular classroom communication, most focused on correctness to the point of stifling learning. Examples of instructional practices that interrupted learning conversations include a teacher who stopped a student mid-response instructing him to restate his answer in a complete sentence (CB1, 3/2010) and another who prompted a student to repeat a word more than three times until the student pronounced it correctly (CA3, 3/2010). In scenarios such as these, researchers observed students to lose their train of thought and struggle to pick up their original line of thinking. The instructional focus on grammar, pronunciation, and the use of complete sentences broke up the authentic conversation and communication in the classrooms between teacher and student and among students. In these instances, instructional practices valued language correctness over the role language can serve for communication among learners.

The fact that researchers observed many teachers using instructional practices that focused on language correction is not surprising, given the emphasis on these practices by ADE.  It is troublesome however, that ADE expects the use of these instructional practices, given research shows that a hyper-focus on language correctness is not always appropriate for language learners (Krashen, 1984; Lightbown & Spada, 2006; Roberts & Griffiths, 2008). ADE’s determination that instruction should focus on error correction, without attention to what kind of error correction is beneficial for students at various stages of acquisition, ignores the intent of the Consent Order: that instructional practices be appropriate to the level of the ELL students.  


Based on the ruling “Arizona law requires English language learners to be grouped together in a structured English immersion setting” (A.R.S. •15-751. Definitions, 5), the Task Force delineated the requirements for the number of ELLs in each classroom and the preference for grouping students based on English proficiency levels (ADE, 2009). While these programmatic features seem to correlate with recommendations from the Consent Order that the language programs monitor class size and provide instruction appropriate to the levels of ELLs, the requirement that ELLs be grouped together for English instruction, materialized as ELLs being segregated from the mainstream student population for a majority, if not the entirety, of the school day, and that the students with the least familiarity with English be grouped together where there would be no opportunity to interact with other students who knew the language.   

Within the high school districts observed, ELLs were in one classroom with one teacher5 for four hours per day and with other mainstream classrooms/teachers for the remaining two periods of the day. At the elementary level, ELLs were in SEI classrooms the entire day, receiving four hours of ELD as well as math instruction from their SEI teacher and with the same group of ELL peers. When elementary ELLs left their classroom for specials such as art or music they remained grouped with the students from their SEI classroom. In short, ELLs in the SEI model were spending their entire day with their fellow ELL peers. Lunch and recess provided the one consistent opportunity across K-12 schools wherein ELLs had opportunities for contact and interaction with native English-proficient students. However, at the elementary levels observed, all cafeterias were arranged so that students were forced to sit with students from their classroom. At the high school level, although allowed to sit wherever they chose, teachers acknowledged that ELLs would “hang out and stick with each other instead of with others” (AA2, p.c., 3/2/2010), noting that ELLs seemed to feel “safer” when clustered in groups based on ethnicity, culture, and language (AA1, p.c., 3/2010). While this is not unusual for students, teachers did not try to ameliorate the clustering or help integrate different groups of students with one another. No effort was made to provide the ELLs with the opportunity to interact with non-ELL students. Outside of the mandated four hours of SEI, whether because of rules for lunchroom seating or student preference, the segregation of ELL students from non-ELLs was complete.

Although teacher training, rules for instruction, and requirements for student grouping may have been intended to promote instructional practices that are appropriate for ELLs, this intent was not evidenced in the majority of classrooms observed. On the contrary, findings illustrated that teachers who held only an SEI endorsement lacked the preparation needed to value students’ linguistic and cultural knowledge as a resource for English acquisition. In addition, “instruction in English” was interpreted by most to mean using “English only” in the classroom. Finally, the requirement that ELLs be grouped with other ELLs for four hours of English instruction, effectively segregated ELLs from the mainstream student body population for the entire school day.


One of the points to the Flores Consent Order is that the curriculum of all language learner programs includes daily instruction in English Language Development (again, ELD). The current four-hour SEI model, established by the Task Force per H.B. 2064, evidenced this by requiring ELLs be placed in SEI classrooms for four hours a day in which they are taught an ELD curriculum. However, the Consent Order also required that language programs provide ELLs with daily instruction in the content areas that is not only appropriate for ELLs, but also comparable in amount, scope, and quality to the curriculum provided to English-proficient students. Findings demonstrated that because the current SEI model requires ELLs to receive four hours of ELD curriculum daily, ELLs’ access to subject area curriculum outside of English was severely limited and not comparable in amount, scope, and quality to content area curriculum provided to their English-proficient peers.

This curriculum segregates ELLs from native English-speaking peers, creates a physical and social isolation so that participation in the mainstream school culture is limited, and is still lacking in resources and materials in order to be sufficient for ELLs thus not addressing that aspect of the Consent Order. Finally, the impact of the curriculum on an ELLs’ chances of a typical four-year high school completion in regard to graduation is severe. These three areas (content, resources, and promotion/graduation), are all those in which one can analyze the components of the Flores Consent Order as it relates to the issue of comparable curriculum in amount, scope, and quality.


An overview of the SEI curriculum’s components is needed in order to compare an ELLs’ curriculum to that of English-proficient students’. The SEI curriculum relegates ELLs to four hours of ELD, with one hour each of “English Reading”, “English Writing”, “Conversational English and Academic Vocabulary”, and “English Grammar”. Core content, such as math, science, and social studies, is excluded from the ELD curriculum. With this strong emphasis on learning language, ELLs are expected to become proficient after one year to be then exited from the program into mainstream classrooms.

While the SEI model does not exclude access to content area curriculum outside of the four hours designated to ELD, findings showed that ELLs’ access to quality content curriculum was severely limited. For elementary ELL students, being designated as an ELL meant that they were in a SEI classroom for the entire day.  In a typical elementary classroom, the teacher provides students with curriculum across all subject areas; however, SEI teachers reported that because they were required to spend four hours each day teaching English, in addition to lunch and the required elective (e.g., P.E. or music), there was very little time for other content area curriculum. In the classrooms observed, elementary ELLs had access to a minimal amount of math instruction (averaging less than 40 minutes a day), and even less access to science and social studies content (some teachers reported having no time). Being designated to SEI classrooms for the entire day meant that ELLs faced barriers to a content-rich curriculum. Since elementary ELLs are placed in their SEI classrooms for the entire year, the absence of core content area curriculum was compounded as the school year continued and the divide between the scope and amount of content in the curriculum which ELLs received, as compared to their grade-level English-proficient peers, is great.

Along with lack of time for content area instruction, differences in curriculum were also evident through visual representations on the walls in SEI classrooms. Posters and visuals depicting the SEI model curricular components were present in six of the eight elementary classrooms observed but not in mainstream classrooms. Signs stated “Practice your English 24/7!” “Tell me in a complete sentence,” and “50/50,” the latter of which refers to the amount of time for teachers and students to speak during the four-hour block. There were also poster-sized copies of the Discrete Skills Inventory (DSI)6 and of the Language Star7. While it is not uncommon for classroom walls to depict curriculum, the wall displays in SEI classrooms illustrated the extreme weight of ELD to the exclusion of other learning. The clear, visible distinction between mainstream classrooms and ELL classrooms created the sense that ELL students were only expected to learn English and not any other subjects. This cannot help but send the message that these students are not capable of learning more. It is a highly deficit perspective. These differences continued at the high school level.

Secondary school ELLs did not necessarily have scheduling barriers like the elementary students, as some were able to attend content courses during their two hours of the day that were not explicitly the four-hour block. However, oftentimes ELLs met resistance from the mainstream content teachers, thus impacting their access to curriculum comparable to that received by non-ELLs. It was reported by some SEI teachers that mainstream content teachers at the high school level viewed ELLs as “others” and “those students,” thus stigmatizing ELLs in the larger school community (AA1, AA2, & BA1, 3/2010). One SEI teacher remarked that among other mainstream teachers there was a “morale” wherein mainstream teachers “just don’t like the ELLs, particularly the Spanish speaking ones” (BA1, p.c., 3/22/2010). At two of the three high school districts, the SEI teachers participating in the study commented that non-SEI teachers “looked down” on ELLs (AA2, BA1, 3/2010) and had mentioned their beliefs that “those students did not want to learn”, were “not able to learn” [content curriculum outside of SEI], or were mostly “just a behavior issue” (AA1, AA2, BA1, 3/2010). As one SEI teacher retorted, “it’s not a behavior issue, but the way [non-SEI] teachers treat them, they become at-risk” (BA1, p.c., 3/22/2010). Research has shown that what teachers believe about students affects the ways in which they teach (Kaplan & Leckie, 2009; Nieto, 2000; Pajares, 1992; Richardson, 1996), as well as the way students learn (Jussim & Harber, 2005). Teachers can have powerful effects on the learning of students from stigmatized groups (Jussim & Harber, 2005). Therefore, the researchers make the claim mainstream teachers’ beliefs about ELLs impacted language learners’ access to content area curriculum.

While all students in high school participate in classes related to reading and writing, the differences in amount and scope between the SEI curriculum and the mainstream language arts curriculum was marked. Physically being in the ELL classroom and participating in the reduced ELD curriculum enhanced this perceived divide between ELL students and non-ELL students. In some of the high schools SEI teachers were required to teach “regular” English classes (e.g., American Literature) to native English-speaking students during periods where they were not teaching ELD. Student work on the walls, as well as the objectives listed on the whiteboard, spoke to the two detached curriculums and highlighted the major differences for any observer to see. In these classrooms, the objectives that were posted for the mainstream English classes were more academically appropriate for a typical high school curriculum compared to those objectives listed for the ELL students. Moreover, the materials and the lessons (again as depicted through observations and the visible objectives) reflected more grade-appropriate content than curricular plans for the non-SEI classes. In one classroom, the researcher observed that the ELL work, displayed in the back corner of the room, was noticeably different in terms of academic content as compared to that of the mainstream students’ work. The ELL work consisted of student-created invitations (not centered on a content theme or connected to grade-appropriate English standards) while the non-ELL student work exhibited essays that showcased the high school’s American Literature curriculum (AA, 3/2010). Further, with the four-hour block emphasizing ELD and not content typical of mainstream high school English courses, each year an ELL is in the SEI classroom, he or she falls further behind in the grade-level content curriculum.

The placement of SEI classrooms at the secondary schools furthered the distinction between ELLs’ daily curriculum and non-ELLs’ classes. Even though the SEI model is considered to be part of the English requirements for school8, researchers noticed that the SEI classrooms were not always located within, or near, other classrooms provided for the English department (or any content area department). For example, a junior high SEI classroom was located on a floor with elective course classrooms (i.e., art), while all the other junior high content classrooms were on a different floor. At two of the three high school districts observed, this separation was heightened. One district had the ELL classrooms in completely different buildings from the content area, mainstream English classrooms. The mainstream classrooms for English had the same department chair as the SEI classrooms (again, since they are part of the English Department and are designated as English classes) and were all located in one building, thus sharing common work areas, walls, and other school components (such as restrooms). The SEI classrooms were across the campus, in a building with the severely disabled, limited functioning (e.g., life skills) special education students, and emotionally disturbed students. These ELLs were completely removed from any core content area classrooms, meaning they were not visible to other non-ELL students or mainstream teachers until and if they went to a non-SEI class.

Across K-12 classrooms, these clear, noticeable distinctions between mainstream curricular content and the SEI curriculum intensified the separation between the two groups of students. Observations showed that ELLs were receiving content that was fundamentally different from the non-ELLs, regardless of the fact that the purpose of the ELL curriculum is to focus on English skills. This emphasis ignores the importance of maintaining a highly motivating, rigorous curriculum, the latter of which is shown to help students “perform better…than those with the same or better language proficiency, particularly at the high school level” (Gándara & Orfield, 2010, p.11). Moreover, as Arizona law allows one grade-level English credit for the entire four-hour block, this begs the question if the ELL students really receive a grade-level appropriate course in English that will prepare them for graduation, postsecondary education, or work. The curriculum, devoid of rich content learning, is not comparable in amount, scope, or quality to that of native English-speaking peers. ELLs are receiving an unequal education.  


Marquez ruled in 2000 that there were many deficiencies with ELL programs, agreeing with the plaintiffs that inadequate funding resulted in the shortcomings he noted, such as inadequate tutoring/after-school programs, poor resources, and not enough funding to provide sufficient teaching materials. Therefore, per the Flores Consent Order, any instruction given to ELLs should reflect academic/grade-level appropriateness that is cognitively/academically equal to that of the mainstream curriculum. It is further stipulated that if ELLs are not progressing academically, as now defined by the Arizona state statutes (see S.B. R7-2-306 as pursuant to A.R.S. •15-741), schools are required to provide them with compensatory education (i.e. tutoring, after-school programs, and/or summer school).

When examining the implementation of SEI in Arizona schools, it became apparent that there was a lack of resources for adequate instruction (particularly at the high school level) and there were disparities in ELLs’ access to academic support outside of the school day. While most of the elementary schools observed did not seem to be lacking in resources, none had summer school or after-school programs in place to support further learning. This was also the case at the high school level, where in addition to not providing the required tutoring and/or summer school, there was a severe lack of other necessary resources. The inadequate resources, such as not being age/grade-appropriate, and the disparity in access to materials revealed the degree to which the Flores Consent Order mandates are (or are not) being met in terms of providing ELLs with a curriculum that is appropriate in amount, scope, and quality to the mainstream curriculum, and the needed resources to access and keep pace with that curriculum.

Lack of Materials  

High school SEI teachers lamented the lack of materials and the appropriateness of the resources at their disposal. Many resorted to using worksheets to instruct their students. The worksheets teachers were able to access (or created themselves) were often reduced in substance, the result being that high school ELLs were working on worksheets more suitable for elementary and middle school students. Some teachers included supplemental reading materials on their classroom bookshelves in order to augment the lack of resources, while another teacher “stole” books from the English Department copy room because they were department sets to be used as needed (AA2, p.c., 3/2/2010). The teacher explained that although it would be a stretch for her ELLs to comprehend the level of English used in Shakespeare, using it in her classroom allowed her ELLs exposure to curriculum more comparable to what students in mainstream English classes were receiving. Unfortunately, in classrooms where teachers supplied their own reading materials, the lack of appropriate materials sometimes left 9th-12th graders with choices such as Clifford the Red Dog (a beginning reader for five to six-year-olds).

A lack of resources at the elementary level affected the content curriculum delivery outside of the four-hours devoted to ELD. In one elementary district, literary and content area resources, which might have been beneficial to the lesson at hand, were unavailable to SEI teachers. Therefore, when SEI teachers attempted to teach content, outside of the four hours of the school day devoted to ELD instruction, they had limited resources from which to do so. For example, one SEI teacher was observed offering a science lesson to the ELLs. The teacher did not have a science text but had managed to gather some materials from another mainstream teacher in the school. Unfortunately, the materials gathered did not correlate to a grade-appropriate science standard. Furthermore, the teacher was only able to gather enough materials to model the experiment, as opposed to students conducting the experiment themselves (CA2, 3/5/2010). In this example, ELLs’ learning was focused on labeling objects while non-ELLs at the same grade level were provided with enough resources (including time and materials allotted for content area learning) to be able to properly learn how to make hypotheses and predictions as per the state standards for Science.

Disparity in Access  

The disparity in access to other supplies and teaching materials was heavily noted at the high school level. What was most alarming about the missing resources was the disparities regarding textbook availability to high school students across districts. In some districts, SEI teachers had no textbooks to use with their students and resorted to using three-year old workbooks, in which many were already written (A & D, 2010). One teacher photocopied pages of an unwritten workbook to distribute to the ELL students instead, noting that there would be no new workbooks in the coming years (AA1, 3/2010). In District A, teachers were limited in their use of paper and were only allowed three reams of paper per year; one SEI teacher used the allowance up by October. Overall, the lack of materials and resources was a practical hindrance in the implementation of the four-hour SEI model, especially at the secondary level. Teachers worked to the best of their ability given the circumstances but were handicapped in their ability to teach ELLs because they were not provided with the materials they needed.

At elementary and high school districts alike, there was not enough funding allocated for summer school or after-school programs to help ELLs catch up with their non-ELL peers, thus meaning the requirement mandated by the Flores Consent Order was not being met. In conjunction with the limited opportunities to get content requirements because of the prescribed curriculum, the lack of summer school and after-school programs meant that ELLs were restricted in being able to satisfy missing course requirements outside of their school day/year. Additionally, the strict focus on ELD meant that ELLs were not making academic progress comparable to their English-proficient peers. More importantly, at the high school level, the absence of this necessary compensatory instruction virtually ensures that secondary ELLs will be unable to gradate with their same-age peers, particularly if they remain in the four-hour SEI model for more than one year.

Graduation/Promotion Effect  

The promotion of ELLs was consistent in both elementary and secondary schools. Participation in the SEI model did not prohibit ELLs from advancing one grade level each year. However, this does not mean that ELLs are on par with their native English-speaking peers in regard to academic content. High school ELLs who fail to exit the block after one year are also behind in the requirements to meet high school graduation and unable to make up these courses if after school or summer school opportunities are not provided.

The findings show that ELLs are not passing out of the SEI model in the intended one-year timeline, per A.R.S. •15-752 (which stemmed from H.B. 2064) and that the four-hour block is placing students at a severe disadvantage for high school graduation. After interviewing more than twenty professionals in all five districts, the answer to whether or not students were exiting the SEI model in one year was a resounding “no.” Almost every educator noted that students took more than a year and most likely three or four years to exit the program. Two high school administrators stated emphatically that “no”, ELLs were not exiting in one year (AA3, 2/17/2010; DA3, 3/5/2010) and that “the time frame depends on the student” (DA1, p.c., 3/5/2010) due to other factors, such as prior schooling, amount of English skills, and age. One administrator remarked that ELLs who came with a strong schooling background and literacy in their native language were able to exit in two years, but not one (AA3, 2/17/2010). In District A, an alarming finding was when an SEI teacher reported that one counselor transferred all would-be senior class ELLs to a fast-track program at a different night school because the ELLs were short on required high school credits and would not be able to graduate (AA1, 3/2010).  

Graduation requirements for content credits are also increasing, meaning ELLs (who only receive one content credit per year for all four classes of ELD) average three content credits a year, assuming they pass all of their classes. When ELLs cannot exit in one year, which most do not, this is compounded exponentially. The achievement gap between ELLs and mainstream students is persistently widened since ELLs are not exiting in one year; so while SEI involvement does not prevent promotion, it severely impedes graduation. ELLs are restricted from accessing grade-level academic content that is necessary for their overall schooling experience and graduation.


The intent of the current four-hour SEI model is that after one year of the language program, ELLs will acquire enough English skills to pass a language exam, exit the model, and be successful in mainstream classrooms. Findings related to the assessment and reclassification procedures used within the current four-hour SEI model show that students do not exit the model after one year. Furthermore, when ELLs do exit they struggle in mainstream classes.


A look at the policy trajectories related to assessment and reclassification is needed before the findings are revisited. Per the Consent Order, Arizona ELLs who pass the state-adopted English language proficiency exam (the AZELLA)9 are labeled reclassified (RC), exited from the language program, and entered into mainstream classes. However, RC students are still considered participants of the language program for a period of two years during which their progress is supposed to be monitored. RC students are re-tested at the end of each year using the AZELLA. If students do not pass, they are no longer considered RC and are re-enrolled in the SEI classroom and relabeled as ELLARs (ELLs after reclassification). Along with monitoring for English proficiency, state board rules describe that students “ will be monitored in reading, writing and mathematics skills and mastery of academic content areas, including science and social studies…Students who are not making satisfactory progress shall…be provided compensatory instruction or…be re-enrolled in an ELL program” (S.B.R. R7-2-306).

Kindergarten students appeared to be the most likely to pass the AZELLA at the end of one year, but when many of these same students were reassessed at the end of first grade (as per the monitor requirement of the Consent Order), they did not pass the AZELLA and were placed back into the SEI program starting in second grade. This was probably due to irregularities in the test (see e.g., Florez, 2010), and the U.S. Department of Justice, noting these problems, demanded that Arizona stop using the AZELLA and develop a more valid test. The instability and fluctuation of a students’ SEI program involvement interrupted their access to grade-level content thereby affecting their ability to meet grade-level academic progress.

While studies have shown that the AZELLA is a poor measurement of proficiency and has issues with reliability and validity (Florez, 2010), this study also found that reclassification rates are a poor indicator of success. Teachers reported that ELLs struggled once mainstreamed, largely due to the lack of language support from mainstream teachers. One teacher and an ELL Coordinator from two different high schools separately remarked that the teachers “do not understand that reclassified ELLs are still ELLs” (AA3, p.c., 2/17/2010; BA1, p.c., 3/22/2010). Similarly, ELL coaches at the elementary level discussed how mainstream teachers misunderstood the amount of support RC students would need (CA5 & EB1, 3/2010).  

High school personnel also noted that high reclassification rates were poor indicators of student success. One school administrator noted that many of the RC students were quickly relabeled as ELLARs at the end of their first year on monitor status (DA3, 3/5/2010). In one high school district (D, 2010), teachers shared that the school had been documenting the academic success of ELLs by comparing the grades ELLs earned while in the SEI model to the grades the same ELLs earned during their first year as reclassified students in mainstream courses. While in the four-hour model, 14 of 29 ELLs received one or more “F” grades in a semester; once these same students were RC, all 29 got at least one F in their classes. Given that the ELL students had limited exposure to the regular curriculum through participation in the SEI model, it would hardly be surprising that ELLs are ill-prepared for the content and rigor of mainstream classrooms.


Analyzing the decade-long policy trajectory affords a picture of the devastating effects on ELLs’ education through the implementation of Arizona’s current SEI model. The convergence of these policies appears to have resulted in a “perfect storm” for ELL children. A major implication of the on-going saga of Flores, over twenty years, is that extended litigation initiated by states can effectively challenge and delay the implementation of equal opportunity and an equal “chance to learn” (Weinberg, 1995) for ELL children. Scholars have continually noted the failure to have or develop a coherent plan in the years immediately following Proposition 203 (Wright, 2004). Now, once there is a plan, it fails to achieve its basic state purpose (see e.g., Davenport, 2011). A generation of kids has been lost. From Flores (1992) forward, we have underserved nearly two generations.  

The initial challenge of Flores was to ensure adequate funding for the instruction of ELL students. Now, after nearly two decades of litigation, serious concerns remain regarding whether ELL students are receiving an equal chance to learn. Proposition 203 mandated English-only instruction in 2000, yet it took the state of Arizona six years to legislate its SEI model through the Task Force’s designed components, the efficacy of which remains difficult to determine. It was two more years before the model was implemented in schools in 2008. Although information on the characteristics of the four-hour SEI model as implemented is limited, during the implementation period, some school administrators noted serious concerns about the model (Grijalva, 2009). Concerns were also registered regarding the minimal teacher preparation provided for the SEI endorsement (Moore, 2008). In the Arizona Auditor General’s first report, significant discrepancies were noted among programs in their implementation of SEI during 2006-2007 (Davenport, 2008), and Lillie et al. (2010) noted that educators implementing the four-hour model expressed tensions between what was expected within the SEI model and what they understood to be best practices in providing ELL students an equal chance to learn.  

Outside of the instruction within SEI classrooms, additional concerns were noted regarding the ways ELLs were physically and socially segregated from non-ELL students. The very design of the four-hour SEI model not only isolates students, but tends to stigmatize them. Furthermore, in order to ensure equal access and opportunities to learn, ELLs must be able to keep pace academically with their peers. The findings of this study through the policy trajectory lens indicate that the four-hour SEI model creates a greater divide in access to the scope, amount, and quality of curriculum to which ELL students are exposed compared to their English-proficient peers. Being physically separate and given a reduced curriculum accentuates the real and perceived divide between ELL students and non-ELL students. As a result, all ELL students lag behind their grade-level peers in content-level achievement, resulting in an unequal chance to learn.

Retention and re-entry into SEI for those in high school further presents major hurdles for ELL students to complete graduation requirements. Unequal opportunity typically ensures that secondary ELLs will be unable to gradate with their peers, especially if they are cloistered in SEI classrooms for more than one year. Arizona also seems to have no plan for what to do with the students who do not exit after one year; students languish in the four-hour block year after year and are therefore continually held back from reaching their full potential and access to a wider school curriculum. Further, given limited opportunities to access content as well as the complete absence of summer school and after-school programs as evidenced here, ELLs are less likely to fulfill curriculum requirements and will fall behind. Arizona has no way for ELLs to “catch up” to their mainstream peers by the elimination of opportunities outside of the typical school day (e.g., compensatory education, summer school). This would appear to not only defy the mandates within the policy trajectory (e.g., the Flores Consent Order), but also may be a blatant violation of Lau v. Nichols (1974).

The inadequacies of the Arizona SEI model and the limitations of SEI teacher-preparation put well-meaning teachers and administrators in a difficult position. In the current “results-based” environment, it is easy for state departments of education to blame teachers and students who have been saddled with poorly conceived policies, inadequate or inappropriate preparation, questionable assessment instruments, limited materials, and segregated school environments. Those students and teachers who struggle under these conditions deserve better. It is ironic, if not tragic, that all the lessons learned concerning the history regarding the unequal treatment of language minority children are being ignored and repeated. The unequal legacies of early and mid-twentieth century of segregation and under-education are being duplicated (Blanton, 2005; Weinberg, 1995), and the struggle for educational rights to equitable education for language minorities continues (Wiley, 2007). Fifty years ago, amidst the victories in the struggle for equal access and rights in education, who would have thought it would still be so in the twenty-first century?


1. The SEI model is also called the four-hour model, the four-hour block, and ELD. For the purposes of this paper, the four-hour model or four-hour block are interchangeable and ELD refers to the instruction ELLs receive in the model.

2. Many people contributed to this study. The original citation can be found in the reference list of this paper, but is also accessible at http://civilrightsproject.ucla.edu/research/k-12-education/language-minority-students/policy-in-practice-the-implementation-of-structured-english-immersion-in-arizona

3. Educators certified after 8/31/2006 must have three semester hours of coursework related to SEI to receive their Provisional SEI endorsement, which is valid for three years. Those certified before that date must complete 15 clock hours, or one semester hour, of SEI training. Full SEI endorsements are only eligible to those who qualify for the SEI Provisional, and then either complete three semester hours or 45 clock hours of professional development in teaching ELLs (see Arizona SBE Rules, R7-2-613.J).

4. This will be scripted as p.c. for the remaining document.

5. In two districts, ELL students had two ELD instructors.

6. The Discrete Skills Inventory (DSI) is defined by the ADE as “the specific teaching/learning objectives derived from the Arizona K-12 English Language Learner Proficiency Standards approved by the Arizona State Board of Education (SBE), January 26, 2004, and refined as needed to remain synchronized with the Arizona K-12 Academic English Language Arts Standards” (ADE 2008, pp. 1-2).  Specifically, DSI “is a sequential series of English language skills that provide a guide to teaching the grammatical foundations necessary for students…and provides the critical grammatical foundation for achieving proficiency in listening, speaking, and writing” (ADE, n.d., p. 1).

7. The Language Star consists of the “five main components of the ELD classroom” (ADE, 2010). These components are syntax, lexicon, semantics, morphology, and phonology, with vocabulary over-arching all five star points.

8. For example, in District A the SEI classes counted for 1 English credit and were considered to be English Department class requirements (albeit only for ELLs), and the teachers were considered English teachers within the English Department.

9. AZELLA stands for Arizona English Language Learner Assessment. The test was found to be invalid for this purpose by the Department of Justice (DOJ) and Arizona is currently negotiating with the DOJ about the future of its testing.


Arias, M. B., & Faltis, C. (2007, April). Critical race theory and adolescent Mexican-immigrant students in high school: Stories of success and silence. Paper presented at the annual meeting of the American Educational Research Association. Chicago, Illinois.

Arizona Department of Education. (n.d.). Discrete skills inventory (DSI). Retrieved from http://www.ade.state.az.us/oelas/

Arizona Department of Education. (2006). Accountability: Office of English language acquisition services. Retrieved from http://www.ade.state.az.us/oelas/

Arizona Department of Education. (2007). Curricular framework for SEI endorsement.  Retrieved from http://www.ade.state.az.us/oelas

Arizona Department of Education. (2008). Structured English immersion models of the Arizona English language learners task force. Retrieved from http://www.ade.state.az.us/ELLTaskForce/2008/SEIModels05-14-08.pdf

Arizona Department of Education. (2009).  English language learner monitoring process for federal and state compliance. Retrieved May 28, 2010, from http://www.ade.state.az.us/oelas

Arizona Department of Education. (2010). Administrator’s model implementation training 06-04-2009 [PowerPoint slides]. Retrieved from http://www.ade.state.az.us/oelas/  

Arizona Revised Statutes. (2000). Article 3.1 English Language Education for the Children in Public Schools. Retrieved from http://www.azleg.state.az.us/ArizonaRevisedStatutes.asp?Title=15

Ball, S. J. (2000). What is policy? Texts, trajectories and toolboxes. In S. J. Ball (Ed.), Sociology of Education (pp. 1830-1839). New York: RoutlegeFalmer.


Blanton, C. K. (2005). The strange career of bilingual education in Texas, 1836-1981. College Station, TX: Texas A&M Press.

Combs, M. C., Evans, C., Fletcher, T., Parra, E., & Jiménez, A. (2005). Bilingualism for the children: Implementing a dual-language program in an English-only state. Educational Policy, 19 (4), 701-725.

Crawford, J. W. (2000). Anatomy of the English-only movement. Retrieved from http://ourworld.compuserve.com/homepages/JWCRAWFORD/anatomy.htm

de Jong, E. J., Arias, M. B., & Sánchez, M. T. (2010). Undermining teacher competencies: Another look at the impact of restrictive language policies. In P. Gándara & M. Hopkins (Eds.), Forbidden language: English learners and restrictive language policies (pp. 118-136). New York: Teachers College Press.

Davenport, D. K. (2008). Baseline study of Arizona’s English language learner programs and data, fiscal year 2007. State of Arizona Office of the Auditor General. Retrieved from http://www.auditorgen.state.az.us/Reports/School_Districts/Statewide/2008_April/ELL_Baseline_Report.pdf

Davenport, D. K. (2011). Arizona English language leaner program, fiscal year 2010 (Report No. 11-06). State of Arizona Office of the Auditor General. Retrieved from http://www.azauditor.gov/Reports/School_Districts/Statewide/2011/ELL_Report.pdf

Erickson, F. (1986). Qualitative methods in research on teaching. In M. C. Wittrock (Ed.), Handbook of research on teaching (3rd ed.) (pp. 119-161). New York: Macmillan.

Florez, I. R. (July, 2010). Do the AZELLA cut scores meet the standards? A validation review of the Arizona English language learner assessment. Los Angeles, CA: Civil Rights Project: University of California, Los Angeles.

Flores v. Arizona. No. 07-15603, No. 07-15605, 1014, United States Court of Appeals for the Ninth Circuit, 577 F.3d 1014; 2009 U.S. App. LEXIS 18092.

Gándara, P., & Orfield, G. (2010). A return to the “Mexican room”: The segregation of Arizona’s English learners. Los Angeles, CA: Civil Rights Project: University of California, Los Angeles.

Grijalva, G. G. Implementing language policy: Exploring concerns of school principals.  (Unpublished doctoral dissertation). Arizona State University, Tempe, Arizona.

Jussim, L. & Harber, K. (2005). Teacher expectations and self-fulfilling prophecies: Knowns and unknowns, resolved and unresolved controversies. Personality and Social Psychology Review, 9 (2), 131-155.

Kaplan, S., & Leckie, A. (2009). The impact of English-only legislation on teacher professional development: Shifting perspectives in Arizona. Theory into Practice, 48 (4), 297-303.

Krashen, S. D. (1984). Principles and practice in second language acquisition. New York: Pergamon Press.

Krashen, S., Rolstad, K., & MacSwan, J. (2007). Review of ‘research summary and bibliography for structured English immersion programs’ of the  Arizona  English language learners task force, Arizona State University. Retrieved from http://bale.wiki.educ.msu.edu/file/view/Krashen_Rolstad_MacSwan_review.pdf/37126027/Krashen_Rolstad_MacSwan_review.pdf

Lau v. Nichols (1974). No. 72-6520, U.S. Supreme Court, 414 U.S. 563; 94 S. Ct. 786; 39 L. Ed. 2d 1; 1974 U.S. LEXIS 151

Lightbown P., & Spada, N. (2006). How languages are learned (3rd ed.). Oxford, UK: Oxford University Press.

Lillie, K. E., Markos, A., Estrella, A., Nguyen, T., Peer, K., Perez, K., Trifiro, A., Arias, M. B., & Wiley, T. G. (July, 2010). Policy in practice: The implementation of structured English immersion in Arizona. Los Angeles, CA: Civil Rights Project: University of California, Los Angeles.

Mahoney, K., MacSwan, J., Haladyna, T., & García, D. (2010). Castañeda’s third prong: Evaluating the achievement of Arizona’s English learners under restrictive language policy. In P. Gándara & M. Hopkins (Eds.), Forbidden languages: English learners and restrictive language policies (pp. 50-64). New York: Teachers College.

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis (2nd ed.). Thousand Oaks, CA: Sage Publications.

Moore, S. C. K. (2008). Language policy implementation: Arizona's structured, English, immersion (SEI) training. (Unpublished doctoral dissertation). Arizona State University, Tempe, Arizona.

Mora, J. K. (2000). Staying the course in times of change: Preparing teachers for language minority education. Journal of Teacher Education, 51 (5), 345-357. doi: 10.1177/0022487100051005003

Nieto, S. (2000). Affirming diversity: The sociopolitical context of multicultural education (3rd ed.). Boston: Pearson, Allyn & Bacon.

Office of English Language Acquisition Services (OELAS), Arizona Department of Education. (2009-2010). English language learner monitoring process for federal and state compliance. Retrieved from http://www.ade.state.az.us/oelas/downloads/2009-2010EnglishLanguageLearnerMonitoringProcessforCompliance.pdf

Pajares, M. F. (1992). Teachers' beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62 (3), 307-332.


Richardson, V. (1996). The role of attitudes and beliefs in learning to teach. In J. Sikula (Ed.) Handbook of research on teacher education, (2nd  ed.) (pp. 102-119). New York: Association of Teacher Educators.

Roberts, M., & Griffiths, C. (2008). Error correction and good language learners. In C. Griffiths (Ed.), Lessons from good language learners (pp. 282-293). Cambridge, UK: Cambridge University Press.

Weinberg, M. (1995). A chance to learn: The history of race and education in the United States (2nd ed.). California State University, Long Beach: The University Press.

Wiley, T. G. (2007). Accessing language rights in education: A brief history of the U.S. context. In O. García & C. Baker (Eds.), Bilingual education: An introductory reader ( pp. 89-107). Clevedon, UK: Multilingual Matters.

Wiley, T. G., & Wright, W. E. (2004). Against the undertow: Language-minority education policy and politics in the “age of accountability”. Educational Policy, 18 (1), 142-168. doi: 10.1177/0895904803260030

Wright, W. E. (2004). Intersection of language and assessment policies for English language learners in Arizona. (Unpublished doctoral dissertation). Arizona State University, Tempe, Arizona.

Wright, W.E. (2005). The political spectacle of Arizona’s proposition 203. Educational Policy, 19, 662-700.

Wright, W. E., & Choi, D. (2006). The impact of language and high-stakes testing policies on elementary school English language learners in Arizona. Education Policy Analysis Archives, 14 (13), 1-75. Retrieved from http://epaa.asu.edu/ojs/article/viewFile/84/210

Wright, W. E., & Pu, C. (2005). Academic achievement of English language learners in post proposition 203 Arizona. Language Policy Research Unit, Educational Policy Resource Laboratory, Arizona State University. Retrieved from http://epicpolicy.org/files/EPSL-0509-103-LPRU.pdf

Wong-Fillmore, L., & Snow, C. E. (2002). What teachers need to know about language. In C. T. Adger, C. E. Snow, & D. Christian (Eds.), What teachers need to know about language (pp. 7-53). McHenry, IL: Delta Systems.

Cite This Article as: Teachers College Record Volume 114 Number 9, 2012, p. 1-33
https://www.tcrecord.org ID Number: 16588, Date Accessed: 5/28/2022 7:00:05 PM

Purchase Reprint Rights for this article or review
Article Tools

Related Media

Related Articles

Related Discussion
Post a Comment | Read All

About the Author
  • Karen Lillie
    SUNY Fredonia
    E-mail Author
    KAREN E. LILLIE is an Assistant Professor in the Department of Language, Learning, & Leadership at SUNY Fredonia, where she works with TESOL graduate students towards their K-12 certification. Her specialization is in language policy and forensic linguistics (language & law). Dr. Lillie’s research interests include language and the justice system, language discrimination, rights for language minorities, and specifically within policy, dropout rates for ELLs, secondary level educational policy, and the effects of English-Only laws and propositions. She is on the advisory committee for the Language Policy Research Network (LPREN) and is Editorial Associate for the Journal of Language, Identity, and Education.
  • Amy Markos
    Arizona State University
    AMY MARKOS is a teacher educator, specializing in the preparation of teachers for linguistically and culturally diverse learners. Her research interests include understanding teachers’ dispositions and beliefs about ELs and the use of critical reflection in teacher learning. Dr. Markos is also interested in participatory action research that explores education policies and pedagogical practices related to language learners’ access to quality education. Her recent publications include a co-authored chapter, “Implementing Structured English Immersion in Teacher Preparation in Arizona,”, in Educational Language Policy in Arizona: An Examination of Legal, Historical and Current Practices in SEI for Multilingual Matters, Clevedon, 2011 and Policy in Practice: The Implementation of Structured English Immersion in Arizona (Lillie et al., 2010) published through The Civil Rights Project at UCLA.
  • M. Beatriz Arias
    Arizona State University
    E-mail Author
    M. BEATRIZ ARIAS is an Associate Professor of Applied Linguistics at Arizona State University. Her research interests focus on the impact of restrictive language policy and the transformation of desegregation policy as they both relate to the needs of English Learners (ELs). Professor Arias is currently conducting research on teachers’ perception of their preparation to work with ELs. She has recently co-edited: Implementing Educational Language Policy in Arizona: An Examination of Legal, Historical and Current Practices in SEI for Multilingual Matters, Clevedon 2011, and Academic Language in Second Language Learning, Research in Second Language Learning, Information Age Publishing, Charlotte, NC. 2011. She is currently the Principal Investigator for a 5-year grant to provide ESL certification for secondary content teachers.
  • Terrence Wiley
    Center for Applied Linguistics
    E-mail Author
    TERRENCE G. WILEY is President of the Center for Applied Linguistics in Washington D.C. and Professor Emeritus at Arizona State University, where he served as Executive Dean of the Mary Lou Fulton Institute and Graduate School of Education, and former Director of the Division of Educational Leadership & Policy Studies. His teaching and research have focused on educational applied linguistics, concentrating on language policy, literacy and biliteracy studies, language and immigration, bilingual education and bilingualism, heritage and community language education, English and globalization, and English as a second and international language with emphases on educational equity and access. Among his publications are: The Education of Language Minority Immigrants in the United States (co-editor, 2009, Multilingual Matters). Literacy and Language Diversity in the United States, 2nd edition (author, 2005, Center for Applied Linguistics), and Ebonics in the Urban Education Debate, 2nd edition (co-editor, 2005, Multilingual Matters). Professor Wiley's editorial service includes co-founding and co-editing the Journal of Language, Identity and Education, the International Multilingual Research Journal, and guest co-editing the International Journal of the Sociology of Language and Bilingual Research Journal.
Member Center
In Print
This Month's Issue