---

As the population of students learning English in the United States grows, education policies at the federal, state, and local levels must address the design and role of standardized assessments used to measure the growth and learning of those students. Yet assessment policies for emergent bilinguals currently utilize inadequate tools, and rather than promote equity for emergent bilingual learners, these policies produce negative consequences. Federal policy regulations give emergent bilinguals limited time to learn English before participating in high-stakes English language tests. The recently implemented Common Core State Standards do not yet include language development standards for emergent bilingual students. State assessment policies may use inadequately designed standardized assessments to assess English learning, leading to potential mismatch between students’ needs and the language services they subsequently obtain. As a result, time and resources are ill-used, even as pressure mounts for emergent bilingual students to perform well on standardized English assessments.

In this article, I will outline the flaws in the assessment tools used to assess the English language development of emergent bilinguals in New York State, and demonstrate how New York State’s assessment policies for emergent bilinguals fail to account for the knowledge and identities of the students, encourage test-centered curricula, and thwart student achievement. I will consider the impact of assessment policies on emergent bilinguals in general, as well as on long-term English Language Learners (LTELLs) specifically. Finally, I will make recommendations for educators, administrators, and policymakers to improve emergent bilinguals’ academic experiences and success.

Overview of standardized assessments for emergent bilinguals In New York State

In New York State, upon a parent’s registration of her child to a neighborhood public school, the Home Language Identification Survey is administered to determine if the child speaks English, and/or languages other than English. The student’s abilities in English are then initially assessed through the Language Arts Battery – Revised (LAB-R), which is used “…for initial identification and determination of eligibility for English Language development support services. The assessment is given only one time, upon entry into the New York State Public School System, and within 10 days of admission into the System” (“Bilingual Ed: Facts,” 2011).

The process of identifying emergent bilingual learners (EBLs) is the responsibility of individual schools. Unlike identifying students for special education support services, or recommending students for promotion to the next grade, both of which are multi-faceted processes involving portfolios, standardized test scores, and teacher evaluation, the process of identifying an Emergent Bilingual Learner is determined by a home language survey followed by one test, the LAB-R. Thus, depending on how the school organizes the process, a student’s evaluation may be determined by only one person, and after merely one day of testing.

The New York State English as a Second Language Achievement Test (NYSESLAT)

If students are initially designated through the LAB-R to be Limited English Proficient, they are given language support services in the form of bilingual or English as a Second Language (ESL) classes. Then, in the following spring, the students take the New York State English as a Second Language Achievement Test (NYSESLAT) annually until they pass at the “proficient” level.  Once proficiency is obtained, students are then labeled “former English Language Learners”.  The test is divided into four sections to test the student’s English ability in each language modality: speaking, listening, reading, and writing. Although students are assessed on each section and deemed beginner, intermediate, advanced, or proficient, the lowest score the student receives determines that student’s overall score. Thus, if students are at the intermediate level in speaking and listening, but at the beginning level in reading and writing, they are labeled as “beginners”.

The NYSESLAT was first administered in May 2003, in response to No Child Left Behind (NCLB) requirements focused on standardized assessment for all students (Perez-Hogan, 2003). Gerald DeMauro of the New York State Education Department (NYSED) explained that the NYSESLAT “… includes the vital linking to the English Language Arts examinations in grades 4 and 8, to eliminate the need for LEP students to take these tests” (2003). Thus, according to the state department, the NYSESLAT was originally created as a more equitable alternative to taking the ELA Exam to meet new NCLB requirements. Before NCLB, emergent bilingual students in New York State were exempt from the ELA Exam for three years; by contrast, NCLB requirements reduced that time frame to one year and one day. As a result, while the NYSESLAT initially eliminated the need for emergent bilinguals to take the ELA Exam, currently, students must take both.

Significantly, after students are initially assessed in English language proficiency, a student’s ELL status in New York State is determined exclusively by their performance on the NYSESLAT. As with their initial language assessment, class work, teacher evaluations, and other standardized test scores do not factor into their state-designated English proficiency level. If students perform at the beginner, intermediate, or advanced levels, they continue as “ELLs” and take the NYSESLAT the following spring. When students perform at the proficient level, they are considered to have “passed” the NYSESLAT. They stop taking the test and no longer receive state funding for services, though parents may opt to keep them in ESL or bilingual classes, and they are labeled “former ELLs”. Students who do not test at the “proficient” level on the NYSESLAT after six years of ESL services are labeled long-term ELLs (LTELLs). Districts are required to continue ESL services for these students, but they no longer receive state funding (“Bilingual Ed: Facts,” 2011).

Federal initiatives affecting the assessment of emergent bilinguals

No Child Left Behind, passed in 2002, is the most recent reauthorization of the Elementary and Secondary Education Act of 1965. Several aspects of this legislation led to significant changes in the assessment of emergent bilinguals. First, emergent bilinguals living in the United States for more than one year and one day must take the English Language Arts (ELA) exam administered by New York State for grades 3-8 in addition to the NYSESLAT. This was a drastic change from New York State’s policy pre-NCLB, which previously allowed emergent bilinguals three years to develop English language proficiency before being required to take standardized tests designed for native English speakers.

At the high school level, all emergent bilinguals must pass the ELA Regents exam to earn a high school diploma in New York State, regardless of their English language proficiency or length of time in the country, in addition to taking the NYSESLAT annually and passing 4 other Regents exams. While the state previously offered a local diploma that required students to pass fewer Regents exams, under NCLB, this option is not available to any student entering 9th grade between 2008 and the present, with the exception of students with Individualized Education Plans (IEPs).

The Adequate Yearly Progress (AYP) measure accompanied NCLB legislation. Schools must show progress through standardized test scores not only for their entire student population, but for four subgroups as well, one of which is English Language Learners. Rather than use the NYSESLAT, a test designed for students learning English, emergent bilinguals are assessed according to their performance on the ELA Exam, a test designed for native speakers. Schools that fail to meet the goals set forth by the federal government are placed on the Schools In Need of Improvement (SINI) list. The advent of the AYP measure, which disproportionately affects schools with large emergent bilingual populations, has brought about significant changes to programming and curricular decisions made by schools for emergent bilinguals.

The more recent federal initiative, Race to the Top (RTTT), has also had significant implications for the assessment of emergent bilinguals. In exchange for federal funding, states had the option of adopting Common Core State Standards (CCSS) developed at the national level. Most states have adopted these standards despite their failure to include standards for emergent bilinguals. Based on the organizations involved in their creation, the CCSS are necessarily tied to standardized assessments. Because the standards are distributed nationally, they are necessarily not tailored to meet the needs of individual communities. The language of the standards cannot be changed by individual states, though states may add up to 15% more standards for areas they find lacking. However, this accommodation positions the learning of emergent bilinguals as peripheral to the core learning standards for US schools, and makes it much more likely that assessments emerging from the standards will not align with their needs.

Critique of state policy: The use of single measures to determine language proficiency

Critique of the LAB-R as a single measure of initial language assessment

The first state-level policy that affects emergent bilinguals in New York State is implemented at the local level, that of administering the LAB-R as the single determiner of a student’s language abilities, without state or district overview of any kind. That only one test determines whether a student receives language services for the remainder of their time in New York State public schools means that one simple error on the administrator’s part, or a nervous day on the student’s part, could preclude that student from ever receiving language services, or result in an erroneous placement, an error that could cost that child significant learning opportunities. The LAB-R is required to be administered once and only once, upon entry to the school system. Thus, recent arrivals to the country take a high-stakes test within ten days of their enrollment in a new school, as well as upon arrival to a new country. The LAB-R, a test administered and graded by individual schools, has the potential to mislabel students’ English proficiency, thus denying these students necessary language support services.

Studies have shown that emergent bilinguals who do not receive services fare much worse than their peers in both ESL and bilingual programs. In a national study by Thomas and Collier (2002), researchers found that students in grades K-1 whose parents had refused language services and were therefore placed in mainstream classes lagged in reading and math achievement by grade 5 as compared with their peers in ESL and bilingual programs. This group had the largest number of high school dropouts and the lowest reading achievement as compared with peers in other programs (Thomas & Collier 2002). This study offers a glimpse at the potential achievement of students who are not properly identified as emergent bilinguals by the LAB-R, and the possible result of losing invaluable language support. Schools that use the NYSESLAT for determining student placement risk misplacing students in programs that do not fit their needs, a mistake that could cost students years of learning. In the next section, I will discuss how the NYSESLAT may potentially influence the education of emergent bilinguals, including programming, curriculum, and language support.

Critique of the NYSESLAT as a single measure of ongoing language assessment

While the NYSESLAT provides emergent bilingual students an entry point into the important conversation of student achievement, the test is an inadequate tool for ongoing language assessment in a number of ways, including: the original design of the test itself, the deficit view of the education of emergent bilinguals implied by the test design, and the manner in which the test measures English learning.

The creation of the test

The NYSESLAT was created by testing corporations without the input or the relevant knowledge of the needs of emergent bilinguals and their communities. Proponents of the NYSESLAT may point to the fact that educators were involved in the creation of the test to make the argument that the assessment is connected to classroom learning and serves the needs of the students. However, according to a job posting on the NYS DOE (2012a) website, teachers’ input is limited to the following three tasks:

1. Range Finding – reviewing and scoring student’s field-tested constructed responses to select anchor papers for the test scoring guides.
2. Item Review – examine, discuss and select the test questions that are most appropriate for each modality.
3. Final Eyes Review – take a last look at the final stages of the current year operational test to determine if any minor revisions are needed prior to printing.

Thus, while educators are involved in the process of test-making and the process of grading, they lack the agency to create test questions or shape the format of the test. Rather, the most educators can do is identify an anchor paper for grading, suggest that a question is biased or unbiased, and determine minor revisions. Even within this limited role, one NYC ESL teacher who participated in the process of evaluating questions was surprised to see one question that her group recommended not be used actually appear on the 2011 NYSESLAT, while another seemingly biased question that had never been presented to the group for review, nevertheless was on the exam². The limited role that teachers play in the creation of the NYSESLAT limits the test’s relevance to classrooms, teachers, students, and their communities.

Deciding what counts as English learning

Though an exclusive focus on English is in itself problematic, deciding what counts as English learning complicates the issue considerably. The NYSESLAT test, being the only accepted assessment used to measure English learning in the state, essentially defines what an English speaker should know, and what it means to speak English in New York State. However, some areas of the test require considerable content knowledge and academic skill, while other aspects of the test rely on cultural understandings. For example, the directions for the essay question on the 2007 sampler NYSESLAT for grades 9-12 read, “Write an essay about how people’s personalities are formed. Discuss both the influence of the characteristics people are born with (nature) and the way people are raised (nurture).” Before writing the essay, students are asked to look at two pictures –one of a happy family in a kitchen, and one of identical twins. Using the pictures, they are asked to reflect on the following three questions: “What personality traits do you think people have from birth? What personality traits do you think result from the way people are raised? Do you think nature or nurture has a greater influence on the development of personality? Why?” (NYSED 2011b).

Such a question requires scientific knowledge of genetics and psychology, and may intimidate a student who has little experience with the topic. In addition, the binary view of “nature versus nurture” fails to account for financial, political, and social factors that inevitably shape one’s identity, especially given that the picture representing “nurture” is a family unit, not a society. Further, the essay prompt fails to consider the possibility that some students’ religious beliefs may preclude either nature or nurture from influencing one’s personality. Familiarity with the content knowledge required to answer this question will surely influence a student’s ability to respond, thereby obscuring the student’s actual writing ability in English.

The essay question explored above is representative of many questions asked on the NYSESLAT whose form is incongruent with other tests the students take. Using concrete pictures to reflect on abstract concepts is a unique academic skill that must be specifically taught in preparation for the NYSESLAT. On the ELA Exam, by contrast, rather than writing in response to open-ended questions based on pictures, students read and respond to fiction and non-fiction passages. The unique format of the test therefore requires specific training and academic skills, again calling into question what counts as English learning.

In one NYC school participating in the CUNY-NYSIEB project, the assistant principal explained that while the pull-out ESL teachers usually follow the general education teachers’ lessons, during NYSESLAT testing, the ESL teachers focused on test prep. In another school in the same study, an after school enrichment program offered activities such as cooking and art, as well as test prep; students that received test prep were unable to participate in the other activities offered³. Thus, not only does the administration of the test itself take away from learning time in the classroom, but students’ experiences and opportunities in school and after school have become saturated with test preparation.

The reading and writing sections of the NYSESLAT also focus progressively more on academic skills as students move through grades. Many questions on the test, starting as early as fifth grade, require that students possess academic abilities such as making inferences, finding the author’s purpose, and identifying literary devices in English, much like the ELA Exam requires of fluent English speakers. Since the NYSESLAT tests both English fluency and academic skills, students with less schooling in their home language are likely to remain “English language learners” long after they become socially fluent in English. This raises the question of how fluency in defined, at what point it is “achieved,” and how it interacts with academic achievement. The present conflation of language fluency, content and cultural knowledge, and academic skill on the NYSESLAT point to the need for a more precise assessment that could address students’ linguistic strengths and needs accordingly.

The propagation of a deficit view

The language surrounding students’ achievement on the NYSESLAT further reinforces the importance of English, and reveals a deficit view of students’ diverse linguistic backgrounds. Rather than viewing bilingualism as a positive emerging asset, the state’s exclusive and immediate focus (as early as kindergarten) on emerging English proficiency ignores the assets students bring to their schooling, such as home languages and practices, emphasizing instead the students’ perceived deficits. In doing so, the state not only cultivates a view of these students as lacking; it also deprives students of opportunities to show what they know, and to feel successful and validated at school.  Even when students reach a satisfactory level of English according to the state, they are labeled “Former ELLs”; this label could be seen as potentially informative to educators, but it also stands as a permanent reminder that the students’ English is perhaps different or “less than” their non-labeled peers.

In the same way that the labeling of “former ELLs” implies a difference between the proficiency attained by those students and the proficiency of “native” speakers, the very idea of a test such as the NYSESLAT determining how much language learning is “proficient” supports the myth of native proficiency, or the idea that there is a fixed and finite point at which a language is fully acquired. The NYSED website, in the section “Bilingual Ed: Facts” (2011) states, “Every LEP/ELL student must take the NYSESLAT until she/he has reached the level of proficient in English on the NYSESLAT, which will allow her/him to exit a bilingual education or ESL program.”

However, language is not a static possession, but rather linguistic resources are used in the process of meaning-making (Larsen-Freeman & Cameron 2008). Thus, a student’s language proficiency is difficult to isolate in a decontextualized standardized test given that authentic language use is embedded in meaningful interaction. Larsen-Freeman and Cameron’s view of linguistic resources also highlights that a student’s education must begin with the student, and the context in which that student is learning. Thus, imposing a language assessment such as the NYSESLAT fails to account for the highly contextual nature of language, and violates a student-centered approach to learning and assessment.

NYSESLAT as Gatekeeper

Finally, in referring to the academic trajectory of emergent bilinguals, the NYSED positions the NYSESLAT as a gatekeeper, as “LEP/ELL” students are not “allowed” to join students in general education until a “proficient” level of English has been attained (“Bilingual Ed: Facts,” 2011). While high quality bilingual programs and high quality language support services are shown to contribute to significant academic gains for emergent bilinguals (García, 2009; Menken, 2010; Slavin, et al., 2011), NY State policy assumes that exiting such programs would be a logical, positive step for all students.

However, while English-only or transitional bilingual programs are more limited in scope, additive and dynamic bilingual education programs seek to educate bilingually even as students are proficient in English. In longitudinal studies and meta-analyses of research on ESL and bilingual programs, developmental bilingual programs that continue to educate bilingually even after students have attained English proficiency have shown the greatest success in educating emergent bilingual students (Rolstad, Mahoney, & Glass 2005; Thomas & Collier 2002). Thus, in its function as gatekeeper, the NYSESLAT not only serves to exclude and segregate students viewed as “Limited English Proficient”, rather it also potentially limits students’ academic and social growth and devalues their experiences in bilingual and ESL classrooms by assuming that their only benefit is English learning, when in fact, the home language is an invaluable resource that has the potential to develop in such programs.

Misplacement of students using the NYSESLAT

The scoring of the NYSESLAT and the dissemination of those scores could in theory hinder the most central goal of the test: learning English. After students take the NYSESLAT, teachers receive a one-word assessment meant to categorize students’ abilities -beginner, intermediate, advanced, or proficient. A breakdown of the four modalities (speaking, listening, reading, and writing) may be made available by the administration, but this is not a guarantee. Even with this breakdown, ascertaining which reading skills a student possesses, or where the student needs more support in listening, is impossible to determine from the results given.

As a result, the best-case scenario is that the school must conduct their own assessment to determine placement in language support services, since the breakdown of the NYSESLAT does not provide the necessary information. At worst, the NYSESLAT has the potential to be used to make important programming and curricular decisions for students that may result in students being placed in settings that are not aligned to their strengths and needs. Thus, while the NYSESLAT provides proof to the state that emergent bilinguals are acquiring English (as defined by the NYSESLAT), it also has the potential to negatively impact the quality and nature of education that emergent bilinguals receive. The NYSESLAT has particular implications for long-term ELLs, who are especially vulnerable to being placed in inappropriate classroom settings due to their scores.

Long-term ELLs are a group of students essentially created by the NYSESLAT assessment by virtue of their not attaining a “proficient” score within 6 years of entering the country. In NYC high schools, one-third of emergent bilinguals are long-terms ELLs. Overwhelmingly, they are students who have consistently received high scores in the Speaking and Listening sections of the test, indicating that they are socially fluent speakers of English, but who tend to struggle in the Reading and Writing sections, suggesting that their academic literacy is below grade level.

LTELLs are especially likely to be negatively affected by the NYSESLAT and its function in New York schools. Menken and Kleyn (2010) show that by high school, many LTELL students are apathetic towards their classes. One reason for the students’ disinterest could be that the classes in which they are enrolled are mismatched to their needs, a mistake that could be easily made if schools use the NYSESLAT to make programming decisions (Menken & Kleyn, 2010). Menken and Kleyn further report that, “At the high schools we studied, LTELLs take the same classes as all other ELLs and no services are specifically targeted to their needs” (2010, p. 407).

Menken and Kleyn also reveal that the English services students received were often too basic, while in the few cases where Spanish enrichment was available, students were often placed in classes with recent arrivals, whose Spanish was much further developed (2010). Being placed in the wrong setting demotivated students, made them feel disconnected from learning and school, and was one reason some students skipped classes (Menken & Kleyn, 2010).

LTELLs felt they had nothing to learn as a result of being programmed for mismatched ESL classes and that the classes were also at odds with their identities as English speakers. As one student, Mariluz, noted of ESL classes, “Like that’s for people that don’t know English” (Menken & Kleyn, 2010, p. 407). These ESL classes not only failed to engage students’ identities, but did not recognize the students’ emergent language assets, revealing the failure of the school to provide challenging and appropriate academic material and to provide a school environment in which students’ diverse backgrounds are validated, respected, and integrated into school learning. The use of the NYSESLAT with LTELLs, many of whom identify as English speakers, underscores the disconnect between schools and students’ identities, a mismatch that contributes to students’ academic disengagement.

In reducing a student’s language abilities to one word (beginner, intermediate, advanced, or proficient), based on one assessment (the NYSESLAT), students with varied strengths and needs may appear to be a more homogeneous group than they are. As a result, decisions in programming that are based on the NYSESLAT scores have the potential to mislabel students, thus alienating them from their school experience and wasting valuable class time with inappropriate programming choices.

Going beyond English assessment to create a rich educational experience for emergent bilinguals: Recommendations for schools and teachers

In the following section, I will suggest ways that individual schools and teachers can shield their students and classrooms from the detrimental effects of high-stakes testing, as well as ways to transform current policy at the state and federal level to shift the focus away from external standards and towards a student-centered approach to learning.

The first recommendation is to build classrooms and programs that welcome and develop the students’ home languages, even outside of designated bilingual programs. The significant work of researchers such as Baker (2006), Krashen and McField (2005), and Thomas and Collier (1997, 2002) confirms that developing and maintaining emergent bilinguals’ home language while they learn English leads to greater academic success. Cummins (2000) also corroborates the idea that skills developed in the home language transfer to English; in this way, home language development and maintenance is an end unto itself, as well as a means to acquiring English. While English testing requirements lead some schools to believe that learning exclusively in English will yield the best test results, learning bilingually not only values students’ identities, it helps students progress in English as well. Assessment policies and data collected from standardized assessments should be used to design curriculum and instructional programs that support the students’ emergent bilingualism.

Second, assessing students’ work through an annual portfolio presentation would give emergent bilingual students the opportunity to showcase their learning and expertise in a formalized way. A portfolio assessment would provide students with the opportunity to experience academic progress, regardless of their generic score on the NYSESLAT (which, due to its broad categories, may remain static from one year to the next). While the NYSESLAT is one tool to track students’ language and academic growth, it does not reflect the many dimensions of student learning, nor does it explicitly identify students’ academic needs. Supplementing the results of this test with portfolio of students’ work would provide schools the necessary information to accurately assess students’ needs.

New York State could also track students’ progress by creating a system that used students’ portfolios alongside NYSESLAT scores to determine funding and program placement. The NYSESLAT is currently a single measure for the continuing assessment of emergent bilinguals. Even if the format of the test were improved, one test will not measure every student’s unique linguistic abilities. Including contextualized assignments in the form of a portfolio, and assessing students over time, rather than on one or two days, is essential to truly assessing students’ language abilities. The portfolio tasks would emerge from classroom learning, but could embed state or national standards, and be considered alongside their standardized test scores. Since the NYSESLAT tests academic skills as well as English learning, students who have recently arrived and are still developing their English proficiency could include evidence of these skills using their home languages.

Creating an assessment process rooted not only in state standards, but also in classroom practices, would give students the best opportunity to show how they use language over the course of the school year to make sense of their learning. Such a process would offer increased engagement and motivation for students, a true valuing of bilingualism and cross-cultural understanding, and increased opportunities for student learning and progress.

Demonstrating academic proficiency through portfolios, rather than standardized assessments, also provides an opportunity to shift away from a test-centered model of education, in which teachers must prepare students for the specific concepts and skills on the ELA and NYSESLAT exams, to a student-centered method of education, in which students might demonstrate their learning through meaningful assignments. Teachers are already capable of providing students with this type of educational experience, but detracting from this important work in the classroom is the need to prepare students for externally developed and imposed measures. Expanding the assessment process for emergent bilinguals could relieve teachers of this pressure, and allow them to fully dedicate their curriculum to not only supporting students academically, but to also addressing students’ diverse home languages and practices -a formidable task when teaching from “culturally neutral” test materials and state-developed units of study.

Third, establishing a reasonable time frame for the academic trajectories of emergent bilingual learners is necessary. Despite ample research citing a minimum of five years for the acquisition of a second language, and as long as ten years, state funding is taken from students who need more than five years of language support services to learn English. The only measure used to determine this funding is the NYSESLAT score. To this end, not only should LTELLs, and all emergent bilinguals, have the opportunity to show their learning through a more comprehensive portfolio, they should also continue to receive state funding for as long as there is a demonstrated need.

At the federal level, NCLB policy requires that emergent bilinguals participate in high-stakes English language testing after one year of language learning.  This is also the case for high school students, regardless of the amount of time spent in the country. These testing requirements set unrealistic expectations for students and create a climate of failure as emergent bilingual students are often unable to meet testing requirements designed for native speakers within the given timeframe. Alternatively, allowing students more time in the school system before requiring them to participate in high-stakes English assessments, and instead using a comprehensive portfolio review process to track their language learning, relieves students of unnecessary pressure and failure, while simultaneously holding schools accountable for the learning of emergent bilinguals.

Conclusion

Assessment measures are accountability systems used by the government to efficiently determine a school’s success. I have demonstrated how the current assessment policies that establish such accountability systems are deeply flawed; as such, they should not be used to design a student’s school experience until the measures are proven to be effective and reliable. Assessment policies need to utilize the best practices that have been proven over time to produce positive educational outcomes for emergent bilinguals. Two consistent findings in the research on emergent bilinguals include the importance of home language use in English development, and a period of five to seven years for language acquisition. As such, emergent bilinguals should not be judged by the same assessment as fluent English speakers until they have been given enough time to acquire the language. Schools must provide students linguistically rich curricula and programming regardless of testing policies and their accompanying pressures. Focusing on a student-centered approach to learning that prioritizes rich curricula, authentic assessment, and linguistic diversity, we might create the classroom conditions that are essential to social change.

References