Abstract
This article presents a set of recommendations that promote a more nuanced, meaningful accountability policy for English learners in the next authorization of the Elementary and Secondary Education Act. The authors argue that the ESEA reauthorization must strengthen the law’s capacity-building purpose so that federal, state, and local leaders support continued attention, direction, and innovation in effectively educating ELs. The recommendations put forth in this article focus on monitoring both current and former ELs, establishing time frames for the attainment of English language proficiency, and setting expectations for academic achievement that are reflective of English language proficiency level and time in the state’s school system.
The Elementary and Secondary Education Act (ESEA) has historically played a central role in building national capacity to meet the educational needs of English learners (ELs). The No Child Left Behind (NCLB) Act of 2001 represented a step forward in federal policy for these students in two ways: (1) The law fostered greater inclusion of ELs in standards-based instruction, assessment, and accountability, and (2) it brought wider attention of policymakers and educators to ELs’ language and academic needs. However, the law’s provisions for ELs—as well as for other vulnerable populations—contained significant shortcomings. To date, NCLB’s test-based accountability and status bar, 100% proficiency targets have been blunt instruments, generating inaccurate performance results, perverse incentives, and unintended negative consequences (Ho, 2008; Linn, 2005, 2008; Koretz, 2008). Notwithstanding these flaws in the present accountability policy and structures, a strong federal framework is still needed to ensure that all educators are given clearer signals about and held accountable for their students’ academic performance and, in the case of ELs, their progress toward English language proficiency. As written, the current law ignores the connection between ELs’ expected progress in developing English language proficiency under Title III and their expected academic progress and proficiency under Title I while they learn English.
The next ESEA authorization must strengthen the law’s capacity-building purpose so that federal, state, and local leaders sustain and sharpen attention, direction, and innovation in effectively educating ELs. As representatives of a working group of researchers committed to fostering ELs’ success, we focus in this article on a subset of our recently published recommendations that promote meaningful accountability for ELs. 1 We argue that a more nuanced, meaningful accountability policy should be promoted in the next iteration of the ESEA. Such a policy will foster systems that provide meaningful information educators can use to tailor instruction to ELs’ linguistic and academic needs, thereby more effectively working toward the law’s original goal of improving outcomes and narrowing achievement gaps for these students. We acknowledge that all students, including ELs, must have access to high-quality curriculum, effective instructional practices and teachers, and supportive school environments to meet challenging academic standards; however, our focus in this article is on the specific needs of English learners and the ways in which accountability policy can be better tailored to address them, as these students take on the unique challenge of learning academic content while also acquiring the English language.
Context
The need for national leadership to effectively serve ELs has become more acute as the numbers of these students increase and as the failure of educational systems to meet their needs becomes more evident. U.S. schools serve 11 million students who speak a language other than English at home, representing approximately 20% of national school enrollment (U.S. Department of Education, 2010). Since the last ESEA reauthorization, a number of states, particularly in the Southeast and Midwest, have seen dramatic increases in their EL populations (National Clearinghouse for English Language Acquisition, 2010). Even states with traditionally high proportions of ELs have experienced such growth that these students are ubiquitous throughout the state and no longer just a challenge for some districts, some schools, or some teachers (Samson & Collins, 2012).
Unfortunately, the capacity to support ELs has not kept pace with the growing need. Thirty percent of schools held accountable for adequate yearly progress (AYP) targets for ELs under NCLB did not make AYP for that subgroup in 2005–2006; in high-poverty schools, this percentage was substantially higher (Taylor, Stecher, O’Day, Naftel, & LeFloch, 2010). Moreover, approximately one third of districts receiving Title III funding reported missing at least one Annual Measureable Achievement Objective (AMAO) for their English learners under Title III during the 2008–2009 school year (Tanenbaum et al., 2012). Additionally, a third of all schools (and half of high-poverty schools) reported that they needed technical assistance to improve services for ELs in 2005–2006 and 2006–2007, but only half of those that needed it reported receiving satisfactory assistance (Taylor et al., 2010).
Policy Considerations
Through research and practice, educators and researchers have come to a better understanding of the strengths and needs of ELs. The key facts for policy consideration are that English language acquisition is developmental in nature, occurs over time, and is influenced by students’ initial proficiency in English and primary language(s), time in the school system, and the type and quality of schooling, as well as other conditions (see Cook, Boals, Wilmes, & Santos, 2008; Genesee, Lindholm-Leary, Saunders, & Christian, 2006; Hakuta, Butler, & Witt, 2000; Linquanti & George, 2007; Thompson, 2012). 2 Given these facts, we cannot mandate that students with limited understanding of English learn subject matter taught in English at the same rate as their English speaking peers. The developmental nature of second language acquisition has implications for defining the EL subgroup for accountability purposes and for setting expectations for their linguistic and academic progress and achievement.
Numerous scholars have raised concerns about the validity and reliability of assessment results included in high-stakes accountability systems (Linn, Baker, & Betebenner, 2002), particularly for English learners (e.g., Abedi, 2002, 2004; Solano-Flores, 2008). As we discuss in more detail below, English language proficiency fundamentally influences students’ performance on content-area assessments delivered in English. For example, students at beginning levels of English proficiency may be unable to demonstrate their math knowledge on a standardized math test administered in English because of gaps in their knowledge of English (Abedi & Gándara, 2006; Martiniello, 2008). Meanwhile, analysis of current English language proficiency assessments has found substantial inconsistencies in the constructs measured by these assessments (e.g., Solórzano, 2008) and has led to concerns about the reliability of the assessments as well (e.g., Stokes-Guinan & Goldenberg, 2010). 3 Despite limitations of the assessments and how results are used, accountability remains an important leverage point for improving EL instructional programs on a large scale. Therefore, although we acknowledge the critical need to improve the validity and reliability of the inferences and uses of assessments included in accountability systems for English learners, we also argue there are other aspects of the accountability system that must be strengthened to provide more meaningful and useful information about EL students’ performance and needs. We turn now to our recommendations for improving ESEA accountability for ELs, which are organized in three sets.
Stabilizing the EL Subgroup
Our first set of accountability recommendations addresses unstable identification and reclassification procedures that produce a “revolving door” effect, as more proficient students exit and less proficient students enter the EL subgroup. Under current policy, the more successful schools are in reclassifying their ELs, the more poorly their EL subgroup performance looks (National Research Council, 2011; Wolf, Herman, Bachman, Bailey, & Griffin, 2008; Working Group on ELL Policy, 2010). This poses a problem for accountability because it provides faulty information about the performance of the EL subgroup on long-term outcomes.
Although this “revolving door” effect has been demonstrated using data from several districts and states (Linquanti, 2001; Parrish et al., 2002; Parrish, Perez, Merickel, & Linquanti, 2006; Saunders & Marcelletti, 2012), we illustrate it with data from two school districts in California. Sanger Unified School District, where two of the authors (Thompson and Hakuta) have worked over several years, is a midsize district on the fringes of a city, whereas the Los Angeles Unified School District (LAUSD) is a large urban district. Both districts have high proportions of ELs from high-poverty backgrounds (see Figure 1).
Illustration of “revolving door” effect for ELs (data from the California English Language Arts assessment, 2004–2009).
The panel on the right side of both graphs in Figure 1 shows the district’s progress for ELs (solid lines) and non-ELs (dashed lines), comparing the district (dark gray) with statewide data (light gray). These data take into account all students in the original cohort of ELs, including both current ELs and former ELs who have been reclassified as fluent in English (RFEPs). In Sanger’s case (Figure 1a), the district went from below to exceeding state performance for ELs. For LAUSD (Figure 1b), the EL population increased overall performance and more closely approximated the state average. In contrast, the panels on the left side of both Figures 1a and 1b show the same data, with former (and, by definition, more successful) ELs excluded. Here, each district’s ELs are portrayed as doing more poorly than ELs statewide because their most successful ELs were removed.
These examples make evident that, as former ELs are systematically removed from the subgroup, it becomes impossible to determine which schools and practices are successful for these students. It is also impossible to track EL performance over time when ELs are redesignated several years prior to high school graduation and are thus not included in subgroup statistics, such as graduation rates and college readiness.
Monitor Current and Former ELs
To increase the fairness and accuracy of the accountability system, we recommend establishing a Total English Learner (TEL) subgroup for Title I accountability purposes. The TEL subgroup would include both currently identified ELs and former ELs. As illustrated in the district cases, establishing this aggregate group and maintaining consistent subgroup designation would yield more accurate information about ELs’ cumulative performance over time.
Continuing to monitor ELs’ progress throughout their schooling recognizes the developmental nature of second language acquisition and allows better service delivery to students at all levels of English proficiency. 4 There is no question that English language proficiency (ELP) influences students’ ability to learn content presented in English and to demonstrate what they have learned through participating in assessments given in English (Abedi & Gándara, 2006; Martiniello, 2008). Even after reaching proficiency on a state’s ELP test, former ELs may continue to gain deeper competencies in English needed to meet grade-level standards set for native English speakers. Moreover, in states that allow school districts to define their own exit criteria, a student considered a former EL in one district may still be considered a current EL in a neighboring district (see Linquanti, 2001; Parrish et al., 2006; Tanenbaum et al., 2012). It is therefore critical that states and districts continue to support and monitor former ELs beyond the current federally mandated 2-year monitoring time frame to ensure that they continue to advance academically and that educators are held accountable for the subgroup’s long-term results, as with all other subgroups.
Monitor and Address Needs of Long-Term ELs
In addition, accountability policy should attend to the needs of long-term ELs, or students who are unable to meet specified exit criteria and remain designated EL after 5 or more full years in a state’s public schools. Middle and high schools across the country have increasingly greater numbers of these students; for example, long-term ELs comprise nearly one-third of the EL population at the secondary level in New York City (New York City Department of Education, 2008) and one-half of the EL population at the secondary level in California (Olsen, 2010). To ensure that these struggling students’ performance is not masked in the TEL subgroup and that they receive additional instructional support, states and districts should be required to monitor and report their percentage of long-term ELs; set annual targets for districts with disproportionately high percentages of long-term ELs to annually reduce this percentage; and delineate what consequences and responses follow if these targets are not met.
Relating English Language Development, Time in System, and Academic Progress Expectations
Our second set of recommendations addresses current accountability provisions for attaining English language proficiency. Analytic methods have recently become available that help states use their empirical data to identify an optimal range of ELP on their ELP assessment and to establish an ambitious yet reasonable time frame for ELs to attain this ELP performance level (Cook, Linquanti, Chinen, & Jung, 2012). However, the current ESEA does not require states to incorporate challenging yet realistic developmental trajectories based on available data.
Set Time Frames for Attaining English Language Proficiency
We recommend that time be incorporated explicitly into ESEA accountability provisions for the acquisition of English language proficiency. We suggest considering a target time frame of 5 years, adjusted by students’ initial ELP on entry to the state school system. Empirical research suggests 4–7 years for academic English language proficiency as a challenging but achievable goal (e.g., Cook et al., 2008; Hakuta et al., 2000; Thompson, 2012). To exemplify this, Figure 2 illustrates the rate of attainment of English language proficiency for a cohort of ELs in the Sanger Unified School District who started kindergarten in 2003 and were observed for 7 years. These results have been replicated in other states and districts, including in LAUSD (see Cook et al., 2012; Thompson, 2012).
Illustration of English language proficiency development (data from the California English Language Development Test from Sanger Unified School District, 2003–2010)
The graph plots the probability with which these students attained different levels of ELP. Most students attained a proficiency level of 3 (intermediate) on the state test after 2 years, but it took up to 7 years for most to attain a proficiency level of 4 or 5 (early advanced or advanced). It takes even longer for substantial percentages of ELs to attain basic academic proficiency on state and local measures required for reclassification and exit from EL status (“Being Reclassified” line).
The time required to acquire English language proficiency is affected by many factors, including how language proficiency is defined and measured, as well as a variety of individual student characteristics and the quality of services provided (Cook et al., 2008; Genesee et al., 2006; Hakuta et al., 2000; Linquanti & George, 2007; Thompson, 2012). Because of these factors, ELs progress in language proficiency at different rates (e.g., ELs entering with lower ELP at higher grades may have more difficulty in meeting this timeline compared to ELs entering with higher ELP at lower grades), and the magnitude of these differences may depend on the type and quality of instruction students receive. For example, students who entered kindergarten in LAUSD with beginning levels of English proficiency had a 50% probability of reclassification after 9 years in the district, compared to an 80% probability of reclassification for students who entered with intermediate levels of English proficiency (Thompson, 2012).
For these reasons, we do not recommend setting a high stakes accountability target that 100% of the EL subgroup meet this 5-year goal for English language proficiency. Rather, we recommend that states establish expectations based on their state data, examine current local education agency performance, and set challenging but achievable targets that will lead to progressively higher percentages of ELs achieving proficiency in English within the specified time frame (see Cook et al., 2012; Linquanti & George, 2007). It is beyond the scope of this article to outline the analytical methods that states could use to determine acceptable ELP performance standards using available data; however, potential methods have recently been described in detail (Cook et al., 2012).
Setting Academic Achievement Expectations
Our third set of recommendations acknowledges that the present accountability provisions of Title I set unattainable achievement expectations that are divorced from ELs’ English language proficiency and time in the state school system. It is clear from existing longitudinal studies of state data that ELs’ likelihood of meeting state academic performance standards improve significantly as they become proficient in English (e.g., Francis & Rivera, 2007). Figure 3 illustrates the academic performance of Sanger’s fourth-grade ELs (Figure 3a) and LAUSD’s third-grade ELs (Figure 3b) in mathematics by ELP level, with clear positive associations between English language proficiency level and mathematics performance. As the figures show, this association is consistent across the two districts and at different grade levels. These patterns are also seen across several other districts and states (Cook et al., 2012).
Illustration of academic achievement by English language proficiency level (data from the California English Language Development Test and the state math assessment, 2003–2010). (a) Sanger Unified School District, fourth grade.
Several factors contribute to this relationship, the most important being the fundamental role that language plays in knowledge acquisition and content mastery in all academic domains. Of course, the type and quality of instruction ELs receive also plays an important role in ELs’ language development and academic performance. A key issue in accountability for ELs is ensuring that districts and schools provide substantive and responsive instruction for these students. Although instructional capacity will vary depending on local resources, the size of a district’s EL population, and distribution across schools, etc., states should offer guidance related to the types of programs and services that districts and schools should offer, as well as foster leadership and support for ensuring that teachers and administrators are adequately prepared to provide these services. Such guidance is essential given that many districts have reported a lack of information on proven programs and curricula for ELs (Tanenbaum et al., 2012). However, research related to effective programs for English learners (Parrish et al., 2006; Williams et al., 2007) and how best to prepare educators to work with these students (e.g., Samson & Collins, 2012; Téllez & Waxman, 2006) is limited, and more exploration in these areas is needed (Goldenberg, 2008).
Set Achievement Expectations Reflective of English Language Proficiency Level and Time
Educators are responsible for ensuring that all ELs become proficient in English, but it takes time for children to acquire English language proficiency even in the best instructional settings. Current federal accountability provisions ignore these developmental factors, and as a result, school systems are required to set unrealistic academic performance expectations for ELs who are not yet proficient in English. This undermines both the meaningfulness and the credibility of the accountability system and acts to demoralize teachers and students. We recommend that ELP progress expectations and percentage targets be incorporated into Title I accountability provisions. States would also be required to hold schools and districts accountable for ELs’ progress toward and attainment of content area proficiency standards that take into account time in the state system and the corresponding expected ELP level (or current level if higher than expected). If schools and districts fail to meet these growth targets and expectations, sanctions consistent with current federal policy should be followed, where additional professional development and any further program improvement efforts must be tailored specifically to meet the needs of ELs.
Several analytic methods are now becoming available that allow states to use their existing data on the relationship between English language development, time in system, and academic performance to set ambitious and reasonable progress expectations and accountability targets. These longitudinal approaches to analyzing existing state data, some of which have been presented in this article, would facilitate the development of state accountability models that are predicated on student growth. State data systems have the necessary information, but have not been used well for these purposes. For example, states might define graduated expectations on the achievement test for ELs at different levels of language proficiency in each grade. Alternatively, states might develop a system for weighting performance on the language proficiency test and performance on the achievement test, with increasing weight assigned to the achievement test as students’ ELP level and/or time in the school system increases. Still other approaches—for example, defining expectations that combine growth and status—are possible (see Cook et al., 2012).
Final Thoughts and Policy Guidance
Our recommendations attempt to correct for two extremes in accountability policy. The current “one standard for all students at all times” accountability provisions define one end of the continuum for ELs. At this current extreme, the system focuses solely on reaching academic proficiency and ignores the growth students make in content area achievement toward that performance standard. Such systems underreport progress and discourage students and educators. (In fact, growth models were introduced during the current authorization of ESEA to address these problems; our recommendations attempt to resolve these problems as they relate to ELs.) At the other extreme, accountability systems could be designed to ignore academic achievement expectations for ELs until they become proficient in English, thereby separating second language development from expected progress in content area achievement. Such a separation may yield negative consequences by creating incentives for educators to focus on English language proficiency development apart from or instead of content area knowledge.
In summary, our recommendations aim to make accountability policies more meaningful for ELs and to foster systems that avoid these extremes through stabilizing the subgroup, and through developing expectations, targets, and a reasonable time frame for moving increasing percentages of ELs to English language proficiency and grade-level academic performance. Critics may suggest that such a system would place an undue reporting burden on districts and schools, yet states are already being asked to break down achievement by multiple variables for many new federal and state programs. Once the requisite data systems and analytical methods are in place—for which guidance is being distilled (see Cook et al., 2012)—the reporting itself will become routine. Yet, this routine reporting and monitoring will provide key information needed for educators to better differentiate instruction and foster instructional settings that better meet ELs’ linguistic and academic needs.
Our recommendations could be incorporated into ESEA accountability regulations in at least two ways (Working Group on ELL Policy, 2011). First, the Title III requirements for progress in learning English and attaining the English proficiency level (currently AMAOs 1 and 2, respectively) could be moved into Title I and related to Title I academic progress and achievement criteria. Alternatively, Title I accountability provisions could require states to utilize ELP progress and attainment data derived from the English language proficiency assessment currently mandated under Title I (Section 1111 (b)(7)) in establishing academic progress and achievement expectations for ELs. Because these ELP progress and attainment results should come directly from Title III accountability provisions, these provisions could remain within Title III. Either approach would notably improve Title I AYP’s current status bar model, which ignores meaningful academic progress below the proficient level, omits defining time frames for attaining English proficiency, and devalues setting reasonable achievement expectations for ELs as they work to acquire proficiency in English. Also, integrating or aligning Title III accountability provisions with those of Title I would enhance the meaning and utility of Title I AYP results for ELs; end the isolation of Title III accountability; and more precisely and quickly target Title I’s substantial resources on the academic and linguistic learning needed for ELs to make steady progress toward proficiency on a prespecified time frame.
We are well aware that accountability provisions represent only a small part of the total capacity-building effort needed to ensure that ELs stand an equal chance of success as the Common Core State Standards are implemented. Clearly, the sophisticated language uses called forth in all content areas of these next-generation standards have enormous capacity implications for educational leaders, content area and ESL teachers, and curriculum and assessment system developers. It is essential that strategic investments be made across the educational system to develop the capacity of school and district personnel to better serve ELs and to help these students develop both their English language proficiency and content area knowledge. We have focused here on accountability provisions for ELs because we believe that federal and state accountability policy can and should foster systems that send clearer, fairer, and more useful signals to educators and the public about English learners’ many strengths and needs.