nclr - stanford universityhakuta/courses/ed205x website... · 2009. 3. 26. · nclb requires that...

2006 No. 16

Improving Assessment andAccountability for English LanguageLearners in the No Child Left Behind ActBy Melissa Lazarín*

* Melissa Lazarín is Senior Policy Analyst for Education Reform in the Office ofResearch, Advocacy, and Legislation at the National Council of La Raza (NCLR). Theauthor thanks other NCLR staff who contributed to the completion of this document,including Charles Kamasaki, Senior Vice President, and Raul Gonzalez, LegislativeDirector, who provided substantive input and edits; Jennifer Kadis, Director of QualityControl, who provided expert editorial guidance; and Rosemary Aguilar, Director ofthe Graphics and Design Unit, and Ofelia Ardón-Jones, Assistant Director of theGraphics and Design Unit, who prepared the document for publication anddistribution.

** The terms “Hispanic” and “Latino” are used interchangeably to identify persons ofMexican, Puerto Rican, Cuban, Central and South American, Dominican, Spanish, andother Hispanic descent; they may be of any race.

*** The terms “English language learners” and “limited-English-proficient” are usedinterchangeably to identify persons whose native language is one other than Englishand whose difficulties in speaking, reading, writing, or understanding the Englishlanguage may hinder the individual from the ability to achieve academically inclassrooms where the language of instruction is English.

Overview . . . . . . . . . . . . . . . .1

The Promise of NCLBfor English LanguageLearners . . . . . . . . . . . . . . . . .4

Challenges to EffectiveImplementation ofNCLB for English LanguageLearners . . . . . . . . . . . . . . . . .9

Recommendations . . . . . . .19

Conclusion . . . . . . . . . . . . . .24

INSIDEINSIDE

NCLRNATIONAL COUNCIL OF LA RAZA

I S S U E B R I E F

OVERVIEWIn 2003, Latinos** accounted for more than 8.8 millionstudents in U.S. K-12 public schools, or 19% of totalschool enrollment, making them the second-largest

segment of the U.S. student population after White students.1

Immigrant and English language learner (ELL), or limited-English-proficient (LEP),*** students are a significant part of theLatino student population. Of the estimated five million ELLstudents who were enrolled in our nation’s schools in the 2003-2004 academic year, more than three-fourths (79%) were nativeSpanish speakers.2 In fact, nearly half (45%) of all Latino publicschool children are ELLs.3 Thus, Latino student outcomes areintrinsically tied to ELL student achievement.

II

More importantly, given the growth of Latinosand ELLs in our nation’s schools, overallstudent achievement in U.S. schools willincreasingly depend on how these groups fareacademically. ELLs represent 10.3% of publicschool enrollment4 and are concentrated inlarge, urban school districts; a quarter of the100 largest school districts have an ELLpopulation of at least 15%.5 While most ofthese districts are located in “traditional” Latinoand immigrant states such as California andTexas, states that witnessed the largestpercentage growth of ELLs between 1994 and2004 include “nontraditional” Latino andimmigrant states such as South Carolina(526%), North Carolina (471%),Tennessee(448%), and Indiana (438%).6 Therefore, forthe nation’s public school system overall to besuccessful, student achievement and graduationrates for Latinos, as well as ELLs, must improvein every region of the country.

Unfortunately, educational achievement andattainment gaps between Latinos and other U.S.students remain wide. The gaps are even largerwith respect to ELL children and their Whiteand African American counterparts. Forexample, in the 2005 National Assessment ofEducational Progress (NAEP), also known asthe Nation’s Report Card, 29% of eighth-gradeELLs scored at or above the basic achievementlevel in reading, compared to 75% of non-ELLs. Similar gaps were demonstrated inmathematics.7 These gaps are also evident atthe state level. In Texas, in the 2001-2002school year, ELLs in grades 7-12 were retainedin grade at twice the rate of English-proficientstudents (13% compared to 6.5%, respectively)and dropped out of school at a rate of 77%greater than non-ELLs.8

The No Child Left Behind Act (NCLB), whichreauthorized the Elementary and Secondary

Education Act (ESEA), provides opportunitiesto narrow these gaps by holding schoolsaccountable for improving academicachievement among all groups, including Latinoand ELL students. Although previous iterationsof the federal education law required theinclusion of all students in state accountabilitysystems, they contained no requirements toclose the achievement gap specifically for ELLs,nor did they require disaggregation ofachievement data to help parents hold theschool system accountable for closing the gap.Moreover, there was little enforcement of theseprovisions. As a result, Latino and ELLstudents, along with students with disabilities,were among the most vulnerable to “gaming” ofthe system.9 That is, the school system couldmask the low achievement of Latino and ELLstudents by reporting aggregate achievementresults. Thus, ELLs were among the most likelystudents to be exempted from stateaccountability systems. For example, in the1998-1999 school year, testing andaccountability policies in at least 46 statesallowed some opportunity for exemption ofELLs.10

In addition, distortion of dropout and pushoutrates helped to artificially inflate test scores andmisrepresent student outcomes, as was the casewith the Houston Independent SchoolDistrict.11 This made tracking of ELL studentachievement difficult, which meant thatdistricts and states, in effect, were not beingheld accountable for improving educationaloutcomes for ELLs. This also meant that it wasunclear how ELLs, or millions of Latinostudents, were faring academically and whetheror not they were receiving high-qualityinstructional services. Consequently, duringthe reauthorization of ESEA, the NationalCouncil of La Raza (NCLR) worked with

Congress to ensure that NCLB would helpdistricts and states to appropriately includeELLs in their assessment and accountabilitysystems. Under NCLB, sanctions are now tiedto ELL and Latino student outcomes and, atleast in theory, there is clear accountability.Now a part of schools’ bottom line, ELLstudents will be more likely to have access torigorous coursework and highly-qualifiedteachers.

The law has unarguably directed substantialnew attention toward ELL studentachievement. Since the enactment of NCLB,educators and policy-makers are grappling withthe challenge of improving Latino and ELLstudent achievement as a means of improvingoverall student academic outcomes, many forthe first time. However, in the years sinceNCLB was enacted, demands for changes andeven outright defiance of the law at the localand state levels have become common. Forexample, in 2004, school chiefs of 14 statessent a joint letter to U.S. Education SecretaryRod Paige requesting permission to modify theaccountability system model set by NCLB.12 Inaddition, state legislation was introduced inapproximately 36 states in 2004 and 2005seeking increased flexibility, or more fundingunder the law, or limiting participation in thefederal law.13 The 2005 passage of Utah’s statelaw allows state education laws to supersedeNCLB,14 and Connecticut and Maine haveconsidered suing the U.S. Department ofEducation over NCLB.15 In response, theAdministration has modified some of the law’saccountability provisions since its enactment,including those related to ELLs. Althoughadvocates have welcomed some of theseadjustments, some of NCLB’s provisions

holding schools accountable for studentachievement outcomes, particularly those ofELLs, have been diluted by the modifications.

While NCLB continues to hold considerablepromise for closing the academic achievementgap for ELLs, the challenges that lie beforeschools are considerable and the political will touphold the rigor of the law is uncertain.Absent immediate, firm leadership and policyinterventions, it is unlikely that ELL academicoutcomes will improve, and as noted above,improving the school system as a whole isdependent in large part on whether or not ELLoutcomes improve. Fortunately, NCLRbelieves that policy-makers are genuinelyseeking answers that will inform how schoolscan appropriately include ELLs in theiraccountability systems.

NCLR has prepared this issue brief to helpinform future dialogue on assessment andaccountability. The brief will examine theprogress and manner in which states haveimplemented the federal law’s accountabilityand testing provisions with respect to ELLs.Specifically, this paper:

◗ Provides an overview of the law’s keyassessment and accountability provisionsaffecting ELLs

◗ Reviews the manner in which local, state,and federal decision-makers haveimplemented these provisions

◗ Presents policy recommendationsinforming present and futureimplementation of the law as it pertains toELL students

THE PROMISE OF NCLBFOR ENGLISH LANGUAGELEARNERSPrior to NCLB, the ELL student populationwas often overlooked. Little to noaccountability for the learning of these studentsexisted. Indeed, most states did not includeELLs in their accountability systems. Forexample, a report by the Citizens’ Commissionon Civil Rights indicates that 22 states did notappropriately include ELLs in their assessmentsystems under NCLB’s predecessor, theImproving America’s Schools Act.16 InMassachusetts, 78% of ELL sophomores wereexempted from the state’s reading/Englishlanguage arts test in 1999-2000.17 Studentperformance data about these students, as aresult, were limited as well, and the little datathat were collected revealed dismal academicoutcomes. For example, only 2% of tenth-grade ELLs in Florida met the state’s standardsin reading/English language arts in 1999-2000,and gaps between ELLs and non-ELLs in otherstates ranged from approximately five to 60points in difference.18

Based on the key premise that all students canlearn, NCLB is intended toaddress these disappointingstatistics through a standards-based reform approach.Standards-based reformincludes three majortheoretical components.First, high standards willmotivate students to improvetheir performance if they arechallenged by rigorousacademic courses. Second,

accurate assessments will be used to measureimprovement and make important decisionsabout students. Third, this reform will lead toschool system accountability by providingparents, policy-makers, and advocatesinformation about the performance of theirlocal schools. Therefore, based on theseelements, NCLB is an opportunity to ensurethat ELL students obtain access to the rigorouscoursework they need to meet challengingstandards, and that their progress is measuredby appropriate assessments. In addition, thereporting of this information required underNCLB can provide parents of ELLs and thecommunity at large with the tools to hold localschools accountable for helping ELLs meetacademic standards.

KEY NCLB PROVISIONS

AFFECTING ELLSUnderlying NCLB is the fundamental andambitious goal of closing the academicachievement gap and bringing all students to100% proficiency in core academic subjects by2014. In an effort to track schools’ progresstoward this goal, states have set yearlybenchmarks for all students and certain

What is AYP?Under NCLB, a school is considered to make AYP only if ELLs,each racial/ethnic group, low-income students, and studentswith disabilities, as well as the overall student population, meetstatewide proficiency targets in math, reading/language arts, andscience. In addition, states must include graduation rates indetermining AYP for secondary schools and one additionalfactor, such as school attendance or grade retention, indetermining AYP for elementary schools. Schools must ensurethat ELL students also meet English proficiency benchmarks,Annual Measurable Achievement Objectives (AMAOs), in orderto make AYP.

subgroups, including Latino and ELL students.In addition, states have set English-languageproficiency benchmarks for ELLs. Assessmentsare the key tool used by states to track schools’adequate yearly progress (AYP) toward thesebenchmarks and to hold them accountable forstudent outcomes. While NCLB allows a greatdeal of state and local flexibility, the law alsoprovides some specificity concerningassessment and accountability.

ASSESSMENTS

NCLB requires that by the 2007-2008 schoolyear states must administer reading/languagearts and math assessments on an annual basis ingrades 3-8 and at least once in grades 10-12, aswell as annual science assessments in grades 3-5, 6-9, and 10-12. The general testingrequirements in NCLB apply similarly to bothELL students and their English-proficient peers.Some of these provisions, however, haveimportant implications for ELLs. Particularly,the federal law stipulates that the assessmentsused by states must be aligned to state academiccontent standards. In addition, the tests mustbe valid, reliable, and of adequate technicalquality for each of the purposes for which theassessment system is used. Finally, NCLB statesthat assessments must be consistent withnationally recognized professional and technicalstandards. If properly implemented andenforced at the state and federal level, thesesafeguards can mitigate the misuse of tests andhelp ensure that data collected from state testsare meaningful to educators, parents, andadvocates.

NCLB also includes specific provisionsconcerning the assessment of ELLs. The federallaw requires states to:

◗ Include ELL students in their annual stateassessments of reading/language arts,math, and science – not exempt them.

◗ Assess ELLs in a “valid and reliablemanner.”

◗ Provide reasonable accommodations,including, to the extent practicable,“assessments in the language and form mostlikely to yield accurate data on what suchstudents know and can do in academiccontent areas,” and, if needed, with theassistance of the U.S. Department ofEducation.

◗ Identify the languages for which studentacademic assessments are not available and“make every effort to develop suchassessments.”

◗ Assess ELL students who have attendedU.S. schools for three or more consecutiveyears in English, though a waiver for amaximum of two additional years may begranted on a case-by-case basis.

◗ Annually assess English-languageproficiency in the areas of reading, writing,speaking, and listening.

These provisions provide important parametersfor assessing ELLs. However, as is discussedlater in this brief, poor or an altogether lack ofguidance for their implementation hasweakened many of these provisions.

ACCOUNTABILITY

As noted previously, federal provisionsrequiring states to test ELLs are not new.However, NCLB strengthens the inclusion ofELLs by linking both their test participationand performance to accountability sanctions.With respect to participation, NCLB requiresthat states assess 95% of their overall studentpopulation, as well as 95% of certainsubgroups, to make AYP (the “95% rule”).ELLs are one of the subgroups in which 95% ofstudents must be assessed. Therefore, schoolsface sanctions (see side box) unless nearly all oftheir ELL student population are tested.*Similarly, schools risk facing sanctions if theirELL student population fails to demonstrateprogress on both state academic and English-language proficiency assessments.

In addition, states must include graduation ratesas an indicator in determining AYP for highschools, though guidance released by the U.S.Department of Education relaxed thisrequirement. This has important implicationsfor Latinos and ELLs for two reasons: 1)Latinos and ELLs have a disproportionately highdropout rate compared to other groups,19 and2) accountability that is principally based ontest scores can have the unintended effect ofcreating incentives for schools to push out low-performing students, which have traditionallyincluded Latinos and ELLs, therebyexacerbating the already high dropout rate inthese groups.20 Most at risk for this type ofgaming are late-arrival ELLs who first enter theU.S. school system at the ninth grade level or

NCLB’s AccountabilitySanctions

A school shall be identified for “schoolimprovement” if it fails to make AYP for twoconsecutive years. Schools must come upwith a two-year school improvement plan thataddresses the specific problem that causedthe school to be identified for improvement,includes professional development, andenhances parental and communityinvolvement. Sanctions are triggered in thefollowing manner after the school has beenidentified for improvement:

Year Two – Students attending schoolsidentified for school improvement canimmediately transfer to another school in thedistrict.

Year Three – Students in such a school cancontinue to transfer and can receivesupplemental services outside of the school ifthe school fails to make AYP one year after ithas been identified for school improvement.

Year Four – If the school fails to make AYP fortwo years after it has been identified forschool improvement, students can continueto transfer and receive supplemental services,and the school can be reconstituted.Reconstitution includes such options asreplacing the staff relevant to the school’sfailure to make AYP, changing the curriculum,and extending the school day or year.

Year Five – If a school fails to make AYP forthree years after it has been identified forschool improvement, the above studentservices and school-level options apply. Inaddition, the school can be reopened as acharter school, turned over to a privatemanagement company, or be subject to astate takeover.

* In a policy letter dated May 19, 2004, Assistant Secretary of Education Raymond Simon clarified that if a school ismaking AYP but falls short of assessing 95% of students in one subgroup, it may meet the “95% rule”requirement by using a three-year test participation average.

above and have limited time to master theEnglish language, keep up academically withtheir peers, and pass the required state exams.

While NCLB employs top-down accountabilityby imposing federally-mandated sanctions onschools, the federal law also expects parentsand community members to enforce theaccountability provisions. Under NCLB,parents and stakeholders have an unprecedentedopportunity – and responsibility – to becomekey players in their state’s accountabilitysystem. By arming parents and the public withinformation about their neighborhood schoolsand state education system, NCLB seeks toincrease the likelihood that parents will engagein, monitor, and hold their child’s schoolaccountable for classroom learning andoutcomes.

The dissemination of school, district, and statedata through “report cards” and ParentAssistance Programs, which aim to help parentsuse this information, are two important ways inwhich the federal law attempts to enhance theroles of parents and community members inaccountability. Every year, states and schooldistricts must disseminate report cards toparents which are “in an understandable anduniform format, and to the extent practicable,provided in a language that the parents canunderstand.”21 The report cards, which mustinclude disaggregated data that illustrate howELL and Latino students compare to theirEnglish-proficient and non-Latino peers onacademic assessments and graduation rates, areintended to aid parents and the community in

identifying significant gaps between studentsubgroups. In addition, the report cards mustinclude disaggregated data on the percentage ofstudents that were exempted from stateassessments to help parents better identify anydisproportionate rates of exemptions amongsubgroups. Although not required, report cardscan also include the progress that ELL studentsare making in learning English.

To get help in understanding and interpretingthe information in these report cards, parentscan turn to community-based Parent AssistancePrograms, or Local Family Information Centers(LFICs).* These centers can also informparents about issues such as curriculum,standards, instruction, and assessments, and canhelp parents of ELLs make informed decisionsabout their child’s education, such as whichprogram of study is best for helping them learnEnglish and master academic coursework.

In addition, NCLB includes provisions toenhance communication and information-sharing between schools and parents of ELLchildren specifically. Within 30 days of a newschool year, schools must inform parents ofELLs of:

◗ The reasons for identifying their child asLEP and as in need of placement in alanguage instruction educational program

◗ Their child’s level of English proficiencyand how it was assessed

◗ The status of their child’s academicachievement

* Local Family Information Centers were established under Title V (Promoting Informed Parental Choice andInnovative Programs) of the No Child Left Behind Act. However, they have never been funded.

◗ The method of instruction used in theprogram in which their child isparticipating, and the methods ofinstruction used in other availableprograms, including how such programsvary in content, instruction goals, and useof English and native-language instruction

◗ How the program will help their childlearn English and meet age-appropriateacademic achievement standards for gradepromotion and graduation

◗ The specific requirements their child mustmeet to exit the language instructionprogram, the expected rate of transitioninto “mainstream” classrooms, and theexpected rate of high school graduation

◗ Information pertaining to the rights thatparents have to remove their child from aprogram and to opt out of a certainlanguage instruction program or to chooseanother available program or method ofinstruction

◗ Information intended to assist parents inchoosing among various programs andmethods of instruction if more than one isoffered

Furthermore, parents must be notified if thelanguage instruction program in which theirchild is enrolled fails to meet the benchmarksfor academic achievement and languageacquisition. All such information should beprovided in an easily accessible format and, ifpossible, in a language that the parent canunderstand. Finally, schools are required to

conduct effective outreach to parents of ELLsand inform them of how they can be activeparticipants in their child’s education.

RESOURCES

Funding is an important component of anyaccountability system. While a number ofprograms and initiatives under NCLB worktogether to close the achievement gap betweenELLs and their peers, funding for assessmentdevelopment, instructional programs for ELLs,and Parent Assistance Programs are ofparticular importance. In discussing funding,this section will focus on these threecomponents.

NCLB authorized up to $490 million tosupport the implementation of the law’s testingprovisions. This funding can be used to supportstates in their development of assessments andstandards, as required by NCLB. If states havealready developed assessments and standardsthat meet NCLB guidelines, states can use thesefunds either to administer assessments or toenhance their already-developed assessments.The law states that these funds can specificallybe used to improve assessments for ELLs.

In addition, targeted funding for ELLinstruction and programs was authorized underNCLB’s Title III, Language Instruction forLimited English Proficient and ImmigrantStudents, at a level of $750 million. However,NCLB changed the way programs for ELLs arefunded in comparison to the old law. First,NCLB consolidated the former BilingualEducation Act and the Emergency ImmigrantEducation Act.* These programs were alsochanged from being competitive-based grant

* In fiscal year 2001, prior to the passage of NCLB, the Bilingual Education Act and the Emergency ImmigrantEducation Act were funded at a total level of $446 million. In fiscal year 2002, programs for ELLs and immigrantstudents were funded at $664 million.

programs to a single state formula program,based on the number of ELL and immigrantchildren, for any year in which the Title IIIappropriations level exceeds $650 million.This modification was designed to ensure thatfederal ELL education funds reach a greaternumber of ELLs, particularly in areas that hadnot previously received dollars from the oncecompetitive-based federal grant program.

Finally, NCLB authorizes the previouslymentioned LFICs under Title V, “PromotingInformed Parental Choice and InnovativePrograms.”* Funding these centers wouldensure that parents of ELLs are in a positionto hold schools accountable for helping theirchildren meet the rigorous academicbenchmarks delineated in the law.

CHALLENGES TOEFFECTIVEIMPLEMENTATION OFNCLB FOR ENGLISHLANGUAGE LEARNERSNCLB has the potential to improve educationaloutcomes for Latinos and ELLs bystrengthening assessment and accountability.However, while NCLB in its current formoffers opportunities for educationalimprovement, implementation of NCLB hasrevealed areas of the law that have fallen shortof their mark. However, these challenges inimplementation can help inform ways in whichthe law can be fine-tuned in the future, throughlegislative or administrative policy changes atthe federal and state levels.

Appropriate implementation of the law’s testingand accountability provisions, in particular, hasproven to be the largest obstacle in effectivelyclosing the gap between ELLs and their English-proficient peers. Inadequate resources for ELLstudents have only exacerbated these obstacles.The challenges in assessment, accountability,and resources, as they pertain to ELLs, arereviewed below.

CHALLENGES IN ASSESSING ELLSAlthough assessment of ELLs is not new, NCLBhas raised the stakes for and called attention tothe lack of appropriate assessments and testingaccommodations available for ELLs. UnderNCLB, assessments are intended to act as thelinchpin to improving nearly every aspect ofeducation – teaching and learning, standards,accountability, and parental and communityengagement. Given the stronger emphasis ontests, it was clear that improving test validitywould need to become a priority, and theexpectation was that strong supports would beprovided. Thus far, however, actualimplementation of appropriate testing systemshas fallen far short of meeting the promise ofthe Act, particularly with respect with testvalidity and accommodations.

VALIDITY OF TESTING INSTRUMENTS

Due to lack of resources and technicalassistance from the U.S. Department ofEducation, most states are using invalid andinappropriate testing instruments to assess ELLsin academic content. These assessments aregenerally invalid because they were notdeveloped for use with ELLs, or because theyare being used for a purpose other than forwhich they were designed.

* Funding for LFICs is based on funding under the Parent Information Resource Centers (PIRCs) section of NCLB.PIRC appropriation levels must exceed $50 million before there can be a grants competition under the LFICsprogram.

In the first case, most tests that assess students’knowledge of academic content areinappropriate for assessing ELLs because theyare designed under the assumption that thetest-taker is a native speaker of the Englishlanguage. A valid testing instrument for ELLsshould be designed for and “field-tested” withELLs and may involve incorporating students’native language or simplifying the language ofthe test.

Research indicates that many English-languageassessments are first and foremost a measure ofEnglish-language proficiency for ELLs and donot accurately reflect content knowledge of asubject.22 While the law, as previouslydescribed, allows states to assess Latino andELL students appropriately by offering theflexibility to use native-language assessments,the U.S. Department of Education reports thatonly 13 states have taken steps to ensure thatnative-language assessments are available forELL students.23 Further evaluation of stateassessments, however, indicates that only 11states offer native-language assessmentsstatewide.24 Many states cite the linguisticdiversity of the ELL population as a keyobstacle in developing native-language tests,noting that more than 130 languages may berepresented in their schools.25* Some English-only states, such as Arkansas, note that native-language assessments run counter to state law.26

The complexity of developing and using native-language assessments appropriately has probablyhindered many states from moving forward onthis front. The translation of English-language

tests into native-language assessments alone isnot sufficient; valid native-language testscapture linguistic subtleties specific to eachlanguage as well as cultural appropriateness. Inaddition, native-language assessments, even ifvalid, may not be appropriate for some ELLs,such as those who have not received classroominstruction in their native language.27

While some states are using assessments thatwere not designed for use with ELLs, otherstates are using tests designed for thispopulation for the wrong purposes. As notedpreviously, tests may also be invalid as a resultof their improper use. For example, a testdesigned to assess an ELL’s English-languageproficiency, while valid for that purpose, maynot be a valid test to assess an ELL student’sacademic content knowledge. Schools inVirginia, with authorization from the U.S.Department of Education, use an assessmentdesigned to measure student progress inacquiring English-language skills to assess ELLs’academic content knowledge in language arts.28

Therefore, while this assessment may accuratelyconvey how Virginia’s ELLs are progressing intheir attainment of English, this same test yieldsinvalid data concerning ELLs’ performance inreading/English language arts.

Clearly, the development of valid testinginstruments for ELLs is a complicated task, butit is one that has grown in importance as theELL population has become a larger share ofthe nation’s school enrollment. It is alsoimportant to note that while NCLB allowsstates to use alternative tests for ELLs who have

* Nearly 80% of ELLs in U.S. public schools speak Spanish as their first language. Thus, Spanish-language contentarea assessments would reach a significant number of ELLs. In contrast, states that choose not to use native-language assessments are choosing not to appropriately assess ELL students.

been enrolled in U.S. schools for less thanthree years, attainment of academic Englishproficiency can take four to seven years, evenin districts that have been identified as mostsuccessful in aiding ELLs in attainingproficiency.29 Therefore, developingappropriate assessments for ELLs beyond thearbitrary marker of three years is critical.Although the U.S. Department of Educationhas provided some support to help statesimplement NCLB’s requirements for ELLs,*better targeting of this assistance and increasedguidance is necessary to ensure that validEnglish and native-language assessments aredeveloped, easily accessible to states, and usedappropriately.

ACCOMMODATIONS

NCLB gives states the flexibility to adopttesting accommodations for LEP test-takers,and most states report using them. Theaccommodations that states** report usingmost frequently are extra assessment time (44states), small group or individual administration(43 states), separate room administration (39states), directions read aloud or explained (39states), reading aloud of questions in English(36 states), use of dictionaries (36 states),

breaks during testing (31 states), and oraldirections provided in the native language (30states).30

However, the most frequently usedaccommodations are not necessarily those thathave been found to be the most effective invalidly reducing the testing gap between ELLsand non-ELLs. While research on appropriateaccommodations for ELLs is thin, existingstudies suggest that simplifying the language oftest items with excessive language demandsalone can improve ELL performance byapproximately 10-20%, regardless of thesubject area, without affecting test rigor.31 Infact, linguistic modification of test items is oneof the few, if not the only, accommodation thatnarrows the test performance gap betweenELLs and non-ELLs.32 Yet, only ten statesemploy this accommodation.33 Meanwhile,providing extra time, the most frequentlyreported accommodation by states, has beenshown to improve test performance for bothELLs and non-ELLs, without narrowing thegap, indicating that it may not be an appropriateaccommodation.34 It is clear that both researchand increased guidance in selecting appropriateaccommodations for ELLs are essential.

* In its March 2005 Biennial Evaluation Report to Congress on the Implementation of the State Formula GrantProgram, the U.S. Department of Education states that its technical assistance to states in this area includes: twosets of non-regulatory guidance, five technical assistance meetings with state educational officials, delivery oftechnical assistance via 20 video-teleconferences, three national summits, 30 presentations at regional andnational conferences, and 35 site visits. In addition, the Department has awarded several consortia of states andother organizations to develop English-language proficiency tests, research appropriate accommodations, anddevelop standards-based assessments for ELLs. Limited information on the status of these projects is available.

** Counts include the District of Columbia and Puerto Rico.

CHALLENGES IN

ACCOUNTABILITYNCLB’s emphasis on holding schoolsaccountable for ELL student outcomes is animportant step in the right direction. However,this added emphasis has unveiled a number ofissues that have hindered states fromappropriately including ELLs in theiraccountability systems. These challengesinclude accurately determining AYP, addressingthe unique challenges facing late-entrant ELLs,accuracy in reporting of data, and targeting ofappropriate supplemental services to ELLs.

ADEQUATE YEARLY PROGRESS

Obtaining an accurate measure of AYP for allsubgroups is an issue of much debate, but theELL subgroup presents some unique challenges.First, inconsistency in the manner in whichstates classify LEP children has made it difficultto draw comparisons across states. NCLB

defines LEP children as those who a) are ages3-21, b) are enrolled or prepared to enroll in aK-12 school, c) were not born in the U.S. orwhose native language is one other thanEnglish, and d) have difficulty speaking,reading, writing, or understanding English tothe extent that it hinders their ability to scoreat a proficient level on state assessments, tosuccessfully excel in English-languageclassrooms, and to participate fully in society.While seemingly comprehensive, this definitionfails to capture the diversity of the ELLpopulation. Not only does this populationrepresent more than 400 languages,35 but ELLstudents also vary in socioeconomic status,length of time in the U.S., and proficiency inboth their native language and academicEnglish. Moreover, because the currentdefinition lacks specificity, states differ in theirinterpretation and classification of LEPchildren, making it difficult to drawcomparisons.

Guiding Principles in Selecting AccommodationsAccording to Jamal Abedi, a leading researcher in the assessment of ELLs, accommodations shouldbe informed by the following:

◗ Effectiveness. An appropriate accommodation is effective in reducing the test performance gapbetween ELLs and non-ELLs.

◗ Validity. An appropriate accommodation makes the assessment more accessible to ELLs withoutaltering the rigor of the assessment or providing ELLs an unfair advantage over non-ELLs.

◗ Differential Impact. An appropriate accommodation weighs the effect of student backgroundcharacteristics on ELL performance, such as length of time in the U.S., overall grades, studentmobility, and academic English-language proficiency. Other experts also add that prior schoolingis significant.

◗ Feasibility. Although most accommodations have significant costs, some are more practicable inlarge-scale assessment and should be weighed carefully against the effectiveness and validity ofthe accommodation.

Source: Abedi, Jamal, “Assessment and Accommodations of English Language Learners,” CRESST Policy Brief 4. Los Angeles, CA:University of California, National Center for Research on Evaluation, Standards, and Student Testing, 2001. Also, see Butler, Frances A.and Robin Stevens, Accommodation Strategies for English Language Learners on Large-Scale Assessments: Student Characteristics and Other Considerations(CSE Technical Report 448). Los Angeles, CA: University of California, National Center for Research on Evaluation, Standards, andStudent Testing, 1997.

Second, NCLB’s flexibility in allowing states todetermine the minimum number of students toconstitute a school’s subgroup for AYPdeterminations, or the “n-size,” waters downthe rigor of the law’s accountability provisions.At risk here is the possibility of excluding asubgroup from the accountability system if ann-size is set too high. While the law requiresstates to determine an n-size that both willyield statistically valid data and will not allowthe identification of individual students forpurposes of privacy, this number variessignificantly across states, from an n-size of fivein Maryland to 100 in California.36 In addition,some states have implemented, with the U.S.Department of Education’s approval, a larger n-size for ELLs and students with disabilities thanfor other subgroups. Such a policy allowsschools to avoid being held accountable for ELLstudent outcomes while other subgroups of thesame size are included in the state’saccountability system. While offering somedegree of flexibility in determining this numberis reasonable given the high concentration ofELLs in some states and a small number inothers, a greater understanding of anappropriate n-size is needed.

Third, the lack of stability within the ELLsubgroup population makes it difficult toevaluate its progress in meeting standards andbenchmarks. Because ELLs exit the LEPsubgroup when they become proficient inEnglish, and students with limited English

proficiency enter the subgroup on an ongoingbasis, ELL academic and English proficiencyresults are masked. In an effort to address thisissue, the U.S. Department of Educationproposed regulations* in June 2004 allowingstates to 1) exempt recently-arrived ELLs whohave attended U.S. schools for less than tenmonths from the reading/language artsassessment, 2) exclude math test scores forrecently-arrived ELLs for AYP purposes, and 3)include ELL students who have attained Englishproficiency and have been reclassified as fully-English-proficient in the LEP subgroup for upto two years.37 The exclusion of any group ofstudents from NCLB’s accountability systemcontradicts the law’s operating principles, and,therefore, the possible exemption of excessivenumbers of recently-arrived ELLs and theirscores is a matter of grave concern. Exclusionfails to address the larger and more substantiveissue of accurately assessing LEP students andensuring their inclusion in a valid accountabilitysystem. The U.S. Department of Education’sthird change, however, is an important step inthe right direction. Still, this issue meritscareful attention. Longitudinal tracking ofELLs, including those who become fully-English-proficient, is critical to gaining a betterunderstanding of how ELLs are progressing inacademics and English-language proficiency.Unfortunately, most states and districtscurrently lack the capacity to collect and reportsuch data.

* The regulations were announced February 19, 2004, by former Education Secretary Rod Paige. The threechanges described above took immediate effect. The Department issued proposed regulations on June 24, 2004,for public comment which reiterated these changes and further elaborated on related implications. At the time ofthis writing, the regulations have not been finalized. For more information and NCLR’s position regarding theproposed regulations, please see http://www.nclr.org/content/publications/detail/28870/.

Fourth, while NCLB requires states to setannual benchmarks for all subgroups which willlead to 100% proficiency in reading/languagearts, math, and science by the year 2014,research on expected gains for ELLs isextremely limited. Therefore, states lackguidance in setting high but attainablebenchmarks for ELLs. Some research has foundthat there is a continuing and wideningachievement gap between ELLs and nativeEnglish speakers, suggesting that unless ELLsare given “more time on task,” such as throughsummer and after-school programs, it may not

be possible for them to keep pace with nativeEnglish speakers.38 This does not imply thatstates should yield to the notion that ELLscannot meet the same standards as theirEnglish-proficient peers; rather, it offers statesan idea of the level of investment needed tohelp ELLs meet similar high standards. Inaddition, this suggests that “growth”accountability systems (see text box), whichmeasure the growth of individual students’performance over time, may warrantconsideration as a way to fairly and accuratelymeasure ELL gains.

Growth and Index ModelsCurrently, in holding schools accountable for academic progress, NCLB compares the scores of agrade-level cohort of students with those of previous cohorts of students in the same grade. Thus,individual student progress is not tracked. In November 2005, the U.S. Department of Educationannounced it would approve proposals from up to ten states to participate in a pilot program thatwould allow for the development and use of growth-based accountability systems that track individualstudent achievement from year to year. To qualify for the pilot program, states must demonstratethat their accountability model meets the following core principles:

◗ Ensure that all students are proficient by 2014 and set annual goals to ensure that theachievement gap is closing for all student groups

◗ Set expectations for annual achievement based upon meeting grade-level proficiency, not based onstudent background or school characteristics

◗ Allow for accountability for student achievement in reading/language arts and mathematics,separately

◗ Ensure that all students who are required by NCLB to be tested are included in the accountabilitymodel, and that all schools and districts, statewide, are held accountable for the performance ofall student subgroups

◗ Demonstrate that the state’s assessment system include annual reading/language arts andmathematics assessments in grades 3-8 and in high school; has been operational for more thanone year; has received approval through a peer review process; and produces comparable resultsfrom year to year

◗ Track student progress

◗ Include student participation rates in state assessments and an additional academic indicator(e.g., graduation rates)

The Department will also allow states that do not have the capacity to use a growth model accordingto the above principles to use, instead, an index model. An index model gives schools credit formoving students from “below basic” to “basic,” even if they are not yet “proficient.” States that useindex models must still abide by the end target of having students proficient in reading/language artsand mathematics by 2014.

AYP is intended to give parents and the publicaccurate information about how well allstudents are faring academically, and itdetermines whether or not schools will faceNCLB’s sanctions. In 2003-2004, all but twostates (Alabama and Michigan) failed to makeAYP for ELLs in reading/English language artsand mathematics, and hundreds of schools arefacing sanctions as a result.39 While severalfactors may be contributing to the largenumber of schools facing sanctions, inaccuratemeasurement of AYP for ELLs is likely onereason. Fine-tuning AYP for ELLs should be apriority to help NCLB achieve its intendedresults.

LATE-ENTRANT ENGLISH LANGUAGE

LEARNERS

High schools face unique challenges in servingthe ELL student population due to the largershare of foreign-born immigrant children in theupper grades (5.7% of grades 6-12) than in thelower grades (3.0% of grades PreK-5).40 Inaddition, immigrant children in secondaryschools are more likely to be late-entrantimmigrants who have been in the U.S. for lessthan five years, in comparison to immigrantchildren in the lower grades.41 Under NCLB,high schools face the enormous challenge ofhelping recently-arrived immigrant and ELLhigh school students acquire English, meetacademic benchmarks, and graduate, and all ofthis must occur in a very short time frame.

In general, secondary schools are largelyunprepared to serve this population.

Interestingly, ELL high school students are lesslikely to be enrolled in language instructionclasses (48%) than elementary school-agedELLs (76%), despite the larger share of late-entrant immigrant students and the limitedtime to master the English language beforegraduation.42 With little support to learn thelanguage and pass the assessments required byNCLB, many of these students have beensteered out of traditional public schools andinto GED or other adult education programs,or out of the education system altogether,contributing to the disproportionately highdropout rate among Latinos. The effects havebeen well documented in New York City wheredropout rates among ELLs have risen steadily.43

In fact, limited English proficiency andimmigrant status are characteristic of asubstantial proportion of Latino dropouts. Forexample, 15% of Latinos ages 16-19 who arefluent in English are high school dropouts,compared to 59% of Latino ELLs of the sameage group.44 In addition, in 2000, the statusdropout rate* of Hispanic 16- to 24-year-oldsborn outside of the United States was 43.4%,or nearly three times as high as the statusdropout rate for first-generation Hispanics ofthe same age group (15.4%). Moreover,foreign-born Hispanic dropouts account for26% of all dropouts in the United States.45

Unfortunately, relaxation of NCLB’saccountability provisions for graduation rateshas masked the dropout problem amongimmigrant ELLs. Under NCLB, states musthold secondary schools accountable for both

* “Status” dropouts are persons who are not enrolled in school and who are not high school graduates. People whohave received GED credentials are counted as graduates. Data are based upon sample surveys of the civiliannoninstitutional population.

improving performance on state assessmentsand increasing graduation rates for each ethnicsubgroup. However, the U.S. Department ofEducation, departing significantly from thelaw’s theoretical underpinnings andcongressional intent, issued regulations inDecember 2002 releasing states from the law’srequirement of disaggregating graduationrates.46 Not only did these final regulations“officially” authorize school officials to usegraduation rates in the aggregate, but “somestate officials suggested that the Department ofEducation’s approval of weaker systems hadencouraged them to employ ‘softer’ graduationrate requirements than they had originallyproposed.”47 As a result, states can reportgraduation rates in the aggregate and hide thefact that few ELLs are graduating from highschool.

While stronger accountability for improvinggraduation rates among Latinos and ELLs iscritical, it is clear that states and secondaryschools need increased support to address theunique needs of late-entrant ELLs. Educationpractices and policy that give such studentsmore time on task, in particular, appearpromising. For example,Virginia’s ArlingtonPublic Schools has made summer school, after-school tutoring, and/or weekend tutoringvirtually mandatory for ELL students. As aresult, these students have had more time tolearn English and stay on track academically.

Late-entrant ELL students pose challenges toeducators and to NCLB’s accountability system.While they are small in number, they deservethe same opportunity as other students tograduate from high school and go on to collegeor enter the workforce. Unfortunately, theirunique situation is not adequately addressedunder NCLB and requires modifications to

NCLB as well as additional supports toaccelerate their academic progress and English-language acquisition.

REPORTING

Lack of effective reporting systems for ELLparents also undermines the law’saccountability provisions. In theory, all parentsshould have access to the data collected throughstate assessments and other means to informtheir decisions about their children’s schoolingand to help them hold their children’s schoolsaccountable. Much of this information is madeavailable to parents and the community in stateand district annual report cards, which arerequired under NCLB to be easily accessibleand “to the extent practicable” in a languageother than English when appropriate.However, most ELL parents do not havemeaningful access to report card data.

In general, most states post their report cardson their state education website, and severalstates, including California, Illinois, NewMexico, North Carolina, and Oregon, post thereport cards in Spanish. Still, aside from thereport card data, the websites themselves mustbe navigated in English, making it extremelydifficult for Spanish-speaking parents to find theinformation they need. Guidance released bythe U.S. Department of Education encouragedstates to disseminate report cards in multipleways, noting that the Internet alone was notsufficient. However, some states, such as NewYork, still do not use other means.48 ForLatinos and parents of ELLs who have lessaccess to the Internet,49 or who may be lessfamiliar with the U.S. school system andculturally predisposed to perceive educatorsand school officials as authority figures,disseminating report cards using other methods

is important. If funded, LFICs, or othercommunity-based Parent Assistance Programs,can fill this information gap for Latinos, whooften turn to such entities for many of theirsocial service needs. In addition, Spanish-language media can serve as a way for reachingand informing parents.

Finally, it is important to note that the U.S.Department of Education’s proposedregulations pertaining to assessment andaccountability practices for ELLs haveimportant implications for reporting of data toparents of ELLs and the community at large.The proposed regulations would not allow theinclusion of former LEP students’ achievementscores as part of the LEP category on state anddistrict report cards “because there is adifference between data used for systemaccountability and data used for providinginformation to parents.”50 This directiveconflicts with NCLB’s principle of informingparents and other community stakeholders sothey can track the progress of students in localschools. While allowing schools to includeformer LEP students in the broader LEPsubgroup for AYP determinations is important,it is equally critical that states and districtsreport achievement results to parents in thesame manner in which they are included forAYP purposes. As key players in schoolaccountability systems, parents and thecommunity should be equally informed of theprogress that their school is making to help ELLstudents learn English.

SUPPLEMENTAL SERVICES

Under NCLB, students attending persistentlylow-performing schools are eligible to receivefree tutoring, either by the school district oranother “supplemental service provider,” which

can include nonprofit community-basedorganizations, for-profit companies, schooldistricts, or institutions of higher education.For ELLs, additional time on task is critical and,therefore, access to supplemental educationservices is important. Unfortunately, ELLs,along with students with disabilities, do nothave equitable access to these tutorial servicesin comparison to their peers. A key challengeis ensuring that parents of ELLs are informed ofthe availability of supplemental services.According to a report by Advocates forChildren, letters informing parents of NewYork City’s students of supplemental servicesare difficult to read and understand, andoutreach to parents varies from school toschool.51

In addition, ELLs are having a more difficulttime finding providers that can adequately meettheir needs, particularly through privatecompanies, which make up 63% of all state-approved service providers.52 Although NCLBrequires states and districts to ensure thatproviders of supplemental educational servicescan serve all eligible students, including ELLs,many private providers are unwilling to serveELLs.53 While private providers indirectlyreceive federal funds to offer such services, theU.S. Department of Education’s guidancerelieves them of the responsibility to adhere tocivil rights protection laws.54 Instead, it placesthe burden on states and districts, the directrecipients of federal funds, to ensure that allproviders do not discriminate. In addition,some providers simply lack the capacity toserve ELLs. For example, a survey of privateproviders in New York City indicated that 60%of respondents did not have services availablefor ELLs.55 In the event that none of the state-approved providers can serve ELLs, the school

district must provide tutorial services directlyto ELLs or through a contracted provider thatmay not be on the state’s approved list ofproviders. Some districts, such as ChicagoPublic Schools, have provided tutorial servicesto ELLs themselves.56 However, in general,little research and monitoring of supplementalservices exist and, therefore, it is unclear towhat extent districts are taking on or even havethe capacity to take on this responsibilitythemselves.

High-quality supplemental services can helpELLs catch up academically with their English-proficient peers and should be focused on thoseleast likely to meet academic benchmarks, suchas ELLs. Currently, this provision of NCLB isnot fulfilling this important role.

CHALLENGES IN OBTAINING

ADEQUATE RESOURCESA comprehensive accountability systemencompasses “checkpoints” from the top downand bottom up. Thus, if all schools are pushedto meet the same high standards, there must bea similar effort to ensure that all schools havethe resources needed to meet these standards.Unfortunately, federal funding to carry out thelaw’s assessment and accountability provisionsand to authentically include ELLs has beeninadequate. For example, funding for ELLlanguage instruction programs under Title III ofNCLB reached its peak in fiscal year (FY) 2003at $686 million. Since then, the program hasbeen cut each year, receiving $669 million inFY 2006. As a formula-grant program,

FY 2002 FY 2003 FY 2004 FY 2005 FY 2006Program Appropriation Appropriation Appropriation Appropriation Appropriation

(in millions) (in millions) (in millions) (in millions) (in millions)

Parent AssistancePrograms 40 42 42 42 40

Language Assistance State Grants 664 686 681 676 669

State Assessments 387 385 390 412 408

Source: “Departments of Labor, Health and Human Services, and Education, and Related Agencies Appropriations Act,” ConferenceReports, Fiscal Years 2002-2006.

Note: The Local Family Information Centers (LFICs) are authorized under Parent Assistance Programs in Title V of the Elementary andSecondary Education Act and receive an appropriation of 50% of the amount allocated to Parent Assistance Programs and that exceeds$50 million. Thus, for example, LFICs would receive $2 million if Parent Assistance Programs were funded at $54 million.

Funding for Select Federal Education Programs ConcerningAssessment and Accountability for ELLs

designed to reach the rapidly-growingpopulation of ELLs in every public school,adequate funding is critical. Otherwise, therisk is that federal monies are spread too thinlyand have little impact.

In addition, the President has failed to request asingle dollar to fund Parent AssistancePrograms and LFICs. Fortunately, Congress hasprovided some nominal funding for parentassistance activities. These dollars, however,have failed to reach LFICs, which are fundedonly if Parent Assistance Programs are funded at$50 million or more.

Finally, while the President and Congress haveprovided funds to states for assessmentdevelopment – $408 million in FY 2006 –some of these funds have been used to improveassessments measuring student progress inattaining English. However, it is unclear howmuch of this money has been spent on thedevelopment of appropriate academicassessments for ELLs. States have the flexibilityto use these funds for ELL academicassessments, but given the testing demandsunder NCLB, little progress in the developmentof native-language or simplified Englishacademic assessments for ELLs has been made.

While federal funding to support NCLBimplementation has been largely insufficient,the law has heightened the significance of fiscalequity and adequate resources at the state level.The responsibility that states bear in fundingthe federal education reforms is becoming clearas the trend favoring those who have challengedstate education finance systems begins toemerge.57 The outcomes of these cases have, atleast in part, been due to the standards set byNCLB, as well as the increased data that havebeen generated as a result of the law’s

accountability requirements. Thus, althoughstates may be looking toward Congress and theAdministration to increase resources for NCLBimplementation, states must also fulfill theirresponsibility given that federal educationdollars make up only 8% of all funding foreducation.58

RECOMMENDATIONSMost states are making efforts to assess andinclude ELLs in their accountability systems asa result of NCLB. However, the mere inclusionof ELLs in statewide assessments is notsufficient. During the reauthorization of ESEA,NCLR worked with Congress to ensure thatNCLB would help districts and states toappropriately include ELLs. While some statesand districts have enhanced their capacity toinclude ELLs under NCLB, implementation ofthe law has been difficult. As a result, ELLshave not fully benefited. Reasonableadjustments to NCLB’s assessment andaccountability provisions, with the neededresources, will help states and schoolsappropriately include ELLs. In addition, it willresult in more accurate assessment data andimproved identification of struggling schools.The following recommendations provide abroad framework for improving NCLB forELLs primarily through modest changes to thelaw, leadership by the U.S. Department ofEducation, and funding of NCLB programs.

ASSESSMENTSUntil the principal tools used to enforceaccountability for ELL student outcomes aredeveloped and made accessible, ELLs willremain on the periphery of NCLB’s proposedaccountability system. In addition, schools willbe unable to accurately demonstrate any gains

that ELLs have made under their watch. Thus,to improve the accuracy of assessment forELLs, NCLR recommends the following:

◗ The U.S. Department of Educationshould increase research andinvestment in the development of arange of appropriate assessmentsand testing accommodations,including native-language andsimplified English tests for ELLs. Arange of assessments must be developedgiven the student population’s variedlanguage needs and instructional settings.While the Department should play aleadership role, dollars should be matchedby states, private philanthropy, and otherstate and regional stakeholders.

◗ The U.S. Department of Educationshould provide firm guidance tostates regarding the law’s directive toassess ELLs “to the extentpracticable, in the language andform most likely to yield accuratedata.” As noted above, more than three infour ELLs in U.S. public schools areSpanish speakers. Thus, native-languageassessments in Spanish are unarguably“practicable.” In addition, simplifiedEnglish tests can likely be used by ELLs of avariety of linguistic backgrounds. TheDepartment should vigorously enforce thisprovision and provide leadership by helpingstates develop such assessments. Moreover,the Department should also invest in thedevelopment of native-languageassessments in other high-incidencelanguages, including Asian languages.

ACCOUNTABILITYImproving the validity of assessments for ELLsis the first step toward an authenticaccountability system. However, there areadditional components that are equally critical.Authentic accountability for ELLs encompassesfederal and state enforcement of inclusion ofELLs in AYP determinations, an accuratemeasurement of ELL academic progress,enforced accountability for pushout rates,improved reporting of data to parents, andequitable access to supplemental services.NCLR recommends the following:

◗ With enforcement by the U.S.Department of Education, states anddistricts must continue to assess ELLsand include them in AYPdeterminations. Exempting ELLs fromNCLB’s AYP system should be a “non-starter” for policy-makers. While assessingand including ELLs in AYP is challenging,the alternative – eliminating accountabilityfor improving ELLs’ academic outcomesand English-language acquisition – woulddo nothing to improve schooling for thesechildren. One initial step that theDepartment can take to ensure that ELLsare assessed and included in AYP is toprovide unambiguous guidance to statesregarding an appropriate n-size andrequiring the same n-size across all AYPsubgroups.

◗ The Administration and Congressshould fine-tune the definition ofAYP for ELLs. For many schools, the ELLsubgroup has presented some of the mostsignificant challenges in their goal to makeAYP and avoid sanctions. Limited

knowledge of expected gains for ELLs,together with the unique nature of thesubgroup, has led to inaccurate measures ofAYP for ELLs. Attempts to exclude orsteer ELLs away from AYP determinationshave been common as a result. To enhancethe definition of AYP for ELLs, NCLRrecommends the following:

● The Department and Congress shouldhelp build capacity to move toward amore comprehensive and unifiedclassification of “LEP” children bytracking ELL students over time.Longitudinal tracking of ELLs, evenafter they become fully-English-proficient, can offer a more accurateassessment of their progress andprovide stability to the ELL subgroup.Although states and districts currentlylack the capacity to do thisappropriately, the Department andCongress can provide some initialsupport by establishing a pilot ordemonstration program with aconsortium of states and districts.

● The Department should supportresearch on expected gains for ELLs toinform the development of meaningfulAYP benchmarks. In addition, theDepartment should ensure that thestates which take advantage of theDepartment’s pilot program on growthmodels include “value-added” measuresfor ELLs, which are rigorous as well asattainable.

◗ The U.S. Department of Educationand Congress should enhanceaccountability measures forsecondary ELLs, particularly late-entrant ELLs. As was evidenced prior toNCLB, ELLs are among the mostvulnerable when schools seek to game thesystem to avoid sanctions. At thesecondary level, this is evidenced inincreased dropout rates. Thus, enforcedaccountability for high school completionrates is critical for ELLs. In addition,because it is unreasonable to expect certainELLs, especially late-entrant ELLs, to beprepared to pass high school exit exams,programmatic and policy responses thatgive late-entrant ELLs more time on taskneed to be explored. Specifically, NCLRrecommends that the Department andCongress take the following steps:

● Enforce disaggregation of high schoolcompletion rates in AYP determinationsand include ELLs as a subgroup.

● Enforce a standard definition orformula for calculating high schoolgraduation rates across states.

● Create a grants program to allow schooldistricts to establish early college highschools for ELL students who havemastered academic content but continueto struggle with English. These schoolswill keep these students from droppingout, give them a chance to improve theirEnglish so they can pass state exit exams,and keep them engaged by allowingthem to earn some college credits. Suchschools could be particularly effective forlate-entrant ELLs.

◗ The U.S. Department of Education,states, and districts should improvereporting of assessment data andother AYP indicators to parents ofELLs. Under NCLB, parents bear asignificant responsibility in holding theirchildren’s schools accountable. Currently,however, parents lack the criticalinformation to play this role capably.NCLR recommends that:

● States and districts improve reportingof assessment data by ensuring that theyare both available and easily accessibleto parents of ELLs in their nativelanguage and are disseminated throughmultiple media, such as the ethnicpress.

● States and districts leverage theknowledge and capacity of community-based organizations, which have gainedthe trust of Latinos and parents ofELLs, to help disseminate report carddata to parents of ELLs.

● The Department identifies anddisseminates exemplary report cardsand/or develops sample report cardsfor parents of ELLs throughdemonstration projects that can beadapted and adopted by states.

● The Department, in its final guidanceregarding ELLs and accountability,ensures that states and districts reportachievement results to parents in thesame manner in which they are

included for AYP purposes.Specifically, states and districts shouldinclude former LEP students in theLEP category both for determiningwhether or not schools are meetingAYP targets and in reporting results toparents.

◗ The U.S. Department of Educationand Congress should ensureequitable access to supplementalservices for ELLs. NCLR recommendsthe following:

● Policy-makers should strengthen NCLBprovisions that require states to overseesupplemental service providers toensure that they provide ELLs withaccess to high-quality services,including services in their nativelanguage.

● Congress should immediately reversethe Department’s regulation which saysthat supplemental service providers arenot recipients of federal funds, thusrelieving them of their responsibilityunder Title VI of the Civil Rights Act of1964, which prohibits discriminationbased on race, ethnicity, and languagestatus, among other categories. Freeingsupplemental service providers, or anyother federal funds recipient, fromtheir responsibilities under Title VI doesnothing to improve education and posesbarriers for children who have thegreatest need.

RESOURCESTo be successful, NCLB education reformsrequire significant investment. By now, policy-makers should understand that this law cannotbe implemented “on the cheap.” Policy-makersshould be mindful of how rapidly the ELLpopulation is growing and of the fact that manystates, such as those in the Southeast andMidwest, lack the supports and expertise toappropriately serve ELLs. NCLR recommendsthe following:

◗ The President and Congress mustincrease the federal investment inEnglish language learner programs(Title III). In the past several fiscal years,federal funding for language instructionprograms has been inadequate to ensurehigh-quality instructional services for thegrowing number of ELLs. While NCLBprovides an authorization level of $750million,Title III has been cut or level-funded, receiving only $669 million in FY2006. To keep pace with the growth ofELLs attending K-12 schools, NCLRrecommends a funding level of $900million for FY 2007 and further increasesin subsequent years to keep pace with thegrowth in the ELL student population.

◗ The President and Congress shouldincrease federal support for ParentAssistance Programs. Testing andaccountability are at the heart of NCLB,but parents are the backbone of the law.Yet, many parents are largely unaware ofthe rights and responsibilities they haveunder NCLB, and the Administration hasrequested zero dollars for Parent AssistancePrograms every year since enactment of

the law. NCLB simply cannot work unlessit provides parents the supports they needto take advantage of the opportunitiesNCLB presents and to fulfill theirresponsibilities under NCLB. Rather thanbe eliminated, this basic tenet of fosteringparental engagement should be reinforcedby funding Parent Assistance Programs at$100 million, including $25 million forLFICs for FY 2007.

◗ The U.S. Department of Educationshould increase its investment in thedevelopment of assessments for ELLs.NCLB authorizes funding for thedevelopment of assessments. While mostof the funding goes directly to states, theDepartment reserves some funds to awardcompetitive grants to enhance assessments.NCLR supports the Department’s effortsto allocate some of these discretionaryfunds toward the development of English-language development assessments forELLs. However, NCLR believes it isimportant to increase this investment andreserve a specific allotment for ELLcontent-area assessments andaccommodations.

◗ States should ensure fiscal equity intheir education finance systems, withadequate inclusion of resources forELLs. While the federal investment inELLs must be increased, states must “stepup to the plate” and ensure that they areproviding adequate resources for theeducation of ELLs.

CONCLUSIONNCLR believes that NCLB is an important stepin the right direction and that the law holdsconsiderable promise for closing theachievement gap between ELLs and theirEnglish-proficient peers. It is imperative thatall key stakeholders work together to perfect,not discard, NCLB’s accountability framework.The alternative is a school system with littleaccountability for student outcomes, and onethat is ill-prepared to help a significantproportion of the current and future studentpopulation master academic coursework andgraduate with a high school diploma thatprepares them for college and the workplace.

The Latino student population, of which ELLsare a significant proportion, is one of the largestand fastest-growing subgroups in our nation’sschools and which lags behind non-Latinos inmost achievement measures. Given thesedemographic and statistical trends, NCLR

believes that state and district accountabilitysystems not only must include ELLs, they mustbe implemented in a way that effectively closesthe existing academic achievement gap forELLs. If accountability and assessment systemsfail to work for ELLs, who make up 10% of thepublic school population, they will fail to workfor most public schools.

Nonetheless, to date, NCLB implementationwith respect to ELLs has failed to live up to thelaw’s promise. The National Council of La Razahopes to work with Congress, the WhiteHouse, community groups, and otherstakeholders to improve public schools bystrengthening NCLB. This issue brief identifiesa series of steps required to realign NCLB’simplementation with its stated goal of leavingno child behind. We hope that it serves to spurdiscussion on reauthorization of this law andthat those conversations focus on ELL students.

ENDNOTES1. U.S. Census Bureau, Current Population Survey, “School

Enrollment – Social and Economic Characteristics ofStudents: October 2003,”Table 5. Washington, DC: U.S.Department of Commerce, 2005. Number calculated byNCLR.

2. Padolsky, Daniel, Ask NCELA No.1: How many school-agedEnglish language learners (ELLs) are there in the U.S.?Retrieved (July 9, 2005) Online:http://www.ncela.gwu.edu/expert/faq/01leps.htm.Washington, DC: National Clearinghouse for EnglishLanguage Acquisition and Language Instruction EducationalPrograms, May 2005. Also, Padolsky, Daniel, Ask NCELA No.5:What are the most common language groups for ELL students?Retrieved (July 9, 2005) Online:http://www.ncela.gwu.edu/expert/faq/05toplangs.htm.Washington, DC: National Clearinghouse for EnglishLanguage Acquisition and Language Instruction EducationalPrograms, December 2002.

3. Percentage calculated by NCLR using 2002-2003 data fromthe National Clearinghouse for English Language Acquisitionand Language Instruction Educational Programs and U.S.Census Bureau, Current Population Survey, October 2003.

4. Ask NCELA No.1: How many school-aged English languagelearners (ELLs) are there in the U.S.?, op. cit.

5. National Center for Education Statistics, Characteristics of the100 Largest Public Elementary and Secondary School Districts inthe United States: 2000-2001,Table 15. Washington, DC:U.S. Department of Education, 2002.

6. Ask NCELA No.1: How many school-aged English languagelearners (ELLs) are there in the U.S.?, op. cit.

7. National Center for Education Statistics, National Assessmentof Educational Progress (NAEP): Reading and Mathematics.Retrieved (January 10, 2006) Online:http://nces.ed.gov/nationsreportcard/nrc/reading_math_2005. Washington, DC: U.S. Department of Education, 2005.

8. Rice, Roger, “Adequacy and Civil Rights for Immigrant andELL/LEP Students,” Presentation at the 2005 EducationAdequacy Conference. Washington, DC: Campaign for FiscalEquity, Inc./ACCESS Network, National School FundingNetwork, Public Education Network, Education Law Center,Rural School and Community Trust, June 2005.

9. Losen, Daniel, “New Research on Special Education andMinority Students with Implications for Federal EducationPolicy and Enforcement,” in Dianne M. Piché,William L.Taylor, and Robin A. Reed, eds., Rights at Risk: Equality in anAge of Terrorism. Washington, DC: Citizens’ Commission onCivil Rights, 2002.

10. Rivera, Charlene, Charles W. Stansfield, Lewis Scialdone, andMargaret Sharkey, An Analysis of State Policies for the Inclusionof Accommodation of English Language Learners in StateAssessment Programs during 1998-1999. Washington, DC:Office of Bilingual Education and Minority Affairs, U.S.Department of Education, 2000.

11. Special Data Inquiry Unit,Texas Education Agency, FinalInvestigative Report: Houston Independent School District.Austin,Texas:Texas Education Agency, August 7, 2003.

12. “14 States Ask U.S. to Revise Some Education Law Rules,”New York Times, March 25, 2004.

13. See,Young, Scott, “No Child Left Behind: Quick Facts 2004-2005.” Retrieved (January 10, 2006) Online:http://www.ncsl.org/programs/educ/NCLB2005LegActivity.htm. Washington, DC: NationalConference of State Legislatures, July 1, 2005; and “StatesRevive Efforts to Coax NCLB Changes,” Education Week,February 2, 2005.

14. “Utah Passes Bill to Trump ‘No Child’ Law,” Education Week,April 27, 2005.

15. See, for example, “Conn. Attorney General Says NCLBLawsuit Still a Go,” Education Week, June 22, 2005; and“Maine Legislative Panel Gives Nod to NCLB Lawsuit,”Education Week, May 18, 2005.

16. Citizens’ Commission on Civil Rights, Closing the Deal:APreliminary Report on State Compliance with Final Assessment &Accountability Requirements under the Improving America’sSchools Act of 1994. Washington, DC: Citizens’ Commissionon Civil Rights, 2001.

17. Albus, Debra, Martha Thurlow, and Kristin Liu, 1999-2000Participation and Performance of English Language LearnersReported in Public State Documents and Web Sites. Minneapolis,MN: University of Minnesota, National Center onEducational Outcomes, 2002.

18. Ibid.

19. Swanson, Christopher B., Who Graduates? Who Doesn’t? AStatistical Portrait of Public High School Graduation, Class of2001. Washington, DC: Education Policy Center,The UrbanInstitute, 2004.

20. Orfield, Gary et al., Losing our Future: How Minority Youth areBeing Left Behind by the Graduation Rate Crisis. Cambridge,MA:The Civil Rights Project at Harvard University, 2004.

21. Elementary and Secondary Education Act, No Child LeftBehind Act of 2001, P.L. 107-110.

22. Menken, Kate, “What are the Critical Issues in Wide-ScaleAssessment of English Language Learners?” NCBE Issue BriefNo. 6. Washington, DC: National Clearinghouse for BilingualEducation, 2000.

23. Office of English Language Acquisition, LanguageEnhancement and Academic Achievement for Limited EnglishProficient Students, Biennial Evaluation Report to Congress onthe Implementation of the State Formula Grant Program.Washington, DC: U.S. Department of Education, 2005.

24. “State Testing of English-Language Learners Scrutinized,”Education Week, June 15, 2005.

25. Washington Area School Study Council, ImprovingAccountability for Limited English Proficient and SpecialEducation Students Under the No Child Left Behind Act,December 12, 2003.

26. Arkansas Department of Education, “Consolidated StateApplication Accountability Plan.” Retrieved (July 13, 2005)Online:http://www.ed.gov/admins/lead/account/stateplans03/arcsa.pdf. April 18, 2003.

27. Abedi, Jamal, “Assessment and Accommodations of EnglishLanguage Learners: Issues, Concerns, Recommendations,”Journal of School Improvement,Vol. 3, No. 1, Spring 2002.

28. Virginia Board of Education, “Consolidated State Application:Amended Accountability Workbook.” Retrieved (July 13,2005) Online:http://www.ed.gov/admins/lead/account/stateplans03/vacsa.pdf. May 26, 2004.

29. Hakuta, Kenji,Yuko Goto Butler, and Daria Witt, How LongDoes it Take English Learners to Attain Proficiency? (PolicyReport 2000-1). University of California: University ofCalifornia Linguistic Minority Research Institute, 2000.

30. Biennial Evaluation Report to Congress, op. cit.

31. Abedi, Jamal and Ron Dietel, “Challenges in the No ChildLeft Behind Act for English Language Learners,” CRESSTPolicy Brief 7. Los Angeles, CA: University of California,National Center for Research on Evaluation, Standards, andStudent Testing, 2004.

32. See, Abedi, Jamal, “Assessment and Accommodations forEnglish Language Learners,” CRESST Policy Brief 4. LosAngeles, CA: University of California, National Center forResearch on Evaluation, Standards, and Student Testing,2001; and Stansfield, Charles and Charlene Rivera, “TestAccommodations for LEP Students,” ERIC Digest. Retrieved(July 13, 2005) Online:http://www.ed.gov/databases/ERIC_Digests/ed458289.html. College Park, MD: ERIC Clearinghouse onAssessment and Evaluation, 2001.

33. Biennial Evaluation Report to Congress, op. cit.

34. “Assessment and Accommodations of English LanguageLearners,” op. cit.

35. Kindler, Anneka L., Survey of the States’ Limited EnglishProficient Students and Available Educational Programs andServices, 2000-2001 Summary Report. Washington, DC:National Clearinghouse for English Language Acquisition andLanguage Instruction Educational Programs, 2002.

36. Pierce, Jason, “Minimum Size of Subgroups for AdequateYearly Progress (AYP),” State Notes: No Child Left Behind.Retrieved (July 13, 2005) Online:http://www.ecs.org/clearinghouse/49/76/4976.htm.Denver, CO: Education Commission of the States, 2003.

37. “Title I – Improving the Academic Achievement of theDisadvantaged; Proposed Rule,” Federal Register,Vol. 69, No.121, June 24, 2004.

38. How Long Does it Take English Learners to Attain Proficiency?,op. cit.

39. “Federal Data Show Gains on Language,” Education Week,March 23, 2005.

40. Capps, Randy et al., “Promise or Peril: Immigrants, LEPStudents and the No Child Left Behind Act,” PowerPointpresentation. Washington, DC:The Urban Institute,December 2004.

41. Ruiz de Velasco, Jorge and Michael Fix, “Limited EnglishProficient Students and High-Stakes Accountability Systems,”in Dianne M. Piché,William L.Taylor, and Robin A. Reed,eds., Rights at Risk: Equality in an Age of Terrorism.Washington, DC: Citizens’ Commission on Civil Rights,2002.

42. Ibid.

43. See, Creating a Formula for Success:Why English LanguageLearner students are dropping out of school, and how to increasegraduation rates. New York, NY: Advocates for Children ofNew York and the New York Immigration Coalition, 2002;Hyman, Elisa, “School Push-Outs: An Urban Case-Study,”Clearinghouse REVIEW: Journal of Poverty Law and Policy,Vol.38, January-February 2005, pp. 684-689.

44. Fry, Richard, Hispanic Youth Dropping Out of U.S. Schools:Measuring the Challenge. Washington, DC: Pew HispanicCenter, 2003.

45. National Center for Education Statistics, The Condition ofEducation 2003.Washington, DC: U.S. Department ofEducation, 2003.

46. “Title I – Improving the Academic Achievement of theDisadvantaged; Final Rule,” Federal Register,Vol. 67, No. 231,December 2, 2002.

47. Losing our Future, op. cit.

48. Miller, Gifford and Eva S. Moskowitz, Keeping Score: Can youJudge a School by Its Report Card? New York, NY: Council ofthe City of New York, 2004.

49. Economics and Statistics Administration and NationalTelecommunications and Information Administration, ANation Online: How Americans are Expanding their Use of theInternet. Washington, DC: U.S. Department of Commerce,2002.

50. “Title I – Improving the Academic Achievement of theDisadvantaged; Proposed Rule,” op. cit.

51. Serving Those Most in Need or Not? A Report on theImplementation of the NCLB’s Supplemental Education Servicesin New York City. New York, NY: Advocates For Children,2003.

52. “Tutoring Aid Falling Short of Mandate,” Education Week,February 25, 2004.

53. Ibid.

54. “Supplemental Educational Services: Non-RegulatoryGuidance.” Washington, DC: U.S. Department of Education,June 13, 2005.

55. Serving Those Most in Need or Not?, op. cit.

56. “Tutoring Aid Falling Short of Mandate,” op. cit.

57. “States on Ropes in Finance Lawsuits,” Education Week,December 8, 2004.

58. U.S. Department of Education, “The Federal Role inEducation.” Retrieved (January 17, 2006) Online:http://www.ed.gov/about/overview/fed/role.html?src=ln.Washington, DC: U.S. Department of Education, April 11,2005.

TO ORDER NCLR PUBLICATIONS CONTACT:The National Council of La Raza (NCLR)

Attention: Office of Publications Raul Yzaguirre Building

1126 16th Street, NWWashington, D.C. 20036

Tel: 202.785.1670 ◆ Fax: 202.776.1794

There is a cost of $5.00 per Issue Brief, with a minimum charge of $10.00, plus the cost of shipping and handling per order.

All orders should be pre-paid. Discounts are available for bulk orders.

w w w . n c l r . o r g

HISPANIC FAMILIES AND THE EARNEDINCOME TAX CREDIT (EITC) ISSUE BRIEFISSUE BRIEF NO. 1

FINANCIAL SERVICES AND HISPANICAMERICANSISSUE BRIEF NO. 2

WELFARE REFORM, TANF CASELOADCHANGES, AND LATINOS: A PRELIMINARYASSESSMENTISSUE BRIEF NO. 3

THE LATINO VOTE IN THE 90’SISSUE BRIEF NO. 4

FINANCIAL INSECURITY AMID GROWINGWEALTH: WHY HEALTHIER SAVINGS ISESSENTIAL TO LATINO PROSPERITYISSUE BRIEF NO. 5

SAFE ROADS, SAFE COMMUNITIES:IMMIGRANTS AND STATE DRIVER'S LICENSEREQUIREMENTSISSUE BRIEF NO. 6

INCREASING HISPANIC HOMEOWNERSHIP:STRATEGIES FOR PROGRAMS AND PUBLIC POLICYISSUE BRIEF NO. 7

THE NO CHILD LEFT BEHIND ACT:IMPLICATIONS FOR LOCAL EDUCATORS ANDADVOCATES FOR LATINO STUDENTS,FAMILIES, AND COMMUNITIESISSUE BRIEF NO. 8

IMMIGRATION ENFORCEMENT BY LOCALPOLICE: THE IMPACT ON THE CIVIL RIGHTSOF LATINOSISSUE BRIEF NO. 9

COUNTERTERRORISM AND THE LATINOCOMMUNITY SINCE SEPTEMBER 11 ISSUE BRIEF NO. 10

PENSION COVERAGE: A MISSING STEPIN THE WEALTH-BUILDING LADDERFOR LATINOSISSUE BRIEF NO. 11

TANF IMPLEMENTATION INPUERTO RICO: A SUMMARY OFDATA ON LEAVERSISSUE BRIEF NO. 12

IMMIGRATION REFORM: COMPREHENSIVESOLUTIONS FOR COMPLEX PROBLEMSISSUE BRIEF NO. 13

ACHIEVING A HIGH-QUALITY PRESCHOOLTEACHER CORPS: A FOCUS ON CALIFORNIAISSUE BRIEF NO. 14

JEOPARDIZING HISPANIC HOMEOWNERSHIP:PREDATORY PRACTICES IN THEHOMEBUYING MARKETISSUE BRIEF NO. 15

NCLR ISSUE BRIEFS

w w w . n c l r . o r g