Carolyn Vincent,
The George Washington University 
Center for Equity and Excellence in Education 

Anne Hafner, 
California State University, Los Angeles

Mark LaCelle-Peterson,
State University of New York, Geneseo 

The standards-based educational reform initiatives of the 1990s call for assessment innovations in support of high standards. The move to standards requires consideration of how assessments, both those currently used and those that states and school districts are currently developing, will enable all students, including LEP students, to demonstrate what they know and can do (LaCelle-Peterson & Rivera, 1994).

The assessment of LEP students' educational progress has long posed a dilemma for state departments of education, which generally use state assessment data to report student achievement. LEP students have typically been either exempted from state assessments or included inappropriately. Because kindergarten through twelfth grade students whose first language is not English constitute a large and growing segment of the U.S. population (Waggoner, 1995), it is critical that state education departments be able to monitor and report on the academic progress of all students, including LEP students, through statewide assessment programs.

To document state assessment policies and to develop policy recommendations, The George Washington University (GW) Center for Equity and Excellence in Education (CEEE) surveyed state assessment directors in 1994 Note 1 . The GW/CEEE survey, Policies and Practices, 1993-1994 Statewide Assessment Programs: Participation of Limited English Proficient Students, was sent to all state assessment directors and was augmented with relevant data regularly collected through the CCSSO/NCREL annual state directors' assessment survey (Bond, van der Ploeg, and Braskamp, 1995).

Survey data from the fifty states and the District of Columbia were aggregated and analyzed to examine state policy/practice questions related to (1) the nature and extent of statewide assessment programs, (2) the extent to which LEP students are included in statewide assessment programs, (3) the extent to which individual state assessments are modified to include LEP students, and (4) the extent to which state assessments are used for accountability purposes.

CONTEXT OF THE STUDY

Although a federal definition of limited English proficiency exists (see Improving America*s Schools Act, Title VII, Part E, Sec. 7501 (8)), there is no common operational definition used by states to identify LEP students. The variation among states in the criteria and instruments used means, in effect, that a student could be considered LEP in one state but not in another. In the school year 1993-94, when the GW/CEEE survey was conducted, 3,037,922 LEP students were identified out of a total of 45,443,389 students enrolled in U.S. public schools in the 50 states and the District of Columbia (Donly, Henderson, & Strang, 1995).

RESPONSE TO THE GW/CEEE STATE DIRECTORS SURVEY

Fifty states and the District of Columbia responded to the survey. In 45 of the states (88%), the directors of assessment or a designated person in the assessment office completed the survey. Six of the states (12%) delegated completion of the survey to the state Title VII (bilingual education) director. New Hampshire provided partial data.

FINDINGS

The nature and extent of statewide assessment programs

Forty-eight states, including the District of Columbia, reported using 117 assessment programs (an average of about 2.3 assessment programs per state), covering a range of grades and subjects. Minnesota, Nebraska, and Wyoming were the only states that reported not conducting statewide assessment programs in 1993-1994. Iowa and Montana reported that they permit local district choice of assessment instruments, although student achievement data are compiled and reported by the state.

The extent to which LEP students are included in statewide assessment programs

Forty-five states indicated including some or all LEP students in at least one statewide assessment program, although only 15 (33%) reported the actual number of LEP students assessed.

Of the 48 states with statewide assessment programs, 44 reported allowing exemptions for LEP students on at least one statewide assessment. The most frequently reported criteria for exempting LEP students from statewide assessments in all states are shown in Table 1.

The extent to which individual state assessments are modified to include LEP students

As shown in Table 2, 27 states (52%) reported that they allowed test modifications for LEP students on at least one statewide assessment. Four states-Arizona, Hawaii, New Mexico, and New York-reported developing, translating or using commercially available instruments for LEP students in languages other than English. California and Texas reported piloting statewide assessment programs in a language other than English Note 2 .

The extent to which state assessments are used for accountability purposes

Seventeen states, about one-third, reported having a policy requiring students to pass a high school graduation test to receive a standard diploma. These states are Alabama, Florida, Georgia, Hawaii, Louisiana, Maryland, Mississippi, Nevada, New Jersey, New Mexico, New York, North Carolina, Ohio, South Carolina, Tennessee, Texas, and Virginia. Eleven of these states (65%) permit a range of modifications in statewide assessment instruments. Four of the states reported having test alternatives to the high school graduation test. New Mexico and New Jersey reported permitting the use of alternative assessments designed for non-native English speakers. Hawaii and New York reported allowing alternative tests (not specifically targeted to LEP students) for all students. Two states--New Mexico and New York--provided versions of the required high school graduation test in one or more non-English languages. Texas reported that it was developing a Spanish language version of the statewide assessment program for the lower grade levels; however, no plan is in place to develop a Spanish language version of the test required for high school graduation.

CONCLUSIONS

The findings provide insights into overall trends in state policies for including LEP students in state assessment programs and suggest directions for further research.

Overall, the findings suggest that states are struggling to develop appropriate policies to include LEP students in statewide assessment programs. The evidence indicates that states have not regularly included LEP students in most state assessment programs. Furthermore, policies for testing LEP students are sometimes inconsistent within and across states. For example, some states have policies to exempt LEP students from taking statewide assessments and at the same time have another policy that allows LEP students to take certain statewide assessments with modifications.

In addition, the findings suggest several areas for the development of policies and extended research. First, states need to refine policies for reporting LEP student data and to include them in state accountability reports. Second, since there is little documentation of their effectiveness, test modifications identified through the survey should be studied further. Implications for using test modifications with LEP students who possess varying levels of English language proficiency should be tracked and evaluated. Third, in states that offer certain assessments in languages other than English, further study is needed to ensure high technical quality of the translated tests (Stansfield, 1996). In addition, studies are needed to evaluate the best strategies for administering translated tests (e.g., test format). And, finally, for the seventeen states that require high school graduation tests, there is a need to study the validity of the assessments for LEP students and to identify strategies for allowing LEP students to participate in these assessments early on and in meaningful ways.

NOTES:

1. This research was carried out with funding from grant # T003H10002 from the U.S. Department of Education. The views expressed in this document are those of the authors and do not represent an official position of the Department of Education.

2. The initiative in California has since been abandoned.

REFERENCES

Bond, L., van der Ploeg, A., & Braskamp, D. (1994). Student assessment programs database. North Central Educational Laboratory and the Council of Chief State School Officers: Oak Brook, IL.

Donly, B., Henderson, A., & Strang, W. (1995). Summary of the bilingual education state educational agency program survey of states' limited English proficient persons and available educational resources 1993-94. Arlington, VA: Development Associates, Inc.

LaCelle-Peterson, M., & Rivera, C. (1994). "Is it real for all kids? A framework for equitable assessment policies for English language learners." Harvard Educational Review, 64 (1), 55-75.

Stansfield, C. (1996). Content assessment in the native language. ERIC Digest. Washington, DC: Clearinghouse on Assessment and Evaluation.