Introduction

 

An important issue in American education is the growing number of students who come from non-English speaking backgrounds. In California, for example, this number has more than doubled in the past ten years. These students account for almost 37% of California's total student enrollment in k-12 public schools (EdSource, 1998). This high population offers a compelling reason for educators to analyze the assessments with which English language learners (ELLs) are identified and subsequently classified. This literature review will describe current methods used in assessing the English proficiency of ELLs. In addition, the review will discuss practical implications of research on how to best identify and assess limited English proficient students.

 

Background

 

The issues of assessing bilingual students' English proficiency has spanned nearly three decades. The 1974 Supreme Court decision in Lau vs Nichols confirmed that schools must take "affirmative steps" to provide equal educational opportunities for students with limited English proficiency. Many school districts have interpreted the practical implication of this decision by providing some form of native language instruction for students--hence the growth of bilingual programs.

With the 1998 passage of proposition 227 in California, bilingual education has undergone massive restructuring in efforts to streamline students' transition to instruction in English only. Due to this change, the state has placed stringent accountability standards on school districts with bilingual programs (EdSource, 1998). This has set a precedent for many other states' bilingual programs and increased the need for classification of students' language proficiency through valid and reliable measures.

 

Defining Proficiency

Determining students' proficiency in English requires a thorough understanding of what this means. Hargett (1998) defines proficiency as being able to "effectively communicate or understand thoughts or ideas through the language's grammatical system and its vocabulary, using its sounds or written symbols" (p. xx). While this may seem simplistic, one can infer that how educators define proficiency will drive what language proficiency tests will measure. Obviously, definitions of language proficiency that are incomplete or unreasonable will distort educators' perceptions of limited English proficient students' abilities and progress.

Proficiency not only determines what instruction a student receives, but what instruments educators use to measure language progress over time. Since language acquisition is not a static event, proficiency is naturally best measured by assessments that can show incremental progress in English. In addition, instruments and procedures used to annually assess students do not have to be the same as those used for initial ELL identification. The instruments should, however, be similar in that they represent a range of difficulty based on age and grade level considerations (Hargett, 1998).

 

Current Practices

A variety of assessments are currently employed to determine students' eligibility for specialized instruction in English. These assessments include:

• home language surveys (HLS) which parents and guardians complete. These survey generally include questions about languages used at home;

• standardized achievement tests (SAT) which are used in every state. The cutoff score for identification of ELLs and exit criteria varies between the 23rd percentile to the 40th percentile (Canales, 1992);

• oral language proficiency tests (OLPT) which most often include the Language Assessment Scales, Oral (LAS-O), the Language Assessment Scales, Reading and Writing (LAS-R/W), and the Woodcock-Muñoz Language Survey.

The above assessments often are used in conjunction with each other to determine students' language ability. In many cases, however, only one of these assessments is used as the screening tool in placing students in instructional programs.

In addition, other criteria have traditionally been used to determine students' placement and English instructional needs. These criteria include parent recommendations, teacher referrals, and oral interviews. In the western United States, for example, 51.5 % of school districts use some type of language exam in their determination of student placement (Han, 1997).

Problems

Perhaps one of the greatest difficulties facing educators is the complete lack of national agreement on the criteria for designating students as ELLs. "Some districts attempt to find out the English skills of incoming students, while others simply place students in classes and wait until language problems appear, and some districts fall in between" (Schwartz, 1995, p. xx). This wide variation in district policies makes it very difficult for educators to provide language instruction based on assessment results from other schools.

Ancillary to the lack of national consensus, choosing an appropriate instrument to measure English proficiency is problematic since no one test will likely predict students' performance in a setting where English is the only language of instruction. Indeed, research on one of the most common proficiency test, the LAS-R/W, indicates that it can be "helpful" in determining English proficiency when combined with the test's oral component, however its users cannot be confident that the test determines students' language ability any better than teacher judgment or other instruments. This particular test's low reliability and inadequate validity contribute to this lack of user confidence (Carpenter, 1996).

In addition to the difficulty in selecting appropriate assessments, there is a general scarcity of tests designed to allow ELLs to demonstrate what they know and can do. Anstrom (1997) attests to the "lack of assessment instruments designed for their {ELLs}particular needs" (p. xx). Unfortunately, no helpful taxonomy exists for language proficiency assessments and the instruments themselves spill over across categories and imperfectly fit others.

Finally, school districts can often demand unrealistic language proficiency goals of ELLs through tests (Oakeley, 1998). By deciding proficiency standards for each grade level, assessments can fail to measure the actual progress students have made from one year to the next. Indeed, students themselves run the risk of becoming discouraged by what is perceived to be substandard performance on assessments. The design of a test that truly measures language ability while simultaneously considering the child's cognitive ability presents a definite challenge to educators (Nelson, 1997).

Additional Concerns

The literature also reveals several issues indirectly related to assessing ELLs. Among these issues it becomes evident that methods used in identifying gifted and talented students emerges as a particular challenge for educators. Since approximately 3% of the population is considered "gifted," it would stand to reason that this percentage holds true for limited English proficient students as well (Bermudez, 1993). In many districts, however, ELL classification precludes classification of students in other categories--significantly limiting gifted students' potential. The implications here clearly indicate the need for carefully designed staff development that targets the identification of such students in the classroom (Frasier, 1995).

Another concern arising in the literature is the obvious difficulty in measuring student progress in language through normed, standardized achievement tests. In California, an ELL student must begin to take the Stanford Nine test in English after twelve months in a California public school--regardless of his or her instructional program (California Department of Education, 1999). It is very evident that after one year of instruction, very few students will have attained the linguistic capacity required to perform well on such a test since language proficiency cannot be dictated by a specific time limitation.

Implications

Careful review of the current identification and assessment methods used with ELLs suggests the need for a more comprehensive method of assessment. While recent tests like the IMAGE in Illinois mark a shift in assessment design to a closer alignment with research in second language acquisition, these tests cannot tell the whole story (NCBE, 1997). A logical assumption would be that data from various sources need to be combined to accurately determine students' language proficiency. In particular, three areas could provide a more ample base of language assessment for students (Canales, 1992). These would include:

1. oral language proficiency data incorporating home language surveys, oral language proficiency tests, oral language interviews, and observation;

2. social data considering socio-economic status, schooling experiences, observations of social interactions at school and home;

3. academic data comprised of achievement test scores, classroom grades, and teacher observation.

The literature clearly expresses the need for assessments to evaluate all the information a school needs to know about an ELL. Essential information would include answers to questions about student capacity, reading and writing levels, appropriate placement, past progress in English, specific grammar and vocabulary problems, and native language ability.

 

Conclusion

Relying on the facts revealed throughout the literature, it can be assumed that assessing limited English proficient students in American schools will continue to be an important issue in the twenty-first century. It is evident that no one test can appropriately determine or predict student success or ideal matching of instruction to student needs. It is logical to infer, rather, that effective identifying and assessing ELLs requires a multi-faceted approach that synthesizes data from oral proficiency tests and observations, social statistics, and observations, as well as academic sources. This assumption leads to the hypothesis that limited English proficient students' performance on such a multi-faceted assessment would more accurately predict students' performance in an English language instructional setting. Ancillary to this, student results on the proposed assessment would probably bear a high concurrent validity to teacher recommendation for student placement.  

Anstrom, K. (1997, Fall) Assessing English Language Learners: A Look at Illinois. NCBE CrossCurrents, pp9-11.

Bermudez, A., & Rakow, S. (1993) Examining Identification and Instruction Practices for Gifted and Talented Limited English Proficient Students. Washington, D.C.: Proceedings of the National Association for Bilingual Education Conferences Annual Journal. (ERIC Document Reproduction Service No. ED 360 871).

California Department of Education. (1999). Questions and Answers about Testing Limited English Proficient Students, 1999. Retrieved September 27, 1999 from the World Wide Web: http://gse.slocs.k12.ca.us/star/star.htm

Canales, J. (1992) Innovative Practices in the Identification of LEP Students. Washington D.C.: Focus on Evaluation and Measurement: Symposium on Limited English Proficient Student Issues. (ERIC Document Reproduction Service No. ED 349 825).

Carpenter, C.D. (1995). Review of the Language Assessment Scales, Reading and Writing. In J. Conoley and J. Impara (Ed.), Twelfth Mental Measurements Yearbook (pp. 550-52). Lincoln, NE: Buros Institute.

EdSource (1998). Backgrounder: Bilingual Education in California. Palo Alto, CA: EdSource Inc. (ERIC Document Reproduction No. ED 423 669).

Frasier, M. (1995). Educators' Perceptions of Barriers to Identification of Gifted Children from Economically Disadvantaged and Limited English Proficient Backgrounds (Report No. RM-95216). Storrs, CT: National Research Center on the gifted and Talented. (ERIC Document Reproduction Service No. ED 402 707).

Han, M. (19xx). A Profile of Policies and Practices for Limited English Proficient Students: Screening Methods, Program Support, and Teacher Training. (Report No. ISBN-0-16-048976-8) Washington D.C.: National Center for Education Statistics. (ERIC Document Reproduction No. ED 403 781).

Hargett, G. (1998). Assessment in ESL and Bilingual Education: A Hot Topics Paper. Portland, OR: Northwest Regional Educational Lab. (ERIC Document Reproduction Service No. ED 425 645).

Nelson, J. (1997). Evaluating Elementary School LEP Language Skills: A Comprehensive Model. Orlando: Annual Meeting of the Teachers of English to Speakers of Other Languages. (ERIC Document Reproduction Service No. ED 409 727).

Oakeley, C.(1998). When Can LEP Students Exit a BE/ESL Program: Predicting Academic Growth Using a Test That Measures Cognitive Language Proficiency. San Diego: Paper presented at the Annual Meeting of the American Educational Research Association, April 13-17, 1998. (ERIC Document Reproduction Service No. ED 423 296).

Schwartz, W. (1995). A Guide to Assessing and Placing Language Minority Students. For Parents/about Parents. (Contract No. RR93002016). Washington D.C.: Office of Educational Research and Improvement. (ERIC Document Reproduction No. ED 396 016).