Introduction
Between 1990 and 1997 the number of U.S. residents who were not born in the United States increased by 30 percent, from 19.8 million to 25.8 million, to reach the largest total in the nation's history (U.S. Census Bureau, 1999:2). Many in this population may be fluent English speakers, and many born in the United States may not be; nevertheless these numbers illustrate a significant challenge facing our public schools. Schools around the country have been struggling to keep up with the responsibility for educating sometimes rapidly shifting populations of students whose command of spoken and written English—and previous academic preparation—vary widely.
This challenge is not a new one, of course, for a country that has been the destination of waves of immigrants from all over the world in the course of its history, but the specific responsibilities facing states and districts today are shaped by current perspectives and circumstances. As the nation enters the twenty-first century, few question its responsibility to provide schooling for all children through high school, though this was not always the case. But in recent decades changing immigration patterns have heightened tensions around the question of how best to make room for new students in systems that are sometimes already highly stressed, and how best to educate students who are not already proficient in English. Most states have witnessed sharp disagreements about the pros and cons of different educational strategies for English-language learners —students some-
where along the road from understanding no English to being completely proficient.1
At a time when educational testing is a factor with ever-increasing impact on students' lives and on the fates of schools, districts, teachers, and administrators, it is not surprising that questions about how and when to test English-language learners, and what to make of their test scores, have been some of the most vexing ones. In discussions of these issues, the Forum on Educational Excellence and Testing Equity identified five specific questions about considerations that affect the testing of English-language learners to explore at a workshop held in October 1999.
-
What is the best way to decide which English-language learners should be included in a given testing program?
-
What is the best way to decide which accommodations are appropriate for English-language learners who are taking a particular test? 2
-
How might we evaluate the effects of accommodations on the results of a particular test?
-
What should reports of test results convey about which students were included and about any accommodations that were provided? How might these reports vary depending on their intended recipients?
-
What factors need to be considered in planning tests for different purposes, such as those for use in making high-stakes decisions about individual students or those used for system accountability?
1 |
The term English-language learner is generally used in this report, following the practice adopted in a previous NRC report “Improving Schooling for Language-Minority Children.” While other terms are used in some contexts, the committee responsible for that report chose the term that emphasizes these students' learning rather than their limitations. The term limited English proficient (LEP) is used when the context requires it. That term is defined in federal guidelines as “national origin minority students who cannot speak, read, write, or comprehend English well enough to participate meaningfully in and benefit from the schools' regular education program.” (Office of Elementary and Secondary Education, 2000) |
2 |
Accommodations are changes “to the testing situation, (e.g.) presentation format, response format, setting, and the timing/scheduling of tests. [They are] a means of enabling [English-language learners] to demonstrate their academic knowledge despite their limited English proficiency.” (Rivera et al., 2000) A list of specific accommodations is provided on page 25. |
These five questions provided the genesis for a workshop that was an opportunity to explore some of the factors that complicate decisions about testing English-language learners, and to hear about both the experiences of several states that have grappled with them in different ways and the perspectives of test publishers. Researchers, state-and local-level policymakers and administrators, representatives from advocacy groups, and others convened to consider the questions the forum members had raised.
When the forum was converted in the spring of 2000 to a standing committee—with expanded powers under NRC guidelines—it took another look at the results of that workshop and identified some key messages from its findings. In the course of several meetings, the committee developed its thinking about the issues raised and identified recommendations it wanted to make to researchers, educators, policy makers, and test developers. Part One is the committee's report of its findings; it describes what the committee saw as major messages and uses them to frame recommendations. Part Two summarizes the key points from the workshop and is organized around what might be described as common-sense questions about the testing of these students.