Immigrant high school students learning English score better on tests when given in their native language, even two years after the district labeled them proficient in English, according to a Miami-Dade County schools study.

The study suggests that standardized tests, which are given in English, do not accurately measure the knowledge of students who have spoken another language for most of their lives. Those exams -- especially the Florida Comprehensive Assessment Test -- can determine a student's advancement and graduation, leading the study's author to contend that recent immigrants may be at a serious disadvantage.

``These students have knowledge that they're not able to exhibit, said Rodolfo Abella, the district's supervisor of evaluation and research, who conducted the study of 2,025 Hispanic 4- and 10th-graders in Miami-Dade. All of them were in English for Speakers of Other Languages classes or had exited the program in the last two years.

Last year, there were 63,520 ESOL students districtwide, according to school records.

DIFFERENT TESTS

Students in the research were given two math tests that cover similar content but with different questions: the widely used Stanford 9 test in English and its Spanish-language cousin, the Aprenda 2, both produced by Harcourt.

In fourth grade, students who had recently completed two years of ESOL performed better on the English version than the Spanish one, scoring about 4 percent higher. But high school sophomores, who must pass the FCAT to graduate, scored 6 percent lower on the English version.

Examined another way, 88 percent of the 10th-graders who completed ESOL -- and were therefore classified as English-proficient -- scored higher on the Spanish version of the test.

Limited English Proficiency students ''for the most part, are not able to exhibit their mathematics skills on English-language achievement tests, regardless of the number of years they receive ESOL instruction,'' the study said.

LIKELY TO GROW

The study said the gap would likely grow on more language-intensive exams such as writing and reading.

Gov. Jeb Bush's office did not return calls seeking comment, and Department of Education officials said they needed to review the report before responding to it.

The study appears to be consistent with other research on bilingual education, saidStanford University professor of education Amado Padilla.

``It probably takes more than a couple of years to develop a second language in a high-proficient way to be successful and competitive academically, said Padilla, who had not reviewed the Miami-Dade study.

Even students who become functional with English usually work more slowly in their new language, Padilla said. That can be a double hurdle on timed standardized tests such as the FCAT, which will be given over the next six weeks.

AGE ADVANTAGE

The success of fourth-graders who finished ESOL is consistent with a wide range of research that suggests students learn new languages far more effectively when they are younger. The subjects on a fourth-grade test are also less abstract, Padilla said, and therefore easier to translate.

But even those students are not presumed to be entirely proficient in English, despite the school system's label.

''Once the children are exited from their ESOL program, they're still on a learning curve for the English language,'' said Deborah Stevens, principal of Hialeah's Palm Springs Elementary, where 28 percent of the students are classified as having limited English proficiency.

While students are in ESOL, their test scores do not count toward the school's accountability grade from the state. After that two-year program, however, they are automatically included.

FCAT FAILURES

''We know that after just two years of English instruction, they bomb horribly on the FCAT,'' said Santiago Corrada, principal of Miami Edison Senior High, which has a sizable Haitian population. Last year the school received its second F in four years. He suggested students be eligible to stay in ESOL classes for more than two years.

``We're going to deny high school students a diploma based on a language barrier, and it's going to affect entire ethnic groups, he said.

The study, funded with a federal Department of Education grant, will be unveiled Saturday before the National Association for Bilingual Education in New Orleans.

Download pdf study URL:

http://oer.dadeschools.net/nabe03paper.pdf

An Examination of the Validity of English Language Achievement Test Scores
in a LEP Student Population
Rodolfo Abella, Joanne Urrutia
and Aleksandr Shneyderman
Miami-Dade County Public Schools
February 1, 2003

Length of Time receiving English Language Instruction. It is generally accepted that language proficiency is a direct function of the extent of time students receive English instruction. In previous years, educators and policy makers have assumed that English language proficiency can be achieved in two to three years if students are exposed to proper language instruction. Nevertheless, research has brought this assumption into question. It is now thought that true language proficiency takes longer to achieve, as long as seven years, depending on whether oral proficiency or academic proficiency is being considered (Hakuta, Goto-Butler & Witt, 2000, Thompson & Collier, 1997).....

In summary, English language achievement tests are, for the most part, not a valid measure of content area knowledge for secondary students who recently exited ESOL and appear to discriminate most particularly against bright students and generally against those who have strong home language skills. At the 4 th grade level the English language achievement test seems to validly assess the performance of most of the recently exited students. The test seems to discriminate only against students with strong home language skills who have not fully developed their English language skills.