When a conservative think tank released a book last year that criticized bilingual education, the group set out to advertise it in the newsletter of the organization best known for promoting such programs, the National Association for Bilingual Education.
After submitting the ad, however, the Pioneer Institute for Public Policy Research received a refund of its $237.50 and a letter saying that the educators’ group wouldn’t print it.
The Boston-based institute trumpeted the refund in promotional materials it sent out with the book, Bilingual Education in Massachusetts: The Emperor Has No Clothes. NABE’s rejection, the institute claimed, was “an indicator of the study’s importance.”
Nancy F. Zelasko, the deputy director of the Washington-based NABE, countered that she turned the ad down because the research was questionable.
“There are organizations out there like Pioneer whose mission it is to prove that bilingual education doesn’t work,” she said. “It bothers me, but at the same time I know there is much more objective and comprehensive research out there.”
The squabble last year over the Pioneer Institute book illustrates the degree to which politics surround the issue of bilingual education. All too often, political and emotional concerns undercut the ability of research to guide educators as they struggle to teach the growing number of language-minority students.
“The educational agenda is taking a back seat to the political agenda,” said Arthur Levine, the president of Teachers College, Columbia University. “We’re fighting an ideological war, and the school is becoming the vehicle through which we decide on the resolution of social or political differences.”
He and other leading educators say those battles have hindered efforts to recruit top-notch researchers on bilingual education, flooded the field with questionable studies, and weakened the credibility of the legitimate research that is produced.
“Bilingual education illustrates the problems of education research in general, but in an extreme form,” said Carl Kaestle, a professor of education at the University of Chicago and the president of the National Academy of Education. “The problems that are worse in bilingual education than education research in general are the politicization and mistrust.”
The battle in recent years has largely focused on whether--and the extent to which--schools should teach students in their native languages or in English. Many view that issue as a part of a larger debate over two competing visions of America--a diverse society of many cultures or a melting pot where those cultures merge.
Each side in the debate typically charges the other with having ulterior motives. Those who advocate a more English-intensive approach are accused of being xenophobic foot soldiers in the English-only movement. Those who promote a bilingual method are branded as self-serving protectors of an education bureaucracy that amounts to an ethnic jobs program.
Even the questions posed in research on bilingual education are often motivated more by politics than pedagogy, Mr. Levine said.
During the 1980s, the primary questions were: Does bilingual education work? And does it work better than other approaches?
Those questions, a National Research Council report concluded in January, are hopelessly oversimplified and virtually impossible to answer in the black-and-white terms in which they are cast. The NRC is the research arm of the Washington-based National Academy of Sciences. (“Politics Found To Hinder Research on LEP Students,” Jan. 22, 1997.)
Research now must focus on slices of those questions or on different questions, many experts say: What components of the various approaches work best? Under what conditions? And for which children? What are the next-best alternatives if those conditions can’t be met?
About 3.2 million limited-English-proficient children attend U.S. public schools, and the number is rising. Educators and researchers say it is more crucial than ever that the field produce credible and relevant research to guide what happens in classrooms.
“Philosophically, we have a belief” about what works, said Tony Vigil, the director of bilingual and English-as-a-second-language programs for the Denver public schools. “Yet we’re not researchers, so we depend on others to support us.”
And it is no longer just the large urban districts like Denver that must effectively educate LEP students. Already, many suburban and rural districts are wrestling with the issue. And in the years to come, experts say, few districts will be left untouched by the country’s changing demographics.
The barriers that impede “good” research on bilingual education are many, researchers say. Among the most often cited are that:
- It is a culturally charged and emotional issue. Bilingual education in the eyes of some advocates is a civil right. Thus, those who question it run the risk of being cast as racist or anti-Hispanic.
- Those who conduct or pay for the research often have a preconceived point to make.
- Money for academic research is often scarce.
- Program labels can be misleading. The variation of approaches clumped under the label “bilingual education” is enormous.
That makes it extremely difficult to accurately compare efforts to educate language-minority students in different states, school districts, individual schools, or even classrooms within schools.
Teachers in what are labeled “bilingual” classes may actually use very little of their students’ native languages, or none at all.
Most bilingual programs today are considered transitional, with the goal of preparing students to enter mainstream English classrooms as soon as possible. In such programs, children study English in classes specially designed for second-language learners and receive a portion of their instruction in their native languages to keep up in academic subjects.
- Instruction varies tremendously. Though some educators are fully certified in bilingual education, there are far too few of them to meet the growing national demand. Often, students are taught by aides or non-certified educators. Some teachers are truly bilingual, some don’t speak English well, and some don’t speak the students’ language well or at all.
- The rules on the subject vary widely. Some states, such as Massachusetts, generally require schools to use transitional bilingual education, at least in name. Others permit more local choice. Therefore, some schools offer bilingual education because they have to, others because they want to.
- The criteria used to classify students as limited in English proficiency vary from state to state.
- It is difficult to make general statements about LEP students. Some are immigrants, others are U.S. born. Some arrive at school with little formal schooling. Many--but not all--are poor. Approaches that prove effective with Mexican-American students in Los Angeles may not work with Chinese immigrants in New York City.
- There is no common definition of what a “successful” bilingual education program is. Some argue the measure should be how quickly students enter the mainstream. Others would measure how well students keep up academically while learning English, regardless of how long that takes. Still others say the test should be whether a program produces a fully bilingual person capable of writing, reading, and speaking in two languages.
- It’s difficult to say what’s working when the experts themselves aren’t sure of the best ways to assess LEP students. For example, if you aim to test an LEP student’s math skills with a test written in English, are you really measuring his math skills or his English?
- LEP children tend to be highly mobile, so it’s hard to track the same group over time to produce good longitudinal studies.
Though difficult, some experts say, those barriers are not insurmountable, especially if researchers are willing to ask the right questions.
“If the research is simply: Does bilingual education work? Then yes, these barriers are huge problems,” said Delia Pompa, the director of the U.S. Department of Education’s office of bilingual education and minority-languages affairs. But if the research takes on smaller pieces of that question, she added, the barriers are less imposing.
Research on bilingual education has suffered a rather troubled history. In the 1980s, the Education Department undertook a few multimillion-dollar evaluations of programs for LEP students that were intended to provide a definitive answer on what model worked best.
Those studies made headlines in later years, but were so riddled with methodological problems that they failed to show conclusively which programs were most effective, this year’s report from the National Research Council concluded.
With so many resources poured into costly large-scale studies, little was left for other basic or applied research. The focus, say researchers like Donna Christian of the Center for Applied Linguistics in Washington, has been skewed by “horse race” studies that pit one program against another.
By zeroing in on how much of the native language is used in teaching and how fast students exit programs, issues such as how well LEP students keep up academically over the long term are often cast aside, many experts say. Each side in the debate can point to studies over the years that bolster its own views of the effectiveness of bilingual education.
“Because advocacy is the goal, very poor studies that support an advocated position are touted as definitive,” the NRC panel reported.
Added Boston University political scientist Christine H. Rossell: “The critics of bilingual education make as many mistakes in interpreting data as the advocates.” Ms. Rossell was the chief author of the Pioneer Institute book on bilingual education, which asserts that there is no proof that bilingual education works any better than other approaches to teaching language-minority students in Massachusetts.
Bilingual education advocates hailed a 1987 report from the U.S. General Accounting Office that evaluated bilingual education programs. The GAO surveyed 10 experts in the field, most of whom supported native-language instruction. The study was not scientific, the NRC panel noted in its report, because “if different experts had been chosen, different conclusions would have been drawn.”
In 1994, the New York City school board released what it billed as a preliminary report on LEP students in bilingual education programs and compared them with children taught only in ESL classes. The study concluded that students in the ESL classes--who spend most of their time learning in English--fare better in the short run than their peers in bilingual programs.
But the authors did not control for socioeconomic status or the fact that most students in bilingual programs were Hispanic while the ESL students came from a host of ethnic groups. Supporters of bilingual education were quick to point out those shortcomings while advocates of other approaches sought to minimize them. (“N.Y.C. Bilingual-Ed. Report Spurs Questions and Complaints,” Nov. 2, 1994.)
The situation is made worse by the lack of a solid infrastructure to support good research. The most notable deficiencies are poor coordination among researchers and within federal and state agencies, and little money to entice premier researchers to the field, pay for research, and support dissemination of results.
The problem is one of quality, not quantity of studies, said Mr. Levine of Teachers College.
“There are a number of think tanks with hired guns who are out there to trash or laud bilingual education,” he said. “These hired guns are far more savvy about dissemination than academics. They make sure their [work] gets to the press, public, and policymakers.”
State education agencies, even those in states with big LEP populations, don’t tend to do a lot of research or evaluation on their own. Nor do the private foundations that support education projects. As a result, research funding is often scarce.
At the federal level, Congress for the past two years has refused to provide money that would have gone to the Education Department’s office of bilingual education and minority-languages affairs, or OBEMLA, for research, evaluation, and fellowships designed to draw new blood to the field.
Historically, the NRC report concluded, OBEMLA often has been the sole voice within the department plugging research on language-minority students. But the office itself has too few people who are well-versed in how to do good research, some researchers say. Like the research on bilingual education in general, the office has suffered a credibility crisis.
“The work that came across my desk was abysmal,” recalled Edward J. Fuentes, who served as OBEMLA’s research director during the later years of the Reagan administration--which was widely viewed as being openly hostile to bilingual education.
The situation since then has improved substantially, said Mr. Fuentes, who currently directs the federally supported National Institute on the Education of At-Risk Students.
Even so, Mr. Levine said, there is so much heat surrounding bilingual education that policymakers and the public may be tempted to throw up their hands in frustration while “truly important research is clouded over.”
For educators like Denver’s Tony Vigil, the task of seeing through those clouds is far from simple.
“We have research that tells us that bilingual education works better, then research that tells us that bilingual is not even better than nothing,” Mr. Vigil said. “Our school board members get confused and then they say, ‘Well, why not just use English then if doing the other thing is going to cost us more?’”
Nearly 20 percent of Denver’s 64,000 students speak a language other than English, overwhelmingly Spanish. Concerned that too few students are shifting successfully into all-English classes--and that the hiring pool holds too few truly bilingual teachers--the district is looking to retool its programs.
“We need some research on those variables that tell us whether a student is ready for the mainstream or not,” Mr. Vigil said. “The research we’re looking for hasn’t been out there.”
For Waldemar Rojas, the superintendent of the San Francisco public schools, the issues are a bit different. Nearly a third of the district’s 64,000 students are considered LEP. Only a third of them, however, speak Spanish.
Mr. Rojas said he needs more research on the students who make up the remaining two-thirds of the language-minority school population, predominantly Chinese speakers. “There’s just no question that the research addresses Spanish better,” he said.
Research on appropriate assessment, he added, is vital. “These tools have to become available. And they don’t exist. ... How can you claim success if you’ve never assessed [the students]?”
In New York City, the nation’s largest school system, LEP children make up about 15 percent of the district’s more than 1 million students. But the LEP label is deceptively simple, said John Acompore, the deputy director of the system’s bilingual education office.
He is particularly concerned about the new students who show up at the city’s elementary and high schools not only with no grasp of English, but also with little formal schooling in any language.
“I just don’t see a lot of research on this group,” Mr. Acompore said.
Kenji Hakuta, a professor of education at Stanford University who has written extensively about bilingual education and second-language learning, agreed that there are many questions from schools that research could--and should--address.
“There’s a lot of improvement needed in what we know,” said Mr. Hakuta, who also served as the chairman of the NRC panel on research and LEP students. “But it’s not like we don’t know anything.”
Studies that most closely approximate “true experiments” indicate that programs that use a student’s native language will produce outcomes slightly better than English-only approaches, Mr. Hakuta said. And, he noted, though modern-day bilingual education is a relatively new field, related research in linguistics, cognitive development, and literacy abounds.
The Center for Research in Education on Diversity and Excellence, a federally funded consortium of more than a dozen colleges and universities based at the University of California, Santa Cruz, is tackling many of the questions educators have and the questions researchers think ought to be pursued:
What program components are most effective with LEP students? Under what local circumstances? What practices are most helpful for different kinds of LEP students? What research is needed to develop accurate assessments for these students?
Making It Better
Given the methodological and design problems of past studies, the NRC panel and other researchers are calling for research to find better ways to evaluate programs. They also urge investigation of new program models such as two-way bilingual education, where speakers of two languages are taught in the same classroom, learn each other’s language, and work in the two languages.
Research must also look at special groups of LEP students: older students, those who speak languages other than Spanish, and LEP students with disabilities--groups that current research often views only peripherally.
To attract top-flight researchers and produce high-quality work in the field, research resources must improve, said Eugene E. Garcia, the dean of the school of education at the University of California, Berkeley, and a former director of OBEMLA. For example, although longitudinal studies in bilingual education are needed, foundations and others are often loath to pay for such studies because they are difficult and costly, he said.
“We have to look not just at the present effects, but at long-term effects” of programs, Mr. Garcia said. “But nobody’s willing to put their dollars on that kind of long-term study.”
And though experts like Mr. Fuentes say the field has matured substantially in recent years, there is still a need to improve and bolster the researcher ranks.
“Many of the people involved in bilingual education were themselves limited-English-proficient at one time in their lives. They have a real emotional stake in it,” Mr. Fuentes said. That in and of itself is not necessarily a problem, he said. “But we also need a cold eye on what works for kids if we want to move the field along.”
‘A Talent Problem’
One way to broaden the research pool is to encourage those in related fields--such as psychology, linguistics, and sociology--to study LEP students and bilingual education, Mr. Hakuta said.
“We’re always having a talent problem,” he said. “And within education, bilingual education is seen as probably more marginalized from the mainstream, so it is harder to get a big pool of good researchers.”
If research in the field is ever to improve, experts say, that marginalization of LEP students and the issues they face must stop.
Rather than being seen as a niche, language-minority students must become an integral part of education policy discussions at all levels--from schools of education and teacher training programs to private foundations and state and federal education agencies, said Roland G. Tharp, the director of the Santa Cruz research center.
“We need to change that and move this research into the big journals and the big rooms at places like the American Educational Research Association,” Mr. Tharp said. “The level of discourse has to be raised above political shouting.”
A version of this article appeared in the March 05, 1997 edition of Education Week as The Politics of Language