• by Oanh H. Maroney, M.A. • IDRA Newsletter • February 1999

Bilingual education continues to be subject to attacks by some individuals, educators and non-educators. Of course, as with most types of programming – particularly initiatives that receive financial support from taxpayer-funded budgets – the stakes are high, and everyone wants to have a say. Often times, some of these voices speak loudly, with misguided, misinformed principles and purposes.

One such voice is that of Silicon Valley millionaire businessman Ron Unz. In June 1998, California voters passed Unz’s “English for the Children Initiative” by referendum vote. This initiative sought to outlaw bilingual instruction for limited-English-proficient (LEP) students and implement the use of sheltered English immersion for LEP and non-English speaking children up to age 10. After one year of intensive English instruction, LEP children would be placed into mainstream classes (Romero, 1998).

Others who are even somewhat familiar with the research on second language learning know that the sheltered English immersion approach – while it may work for a few students – is not the best way for the majority of students to learn English in our public schools.

Simply looking at the way Unz approached the basic premise of his initiative sheds light on how misinformed his efforts are:

“I just asked friends how long it took them to learn English.” As a child, Unz heard the stories of how his Yiddish-speaking mother, the daughter of Russian immigrants was steeped in English before entering kindergarten. He concluded that if she could do it, then why not immigrant children today? Unz disregards the extensive academic research on the topic because “it’s funded by pro- or anti-bilingual supporters” [emphasis added] (O’Brien, 1998).

While much could be said about this particular initiative (now law) and what is wrong with it, Ron Unz is not the only ill-informed voice out there. Many people see fault with bilingual education. And, yes, there are some problems that exist with bilingual education as it has been implemented in some schools. However, the solution is not to implement radical approaches and practices for which there is research evidence proving their inability to be effective as a common practice for teaching all children.

This article looks at some of the evidence for only a few of the many approaches available for use in teaching English language learners. While it is not intended to present every available perspective on the approaches discussed, it is intended to highlight a small fraction of a vast body of evidence and, more importantly, to illustrate the fact that it is not acceptable to pick and choose or to simply disregard existing evidence when proposing to implement change. This is especially true concerning bilingual education because, in this case, change impacts millions of school children – for the rest of their lives.

Dr. José A. Cárdenas, IDRA founder and director emeritus, notes three important facts about research as a concept:

  • research is the scientific determination of that which common sense tells us;
  • research methodology is easy to manipulate, therefore one can readily prove or disprove any assumption; [and]
  • no research is many times better than poor research. It is much preferable to make educational decisions in the absence of information than to make them on the basis of erroneous information (1995).

The debate about bilingual education is decades old and still raging. One factor that continues to fuel the debate is related to Cárdenas’ observation that research methodology is easily manipulable. Opponents and proponents alike have performed research studies and identified research findings that support their point of view. Because of this, it is important to discern which evidence is valid and credible and which is not.

The body of research that exists on second language learning (and countless other topics) serves as a collection of lessons learned about what works and what does not work. It is an active collection of formulas and anecdotes that often, at minimum, point us in the right direction in terms of finding the most plausible solution for the challenge at hand.

Our schools are laboratories in which research is conducted every day on millions of school-age children. The differences that students bring with them to school – including cultural and linguistic differences – and how schools approach those differences are critical factors in shaping the schooling experience.

Language is inextricably linked to culture The striping away of students’ native language and culture is usually done for what teachers and schools believe are good reasons. Schools often make a direct link between students’ English assimilation and their economic and social mobility. As a consequence, students who speak a language other than English are frequently viewed as “handicapped” and urged, through both subtle and direct means, to abandon their native language (Nieto, 1996).

Because each child is unique and different, each teacher must find the most effective way to help every child to learn. This is the challenge that teachers must meet for every subject they are required to teach, including helping LEP children to learn the English language.

Bilingual education opponents argue that research evidence on bilingual education is inconclusive, inconsistent and contradictory. However, this is not the case. Crawford notes:

The most sophisticated evaluation study to date [by Ramirez, Yeun and Ramey, 1991] – a four-year, longitudinal study of 2,000 Spanish-speaking students in five states – found that “late-exit,” developmental bilingual programs proved superior to “early-exit,” transitional bilingual and English-only immersion programs…in programs that stressed native-language skills, students’ growth in English reading and mathematics continued to increase long after it had leveled off among their peers and in the other programs (1998).

For many, it is inconceivable to support an educational program that requires students to be given native language instruction support in school for several years. However, much of the research has documented that students generally need a minimum of five to seven years to develop the level of English proficiency needed to succeed academically in school:

With most programs permitting students to remain a maximum of only three to four years, only partially positive results can be expected. The research evidence is in direct contrast to program implementation; nevertheless, many programs are successful (Nieto, 1996).

One of the major issues that affects bilingual education being recognized as an effective method for helping students acquire English is the somewhat loose definition and implementation of the program concept. There are many programs that are called “bilingual education” that aren’t really bilingual education programs or that may not incorporate all of the principles of a good and effective bilingual education program model. This is an important consideration when analyzing a bilingual education program for its effectiveness. In most instances, to take any given bilingual education program and perform a simple comparison of its outcomes to another bilingual education program or to an immersion program would be like comparing apples and oranges.

Nieto notes that reviews of research in first and second language acquisition show that students’ native language proficiency level directly influences the development of their second language proficiency (1996). IDRA is concerned about ensuring that children who speak a language other than English benefit from quality instructional programs that capitalize on students’ language and culture. Based on research and appropriate pedagogy we know that the most effective way to teach English to children who speak another language is through an adequate bilingual education program (1996).

Cárdenas says that three characteristics of a good bilingual education program are that it:

  • allows students’ learning to continue to occur and cognitive stimulation, intellectual stimulation takes place without having to interrupt the cognitive development or the intellectual development while the kid sits in a classroom for any extended period of time while acquiring a new language system;
  • facilitates students’ acquisition of a new language system; and
  • further extends students’ original language system (1994).

Simply grouping limited-English-proficient students together in a classroom to learn English and allowing them to sometimes speak their native language does not constitute a bilingual education program.

According to IDRA, “a good bilingual education program…enhances the learning of English and subject matter. Bilingual education teaches English to children and gives them a chance to practice it while they also learn subjects like math and science” (1996).

Stephen Krashen notes that “the principles underlying successful bilingual education are the same principles that underlie successful language acquisition in general” (1993). These principles include the following:

  • We acquire a second language by understanding messages, by obtaining comprehensible input.
  • Background knowledge [i.e., the first language] can help make second language input more comprehensible, and can thus assist in the acquisition of the second language.
  • The development of literacy occurs in the same way as second language acquisition does (Krashen, 1993).

Bilingual education involves more than simply translating words from one language to another or giving students a “sink or swim” course in the English language (Maroney, 1998).

In its simplest form, bilingual education is the use of two languages. In a more sophisticated context, such as an educational response to the problems of limited-English-proficient students in American schools, bilingual education is the use of the native language for instructional purposes while English is being learned as a second language. Interrupting and delaying cognitive development and the acquisition of skills and knowledge until a new language is acquired is a waste of time and produces academic retardation and overagedness, which in turn produces underperformance and a propensity for dropping out of school (Cárdenas, 1994).

The end goal of bilingual education is to produce bilingual and biliterate students, who will in turn become productive citizens.

In spite of the enormous impact that language has on children’s schooling, lack of English skills alone cannot explain the poor academic achievement of students. It is tempting to fall back on this explanation and thus count on simple solutions like English “sink or swim” programs to solve the problem (Nieto, 1996).

In examining some of the research evidence used by opponents of bilingual education, researchers have cited poor methodology as an explanation for the resulting “proof” that bilingual education programs do not work. For instance, the 1996 study by Rossell and Baker is often cited by bilingual education critics as evidence against bilingual education programs. The authors of this study “concluded that there was no evidence that bilingual programs were superior to English-only options for limited-English-proficient children” (Krashen, 1998).

Critiques by Cummins and Krashen show that the Rossell and Baker study is flawed in many ways, including the fact that the researchers strongly rely on studies of French immersion in Canada as evidence (Crawford, 1998).

There are several approaches that fall under the general umbrella of immersion programs. There is Canadian-style immersion (CSI), which is frequently offered as an alternative approach to bilingual education for limited-English-proficient students in our schools. CSI has been used quite successfully in Canada with middle-class students who receive much of their content area subject matter in the second language, with efforts being made to ensure comprehensibility (Krashen, 1993).

Sheltered subject matter teaching or content-based language teaching is an approach that provides students opportunities to learn content area subject matter in the second language, provides them with ESL methodology, and also takes advantage of the background knowledge and literacy provided by the first language (Krashen, 1993).

Third, there is structured immersion, which has four primary characteristics: comprehensible subject matter teaching in the second language; minimum use of the first language when needed for explanation; direct instruction of grammar; and pre-teaching of vocabulary. Krashen notes that there is little evidence which supports the effectiveness of direct grammar instruction or pre-teaching vocabulary (1993).

A fourth immersion approach is submersion. According to H. Douglas Brown, this way of treating second language learners is “really no treatment at all”; it “submerges” students in English-speaking content-area classes with no special foreign language instruction, assuming the students will “absorb” English through content-area learning (Brown, 1994).

The popular arguments for English-only immersion rest on the weak theoretical assumption that there is a direct relationship between the amount of exposure to English and academic learning in English (Cummins, 1988). The aim of immersion programs, as they are proposed for students in U.S. public schools, is to teach children English in a brief period of time, with no regard for developing and maintaining the first language. However, the well-known Canadian-style immersion program, while seen as an alternative to bilingual education, is, in fact, a bilingual education program (Krashen, 1993). CSI promotes and fosters bilingualism, rather than the replacement of the first language.

One key factor that must be considered is the population of students for whom immersion has proven to be an effective alternative. In Canada, immersion has been successful with middle-class students whose first language is the highly valued majority language. Contrary to the language learners in the Canadian-style immersion program, most of the bilingual education programs in our country serve many children from economically oppressed communities whose native languages are not highly valued by our society.

Nieto observes that if we wished to implement the CSI model in the United States as it has been successfully implemented in Canada, it would require that we place native English-speaking students in another language setting, rather than placing limited-English-proficient students in English-speaking classrooms. “Even the researchers associated with Canada’s immersion program have cautioned against using it as a model with language-minority students in the United States” (Nieto, 1996).

While immersion programs may have proven to be successful in other countries and in working with adults in the United States, attempting to apply those same concepts and practices to children in U.S. public schools doesn’t necessarily work because the change of context changes the effectiveness of the program: ”the immersion model, if applied to language-minority students, would properly be called submersion” (Nieto, 1996).

According to Krashen, “submersion is rejected by all professionals as an option for limited-English-proficient children” (1993). Martin Carnoy, a professor of education at Stanford University, notes

It is a fact that European kids with highly educated parents who are immediately immersed in English are able to learn to break the code. And so if it were those kinds of children whom we were dealing with primarily, there would be no problem with this initiative [Proposition 227]. The real problem is that those aren’t the majority of children (Stanford, 1998).

Because of the number of limited-English-proficient students attending our nation’s public schools, it is critical that policies and practices be adopted that will benefit the greater population of students by increasing their opportunities to learn.

Webster defines research as close, careful study (1996). Such has been the case with second language learning; a vast body of evidence exists regarding practices and approaches to use that are most successful in helping learners acquire a second language. Many individuals have and will continue to encourage others to disregard what some of the best research in the field says. It may sound a lot easier and less expensive to provide students with intensive English instruction for a year or so and then mainstream them into English-speaking classrooms, but as Dr. Nancy Zelasko notes, “Learning a language is a difficult task which takes time. Even native English-speakers study English for 12 years of elementary and secondary school” (1998).

What is difficult for many to understand is that there is still a lot of stigma attached to limited-English-proficient students, especially students from particular ethnic minority groups.

Research shows that bilingual education is the best way to teach children English. Not only does it foster cognitive and linguistic growth in their native language, but bilingual education also fosters cognitive and linguistic growth in English. At the same time, bilingual education provides students with a sense that their native language and culture are valued in the school setting. Valuing students while providing them with proper and adequate resources in the learning environment will foster success better than any quick-fix remedy can.

It is rather ironic that, in a country that emphasizes the value of bilingualism during the secondary and post-secondary schooling years, many are emphatic about forcing school children to immediately acquire English, even at the expense of their native language. There is absolutely no question about the fact that mastering the English language is important for students to be able to participate fully and successfully as citizens in the United States. However, it is not good educational practice to implement a particular approach on a large scale simply because it has been proven to work in a particular context or for a particular few individuals. If this nation continues to implement educational policies and practices using a “bandwagon” approach, we will continue to deny opportunity to millions of children who will miss the boat.


Brown, H.D. Teaching by Principles: An Interactive Approach to Language Pedagogy (Englewood Cliffs, New Jersey: Prentice Hall Regents, 1994).

Cárdenas, J.A. Multicultural Education: A Generation of Advocacy (Needham Heights, Mass.: Simon and Schuster Custom Publishing, 1995).

Cárdenas, J.A. “Bilingual Onions” in Compendium of Readings in Bilingual Education: Issues and Practices, Rodolfo Rodríguez, Nancy J. Ramos and José Augustín Ruiz-Escalante (Eds.) (San Antonio, Texas: Texas Association for Bilingual Education, 1994) pp. 301-307.

Crawford, J. Language Policy Web Site, http://ourworld.compuserve.com/homepages/jwcrawford (1998).

Cummins, J. “Second Language Acquisition Within Bilingual Education Programs” in Issues in Second Language Acquisition: Multiple Perspectives, Leslie M. Beebe (Ed.) (New York: Newbury House Publishers, 1988) pp. 145-166.

Intercultural Development Research Association. “America needs bilingual education to produce educated, well informed citizens,” Class Notes (San Antonio, Texas: Intercultural Development Research Association, 1996) Issue number 1.

Krashen, S.D. “Bilingual Education Works” in Methods That Work: Ideas for Literacy and Language Teachers, Second Edition, John W. Oller, Jr. (Ed.) (Boston, Mass.: Heinle and Heinle Publishers, 1993) pp. 89-98.

Krashen, S.D. “A Note on Greene’s ‘A Meta-Analysis of the Effectiveness of Bilingual Education,’” Internet posting (March 1998).

Maroney, O.H. “Who is Teaching the Children? More Trained Bilingual Teachers Are Needed for Excellent Education,” IDRA Newsletter Notes (San Antonio, Texas: Intercultural Development Research Association, January 1998).

Nieto, S. Affirming Diversity: The Sociopolitical Context of Multicultural Education, Second Edition (White Plains, New York: Longman Publishers USA, 1996).

O’Brien, T. “Seizing The Initiative,” Stanford (May-June 1998) pp. 86-89.

Stanford. “The Bilingual Battlefield,” Stanford (May-June 1998) pp. 90-91

Romero, A. “Bilingual Education Under Attack: Misconceptions Fuel The Fire,” IDRA Newsletter (San Antonio, Texas: Intercultural Development Research Association, January 1998).

Webster’s II New Riverside Dictionary. Revised Edition (Houghton Mifflin Company, 1996).

Zelasko, N. “The Benefits of Bilingualism,” Internet Posting (1998).

Oanh H. Maroney, M.A., is a research assistant and administrative assistant to the IDRA executive director. Comments and questions may be directed to her via e-mail at feedback@idra.org.

[©1999, IDRA. This article originally appeared in the February 1999 IDRA Newsletter by the Intercultural Development Research Association. Permission to reproduce this article is granted provided the article is reprinted in its entirety and proper credit is given to IDRA and the author.]