TWO-WAY (DUAL) IMMERSION

Two-way (dual) immersion programs first started at the Coral Way Elementary School in Miami in 1963, when political refugees fleeing Fidel Castro’s Cuba created a program to educate their children in both Spanish and English (Baker, 2006). The Cubans were motivated to teach their children Spanish because many of them believed that the Castro regime would not last and they would be able to return to Cuba. The bilingual program at Coral Way Elementary was special in that it was open to both Spanish-speaking Cuban children and local English-speaking children with the goal of bilingualism for both groups. Indeed, the program had very good results. Crawford (2004) explains that in English reading, both the Spanish-speaking and Englishspeaking groups did as well as or better than their counterparts in monolingual English schools, and the Cuban children achieved equivalent levels in Spanish. Although the English-speaking group did not quite reach national norms in Spanish reading achievement, the program was largely deemed successful and became a model for other school districts.

Since the establishment of the Coral Way program, the number of two-way immersion programs has grown considerably in the U.S. There are now 400 dual immersion programs in 30 states and Washington DC (Center for Applied Linguistics, 2011). Three hundred and seventy-five of them (94%) pair Spanish with English while the remaining programs have French, Chinese, Japanese, and Korean as the target languages. Two-way immersion typically involves about equal numbers of language majority and language minority children learning both languages and academic content in the same classroom. Each group serves as native language models for the other group, and students learn each other’s languages through social interaction. According to Christian (1994), a typical two-way immersion program has the following goals:

• • Language development. Students will develop high levels of proficiency in their first language and in a second language.
• • Academic development. Students will perform at or above grade level in academic areas in both languages.
• • Social development. Students will demonstrate positive cross-cultural attitudes and behaviors and high levels of self-esteem.

In terms of amount of instructional time spent in each language, most twoway immersion programs employ either a 50/50 model or a 90/10 model. In the 50/50 model, 50% of the instruction is in one language and the other 50% is in the other language. This ratio is maintained more or less throughout all grades. In the 90/10 model, the minority language is given much more weight toward the beginning of the program in kindergarten and first grade—about 90% of instruction is provided in the minority language and 10% in English. Instruction in English gradually increases through the grades until the program becomes about 50/50. Research shows that students in 90/10 programs tend to become more fully bilingual than those in 50/50 programs (Lindholm-Leary, 2001). In terms of which language is used when, some programs may use one language for instruction in the morning and the other language in the afternoon, while other programs may use the languages in alternating days or weeks. Some programs may teach certain subjects (e.g., math, science) in one language and other subjects (e.g., social studies, art, music) in the other language.

Two-way immersion offers a number of benefits. First, by having both language majority and language minority children learn side by side in the same classroom, it is more inclusive than traditional bilingual education programs serving only minority language speakers. Crawford (2004: 289) states that “a program model open to all children could hardly be branded a ‘special interest’ subsidy” and that English-speaking parents, who are “more accustomed to activism and more likely to influence politicians, [could] become effective ambassadors for bilingual education.”

For English-speaking students who would not otherwise have significant exposure to another language, two-way immersion gives them an immersion experience. An important difference between two-way immersion and one-way immersion (e.g., Canadian French immersion) is that English speakers in twoway immersion have the additional benefit of interacting with classmates who are native speakers of the target language. For language minority children, twoway immersion provides them with an educational environment where their native language is valued as a resource. This is in sharp contrast to English-only programs where it is often treated as a problem. In addition, by promoting the development of both basic and advanced literacy in the native language and in English, dual immersion allows language minority students to gain important content knowledge that in turn will make English more comprehensible (Christian, 1994).

Fred Genesee, as cited in Crawford (2004: 290), states that

including students from both language groups creates a learning environment that can be truly bilingual and bicultural. Sustained contact with members of the target language group of the same age as the learners may be necessary if students are to develop fundamentally more tolerant and positive attitudes toward each other.

The society ultimately benefits from having bilingual citizens who are positively disposed toward people of other cultural backgrounds and who can meet national needs for language competence (Lindholm-Leary, 2001).

A growing body of research shows that two-way immersion is effective in meeting the language and academic needs of both language majority and language minority students (e.g., Genesee et al., 2006; Lindholm-Leary, 2001; Lindholm-Leary & Block, 2010; Oller & Eilers, 2002; Thomas & Collier, 2002). Lindholm-Leary (2001) conducted one of the most comprehensive evaluations of two-way programs. Analyzing data from 18 schools, she analyzed student outcomes in different program types including both 90/10 and 50/50 dual immersion models, transitional bilingual education, and mainstream English- only programs. She examined students’ proficiencies in English and Spanish, academic achievement, and attitudes.

In terms of L1 development, Lindholm-Leary (2001) found that both English-dominant students and Spanish-dominant students in two-way immersion programs scored high in oral and academic skills in their respective native languages. In terms of L2 development, she found that English speakers in 90/10 programs did better in oral Spanish than English speakers in 50/50 programs. But English speakers failed to acquire the levels of Spanish proficiency that the native Spanish speakers did. This may be due to the fact that the main source of Spanish input for these children was in school whereas the

Spanish speakers had Spanish input at home as well. For Spanish-speaking students, no difference in English language proficiency was found between the 90/10 and 50/50 programs, leading to the conclusion that more English instruction does not necessarily lead to better English proficiency. Compared to Spanish speakers in transitional bilingual education and English-only programs, Spanish speakers in dual immersion did better in English despite having received less instruction in English. Dual immersion students also did better in math than English-speaking students in English-only programs. Furthermore, two-way immersion students had very positive attitudes toward the programs, teachers, the classroom environment, and other cultures.

A typical two-way immersion program is composed of two populations: (1) Latino, low-income, native Spanish speakers, and (2) White, middle-class, native English speakers (Howard & Sugarman, 2001). However, where dual language programs are implemented in schools that are predominantly Hispanic, English-speaking Hispanics, often also from lower income backgrounds, constitute the “English-speaking group” (Lindholm-Leary & Block, 2010). Lindholm-Leary & Block (2010) asked the question of whether the positive outcomes of dual language programs found with the middle-class White English-speaking populations can be generalized to dual language programs located in low-income, predominantly Hispanic schools. Their analysis of students’ standardized test scores showed that the answer is yes—Hispanic students participating in dual language programs in predominantly Hispanic/ low-income schools achieve at similar or higher levels compared to their mainstream peers in tests of English. In addition, the students achieve at or above grade level in Spanish.

Thomas & Collier (2002) report that two-way immersion programs are the most successful in promoting language minority students’ long-term academic achievement, as measured by standardized tests across all subject areas. As a group, students in two-way programs maintain grade-level skills in their first language throughout their schooling and reach the fiftieth percentile in their second language after four to five years of schooling in both languages. Similarly, Oller & Eilers (2002) compared two-way immersion with English immersion in Miami using 952 bilingual and monolingual students from kindergarten to grade 5. They found that while the two-way immersion students initially showed a lag in English performance, this gap was minimal by the time the students reached fifth grade.

Other evidence of the effectiveness of dual immersion can be seen in a study by Sohn & Merrill (2008), which compared the academic performance of Korean speakers in three types of programs—(1) two-way immersion, (2) modified bilingual with minimal Korean language support, and (3) English- only. They found that two-way immersion programs provide the greatest benefit to participating students. Students in Korean-English two-way immersion performed equally well in English language skills, as measured by standardized test scores, as Korean children in English-only programs. They also found that students in dual immersion scored higher in Korean language and general academic skills than students in modified bilingual programs. Similarly, Bae (2007) evaluated narrative English writing skills of Koreanspeaking students and English-speaking students in a Korean-English two-way immersion program in Los Angeles during an early phase of elementary immersion (kindergarten to grade 2). She found that both the Korean-speaking and English-speaking groups in the two-way program had comparable writing skills to English speakers in English-only classes.

Despite the many benefits however, two-way immersion programs face a number of challenges. A major difficulty stems from asymmetrical power relationships between the languages. Although two-way immersion programs strive for a balance between the languages, classroom-based research shows that use of the minority language is frequently undermined by the presence of English. For example, Sra. Soto, a kindergarten teacher in Renee DePalma’s (2010) study of language use in a two-way immersion program, reflects on the low status of Spanish in the surrounding community: “You have to remember environmentally where they are living ... We are in a very depressed economic area, our school is. And they see Spanish as the lower echelon language. The language of people they do not like” (p. 62).

As one of two dual immersion kindergarten classes in an otherwise English- dominant school, Sra. Soto’s classroom followed a 50/50 model with Spanish used in the morning and English in the afternoon. All of the other classrooms in the school were either English-only or transitional bilingual, where the use of Spanish was minimal. In this English-dominant school environment, Spanish time was often disrupted by announcements, specials, and occasional auditorium events, all of which were in English. DePalma (2010: 65-66) provides the following example of an intercom interruption during official Spanish time, which happened just as Sra. Soto finished scolding Rashid, an English speaker, for speaking in English:

Sra. Soto: [In a sharp voice] Mira, Rashid. ^Por que sigues td

hablando de eso en ingles? Si td vas a hablar conmigo en ingles . .. (Look, Rashid. Why do you keep talking about that in English? If you are going to talk with me in English . ..) Unknown woman: [Voice cuts in over the intercom] Mrs. Soto?

Sra. Soto: Yes?

Unknown woman: The number you gave for Kathleen, is it a new number? Sra. Soto: No, it’s her old number.

Girl: [to Sra. Soto] What do I have to do?

Sra. Soto: Tu alfabeto. (Youralphabet) [Saying the letters in Spanish]

A, B, C, D, E, F ...

DePalma (2010) notes that the intercom interruption was an unnecessary request for information that was not particularly urgent. The fact that Mrs. Soto was required to respond in English (since the speaker from the office did not speak Spanish) undermined her efforts to encourage her students to use Spanish during Spanish time.

Another difficulty for dual immersion programs is recruiting proficient speakers of the minority language. DePalma (2010) states that many of the Spanish speakers in Sra. Soto’s classroom were actually more English-proficient than the two-way model assumes. Of the 11 children who had been officially designated as Spanish speakers, Sra. Soto only identified three as good “Spanish role models.” Of the remaining children who spoke mostly Spanish at home, four of them preferred to speak English. Similarly, in a study of a KoreanEnglish dual immersion program in California, Lee (2007) states that some of the Korean students who grew up in Korean homes did not speak Korean as their first language. Her observations of students in the classroom and discussions with the teacher revealed that only four of the 11 students designated as Korean speakers could be considered proficient in Korean. Naturally, English became the language predominantly used by the children in classroom interactions, despite the teacher’s strict implementation of the program’s 70(Korean)/30(English) split of instructional time.

Meeting the linguistic and academic needs of both groups can be a problem in two-way immersion programs. With both Spanish-speaking students and Spanish learners in the same classroom, teachers face a dilemma (Crawford, 2004). For instance, if teachers simplify their Spanish too much to accommodate English-speaking students, they may be shortchanging the Spanish speakers. If they do not simplify enough, then the English speakers could be lost. This leads to the question of whether Spanish speakers’ needs might be better served in a classroom with only Spanish speakers. Because English speakers in Spanish-English dual immersion programs in the U.S. have far less exposure to Spanish than their Spanish-speaking peers have to English, teachers often lower their expectations for English speakers’ progress in Spanish (DePalma, 2010). In the kindergarten classroom in DePalma’s (2010) study, English speakers were often praised for minor achievements in Spanish while Spanish speakers had to perform at a higher level in English to earn the same level of appreciation. Crawford (2004: 306) aptly describes the different expectations for the two student groups as follows:

Bilingual skills, for example, are usually seen as a praiseworthy accomplishment by English-speaking students, while taken entirely for granted among English learners. There is a fine line between treating Spanish, say, as a valuable resource and treating Spanish-speaking students as a resource who are there to “service” the needs of English-speaking classmates. While two-way programs are never consciously organized in this way, the potential for inequity is something they must constantly guard against.

English-speaking children in two-way programs have been shown to dominate classroom discourse and take teacher time and attention away from the Spanish speakers (Valdes, 1997). Likewise, well-educated White parents tend to dominate parent councils in dual immersion schools (Crawford, 2004). Palmer (2008) argues that while diversity in student’s race, class, and culture is indeed a benefit of two-way immersion, two-way programs must contend with race and class diversity in more than superficial ways in order to be successful in meeting all students’ goals.