• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Self-Designated Dual-Language Programs: Is There a Gap Between Labeling and Implementation?

Posted on: Sunday, 23 October 2005, 03:01 CDT

By Torres-Guzmn, Mara E; Kleyn, Tatyana; Morales-Rodrguez, Stella; Han, Annie

Abstract

The United States has seen a tremendous growth in dual-language programs over the last decade. This rapidgrowth brings up questions about the congruency between labels and implementation. Our experiences observing some of the programs and listening to teachers talk about the programs in New York City led us to propose that many of the self-designated dual-language programs stray from even the minimal guidelines for such programs. Our study investigated this, and within this paper we will juxtapose those programs that are congruent with the minimal standards with those that are less faithful to the model in order to gain a better understanding of how schools and districts can work to create dual-language programs that foster an authentic transformation of the instructional environment.

Introduction

Dual-language programs have existed since the 1960s in the United States, with the first being established in Miami, Florida, in 1962 and in Las Cruces, New Mexico, in 1967 (Freeman, Freeman, & Mercuri, 2005). Dual-language programs increased at a slow and steady rate until the 1990s, when their popularity soared due to the greater availability of funding, the inclusion of White middle- to upper- class students, and the increased importance of second-language attainment post-September 11. By 2003, the United States had more than 292 dual-language programs, with 86% of them having been established during the years 1991-2001 (Center for Applied Linguistics, 2004). In the state and year of service we will report on, 66 schools and 20 school districts reported that they were either implementing or planning a duallanguage program.

Dual-language education is emerging as the general term to cover a variety of labels (e.g., developmental bilingual education, two- way bilingual education, one- and two-way immersion, dual immersion, and enrichment education) used within the literature to refer to educational programs that share the essential components of using "two languages for instruction" with explicit goals of "full conversational and academic proficiencies in the two languages" (Freeman et al., 2005 p. xvi). Within the continuum of variations possible within the dual-language education label, there are specific guidelines articulated in the literature as to what constitutes a two-way dual-language program (Caldern & Minaya-Rowe, 2003; Cloud, Genesee, & Hamayan, 2000; Freeman et al.; Lindholm- Leary, 2001; Prez, 2004). Minimally, two-way duallanguage programs must attend to the numbers of minority- and majority-dominant speakers as well as to the time allotted to each language. In addition, in the United States, students in these programs are expected to perform academically in both languages as well as or better than similar students in English-only programs (Lindholm- Leary; Thomas & Collier, 2002).

Our study compares the theoretical constructs with the implementation of self-designated two-way dual-language programs. We focus not on the subtle "discrepancy between ideal plan and actual implementation" that Freeman refers to (1998, p. 246) when there is a commitment to developing a program within the bounds established by the literature. Nor do we focus on the asymmetrical social relationship of languages as they appear in the implementation of programs (Amrein & Pea, 2000). Rather, we focus on the even more fundamental issues of inappropriate labeling and poor implementation. We believe this to be a necessary task if we are to ensure that the growth in dual-language programs leads to quality enrichment programs that serve language-minority and language- majority students alike, and that parents and the public are not misled.

Our theoretical construct stems from the belief that what occurs in schools mirrors societal values but that decisions will be locally shaped by the values of those who participate in the process of decisions as they exercise some discretion "because organizational rules cannot cover every contingency and because organizational socialization is rarely total" (Meier, Wrinkle, & Polinard, 1999, p. 1026). Thus, to the extent that there is representation of the diverse community interests within the bureaucracy, the more the diverse community interest represented will benefit. We posit that there is a relationship between how those who participate in decisions, directly and indirectly, see and value the children and the speech communities they represent within the school, and that this, in turn, influences the thoughtfulness and devotion with which educational programs are implemented to serve them. We have made use of sociobureaucratic representation theory (Meier et al.) because it permits us to postulate that in the implementation of dual language, the inherent flexibility in decision making can be used to respond more sensitively to the different educational needs of students within a given community and can serve as a space where the transformational values held by school personnel, administrators, and teachers may be brought into the schools. Or it can serve to undermine the entire endeavor. Thus, locally held social values are critical to the implementation process and often determine whether these programs undergo changes in more superficial or more transformational ways.

The growth of dual-language programs is occurring in a political context of opposition to bilingual education, even though these programs are, technically, bilingual education programs. The social values embodied in duallanguage programs reflect the populations being served, that is, in many cases language-majority and language- minority students, whereas bilingual programs are often associated solely with ethnolinguistic populations. Thus, dual-language programs are regarded as socially worthy, while bilingual education programs are not (Valds, 1997). Dual-language programs are seen as being more inclusive of different populations and, therefore, as promoting the values of bilingualism and diversity within our society. In addition, for minority populations, there is the possibility of contesting the constraints that have perpetuated their social status and expanding the options available for equity and inclusion.

We are trying to get at two parallel phenomena that are occurring within dual-language programs. Some programs are engaging in transforming educational environments and have an impact on the educational attainment and achievement as well as the social value of minority populations, while others are involved in superficial activities such as changing labels, and perhaps faces, with very little regard for altering the essence of the educational environment.

Thus, our examination of dual-language programs looked at how flexibility in decision making has affected such programs. This will be examined from the perspective of the teachers teaching in the programs, since their views are likely to be more revealing about how programming decisions are either attended to or ignored rather than by examining the views of school personnel in charge, who are more invested in defending the label given to the programs. Our focus is on program implementation in terms of time allotted to English and the language other than English (LOTE) and in terms of student enrollment of language-majority vis--vis language-minority students.

Review of the Literature

According to the research and theoretical literature, dual- language programs ought to adhere to certain criteria in order to achieve the outcomes promised by these models. However, as Lindholm- Leary notes, "not all dual language programs have, or are aware of these characteristics" (2001, p. 75). The key features, consistent across the literature, that any dual-language program ought to embrace are delineated below.

Program Goals

There are a number of common goals for dual-language programs. First, students are expected to reach a high level of academic achievement in both languages (Christian, 1996; Lindholm, 1990; Lindholm & Fairchild, 1990; Mora, Wink, & Wink, 2001 ; Torres- Guzmn, 2002). second, students should develop bilingualism and biliteracy as they move through the grade levels' (Lindholm & Fairchild; Mora et al.; Torres-Guzmn). Third, dual-language programs should promote positive cross-cultural attitudes that work to combat racism and linguicism while promoting social change (Lindholm- Leary, 2001 ; Mora et al. Torres-Guzmn & Frez, 1996; Valverde & Armendariz, 1999). The combination of students with different backgrounds and their integration in most content instruction makes this goal more feasible. Although dual-language programs may have additional goals, these three goals are prominent throughout the literature.

Language Allocation

Dual-language programs must meet a minimum language division of 50-50 in English and the LOTE to ensure that students have ample opportunities to develop and learn through both languages in becoming bilingual, biliterate individuals. Programs can vary, with as much as 90% of instructional time in the LOTE and 10% in English. A 90-10 program typically begins in kindergarten and gradually works toward a 50-50 division by the fifth grade, if \not sooner. Therefore, it is not uncommon to see an 80-20 or 70-30 program in Grades 1-3. Decisions concerning the appropriate model to follow are usually made collaboratively by school and/or district administration, instructional staff, and parents.

Language Separation

Dual-language programs must plan carefully to ensure that there is not any mixture of the two languages (Calderon & Minaya-Rowe, 2003; Kirk-Senesac, 2002; Torres-Guzmn, 2002). There are several ways to ensure that only one language is being used at a time-by content area, time, day, or even week. Some programs also divide the languages between two teachers, while others have a bilingual teacher who consciously divides the languages by time and content area. If languages of instruction are divided by content areas, they may be switched or remain the same throughout the year. For example, a unit in social studies may be taught in English while a science unit is being taught in Spanish. Then, when the next units begin, the languages will be reversed. Or social studies may always be taught in English and science in Spanish. The languages may also be divided by time. The most common approaches are alternate days or half days. One of the half-day models is called the serpent, or roller-coaster, method: if the morning is in English on the first day, the afternoon will be in Spanish, and the following day will begin with a morning in Spanish. Regardless of the method of language separation, the time devoted to each language must adhere to the language allocation ratio (e.g., 50-50,90-10, or any ratio in between).

Student Population

The literature on the proportions of language mix of students within a class varies significantly. The most effective model strives to mix English- and LOTE-dominant students so that the two groups can learn from one another through language modeling (Heath, 1986). Ideally, this division should be as close to 50-50 as possible (Lindholm-Leary, 2001; Valverde & Armendariz, 1999). However, variations are feasible. Some scholars suggest that the ratio between the two linguistic groups be no more than 2:1 (Lindholm-Leary), while others believe that a majority of children can be from the same ethnolinguistic group, although their proficiencies in each language will differ (Frez, 2004; Torres- Guzmn, 2002). If the latter is the case, then the program takes on a one-way developmental approach and the former is referred to as a two-way model.

Method

The data used for this study were collected from survey questionnaires designed by one of the researchers in collaboration with New York City public schools. The questionnaire consisted of items about teachers, students, program goals, classroom practices, and scheduling for both languages and content areas. Content validity was established, since a number of experienced educators and administrators in the field of bilingual education were consulted. Although the majority of the survey questions were either multiple choice or fill in the blank, spaces were provided for teachers to write detailed comments. The questionnaire was distributed to all prekindergarten to eighth-grade teachers in schools that had officially declared having a dual-language program during the data-collection period, a total of 60 schools, of which 56 were elementary schools and 4 were middle schools. There was an 85% return rate, which establishes the survey data as reliable. The survey was supplemented with visits and classroom observations to all the schools, at least once during the period in which the study was conducted. These observations provided us with some assurance that the teachers' responses were genuinely theirs and not those of the administrators in charge.

We used the NVivo computer software program (Qualitative Solutions and Research Pty. Ltd., 1999) to do a content analysis of all the written comments associated with enrollment, goals, labeling, and language allocation. We will include quotes from the teachers' written comments that illustrate typical perceptions. The selection of typical comments was based on a quantitative analysis of the frequency of various kinds of responses.

Analysis and Findings

The dual-language programs surveyed were mostly strands within a school, since schoolwide implementation was rare. In terms of LOTE, 59 of the programs used Spanish and one used Chinese. Nearly two thirds (66%) of the teacher respondents held master's degrees in education, and the majority specialized in the appropriate level, with 44% having a bilingual specialization and 38% an elementary education focus. Teachers with less than 5 years' experience accounted for 44% of the group, while 77% had less than 5 years' experience in a dual-language program. Slightly less than half of the teachers (45%) volunteered to teach in these dual-language programs.

Our analysis initially focused on the two characteristics that distinguish programs that follow the central tenets established in the literature as fundamental to dual-language programs-the use of the LOTE as the language of instruction at least 50% of the time, and a distribution where at least one third of the student population is from the LOTE-speaking community (Lindholm-Leary, 2001), as these two variables will help us understand the integrity of the decisions attended to locally vis--vis the labels the programs carry.

With respect to the amount of instructional time in the languages and content areas, we used program-level aggregates of the teacher responses to the scheduling item on the survey, which consisted of a blank schedule with 8 periods per day. The teachers were asked to fill in the subject or activity and the medium of instruction for each period. Where team teaching occurred, we first coded the data at grade level, followed by program-level aggregates, as we were trying to get at the consistency of the program-level implementation.

Out of the 60 programs in the study, 28 had to be excluded because of an incomplete or incomprehensible schedule. Of the remaining 32 programs, only 2 managed to provide at least 50% of instruction in the LOTE.2 This was somewhat unexpected, as we were aware of about half a dozen programs that we would have thought were within the 50-50 model, and we were aware that 54 programs had been in operation for less than 3 years. It seemed that they were still experimenting, and the teacher responses may have been more representative of the moment than of what actually occurred over time. Another contributing factor to the amount of LOTE used during instruction, as was reported, was that many schools lacked bilingual cluster teachers (art, music, physical education, etc.), and this inevitably shifted the language distribution toward English. Although our finding may be the effect of examining a moment in time, it does poignantly highlight the distance between how schools label their programs and the implementation of these programs.

The second characteristic of dual-language programming that we wanted to highlight was the ratio of majority language- to minority language-dominant students. We decided to report on the teachers' responses to the item rather than using the school district testing data. The school district's language proficiency test has been under scrutiny for some time now. Since these survey data were from the teachers' perspective and how the teacher classifies the student influences how he or she works with him or her in the classroom, to use the teachers' classification of students was considered consistent and appropriate.

A total of 4,964 students were identified by the 282 teachers in the 32 programs. The number of students in individual programs ranged from 14 to 265. The lower numbers usually indicated new programs, since many programs are started with one or two grade levels and add more levels as the students who started the program move up. The teachers classified the students as English language learners (ELL entitled students), former ELLs (those who were previously entitled and had gained English-language proficiency), and English-dominant (ED) students, who had never been ELLs.3 Of the ED students, some were from bilingual households and some from English-only households. Although the number was low, there were some children from Euro American, African American, or other ethnic backgrounds that had bilingual home environments that fell into the category of English dominant with a bilingual presence in the home. This was due to either mixed marriages, third- or fourth-generation language minorities with extended family members who were proficient speakers of a LOTE, or caretakers who had been consciously chosen because of their proficiency in a LOTE (Marquez-Lopz, 1998). We treated ELLs and former ELLs as language-minority speakers and the EDs as language-majority speakers.

At the district level, 48% (2,385) of the students enrolled in the duallanguage programs were language-minority speakers, and 52% (2,579) were language-majority speakers. Thus, taken as a whole, the population of the programs that labeled themselves as dual language was within the guidelines established in the literature. When we examined the student population at the program level, we found that 29 programs had a student population that was more than 30% LOTE dominant, and thus were essentially within the guidelines established in the literature.

We then divided the programs into those providing 40% or more of instructional time in the LOTE and those providing less than 40%. The former are technically not the 50-50 model but closely resemble it. Through our school visits and school district information, we knew that there were more than two programs doing a good job implementing the 50-50 model, and we also considered that our results might have been an artifact of the one-time measure, a frozen moment in time. Hence, a cutoff at 40% of instructional time in the LOTE s\eemed reasonable, and a comparison of these two types of programs with respect to student and program variables might provide us with a better understanding of the disconnects between label and implementation.

Table 1 combines the student population and language allocation standards. Based on responses from 229 teachers, it shows that just 10 programs had a student population that was more than 30% LOTE dominant and allocated 40% or more of instructional time to the LOTE. Of the remaining programs, 19 satisfied only the population standard, and 3 satisfied only the language standard. All of the programs satisfied at least one of the standards. The 10 schools that had greater LOTE involvement in both student population and language allocation were designated G-LOTE schools (the shaded area in Table 1), while the 22 schools that had lesser LOTE involvement were designated L-LOTE schools. It was our hunch that by examining the practices as reported by the teachers we could get at the decisions made in these schools and, thus, we could gauge the levels of integrity of the linguistic model.

Table 1

Dual-Language Programs by Student Population and Language Allocation (N = 32)

In the following sections, we will examine in more detail how the G-LOTE (the 10 schools that met both criteria) and L-LOTE groups (the 22 schools that did not meet the two criteria) differ with respect to various program characteristics and look further at the relationship between labeling and implementing dual-language programs.

Student Enrollment

Teachers' responses on admissions policies and student selection items on the survey helped us understand the attention paid to the student enrollment process. The item on student enrollment included a variety of methods for selecting students: neighborhood; magnet; first come, first served; language-proficiency balance; opting in; entitlement; and combination. Other was also provided as a category, for there may have been ways unknown to the researchers in which students were selected. Teachers were able to select all the items that described their situation. As shown in Table 2, the four most frequently selected methods were opting in, neighborhood, entitlement, and language-proficiency balance. Among these, we found a statistically significant difference between the two groups in language-proficiency balance (χ^sup 2^ = 38, df = 1). G-LOTE programs paid greater attention to this factor than L-LOTE programs.

Table 2

Method of Enrollment in G-LOTE and L-LOTE Programs

The data indicate that in the G-LOTE programs, more attention is given to the enrollment of entitled students than in L-LOTE programs. And in the G-LOTE programs, these students-ELLs (48%) and former ELLs (22%)-have the opportunity to continue to develop their native language, even though the latter have reached the designated district level of English proficiency for reclassification. The remaining 30% are EDs who are learning a heritage or second language. In contrast, the L-LOTE programs have 31% ELLs, 7% former ELLs, 31 % EDs from homes where bilingualism is practiced, and 30% from ED homes where there is a total absence of any other language than English. What these enrollment data suggest is that L-LOTE programs are actually second-language enrichment or heritage- language programs catering to English-dominant speakers-programs that include both teaching the language as a subject matter and teaching at least one content area in the second language (Genesee, 1987)-rather than dual-language partial (50-50) or total (90-10) immersion programs.

In the written comments, the L-LOTE teachers spoke most about the optin nature of their programs, about the need for children to be doing well in school in order to select them for such programs, and the space availability as associated with the selection of their students. The G-LOTE teachers expressed a subtle difference in their discourse about the enrollment of students. Many stressed that "parents must apply to the program and express their interest and belief in the program and philosophy of dual language [and], most students attend being that this is their neighborhood school, and are ELL as determined by home language survey and Language Assessment Battery (LAB) scores." There was also a difference in the way each of the groups spoke about linguistic balance. The G-LOTE teachers spoke about "those who opted in as being chosen, in part, to obtain a linguistic as well as a gender and ethnic balance" and about "the main concern being the ELL and language minority child," while the L-LOTE teachers wrote about "the ideal class as being 50/ 50 SSL [Spanish as a second Language] and ESL [English as a second Language] learners and admitted to there being no class at present with this ratio."

Language Allocation

With respect to language allocation, we gave special importance to the schedules teachers provided us. The asymmetrical treatment of the languages and the inferior positioning of the LOTE are self- evident in both groups but, more poignantly, in the L-LOTE programs, which averaged 5.02 more periods in English and 4.42 fewer periods in the LOTE than G-LOTE programs (see Table 3).

Table 3

Mean Periods per Week of Instruction in G-LOTE and L-LOTE Programs by Language

Given that this was one of the program characteristics we used to divide the program types, we expected that there would be a statistical difference in the amount of LOTE use. The Pearson Chi- Square tests indicated significant differences between the two groups. The L-LOTE programs, of course, instructed 5.02 more periods in English (χ^sup 2^ = 7l,df= 28) than the G-LOTE programs and 4.42 fewer periods of Spanish (χ^sup 2^ = 52, df = 25).

Given the mandates of the state, it was not surprising to find that the two subjects receiving the most instruction were language arts and math. As shown in Table 4, both groups devoted a great deal of time to language arts in both languages, with the G-LOTE programs spending 15.8 periods per week and the L-LOTE programs 15.85 periods per week. English was the more prominent language in the L-LOTE programs, while the LOTE was slightly more prominent in the G-LOTE programs. The G-LOTE programs spent 8.1 periods on LOTE language arts, nearly double the 4.95 periods of the L-LOTE programs. These differences were found significant at the .000 level (χ^sup 2^ = 3S,df= 17).

While both groups spent over 3 periods teaching math in English, the G-LOTE programs spent 2.17 periods teaching math in LOTE, and the L-LOTE programs spent 1.76 periods on math in LOTE. The fact that G-LOTE programs spent additional time (half a period) on math may be related to the belief of the district's dual-language leadership that math should be taught in English, since it is less dependent on students' proficiency in the language.

Neither group devoted a great deal of time to science, with the G- LOTE programs spending slightly more-3.05 versus 2.64 periods per week. The G-LOTE group taught science more in the LOTE, while the L- LOTE group taught it more in English. There is a statistical significance at the .000 level (χ^sup 2^= 16, df = 5). Based on our observations and school visits, this finding is likely related to the availability of science materials in Spanish, that is, the district's science inquiry program was available in Spanish.

Table 4

Mean Periods per Week of Instruction in Content Areas in G-LOTE and L-LOTE Programs by Language

Both groups devoted less than 3 periods per week to social studies and allocated the languages in a similar way.

The Other category is comprised of instructional periods that do not fit under the four content areas described above. They range from "specials," such as physical education, art, music, and technology classes, to learning centers. The L-LOTE programs spent more time in the less central content areas than the G-LOTE programs. For both groups, the LOTE was not as valued as a medium of instruction for areas outside of the core content areas. This dominance of English speaks to the monolingualism of schools outside bilingual classrooms and the fact that cluster and other supplemental teaching staff are primarily English-only speakers. We heard complaints about this when we visited the schools, and we also found them in the teachers' written comments: "All day the students are in the assigned language, with exception of the cluster teachers for 1 period. Students do computer, library, science and art with the cluster teacher." Another teacher wrote about an adjustment that was made in response to this situation: "The art teacher does not speak Spanish, so the social studies and science are in Spanish to compensate."

Language Separation

The teachers in both program types use content area most frequently to determine language allocation, with 55% of the G-LOTE program teachers and 66% of the L-LOTE program teachers reporting the use of this method. Based on our observations, this decision was often guided by the availability of instructional materials in the content areas, as alluded to earlier in relation to the science curriculum. Time was the second most common factor, with over half of the teachers in both groups selecting this feature. This seemed to be guided by program policy and the ease of language allocation by periods. Neither type of program uses homogeneous language groups to determine language allocation frequently, with the exception of the literacy grouping. The largest discrepancy emerges in the language allocation by teacher category. The L-LOTE teachers selected this category 43% of the time, while the G-LOTE group only selected 33% of the time. This finding may reflect how L-LOTE programs fall back on the teacher's linguistic ability as a way of allocating the language of instruction. However, it was not statistically significant.

Class Structure

When we looked at the class structures reported bythe two groups (see Table 5), we found that there were statistically significant differences in the use of single-grade structure (χ^sup 2^ =16, df=\) and team teaching (χ^sup 2^ = 21, df=1). The G-LOTE programs used team teaching much more than the L-LOTE programs. This may suggest that the G-LOTE programs are more creative in how they organize teachers and students to better meet the linguistic and academic needs of the students. Together, the class structures and the method of language separation, as reported above, seemed to be stronger evidence of our claim that the L-LOTE programs' choice of medium of instruction might be related to the teachers' linguistic abilities. However, we did not have data to test this interpretation directly.

Program Descriptions

The labeling item on the survey provided eight suggested labels, taken from both the literature and those used in schools, and Other. The respondents were allowed to pick more than one label (see Table 6). Most teachers selected the two-way/dual-language label, as would be expected, and there was no significant difference in the use of this label between the two program types. Sixty-three percent of the teachers in the L-LOTE group and 40% of the GLOTE teachers selected the two-way/dual-language descriptor.

Table 5

Class Structure in G-LOTE and L-LOTE Programs

Table 6

Program Labels in G-LOTE andL-LOTE Programs

However, we did find a statistically significant difference in the use of the second-language enrichment label (χ^sup 2^ = 8,df= 1). That the L-LOTE teachers tended to choose this label suggests that they did know that their programs, in practice, are second-language enrichment programs. The G-LOTE teachers, on the other hand, clustered around the 50-50 labels. There was also a statistically significant difference in the use of the alternating days label (χ^sup 2^ = 54, df=1). The fact that the G-LOTE teachers selected this label in such greater numbers than the L- LOTE teachers suggests that within the L-LOTE programs such a model differentiation did not exist.

With respect to written comments, G-LOTE teachers tended to describe what they did as alternate-day, half-day, or roller- coaster approaches, while L-LOTE teachers referred to the second- language enrichment approach more often. As can be seen in Table 7, where the teachers' written comments are displayed, we found a great distance between the two groups in how they described their programs. We chose the numerically salient responses but would like to caution the reader that our examples of typical teacher responses may give a misleading impression about the range of opinions expressed; there were more responses, and they varied. A full discussion of the teachers' written responses may be a subject of discussion on its own, which we will not take up within this article.

What these findings may indicate is that teacher labeling beyond the dual-language umbrella and into more specific features of the programs is a more accurate indicator of what is occurring in the programs. The L-LOTE programs seem to be reflecting a widespread theoretical confusion, and needless to say, this is reflected in implementation. Our own conversations and observations in these schools confirm that there is a widespread belief that a 90-10 model is implemented as 90% instructional time in English and 10% in the LOTE. In other words, teachers do not know the basic tenets of dual- language programming and are making decisions about medium of instruction in ignorance of what they ought to be doing under the duallanguage label.

Program Goals

The program goals listed in the survey were selected from the literature (see Table 8). When we asked the teachers what the goals of their programs were, we found that there was general agreement on the five goals presented. When we analyzed the goals with respect to program types, we found general agreement with one pedagogically, although not statistically, significant exception: the goal of high academic achievement in both languages was selected substantially more often by the G-LOTE teachers.

This finding, we would suggest, speaks to the expectations that teachers set in each of these program types and, thus, to how the program itself is implemented. The literature on dual-language programs makes an effectiveness claim about academic achievement in both languages.

Table 7

Typical Teacher Descriptions of G-LOTE and L-LOTE Programs

Academic achievement in the second language is not a distinct goal within the L-LOTE programs, since one would not expect a student going through a second-language or heritage-language enrichment program of this sort to acquire sufficient academic language to be able to pursue higher learning.

The number and the nature of the comments that teachers wrote as part of the survey are even more telling. In both groups, the majority of the comments were in reference to Goal 2, which refers to second-language learning, followed by Goal 1, which has to do with achievement in the first language. The G-LOTE teachers also showed significantly more concern for Goals 3 and 4, which are concerned with academic achievement and crosscultural understanding. The comments have embedded discourses that reveal their differences in emphasis. The G-LOTE teachers speak to the "students performing at or above grade level in both languages" (our emphasis) whereas the teachers from the L-LOTE do not refer to the languages in which they expect their students to succeed. The latter group is not inclusive of the language acquisition goals when they speak about academic language. Table 9 provides typical comments with respect to each of the goals.

Table 8

Program Goals in G-LOTE and L-LOTE Programs

The teachers' educational and experiential background (see the first paragraph of the Analysis and Findings section) was similar across the G-LOTE and L-LOTE groups. Therefore, we propose that mislabeling and implementation issues are not due to teacher quality or experience, but are related to problems at the district, program, and administrative levels.

Conclusions

Initially, we claimed that it was necessary to examine the differences between the labeling and the implementation of dual- language education programs because many schools may be self- designating as dual language but not meeting the minimal standards for such programs. We also stated that in dual language or any other educational program, there is a discretionary space that permits local schools to respond more sensitively to the different educational needs of students. Given this discretionary space, we saw aspects of the programs that required local decisions, such as admissions and time allocation, that were as critical in determining whether schools were implementing dual-language programs with integrity and whether the implementation was done in superficial or more significant ways.

Table 9

Typical Teacher Comments About Program Goals

What we have found from teachers' labeling is that many programs within the district we studied were, as we suspected, second- language enrichment programs borrowing the label of dual language. This finding suggests that teachers, as representative of one of the groups involved directly and indirectly in decisions about children, believe that what they do is dual language. In other words, teachers did not know the basic tenets of dual-language programming and are participating in decisions or allowing decisions to be made in ignorance of what they ought to be doing.

Our findings also show that there is a small minority of teachers, those within the G-LOTE group, who show a stronger foundational understanding of the model. This is shown in their remarks on how the program is implemented, a greater likelihood that more ELLs are enrolled in the program, and the focus on instruction, independent of language. Hence, the well-implemented duallanguage programs (Thomas & Collier, 2003) are more likely to cause an environment of possibilities of transformational educational environments that can help to ensure equity for language-minority students while including language-majority students (Valds, 1997). There are probably more decision-making points within programs that need to be identified for further study to identify how to ensure the transformational potential of dual-language programs. As de Jong (2002) proposes, no matter what label is given to an educational program and no matter how a program is held up as a panacea, much of its potential will be centered in the implementation process.

We fear that the sustainability of dual-language programs, in the long run, is compromised by the labeling of second-language and heritage-language enrichment programs as dual language, as we documented in the case of L-LOTE programs. Based on past experience, we can foresee that there will be studies of the effectiveness of dual-language programs that will include all programs that self- designate as dual language, irrespective of the actual characteristics of such programs. As we know, the mislabeling and the comparing of apples and oranges within many studies of bilingual education have damaged its reputation among the public, even though we have ample evidence that, when programs are well implemented, the use of the native language with language-minority students is a positive factor associated with higher levels of achievement.

If what occurs in schools and the decisions made are not challenged, it is likely that the educational environment will continue to mirror governing societal values and the implementation of the programs will be in step with these values. The power structures that dictate linguistic value force schools to create environments that cause language-minority students to feel devalued until they have mastered English and have forgotten or abandoned their home language (Skuttnabb-Kangas, 1991) when language-majority students a\re praised for learning a second language. Even with the best intentions of equalizing opportunity and learning for students of all backgrounds, the reality is that schools, more often than not, tend to mirror social and political injustices, leaving minority speakers on the periphery (Amrein & Pea, 2000; Freeman, 1998; Freire & Macedo, 1987; Valds, 1997). Yet, this should not stop us from pushing for the ideal. Dual-language programs that aim to reposition the LOTE and its speakers within the context of schooling and society at large view the attainment of a second language as a resource (Ruiz, 1984) for all students, regardless of their dominant language. Such an enrichment model can translate concerns about language development and cultural pluralism into an equitable linguistic and cultural education for all students (Crawford, 1997; Hornberger, 1991; Lindholm, 1990). Reaching out to language- majority and language-minority students alike, "dual language programs can be understood as contesting the legitimacy of monolingualism in Standard English as the unquestioned norm in mainstream U.S. schools" (Freeman, p. 11). We must not permit this potential transformational value to slip through the cracks by failing to point out the lack of care and attention that characterizes the implementation of many dual-language programs.

Endnotes

1 Although this is a goal, it has been found that students do not usually reach full bilingualism. In other words, they remain dominant in one language, but significant growth can take place nevertheless (Lindholm-Leary, 2001).

2 Since the initial report to the school district, there has been a change in policy and practices. Programs that call themselves dual- language must now have at least 50% of the instruction in the LOTE.

3 In negotiating the terms of the original study, Thomas & Collier's (2002) categories were the base from which we started: ELLs, formerly classified ELLs, language minorities who were never ELLs, and English-dominant students. The terminology used in this paper was the final outcome of the negotiations.

References

Amrein, A., & Pea, R. (2000). Asymmetry in dual language practice: Assessing imbalance in a program promoting equity [Electronic version]. Education Policy Analysis Archives, 8(8). Retrieved May 31, 2005, from http://epaa.asu.edu/epaa/v8n8.html

Bourdieu, P. (1993). Sociology in question. London: Sage.

Caldern, M. E., & Minaya-Rowe, L. (2003). Designing and implementing two-way bilingual programs: A step-by-step guide for administrators, teachers, and parents. Thousand Oaks, CA: Corwin Press.

Center for Applied Linguistics. (2004). Growth of two-way immersion programs in the US, 1962-Present. Retrieved May 31, 2005, from http://www.cal.org/twi/directory/

Christian, D. (1996). Two-way immersion education: Students learning through two languages. Modern Language Journal, 80, 66-76.

Cloud, N., Genesse, F., & Hamayan, E. (2000). Dual language instruction: A handbook for enriched education. Boston: Heinle & Heinle.

Crawford, J. (1997). Best evidence: Research foundations of bilingual education. Washington, DC: National Clearinghouse for Bilingual Education.

de Jong, E. J. (2002). Effective bilingual education: From theory to academic achievement in a two-way bilingual program [Electronic version]. Bilingual Research Journal, 26(1). Retrieved March 31, 2005, from http://brj .asu.edu/content/vo!26_nol /html/art5 .htm

Freeman, R. (1998). Bilingual education and social change. Clevedon, England: Multilingual Matters.

Freeman, Y. S., Freeman, D. E., & Mercuri, S. P. (2005). Dual language essentials for teachers and administrators. Portsmouth, NH: Heinemann.

Freire, P., & Macedo, D. (1987). Literacy: Reading the word and the world. South Hadley, MA: Bergin & Garvey.

Genesee, F. (1987). Learning through two languages: Studies of immersion and bilingual education. Rowley, MA: Newbury House.

Heath, S. B. (1986). Social cultural context of language development. In Bilingual Education Office, California State Department of Education (Ed.), Beyond language: Social and cultural factors in schooling and minority students (pp. 141-186). Los Angeles: California State University Press.

Hornberger, N. (1991). Extending enrichment bilingual education: Revisiting typologies and redirecting policy. In O. Garcia (Ed.), Bilingual education: Focusschrift in honor of Joshua A. Fishman. (Vol. 1, pp. 215-314). Philadelphia: John Benjamins.

Kirk-Senesac, B. V. (2002). Two-way bilingual immersion: A portrait of quality schooling [Electronic version]. Bilingual Research Journal, 26(1). RetrievedMay31,2005,fromhttp://brj.asu.edu/ content/vol26_nol/html/ art6.htm

Lindholm, K. J. (1990). Bilingual immersion education: Criteria for program development. In A. Padilla, H. H. Fairchild, & C. M. Valadez (Eds.), Bilingual education: Issues and strategies (pp. 91- 105). Newbury Park, CA: Corwin Press.

Lindholm, K., & Fairchild, H. H. (1990). Evaluation of an elementary school bilingual immersion program. In A. Padilla, H. Fairchild, & C. Valadez (Eds. ),Bilingual education: Issues and strategies (pp. 126-136). Newbury Park, CA: Sage.

Lindholm-Leary, K. (2001). Dual language education. Clevedon, England: Multilingual Matters.

Marquez-Lopz, T. (1998). Parental views on participation, dual language education, and bilingualism. Unpublished doctoral dissertation. Teachers College, Columbia University,New York.

Meier, K. J., Wrinkle, R. D., & Polinard, J. L. (1999). Representative bureaucracy and distributional equity: Addressing the hard question. The Journal of Politics, 61(4), 1025-1039.

Mora, J., Wink, J. & Wink, K. (2001). Dueling models of dual language instruction: Acritical review of the literature and program implementation guidelines [Electronic version]. Bilingual Research Journal, 25(4). Retrieved June 15, 2005, from http://brj.asu.edu/ content/vol25_ no4/html/art3 .htm

Prez, B. (2004). Becoming biliterate: A study of two-way bilingual immersion education. Mahwah, NJ: L. Erlbaum Associates.

Qualitative Solutions and Research Pty. Ltd. (1999). NVivo: NUD.IST for Qualitative for Research [Computer software]. Doncaster, Australia: Author.

Ruiz, R. (1984). Orientations in language planning. NABE Journal, 7, 15-34.

Skuttnabb-Kangas, T. (1991). Multilingualism and the education of minority children. In O. Garcia & C. Baker (Eds.), Policy and practice in bilingual education: Extending the foundations (pp. 40- 62). Bristol, PA: Multilingual Matters.

Thomas, W. P., & Collier V. (2002). A national study of school effectiveness for language minority students ' long term academic achievement: Final report: Project 1.1. Retrieved June 6, 2005, from http:// crede.ucsc.edu/research/llaa/1.1_final.html

Torres-Guzmn, M. E. (2002). Dual language programs [Directions in Education and Language Series]. Washington, DC: National Clearinghouse for Bilingual Education.

Torres-Guzmn, M. E., & Prez, B. (1996, April). Access to language and literacy in two-way bilingual classrooms. Paper presented at the Annual Conference of the American Educational Research Association, San Diego, CA.

Valds, G. (1997). Dual-language immersion programs: A cautionary note concerning the education of language-minority students. Harvard Educational Review, 67(3), 391-429.

Valverde, L. A., & Armendariz, G. J. (1999). Important administrative tasks resulting from understanding bilingual program designs [Electronic version]. Bilingual Research Journal, 23(1). Retrieved June 14, 2005, from http://brj.asu.edu/v231/articles/ art3.html

Mara E. Torres-Guzmn and Tatyana Kleyn

Teachers College, Columbia University

Stella Morales-Rodrguez

New York City Department of Education

Annie Han

Borough of Manhattan Community College, City University of New York

Annie Han

Annie Han is a professor of mathematics at the City University of New York. She received her doctoral degree in mathematics education from Teachers College, Columbia University. She was a Title VII Fellow and a recipient of a Fulbright-Hays grant. Her research interests include mathematics and language, bilingual mathematics teacher education, history of mathematics education, and technology use in the mathematics classroom.

Tatyana Kleyn

Tatyana Kleyn is an Ed.D. candidate in international educational development at Teachers College, Columbia University, where she specializes in bilingual/bicultural education. She is also an instructor in the bilingual/ TESOL program at the City College of New York. Her research interests include multicultural education in bilingual classrooms and education for indigenous groups in Central America.

Stella Rodrguez-Morales

Stella Rodrguez-Morales received her Masters of Science in Elementary Education from Bank Street College in New York City. She served as dual language coordinator for the Department of English Language Learners of the New York City Department of Education from 2000-2003. She is currently mentoring beginning teachers.

Mara E. Torres-Guzmn

Mara E. Torres-Guzmn is an associate professor of bilingual/ bicultural education in the Department of International and Transcultural Studies at Teachers College, Columbia University. Among her latest publications is an article with E. Hollin The Preparation of Candidates for Teaching Diverse Student Populations, published in M. Cochran-Smith & K. Zeichner (Eds.), Synthesis of research on teacher education.

Copyright National Association for Bilingual Education Summer 2005

Source: Bilingual Research Journal

More News in this Category