Washington State operates the Transitional Bilingual Instruction Program for students in its public schools who have a native language other than English and who have English-language skill deficiencies which impair their learning in regular classrooms (Office of Superintendent of Public Instruction, 1998c). This program operates under the authority of RCW 28A.180.060, and is detailed in chapter 392-160 WAC (Office of Superintendent of Public Instruction, 1998c). The main objective of the program is student competence in English-language skills (Office of Superintendent of Public Instruction, 1998c). During the 1996-1997 school year, the program's services were provided by 181 school districts to 54,124 students enrolled in pre-kindergarten through twelfth grade; of these students, more than half spoke Spanish as a first language (Office of Superintendent of Public Instruction, 1998c).
Although the program is broadly termed the Transitional Bilingual Instruction Program, it nonetheless has several possible manifestations, not all of which conform to transitional bilingual education as defined within this document. According to the Office of Superintendent of Public Instruction (1998a, 1998c), there are four main program models provided under the Transitional Bilingual Instruction Program:
The Primary Language Development Program: This program provides long-term dual-language instruction in both English and the first language. The goal of the program is to enable students to become academically and socially fluent in both languages.
The Academic Language Development Program: Initially, this program provides intensive ESL instruction, as well as first-language instruction to support academic skills and literacy. However, academic instruction in the first language is discontinued once the student reaches moderate English reading competency.
The Limited Assistance in the Primary Language Program: This program provides intensive ESL instruction and minimal support in the first language. First-language support may include academic tutoring provided by noncertificated staff, translators, interpreters, etc.
The No Primary Language Support Program: This program provides intensive ESL instruction, but offers no first-language support. However, other special instructional services may be provided which enable the student to participate in the regular, mainstream classroom.70
In the 1996-97 school year, these programs served 4.3 percent, 21.4 percent, 41.6 percent, and 27 percent, respectively, of the students receiving support through Transitional Bilingual Instruction Program. In addition to these four program models, districts may also design Alternative Instruction Programs to deliver services, though the Office of Superintendent of Public Instruction must approve them. This latter option served 5.7 percent of the students receiving support through the Transitional Bilingual Instruction Program in the 1996-97 school year.
There are also several classroom delivery models under which ELLs can receive services through the Transitional Bilingual Instruction Program. Whereas program models describe the instructional strategies employed, classroom delivery models describe the setting or circumstances in which services are delivered (Office of Superintendent of Public Instruction, 1998c). The five main classroom delivery models are:
Self-contained classroom: In this model, students are scheduled to attend an all-bilingual classroom that offers instruction in English and English language arts that is appropriate to their levels of English competence; and sometimes provides academic instruction in the first language. The reading and language arts instruction that is provided in these classrooms is parallel, not supplementary, to that offered in the regular classroom.
Center approach: In this model, non-English speaking students are scheduled for a large portion of the day in a bilingual center offering intensive English-language development and, in some cases, instruction in the first language. Students return to the mainstream classroom only for those subjects not requiring significant English-language interaction.
In-classroom model: In this model, students who have attained some English-language proficiency are provided, in the mainstream classroom, with ESL instruction by a specialized instructor. In some cases, students are also provided with academic instruction in the first language.
Pull-out method: In this model, students are taken from the mainstream classroom in order to receive ESL instruction and, in some cases, academic instruction in the first language. Instruction is delivered in small groups or on an individual basis.
Tutoring: In this model, students are provided with a bilingual tutor who assists individuals or small groups in completing class assignments, or who provides limited assistance in ESL.71
In the 1996-97 school year, these classroom delivery models served 14 percent, 5 percent, 29.2 percent, 28.2 percent and 9.3 percent respectively of the students receiving support through the Transitional Bilingual Instruction Program (Office of Superintendent of Public Instruction, 1998a). In addition, 14.3 percent of the students were served using other, unspecified methods of service delivery (Office of Superintendent of Public Instruction, 1998a).
Return to Table of Contents
Model of Program Development in
Relation to Language of Instruction
This appendix provides a framework for the development of an effective educational program for English-language learners (ELLs) in terms of language of instruction.72 We have based this framework substantially on the one endorsed by Krashen (1998) and Krashen and Biber (1988) because:
it is consistent with Cummins' model of language development advanced in Chapter One;
it is supported by the research on program effectiveness discussed in Chapter Three; and
it is flexible and can be tailored to suit the local context of an individual school.73
It should be remembered however that effective practices in regard to language of instruction are just one consideration in the development of an educational program that meets the needs of a school's language-minority students. Researchers have identified a number of practices and attitudes that are characteristic of schools and classrooms that are successful in the education of these children (see Chapter Three). Hence, incorporation of the framework provided in this section should be seen as merely one step (albeit an important one) in the overall construction of an effective program.
The Ideal Case
Krashen (1998) and Krashen and Biber (1988) argue that successful programs for ELLs have the following four components:
comprehensible input in English that is provided directly in the form of content-based ESL and sheltered subject matter classes;
subject matter teaching done in the first language (this does not include the practice of providing instruction in English augmented with concurrent translation);74
literacy development in the first language; and
continuing development of first-language proficiency when possible.
Table B:1 graphically depicts the ideal manifestation of such a program. Termed the "Gradual Exit Plan" by Krashen (1998), this model
…has three components and four stages. The stages, however, are very flexible. In the beginning stage, all children - limited English proficient and native speakers of English - are mixed for art, music and physical education. This makes sense for two reasons: It avoids segregation, and much of the English the minority-language children will hear will be comprehensible, thanks to the pictures in art and movement in PE. Also at this stage, children are in high quality comprehensible input-based ESL classes, and are taught all other subjects in the primary language.
The intermediate stage child is defined as the child who understands enough English to begin to learn some content through English. We begin with sheltered subject matter instruction in those subjects that, at this level, do not demand a great deal of abstract use of language, such as math and science. Subjects such as social studies and language arts remain in the first language, as it is more difficult to make these subjects comprehensible to second language acquirers at this level.
At the advanced level, limited English proficient students join the mainstream, but not all at once: they begin with one or two subjects at a time, usually math and science. When this occurs, social studies and language arts can be taught as sheltered subject matter classes.
In the mainstream stage, students do all subjects in the mainstream, and continue first language development in classes teaching language arts and social studies in the first language. These continuing first language classes are not all-day programs. Rather, they can take the place of (or supplement) foreign language study. (Krashen, 1998, p.200)
Return to Table of Contents
THE GRADUAL EXIT PLAN75
Art, music, PE
All core subjects
Art, music, PE
ESL, math, science
Language arts, social studies
Art, music, PE,
ESL, social studies, language arts
Continuing L1 development
Continuing L1 development
The Gradual Exit Plan's Grounding in Theory and Research
This model is consistent with Cummins' threshold hypothesis discussed in Chapter One; as children reach the linguistic threshold necessary to benefit from English instruction in a particular subject, they transition first to sheltered instruction in that subject, then to mainstream classes. This model is also supported by the research on program effectiveness summarized in Chapter Three. In its ideal manifestation, it conforms to the program identified by Thomas and Collier (1997) as "late-exit bilingual education, including ESL taught through academic content." As was discussed in Chapter Three, Thomas and Collier (1997) found that, of the six program types they studied, only the two-way developmental bilingual program succeeded in producing higher achievement among ELLs than the late-exit program. Significantly however, this Gradual Exit Plan could also easily be incorporated into a two-way developmental bilingual program model.76
Modifications Consistent with Local Contexts
As was mentioned, a strength of the model presented in this appendix is that it is flexible in regard to the local contexts of individual schools. Although the Gradual Exit Plan, as presented above, is the ideal, its practical application may be limited due to school contextual factors. Specifically, its application may be limited in situations where: the ability to provide first-language instruction is restricted due to constraints on resources (i.e., staff and materials), or there is not a large concentration of children who speak the same first language. In such situations, the Gradual Exit Plan can be modified to provide the maximum benefit to ELLs given the local context.
However, before making such modifications, it should first be ensured that the linguistic resources of bilingual teachers are being utilized to their fullest potential. Krashen (1998) argues that one way of doing this is through team-teaching. Team-teaching allows those who speak the child's first language to teach in that language, while those who do not speak the child's first language provide instruction in the mainstream and sheltered/ESL sections of the model presented above.
If the re-allocation of teachers' linguistic resources does not succeed in allowing for a full implementation of the ideal model, program effectiveness can still be maximized by incorporating, to the degree possible, the four key programmatic components listed above. This involves emulating the ideal model, with the exception that sheltered instruction is provided in the least linguistically demanding of the subjects that would otherwise be provided in the student's first language (i.e., math and science for the "beginning stage" child). Further, if no native-language instruction can be provided at all, sheltered instruction is provided in all subjects that would otherwise be provided in the first language. Although this kind of program "lacks the advantages of developing literacy in the first language and using the first language to supply subject matter knowledge," it at least has the advantage of being comprehensible all day long (Krashen, 1998, p.201). Therefore, it is superior to a program of submersion augmented by ESL "pull-out," which results in exposure to incomprehensible input most of the day.
If even this modified plan is not possible due to the presence of only a few limited English students in the school, there are still a number of things that can be done to maximize program effectiveness. According to Krashen (1998), these include:
ESL pull-out, if provided, should be scheduled for times when more proficient English speakers are engaged with subject matter that requires the most abstract use of language and that will be the least comprehensible to ELLs (i.e., language arts and social studies).
First-language development should be promoted by providing books in the student's first language and by encouraging the use of the first language in the student's home. Although the latter may seem an unusual suggestion, as Gandara (1997) notes, "Traditionally, Limited English-speaking parents have been admonished to give up the use of the native language in the home and help their children to transition to English by providing English language models" (no page number). Unfortunately however, research by Wong Fillmore (1991a, 1991b) has shown that such efforts to rapidly shift ELL children to the use of English can result in the loss of the first language and, subsequently, engender a breakdown in parent-child communication. This breakdown in communication can, in turn, have a number of seriously negative effects on parents' ability to raise their children.77
Instructional aides who speak a student's first language should be used to support the student's background knowledge and literacy in the first language. Krashen (1998) argues that these instructional aides are too often used in the less productive task of drilling ELLs in English spelling and vocabulary. Return to Table of Contents
Reading-Related Programs That Influence the
Reading Achievement of English-Language Learners
As part of our investigation of the research on second-language learners and learning to read in English in our public schools, we searched the published literature for quantitative research on the degree to which widely used reading-related programs in Washington State's public schools are effective with English-language learners (ELLs). This search was limited to studies of program effectiveness involving elementary or middle school students, and to programs in which instruction is provided in English. The programs encompassed by our search were: the Higher Order Thinking Skills (HOTS) and Reading Recovery special intervention programs, the Direct Instruction System for Teaching Arithmetic and Reading (DISTAR) instructional program, Open Court's basal reading program, the Success for All school restructuring program, the Reading Mastery reading program, and the Helping One Student To Succeed (HOSTS) structured mentoring program in language arts. Of these, quantitative research that directly addressed program effectiveness for ELLs could be found for only two: Success for All and Reading Recovery. This appendix provides brief descriptions of these programs and summarizes the research on their effectiveness with ELLs.
Reading Recovery is an early intervention program designed to reduce student reading failure. Originally developed by New Zealand psychologist Marie M. Clay, the program has been adapted for implementation in U.S. schools. The program targets those first grade students who constitute a school's lowest achievers in reading, and provides them with temporary (12 to 20 weeks), one-on-one tutoring during daily, 30 minute pull-out sessions. The goals of the program are (a) to bring these students to a reading level that is comparable to the level of the average first-grade readers in the school or classroom, and (b) to develop in them a self-improving system of reading and writing that allows for continued progress in the regular classroom.
The Reading Recovery program has three key components. One is the intensive, year-long inservice program that Reading Recovery teachers are required to participate in. Another is the use of an observational survey that has been developed to help identify low-achieving students for program participation, as well as to monitor their progress in the program. The third component is the highly structured framework within which tutoring sessions occur.
Regarding the latter, although instruction is adjusted to meet the needs of the individual child, Reading Recovery is nonetheless highly prescriptive in regard to how tutoring sessions are to be structured. After an initial 10-day period in which no new learning is introduced (during which the teacher instead focuses on supporting the child in becoming fluent and flexible with what he or she already knows), each tutoring session includes:
engaging the student in the rereading of previously read books;
student independent reading of a book introduced during the previous lesson, during which the teacher records (and later analyzes) the student's reading behavior with a kind of shorthand called a "running record;"
letter identification exercises, if necessary;
student writing and reading of his or her own sentences, during which the student's attention is called to hearing the sounds in words;
cutting-up these written sentences into individual words or phrases for the student to reassemble;
the introduction of a new book; and
the reading of the new book with teacher support.
Marie Clay concluded from her original research on Reading Recovery that it is "an effective programme for reducing the number of children with reading difficulties in New Zealand schools" (Clay, 1993, p.96). However, the program evaluations she conducted have been criticized as being methodologically flawed (for a review of these critiques, see Center, Wheldall, Freeman, Outhred, & McNaught, 1995; see also Shanahan, 1987). Furthermore, the relevance of Clay's findings to the question of Reading Recovery's effectiveness in U.S. schools is limited, due to differences in the program's implementation in New Zealand and the U.S., and due to differences in the educational contexts of these countries (Hiebert, 1994; Shanahan & Barr, 1995).
Nevertheless, studies conducted by the implementers of Reading Recovery in the U.S. have supported Clay's findings (e.g., DeFord, Pinnell, Lyons, & Place, 1990; Pinnell, Lyons, DeFord, Bryk, & Seltzer, 1994). These studies suggest that Reading Recovery, as implemented in the U.S., has substantial positive effects on student reading and writing ability by the end of the first grade, and that a portion of these effects are maintained through the fourth grade. However, secondary analyses of these studies suggest that they may greatly overestimate the effectiveness of Reading Recovery (Hiebert, 1994; Shanahan & Barr, 1995; Wasik & Slavin, 1993), a significant issue given the high cost of the program (Hiebert, 1994; Hiebert; 1996; Rasinski, 1995).78
Regarding the program's effectiveness with ELLs, program evaluations have been conducted of Reading Recovery's efficacy when provided to ELLs in the students' native language (Escamilla & Andrade, 1992; Kelly, Gomez-Valdez, Klein, & Neal, 1995). However, little research has been conducted on the program's effectiveness with ELLs when Reading Recovery tutoring is provided in English. Although Pinnell, Lyons, and Jones (1996) claim that, of the 33,243 children who were successfully exited from the program in the U.S. in 1994, 9% were ELLs who received Reading Recovery in English, Kelly, et al. (1995) is the only study that provides de-aggregated data on the effectiveness of English Reading Recovery for ELLs. These researchers found that ELLs receiving Reading Recovery in English were successfully exited from the program at a comparable rate to native-English speaking students, and also made comparable reading and writing gains. However, the effectiveness of Reading Recovery for these students in comparison to other forms of intervention is not addressed. A further shortcoming of Kelly, et al. (1995) is that the study provides no discussion of how the program was modified (if at all) to fit the needs of ELLs. Although Reading Recovery tutoring is intended to be tailored to fit the needs of individual students, nowhere in the scholarly literature were we able to find a discussion of what this entails for ELL students receiving Reading Recovery in English.
Success for All
Success for All is a school restructuring program that emphasizes prevention and early intervention for students in kindergarten through third grade who are at risk of early reading failure. Developed by Robert Slavin and his colleagues at John Hopkins University, the program is an attempt to
…use everything known about effective instruction for students at risk [in order] to direct all aspects of school and classroom organization toward the goal of preventing academic deficits from appearing in the first place, recognizing and intensively intervening with any deficits that do appear, and providing students with a rich and full curriculum to enable them to build on their firm foundation in basic skills. The commitment of Success for All is to do whatever it takes to see that every child makes it through third grade at or near grade level in reading and other basic skills, and then goes beyond this in the later grades. (Slavin, Karweit, Wasik, Madden, & Dolan, 1994, p.176)
The most salient aspects of the reading component of the program are:
Reading tutors: Certified teachers work individually with those children having the most difficulty learning to read. This is done during daily, 20-minute tutoring sessions. These sessions provide students with individually tailored instruction that is supportive of, and closely coordinated with, the regular reading curriculum.
Grouping: Although homeroom classes are heterogeneous, students are regrouped according to reading level for 90-minute reading classes. The size of these classes is reduced through the utilization of tutors as reading teachers.
Frequent assessment: Students are assessed at eight-week intervals to determine reading progress. This assessment is used to evaluate the needs of individual students in relation to tutoring and reading group assignment, as well as to identify students in need of other types of assistance to support their academic achievement (e.g., screening for vision or hearing problems).
Children's literature: At every grade level, reading periods begin with teacher oral readings of children's literature. Students are engaged in discussions about the story in order to enhance their understanding of the story, their listening and speaking vocabulary, and their knowledge of story structure. Additionally, through mid-first grade, children listen to, retell and dramatize stories in order to develop language and comprehension skills.
Reading Roots: Success for All's "Reading Roots" reading program is usually introduced in the second semester of kindergarten. The program utilizes mini-books with phonetically regular words and interesting stories for repeated student oral readings to partners and the teacher. Letters and letter sounds are introduced in a predetermined sequence and integrated into the context of words, sentences and stories.
Reading Wings: "Reading Wings" is the reading program employed when students attain the primer reading level (usually in the second semester of first grade). The district's basal series or tradebooks are used with cooperative learning activities that emphasize decoding practice, vocabulary building, reading comprehension skills and story-related writing. Reading Wings also emphasizes student home readings of self-selected books; students share their home readings through presentations, summaries, puppet shows and other formats.
Success for All also has a number of integral components that indirectly support student reading achievement:
Kindergarten and pre-kindergarten programs: Whenever possible, Success for All schools provide half-day pre-kindergarten and whole-day kindergarten programs. These programs are intended to give children a needed foundation for success in elementary school by providing a curriculum that supports oral language development, phonemic awareness and concepts of print.
Family support team: The family support team, a group composed of various school staff, has two primary functions. The first is to increase parents' involvement with the school and their children's learning. The second is to help families access needed social services when children have health or home problems that interfere with their academic performance (e.g., needing glasses or not receiving adequate sleep or nutrition).
Program facilitator: Each school has a program facilitator who works to oversee the operation of the Success for All model. Among other duties, the facilitator helps plan the program, assists teachers having difficulties, oversees the 8-week assessments, and coordinates the efforts of the family support team with those of the instructional staff.
Teacher training: Success for All emphasizes long-term professional development that includes both inservice focused on the implementation of the reading program, and the sharing of knowledge between teachers through professional collaboration.
Slavin and his associates have extensively evaluated Success for All's effectiveness. Utilizing quasi-experimental designs, these researchers have found Success for All to have significant, positive effects on student reading outcomes through the fifth grade (Slavin, Madden, Dolan, & Wasik, 1996). Furthermore, these positive effects are sustained as students exit Success for All schools and enter middle school; a comparison of treatment- and control-group cohorts shows the former to maintain higher reading outcomes through the sixth and seventh grades (Slavin, et al., 1996).
An independent evaluation of Success for All conducted by Smith, Ross and Casey (1996) also found the program to have positive effects. The purpose of the study was to conduct an independent examination of Success for All's effectiveness that was "separate from the program's developers and from the school district in which it was first implemented" (Smith, et al., 1996, p.329). Although the study's findings regarding the benefits of Success for All were not as strong and consistent as those found by the program's designers, the study nonetheless found the program to have generally positive effects on student reading outcomes.
Slavin, et al. (1996), Slavin and Yampolsky (1992), and Dianda and Flaherty (1995) have also found Success for All to have positive effects on the English reading outcomes of ELLs in English-only programs. To meet the needs of these students, Success for All was modified in two primary ways:
A cross-age tutoring program was implemented in which older, bilingual students read to, and with, kindergarten students who spoke the same first language as the tutor.
ESL instruction was focused on supporting student's success in the regular reading program. ESL teachers utilized the materials and techniques of the regular reading program to help students with specific difficulties.
Although the researchers acknowledge that Success for All is not a substitute for quality bilingual education, their findings demonstrate that when the program is modified in these ways, the benefits are even more significant than those found with native-English speakers.