A r c h i v e d  I n f o r m a t i o n

Model Strategies in Bilingual Education: Professional Development - 1995

Descubriendo La Lectura

Tucson Unified School District and the University of Arizona
Tucson, Arizona

Goals and Context

Tucson Unified School District (TUSD) serves approximately 60,000 students, of whom about 40 percent are Hispanic. The district has adopted a policy of home-language maintenance to ensure the academic success of students who have limited English proficiency while building a strong foundation for learning English. This district policy provides a welcoming environment for a new program developed by the TUSD Title I staff called Descubriendo La Lectura (DLL). DLL is a Spanish-language reconstruction of the Reading Recovery (RR) program, which has a well-documented record of supporting the early success of at-risk beginning readers in English. DLL was developed to provide strong support for early Spanish literacy by using the same proven enrichment strategy with Spanish speakers that was being used with English speakers.

TUSD Title I staff became interested in a Spanish version of RR because one of their goals was to support children who were having difficulty learning to read, and many in their district learned to read first in Spanish. Title I specialists felt that Reading Recovery's premise that the most powerful teaching builds on children's competencies was consistent with the district's policy of using a student's primary language for initial literacy instruction. TUSD hired a Teacher Leader trained at the Reading Recovery center at The Ohio State University to begin training local bilingual teachers in the basic approach. These teachers then became part of a team that created DLL. The team met some early resistance; dissemination of Reading Recovery is tightly controlled to ensure fidelity to its core concepts and strategies and especially to its insistence on intensive professional development. However, the concerns were resolved, and since 1988, TUSD Title I has been the major player in DLL development.

One of the Title I teachers attended Texas Women's University in 1990-91 to be trained as a Teacher Leader, and she has been instrumental in the teacher training program ever since. In the 1991-92 school year, the first full training class for DLL teachers was conducted, and the TUSD Bilingual Education Department sponsored a second bilingual teacher to receive the Teacher Leader training, which by then was offered at the University of Arizona (UA). During this time, Texas and Chicago districts joined the DLL collaborative. Fifty students received instruction from DLL-trained teachers, and the U.S. Department of Education's Office of Educational Research and Improvement (OERI) funded research on the DLL program.

Project Description

Participants. Only certified bilingual teachers with at least three years of experience in a bilingual primary classroom and/or resource position are accepted for DLL training. They commit to completing a one-year intensive training experience, for which they may pay for and receive graduate credit. About half of the present DLL trainees are working toward a master's degree. An additional, more comprehensive training program for Reading Recovery and DLL Teacher Leader candidates is offered at UA by a Teacher Leader trainer supported by a consortium of Tucson-area districts for a year's study at Ohio State--a requirement of Center-approved RR trainers. For bilingual teachers who have had RR training and who want to become DLL teachers, TUSD and UA offer a "bridging" course that uses RR as a foundation for DLL.

Target students. The students who receive DLL instruction are low achieving children in first-grade Spanish bilingual classrooms. When they exit the DLL program--meaning they have caught up to grade-level expectations--their reading instruction continues in Spanish, adding instruction in English when they have a strong literacy base. TUSD's policy is to maintain literacy in both languages, although other sites will be using a bilingual transition model that shifts to English at the intermediate grade levels. The Spanish Observation Survey used for DLL student selection consists of six observational tasks that provide a profile of a student's reading repertoire. Students scoring in the lowest 20 percent are recommended for DLL, although the program is not yet able to serve all identified children.

Training. The three-year training of DLL teachers is most intensive in the first year, during which teachers attend a three-hour training session once a week held at a local Title I mobile unit. They also observe and/or participate once a week in a "Behind the Glass" lesson where one teacher tutors a student while the trainer and other teachers watch from behind a one-way mirror. As the lesson progresses, the observers formulate questions and hypotheses about the instructional decisions and practices they see. After the lesson, the tutor, trainer, and observers discuss the application of DLL principles. Like RR, the focus of DLL training is to expand teachers' knowledge of cognition, instructional repertoire, and ability to perceive and analyze a student's reading behavior accurately. "Behind the Glass" sessions provide opportunities to learn how to apply new skills and concepts effectively. In the second year, trainees observe and/or participate in a minimum of six "Behind the Glass" sessions during the school year. In addition, the Teacher Leaders conduct on-site visits and provide coaching and guidance on an ongoing basis. What makes RR such a powerful staff development model is that teachers learn how to be informed observers, practice what they have learned, and analyze each lesson carefully with the aim of improving instruction. They become keen observers of the children they teach, and they learn to build on the children's individual strengths as they embark on the path to becoming effective readers and writers.

Development of DLL Teacher Leader training at the University of Arizona has supported expansion of the program throughout the United States. Several districts in Arizona helped fund the education of a Teacher Leader trainer at Ohio State to staff the UA program.

Implementation. DLL implementation takes two forms in Tucson. Both involve teachers working with four first-grade students daily for part of the school day. In one arrangement, two teachers share a first-grade classroom, one tutoring individuals while the other teaches the class. In the other arrangement, the teacher provides bilingual assistance to other (non-DLL) students during the other part of the day.

Funding and support. DLL was conceived by TUSD Title I personnel, and Title I monies have been the major source of funding. Now others, such as the bilingual education department, have collaborated to sponsor teacher training, purchase trade books in Spanish, and underwrite general implementation costs. A consortium of school districts supports Teacher Leader training.

Project Outcomes

Like RR, DLL operates under rules of the National Diffusion Network, which requires evaluation overseen by Ohio State. DLL has also received a grant from the OERI to conduct further independent research and development. The early results are promising: In 1992, participating DLL students who were discontinued from the program on the basis of their strong achievement outperformed comparison groups on a battery of reading tests. Teacher accountability is an integral part of DLL--teacher success is measured in terms of the rate at which students exit the program reading at or above grade level. By this standard, DLL training has so far been almost entirely successful. The careful evaluation of the program is ongoing, and the DLL proponents are hopeful that, as the reputation of the DLL approach gains momentum, a greater number of educators will be inspired by the wisdom of the approach.

A DLL Teacher Leader commented that while evaluation of teachers beyond looking at student outcomes is not well documented, self-evaluation and evaluative peer feedback are built into the training approach. The "Behind the Glass" sessions provide particularly important opportunities to analyze problems and discuss strategies for improving a teacher's performance.

Lessons from Experience

One of the biggest challenges for the expansion of the DLL program is that it does not fit neatly into existing school programs. First, as a pull-out approach, it runs against the tide toward mainstreaming. Second, labor-intensive individual tutoring does not seem cost-effective to those whose accounting focuses on short-term benefits and whose budgets are very tight. Because of these perceived obstacles, DLL program developers so far have been able to convince only a handful of principals to adopt the model. As might be expected, the principals are more prone to support the model if Title I resource teachers are used because their work often involves serving fewer children in pull-out sessions anyway. One DLL Teacher Leader reported that many more teachers want DLL training than there are sites willing to adopt the model. Project leaders cultivate opportunities for expansion by publicizing the documented successes of the DLL approach.

The DLL program has worked well in TUSD because of the district's strong policy of maintaining bilingual literacy. Other sites that shift instruction entirely to English before Spanish literacy is fully developed have found it difficult to realize the full benefits of investing in DLL. To bolster support for programs that promote long-term student success, DLL promoters have learned to cite recent research in early language acquisition, which shows the correlation between primary language development and second language acquisition, and to disseminate evidence from their research on DLL widely.
-###-

[Cooperative Learning in Bilingual Settings and Teachers' Learning Community Center] [Table of Contents] [Educational Leadership in Bilingual Education]