University of California, BerkeleyGSE Home

how to apply faculty news events
programs courses research administration resources

prospective students
alumni & visitors
current students
faculty & visitors

Development and Maintenance of Two-Way Immersion Programs: Advice from Practitioners

Julie Sugarman and Elizabeth R. Howard,
Center for Applied Linguistics
(March 2001)

As an effective and increasingly popular educational approach, two-way immersion (TWI) programs have grown remarkably in the past 15 years. CREDE's "Two-Way Immersion" project, conducted at the Center for Applied Linguistics, has kept track of TWI growth and determined effective program implementation practices. Information on TWI programs is published online in the Directory of Two-Way Bilingual Immersion Programs in the United States (2000) at

The Directory provides detailed program and demographic information for 250+ programs that a) provide content and literacy in English and a minority language, b) integrate students from the two language groups for at least half of the school day, and c) have a roughly equal balance of language majority and language minority students. To collect this information, CREDE researchers sent out questionnaires to the primary contacts for these TWI programs (principals, bilingual coordinators, or teachers), which included open-ended issues, such as:

  1. What are the most important factors in the success of your program?
  2. What challenges have you faced while establishing your program? How have you dealt with them?
  3. What advice can you offer new two-way programs that are starting up?

This brief will summarize the responses given to these questions and provide recommendations supported by CREDE's research and technical assistance.


Most respondents suggested allotting at least one year to plan for the program. Specific suggestions included establishing a planning committee with representatives of the administration, teachers, parents, and community; and keeping the overall goals of the program and the long-range program design in mind throughout the planning process. During the planning phase, the following tasks need to be accomplished:

  • Studying the area's demographics and identifying the needs of parents and children in the community
  • Familiarizing staff, administration, and prospective parents with research on TWI and other relevant topics, such as second language acquisition, biliteracy development, and sheltered instruction
  • Choosing a program model, making curriculum decisions, and developing or translating materials if necessary
  • Choosing student assessment instruments and setting schedules for state-mandated and local tests
  • Establishing criteria for accepting students into the program
  • Selecting staff and planning staff development

Other activities are not essential but certainly aid in program design and implementation, such as:

  • Visiting and forming partnerships with other TWI sites to share ideas for curriculum, scheduling, and materials
  • Attending conferences, such as those sponsored by the National Association for Bilingual Education
  • Marketing the program in preschools, community centers, and at community events using professional brochures

Furthermore, successful programs incorporate ongoing follow-up and flexibility to improve and refine the original plan. Therefore, mechanisms are needed to evaluate the outcomes of planning decisions, so all stakeholders (teachers, administrators, parents) agree on the procedures to follow if part of the program does not work as planned.

Curriculum and Instruction

One basic decision all new TWI programs face is the selection of a program model–in particular, the amount of instruction delivered in each language at each grade level. Several factors to consider are the needs of the student population, the language capabilities of the teachers and support staff, the interests and concerns of parents, and the political climate in the community. Equally important is model fidelity–the extent to which all TWI teachers understand and comply with the model. Faithfully applying the selected model to classroom instruction "ensures articulation between grades and continuity, allowing for development of skills," according to one respondent. Regardless of the program model, it is recommended that the minority language be used for at least 50% of instructional time.

Within the classroom, separation of languages is essential: by time block, subject, day, teacher, or some combination. English is the predominant language in the U.S., so promoting the use of the minority language is often challenging, but can be done through after-school activities and homework help networks that communicate in the minority language.

Finally, it is important to invest time and money in procuring high quality materials in both English and the minority language. These should reflect a multicultural curriculum to promote the TWI goal of cross-cultural understanding.

Student Assessment and Program Evaluation

Respondents suggested the following guidelines for assessing and evaluating the students and the program. First, be mindful of state and district standards and testing requirements, and always set high academic standards for students of both language groups. Second, establish clear program goals, policies, and outcome objectives, and explain these to students, parents, and the larger school community. Finally, measure the success of these goals and objectives by using multiple criteria (e.g., standardized tests and teacher-developed assessment measures).

It is important to collect these measures longitudinally, so the development of students' abilities in various domains can be assessed accurately. It will take several years of data collection, however, to determine program effectiveness and student achievement, or to compare performance at a given grade level. Moreover, the program model may impact initial test scores. For example, in a 90/10 (90% of instruction in minority language, 10% in English) program, it would not be surprising to find low scores on standardized English assessments in the primary grades. In such a case, standardized English test scores from the upper elementary grades would be a better indicator of program effectiveness.


Teachers with a strong understanding of and commitment to the program's philosophy and goals are the backbone of TWI programs. They should be well trained in second language acquisition and sheltered instruction. If teachers with ESL or bilingual endorsements are in short supply, the district could offer credentialing classes. A teacher exchange program with other countries can also help locate teachers proficient in the minority language.

It is beneficial to identify at least two teachers for each grade level, and to have a teacher and an instructional assistant in each classroom. If bilingual specialty teachers (as for music or art) cannot be hired, bilingual instructional assistants can provide support in those monolingual English settings. Because of teacher turnover, it is ideal to hire teachers who are qualified to teach in both languages, so that they can provide instruction in either language as needed.

Prior to implementation, intensive staff development that covers the philosophy and theory of TWI education and effective teaching strategies should be provided. Professional development should be ongoing. Every year, the program must invest time and funds for staff to learn new teaching methods, review current research, design curricula, and plan for articulation. TWI staff should meet at least once a month to discuss program-specific issues.

If the program is a strand within a school, there are additional considerations. Maintaining staff cohesion across the TWI program and other school programs is critical. Providing all staff with frequent updates on the TWI program and helping TWI teachers integrate with the school community are ways to promote staff unity.

Size and Growth of the Program

Many respondents suggested starting a program at kindergarten (K), or K and first grade, and adding one grade level each year. It is important to recruit at least two classes of students at the K level, so if attrition occurs, there will be enough students for at least one class in the upper elementary grades. If maintaining a balance of language groups is a problem because of neighborhood demographics, consider becoming a magnet program, busing students from other areas, or establishing multi-age classrooms. It is also important to consider planning for a program that continues beyond the elementary level.

Because it is challenging to run a program-within-a-school, phasing the TWI program into the school site can help administrators reduce miscommunication and unequal allocation of resources. A program coordinator (full-time and on-site, if possible) can be instrumental in coordinating program policies, liaising with other programs and the administration, developing curriculum, working with parents, and so forth.

Many respondents also advise patience: planning and getting the support of school officials and parents, invaluable to program success, can take a year or longer. However, one respondent from a Navajo program cautioned that when languages are endangered, it is necessary to press ahead quickly for fear of language loss in a community.

Parental Involvement and District Support

Parents are vital to TWI program success, as supporters of their child and as advocates of the program. It is important to involve parents from the start, and to encourage them to volunteer in the classroom and learn as much as they can about TWI. Some schools offer language classes to parents to help them develop bilingualism along with their children. Many respondents suggested empowering parents as spokespeople to reach out to parents of prospective TWI students.

Parents and teachers should meet frequently to discuss program design and theory, performance expectations, and the children's progress. Although parents should be encouraged to express their concerns, one project director wrote, "Be fair and firm in establishing guidelines for this program and maintaining the program integrity," and recommended weighing parents' requests against what is possible or necessary for the program's overall well-being.

Administrative and community support were mentioned frequently as important factors in TWI success. Not every program exists in a highly supportive district, so practitioners should evaluate whether enough administrative and community support exists to sustain the program. As one respondent from a decade-old program wrote, "Carefully consider long-range implications and ponder long-term odds of program stability with or without 'system' support."

Representatives of the district, the site administration, and the community should participate in the planning process. This is the best way to gauge what the constituents will support in the future. Regardless of the level of enthusiasm, all parties should commit to helping the program succeed and to supporting it with similar resources given to other district programs. It is also helpful to hold regular meetings with these constituents to maintain open and honest communication.


When establishing a two-way immersion program, designing curricula, planning for assessment and evaluation, identifying and training teachers, recruiting and involving families, garnering district support, and fostering long-term growth are all tasks that implementers face in the planning stages and beyond. The number and complexity of these tasks may seem daunting, but the lessons learned from established TWI programs will facilitate the process for those who are just starting out.

For more information, contact Elizabeth Howard, Center for Applied Linguistics, 4646 40th Street NW, Washington, DC 20016, (202) 362-0700. For a description of this CREDE project and links to other documents, visit or

This work is supported under the Educational Research and Development Center Program (Cooperative Agreement No. R306A60001-96), administered by the Office of Educational Research and Improvement (OERI), U.S. Deparment of Education. The findings and opinions expressed here do not necessarily reflect the position or policies of OERI.

For a printable version of this Practitioner Brief in Adobe's PDF format, click here.
(Download and install Adobe's free Acrobat Reader here.)