Two-way dual language immersion programs — which bring native English speakers and students who speak another language at home together in the same classroom — may modestly reduce school segregation, according to a new study led by a Penn State College of Education researcher.
In a large-scale analysis of elementary schools in the Los Angeles Unified School District, Erica Frankenberg and her colleagues found that dual-language programs can, in some cases, disrupt the tight link between residential segregation and school enrollment patterns. The findings, published in Social Science Research, suggest that program design — particularly how students are admitted and whether enrollment extends beyond neighborhood boundaries — plays a meaningful role in shaping who learns together.
“Most public schools are still deeply shaped by residential segregation,” said Frankenberg, professor of education and of demography, and associate director of the Social Science Research Institute. “The question is whether there are voluntary, educationally beneficial strategies that can help loosen that link. Dual-language immersion is one possibility.”
In two-way dual language (TWDL) programs, students receive instruction in both languages, with the goal of developing bilingualism and biliteracy. Unlike traditional bilingual models that focus primarily on English learners, TWDL programs are designed so both groups learn with and from one another.
Beyond cultural and academic benefits, Frankenberg explained, bilingualism is increasingly viewed as an economic asset. Students who develop fluency in multiple languages may gain advantages in a global workforce where employers value cross-cultural communication skills and multilingual competence.
“Families are often drawn to these programs because of the long-term benefits,” she said. “There’s a recognition that bilingualism can open doors — academically, professionally and socially.”
To examine how these programs shape school demographics, the research team analyzed enrollment records for more than 221,000 elementary students across 498 neighborhood-based schools in Los Angeles. The sample included 146 dual-language programs spanning 142 schools. The researchers combined student-level enrollment data and attendance zone boundaries with U.S. Census and American Community Survey data to compare who lives within a school’s attendance boundary with who actually enrolls. They also identified whether students attended their assigned neighborhood school or enrolled from outside the attendance zone, determining in- or out-of-zone status for roughly three-quarters of the student sample.
The results were nuanced.
Schools with dual-language programs tended to be located in neighborhoods with high proportions of Hispanic residents, non-English-speaking households and low-income families — communities that have historically been the intended beneficiaries of bilingual education.
At the same time, many programs enrolled students from outside their attendance zones. Those out-of-zone students were, on average, more economically advantaged and more likely to come from English-only households. That enrollment pattern modestly increased the racial, linguistic and socioeconomic diversity of the programs compared to what would be expected if they enrolled only neighborhood students.
In practical terms, this meant that in neighborhoods where Hispanic and low-income families made up a large majority of residents, dual-language programs often enrolled a somewhat more mixed student body than the surrounding attendance boundary alone would predict. In some cases, out-of-zone enrollment brought in students whose racial or economic backgrounds differed from those of neighborhood peers, slightly narrowing gaps that would otherwise have remained unchanged.
“We see modest but suggestive evidence that these programs can help overcome some of the effects of residential segregation,” Frankenberg said. “It’s not a dramatic transformation. But it does show that structure matters.”
In Los Angeles, most dual-language programs operate within traditional neighborhood schools. Students living in the attendance zone receive priority enrollment, and families from outside the zone may apply if seats remain. Transportation is generally not provided.
“That design has consequences,” Frankenberg said. “If families need to know about the program, apply on time and provide their own transportation, that shapes who is able to participate.”
The study also identified inequities in participation. The researchers found that Black students were under-enrolled in some dual-language programs relative to their representation in attendance zones. In addition, Black students enrolling from outside attendance zones traveled farther, on average, than other student groups — suggesting that access barriers may be unevenly distributed.
“If districts want equitable participation, they have to think intentionally about outreach,” Frankenberg said. “That includes how information is shared, who feels invited to apply and whether transportation or application requirements create barriers.”
The type of language offered also influenced enrollment patterns. Spanish-language programs — the most common type in Los Angeles — tended to reflect the demographics of their surrounding neighborhoods more closely. Programs offering languages such as Mandarin or Korean were more likely to draw families from outside attendance zones, including some students from more economically advantaged backgrounds.
“There are perceptions about the status of certain languages,” Frankenberg said. “That can shape which families see a program as desirable and whether it becomes a tool for integration or stratification.”
Frankenberg cautioned that dual-language immersion is not a cure-all for segregation. Broader structural forces — including housing patterns and district boundary lines — continue to shape public education.
“These programs alone won’t solve segregation,” she said. “But they can be one piece of a broader strategy. They’re voluntary options for families, they offer real educational benefits, and they can, under the right conditions, modestly shift enrollment patterns.”
Ultimately, she said, the study highlights the importance of program design in determining outcomes.
“We can’t assume that school choice automatically leads to integration,” Frankenberg said. “But dual-language programs give us an opportunity to examine how educational innovation and equity goals might align.”
The study was co-authored by Sarah Asson, who earned her doctorate in education policy studies from Penn State in 2024 and is now a researcher at Education Northwest; Clémence Darriet of the California Department of Education; Lucrecia Santibañez of the University of California, Los Angeles; Francesca López of the University of Wisconsin–Madison; and Claudia Cervantes-Soon of Arizona State University. The research was supported by the Spencer Foundation.