Among the most important components of a high-quality early childhood education classroom is the quantity of responsive interactions between teacher and the children in their care. These interactions, called conversational turns, play a critical role in building young children’s brains, boosting their social-emotional development, and giving them the preliteracy skills they need to learn how to read. Unfortunately, dual language learners may be missing out on these positive outcomes associated with conversational turns.

That’s according to new analysis of data from LENA’s “talk pedometer” technology, which suggests that toddler and preschool classrooms across the U.S. provide significantly fewer conversational turns for dual language learners than for their monolingual, English-speaking peers.

“In toddler and preschool classrooms where teachers don’t speak a child’s heritage language, we are seeing discrepancies in language interactions between children who speak heritage languages and children who only speak English,” said Dr. Jill Gilkerson, Chief Research and Evaluation Officer at LENA. “When it comes to addressing inequitable learning opportunities, promoting language development for children who use heritage languages is among the most important things we can do.”

A mismatch between science and practice around dual language learners

The science around early childhood development and multilingualism couldn’t be clearer:

• Children are perfectly wired to learn multiple languages simultaneously from birth.
• Multilingualism is an asset, not a deficit. The concept of language as a “barrier” is, simply put, wrongheaded and harmful.
• Children learning multiple languages show benefits in their cognitive abilities, executive function skills, and social-emotional development.
• Children’s English language development benefits even when their caregivers talk with them in a language other than English.
• One child’s bilingualism is not an impediment to another child’s learning — in fact, it’s quite the opposite. Children from monolingual families benefit as well when they interact with multilingual peers.

The science may be clear, but is it guiding real-world practice?

Dr. Dina Castro, director of the Boston University Institute for Early Childhood Well-Being and an authority on young dual language learners, has a clear-cut answer: No. In a webinar hosted by The Hunt Institute, she said, “It seems like the practices and policies in early childhood care and education are not necessarily supporting or following the science in the way that it should.”

Or as the U.S. Department of Health and Human Services and the U.S. Department of Education put it in a joint policy statement about dual language learners: “[T]here is a mismatch between the learning experiences these children need to meet their potential, and the quality of experiences they are currently receiving.”

The new data analysis from LENA’s researchers paints a clearer picture of just how those current practices and policies are failing DLLs. Most studies examining language development in school have been conducted in elementary settings (at which point students who are learning multiple languages are usually referred to as English learners). Combine that with the fact that most research on language development in preschool classrooms has focused on monolingual English speakers, and a hole in the research becomes apparent.

We need to know more about how dual language learners experience early childhood education classrooms, especially as they compose a greater and greater part of the U.S. population. Approximately one in three children aged 0-5 live in homes where at least one caregiver uses a heritage language(s), often alongside English. In some states, such as California, as many as one in two children who attend publicly funded early education programs, such as Head Start, are DLLs.

The data: DLLs in early childhood classrooms

For the present analysis, LENA’s researchers focused on 322 daylong audio recordings from 37 toddler and preschool classrooms. To be included, a classroom’s teachers had to be monolingual English speakers, and the class had to consist of both monolingual English-speaking children and dual language learners.

The researchers found that DLLs experienced significantly fewer conversational turns (-5.3 turns per hour) and vocalized significantly less (-29.2 vocalizations per hour) than their monolingual peers in the same classroom. In toddler rooms, DLLs experienced marginally lower conversational turn rates (-2.5). Our analysis does not allow us to know exactly what is causing these inequities, but other research suggests as solutions the expansion of access to professional development focused on DLL support and greater investment in the recruitment of multilingual educators.

The gap was widest in preschool classrooms, where DLLs experienced 7.6 fewer turns and produced 39.7 fewer vocalizations per hour.

The differences were especially stark in classrooms with just a few DLLs, where the majority of the children were monolingual English speakers. Around two-thirds of the classrooms analyzed included just one or two DLLs. On average, those one or two children experienced 6.7 fewer conversational turns per hour than their monolingual classmates. In the remaining rooms, where DLLs made up a larger percentage of total children, the average difference was smaller, at 2.6 turns per hour. Again, our analysis does not lend itself to identifying causal factors, but it’s possible that schools with a low proportion of DLLs do not adequately invest in professional development around linguistic responsiveness and inclusivity.

Seventeen states across the U.S., along with Washington, D.C., were represented. Among the DLLs in the sample, the most common heritage language was Spanish, while other children were learning English and one of Arabic, French, Chinese, Hmong, or Portuguese. Looking at both groups combined (monolingual and DLL), the majority of children were identified as African American (30%), white (24%), or multiracial (20%). Compared to the overall sample, a higher proportion of DLLs were identified as Hispanic (29%) or Asian (20%).

Where do we go from here?: Creating more inclusive classrooms

On a national scale, long-term academic outcomes for DLLs lag behind those of their monolingual peers, pointing to a need for greater support and training for teachers, greater linguistic diversity in the workforce, continued rebuttals of misguided myths, and more inclusive, linguistically responsive curriculums and assessments. And, as LENA’s analysis of language environments in early childhood education classrooms suggests, these long-term outcomes also point to a need for greater emphasis on equitable experiences for linguistically minoritized children in early childhood education settings.

It’s important to remember that each child’s individual experience of the classroom is unique.

DLLs are far from a homogeneous group, and one DLL’s experiences with English and with their heritage language(s) are sure to be different from another’s. While the current data analysis reveals a troubling trend in aggregate, it’s vital that any proposed solutions account for children’s individual experiences.

Part of what makes LENA’s technology unique is its capacity to deliver insights into those unique experiences. The question is: How do we translate those insights into real-world benefits for dual language learners and the educators who support them?

• Research-informed policy recommendations must be considered and implemented. As one report has put it, “Educational equity for DLLs requires including these students in the design and implementation of all early education programs.” Research suggests that Head Start’s Program Performance Standards for DLLs have yielded great outcomes for DLLs.
• In advance of slow-moving policy improvements, professional development opportunities must be made available to early childhood educators in the here and now. For example, the Institute for Racial Equity and Excellence’s Language Justice Initiative offers a Multilingual Learner Teaching Certificate, and Boston University’s Nuestros Niños project will offer an academic certificate in Teaching Young Dual Language Learners.
• Research must continue to be conducted on the instructional practices that lead to the best early learning experiences and outcomes for DLLs. First 5 California’s Dual Language Learner Pilot Study is a model example.

In addition, returning to the implications of the current data analysis, specialized professional development programs like LENA Grow can help educators develop data-informed strategies for better connecting with each and every child in their care. LENA Grow puts the “talk pedometer” technology into the hands of teachers themselves. Combining child-specific data feedback with weekly coaching enables educators to develop strategies for increasing their responsive interactions with the children who need them most.