Imagine two students entering kindergarten: Rosi and Maya. Both have grown up in families that speak a non-English language around the house. However, both come to their new classrooms with relatively developed English skills, perhaps due to experiences in high-quality early education and pre-K programs or exposure to English in other settings or from other family members.
At this point, federal law legally requires their school to screen for and identify whether they are “English learners” (ELs), students who will receive specialized English language services. So, school staff assess both Rosi and Maya in English. Their scores are nearly identical. However, Rosi scores just barely above the threshold to be considered “initially fluent” in English. Maya scores just barely below it. As a result, though both are emerging in their multilingualism, Rosi becomes a “mainstream” student, the same as her monolingual, English-speaking peers. In contrast, Maya becomes an “EL,” which guarantees her a host of educational services and treatments.
Who lucks out? In theory, the EL designation should help students like Maya over time. Since the 1960s, a long legacy of civil rights legislation and court rulings at the federal level make clear that U.S. schools must serve ELs in equitable ways that support academic success. Over the decades, states and districts have responded by developing systems of identification, assessment, and specialized services intended to support ELs in the pursuit of that goal. Some research suggests that several of these strategies can benefit ELs. For example, many schools make use of specially trained teachers, dedicated classes for English development, instructional “scaffolds” (such as graphic organizers and visual aids), and the integration of oral language and vocabulary development into mainstream teaching.
And yet, despite these efforts, the EL label can be a double-edged sword. Other studies suggest that well-intentioned differences in programming for ELs can also negatively impact students’ perceived status and their integration as learners. EL classification, and the services that come with it, can lead to lower academic expectations and carry a stigma from teachers and peers who associate the label with less able, motivated, and socially integrated students — a mindset that ELs’ can themselves internalize, particularly at the secondary level. Furthermore, some instructional models can segregate ELs from their monolingual English-speaking peers. English as a second language (ESL) “pull-out” models can displace ELs’ time in mainstream, academic instruction while other programs can track students into self-contained “ESL ghettos,” as ELL expert Beatriz Arias puts it.
These consequences — while unintended — have potential to undermine the whole point of the EL label. This reality is not lost on many EL parents. Mario Koran, Voice of San Diego reporter and New America California fellow, recently wrote in an email that this theme has increasingly emerged in his conversations with dozens of California families:
"A good number of parents have expressed concern about English learner as a label, and take steps to avoid it, even if their kids should technically be classified. The most connected and resourced parents sort of hide the fact that the child speaks Spanish at home. They don't list it on the initial survey everyone has to fill out. Their reasoning is that once children are listed as ‘English learners,’ they get caught in a system that works against them."
Now, a new analysis from University of Oregon researcher Ilana Umansky sheds further light on the potential for this damning irony. Umansky examined the longitudinal data on students from a large, urban school district in California in scenarios similar to the Rosi and Maya example. That is, the study “takes advantage of a natural experiment that occurs at the cusp of [initially fluent English] classification” that yields “essentially random assignment.” She refers to students at the threshold (like Maya) who are classified as ELs as “cusp” students. For these students, the study found that EL classification negatively impacts the reading and math test scores by second grade. A substantial gap continues to grow slowly throughout elementary and secondary school. It was better not to be an “EL.”
Some evidence suggested that bilingual instruction can buffer students from this effect. To distill this insight, Umansky disaggregated data by the district’s four programs, English immersion with ESL pull-out and three bilingual models: transitional bilingual, developmental bilingual, and two-way dual immersion. (Click here for more on the differences between instructional models.) In contrast to the negative impact of being an EL in English immersion programs, which served around sixty percent of the district’s ELs, Umansky found that being an EL had no deleterious effect on students in two of the three bilingual models, transitional and developmental. She postulates that this could reflect the fact that, unlike most mainstream English settings, ELs constitute the vast majority of the students in these bilingual models. She hypothesizes that ELs’ may be less stigmatized in these settings and benefit from teachers who share their cultural and linguistic background.
But, it’s not just a simple equation of “bilingual = better.” Because the two-way dual immersion model — which integrates monolingual English speakers and native speakers of the target language — did not protect “cusp” students from the detrimental effects of the EL label; while not statistically significant, this data more closely paralleled the patterns of English immersion. It’s a somewhat surprising finding given the buzz of recent research on dual immersion’s promising potential. Umansky speculates that certain “status hierarchies” may favor English-dominant students even within dual immersion programs. But the explanation is not altogether satisfying, raising questions for further research.
There are several other cautions to keep in mind. This study should not be taken to mean that the EL label is all bad for all students, all the time. The impact of EL classification depends on a complex web of factors: differences in district and school policies, student demographics, and other social or cultural contexts. For example, a recent study underscores this local variability, finding that EL services in some districts correlated with higher graduation rates but, in other districts, correlated with lower ones. In California’s landscape, the English immersion model of instruction is dominant due to an English-only law passed in 1998; parents can only opt their children into bilingual programs through a waiver. So, being an EL in states where bilingual models are more prevalent (or even mandated, such as in New York or Texas) will affect how the EL label impacts students. Moreover, Umansky only studied “cusp” students — in this case, multilingual children entering kindergarten with middle- to relatively high-level English proficiencies. Findings can’t necessarily be generalized to all ELs, especially those with low English levels, who may be more likely to benefit from EL status.
But even with these caveats, the new findings are no doubt troubling. While much research has focused on exiting students from the EL status, this study casts new light on the benefit of putting the EL label on young children in the first place. As the study states, our “current system of classifying students learning English is intended to avoid inequity in educational opportunity…Yet, this study finds that for some students, EL classification may in fact be contributing to educational inequity.” That is the very definition of counterproductive. For the sake of the millions of potential ELs entering the U.S. school system — and the millions of dollars targeted to help them — local leaders should press into uncomfortable questions of whether their status quo actually “serves” their ELs or not.
This post is part of New America’s Dual Language Learners National Work Group. Click here for more information on this team's work. To subscribe to the biweekly newsletter, click here, enter your contact information, and select "DLL National Work Group Newsletter."