High school students may sink rather than swim if it is too easy to exit English-learner status. By contrast, when exit standards make like Goldilocks and increase to a level of difficulty that is just about right, the transition from English-learner to “fluent English proficient” status has little impact on subsequent achievement levels or graduation rates.
So suggest the results of a new study by Joseph P. Robinson-Cimpian, an assistant professor of quantitative and evaluative research methodologies at the University of Illinois at Urbana-Champaign and Karen D. Thompson, an assistant professor of education at Oregon State University.
Robinson-Cimpian and Thompson say their findings have implications for policymakers who may be adjusting reclassification criteria to align with new common-core standards and exams.
“Policy considerations tend to focus on increasing the speed with which [English-learners] are reclassified,” the authors write in a paper presented at the fall conference of the Association for Public Policy Analysis & Management held earlier this month in Albuquerque, N. M. “While attaining English proficiency is indeed important, it is also important to keep in mind that [English-learners and reclassified fluent English proficient students] often receive different instructional bundles (e.g., services, settings, teachers, peers), and thus it is imperative to evaluate reclassification effects on academic outcomes.”
For the study, the researchers analyzed 609,431 student data records from Hispanic English-learners enrolled in the Los Angeles Unified School District, which is home to the nation’s largest population of ELLs. The study included students in grades 3 through 10 who were enrolled in the district for at least three consecutive years between 2003 and 2010.
During that period, in the fall of the 2006-07 school year, the California Department of Education rescaled the state’s test of English-language proficiency (known as the CELDT) to make the scoring tougher. Although the test itself didn’t change, the change to its scoring made it more difficult for English-learners to reach the threshold for reclassification to “fluent English proficient” status. CELDT results are among the criteria used to determine whether students are fluent enough to exit English-learner status.
The researchers compared what happened to two sets of students: those who were reclassified “fluent " because they scored just at or above the proficiency line and those who remained in the “English-learner” category because their scores were a hair short of that line. These two sets of students were pretty much identical except for the fact that one group happened to fall just over the proficiency line and one group happened to fall just under it. The researchers did this analysis for students who were enrolled before the reclassification standards changed. For comparison’s sake, they repeated the analysis for students who were enrolled after the standards got tougher.
Prior to the 2006-07 changes, high school students who met the reclassification standards by the skins of their teeth were 11 percentage points less likely to graduate than their otherwise similar peers who remained English-learners. Exited high school students also earned lower English/language arts achievement scores than their otherwise similar peers: If the reclassified student scored in the 50th percentile, the otherwise similar student who remained an English-learner would score in the 57th percentile.
“These negative effects suggest that the reclassification criteria were not stringent enough and that [English-learners] were being reclassified as [English proficient] when they were still benefitting more from the [English-learner] instructional settings and services than the reclassified settings and services,” Robinson-Cimpian and Thompson write.
Those negative outcomes evaporated in 2007, when it got tougher to exit English-learner status. From that point onward, reclassification unto itself no longer helped or harmed high school students.
Interestingly, researchers found that younger students were not affected by the 2007 change. Both before and after the standards rose, academic outcomes were similar for reclassified students and their otherwise similar ELL peers.
One possible explanation is that reclassification standards and thresholds might have been appropriate for younger students, even prior to the 2007 change, the researchers proposed. By contrast, the high school standards might have been too low. Robinson-Cimpian and Thompson also offered evidence that a student’s life may change more dramatically if she is reclassified in high school than if she is reclassified in an earlier grade. That’s because high school-aged English-learners are often taught in classes separate from their English-proficient peers. By contrast, the researchers note that a 2011 investigation by the U.S. Department of Education’s Office for Civil Rights found that younger ELLs did not always receive the specialized services mandated by district policy.
Robinson-Cimpian and Thompson urge policymakers to carefully consider the needs of English-language learners as they roll out common-core exams.
California’s reclassification criteria is currently in a state of flux due to the forthcoming common-core tests and a redesign of the CELDT to align with new English-language development standards.
“Policymakers must determine what an appropriate English-proficient performance standard is on new [Common Core State Standards]-related [English language proficiency] assessments,” they write. “In addition, policymakers must also consider whether to include results from [Common Core State Standards] content-area assessments as part of the reclassification criteria, and if so, determine appropriate thresholds on those assessments. Meanwhile, given the increased rigor of curriculum and assessments under the Common Core State Standards and Next Generation Science Standards, researchers have highlighted the need for specific scaffolds so that English learners can access and be successful with the new curriculum and assessments.”