������Ƶ

One of the biggest lessons we’ve learned from the No Child Left Behind era of American education policy is that end-of-course assessments matter a great deal. Of course, this isn’t really a revelation. It’s implicit in the project of standards-based reform. If we set clear guidelines about what students need to know and do, and then assess their ability to meet those benchmarks, that ought to guide what happens in classrooms and schools.

In a new report released today, I explored how this logic works for American dual language learners. The report, Chaos for Dual Language Learners: An Examination of State Policies for Exiting Children from Language Services in the PreK–3rd Grades, includes a 50-state scan of states’ policies for determining when DLLs no longer need language support services.

These “reclassification” policies represent states’ attempts to answer a key question: When are DLLs fully proficient in academic English? In Chaos for Dual Language Learners, I found that states have come up with at least several dozen answers. The variance in states’ policies takes a number of forms. States use different:

• assessments for measuring English proficiency;
• benchmarks to determine when students have reached proficiency; and
• procedures for incorporating additional data into reclassification decisions.

Given that researchers are improving their knowledge of what language supports DLLs need—and for how long—this proliferation of inconsistent reclassification standards seems to be arbitrary. Why should a DLL in Michigan face different standards than a DLL in South Dakota? This also undermines efforts to compare state data on serving these students: A DLL who is reclassified and incorporated into accountability systems as a former-DLL in Mississippi might remain classified in Alabama.

Most importantly, however, just as with end-of-course academic assessments, states’ end-of-language-support policies influence DLLs’ educational experiences throughout their time at school. No state has a robust, coherent system for providing language supports to its DLLs. Varying reclassification policies between states make it even harder for policymakers and educators to build support systems that would harmonize with how other states serve these students.

Fortunately, two state consortia are working to build new English language proficiency assessments and standards that would go towards developing a “common definition” of DLL. This is exciting, since, :

Reclassification policies are a key part of the project of knitting this mess into a coherent system. As states work towards a common system for determining when DLLs no longer need language support services, they are also taking the first step towards a more coherent approach to DLLs’ education.

I should note that this brief is, in conjunction with the publication of , part of a concerted ������Ƶ effort to focus more attention on DLLs and the policies that affect them. So: Stay tuned for more work along these lines.”