Convergent validity of the New York State English as a Second Language Achievement Test and the Woodcock-Munoz Language Survey - Revised in a sample of English Language Learners.

摘要

Many states rapidly changed their English language assessment policies in response to requirements of the No Child Left Behind Act of 2001, yet they have been left without the time, expertise and resources to systematically evaluate and document validity issues raised by the use of new assessment instruments. In particular, there is a dearth of research examining criterion-related validity. The present researcher evaluates the comparability of the New York State English as a Second Language Achievement Test (NYSESLAT) and the Woodcock-Munoz Language Survey--Revised (WMLS-R) across broad, written and oral language domains. Both English Language Proficiency (ELP) tests purport to measure the language skills necessary to succeed academically. The sample consisted of 50 English Language Learners (ELLs) from an elementary school in Port Chester, NY. Results found moderate to strong correlations across language modalities, but poor correspondence between classifications of proficiency. The lack of comparability between proficiency classifications suggests that one, or perhaps both, instruments do not meet their stated goal of measuring academic ELP. In addition, classification discrepancies call into question the use of the NYSESLAT as the sole criterion for redesignation decisions in New York State. The results of this study are also discussed in relation to NCLB-mandated Adequate Yearly Progress (AYP) performance targets. Directions for future research are suggested, and practical implications for school-based practitioners working with ELLs are highlighted.