Examining the Validity of Classifications from an English Language Proficiency Assessment for English Language Learners and Native English Speakers in Fifth Grade.

摘要

English language proficiency assessments are used to report English Language Learner (ELL) proficiency and progress, and classifications from these assessments help determine what educational services students receive. While classification validity is routinely reported by test vendors, empirical evidence is rarely used to justify the use of a certain classification model or scheme. In this study, fifth grade ELL and native English speaker (non-ELL) performance data on a State's Standards-Based Achievement Assessment (SBAA) and the State's English Language Proficiency Assessment (ELPA) are used to evaluate "proficient" and "non-proficient" classifications on the ELPA. Findings indicate that the currently-used conjunctive model, compared to compensatory and mixed models, creates classifications that are the least congruent with other sources of proficiency evidence for ELL and non-ELL students. Findings from this study can directly inform how policy, practice and research communities define and verify ELPA classifications for use in high-stakes decision making.