iReasearch学术科研网

摘要：Test accommodations have been proposed to help overcome the unfair challenges faced by English Language Learners (ELLs) due to their relatively low English proficiency. A test accommodation is regarded as effective when it improves the test performance of ELLs. However, this improvement raises the question of whether such accommodations give ELLs an unfair advantage. One criterion used in determining a test accommodation's fairness is that it should only remove the disadvantage that ELLs face in regard to their low language proficiency, without giving ELLs any additional advantages. This criterion is met when the test accommodation does not improve the test performance of the non-ELLs when the same accommodation is applied to them. To determine the fairness and, thus, the validity of test accommodations for ELLs, a meta-analysis using hierarchical linear modeling was conducted to compare the effects of test accommodations on the test performance of ELLs and on that of non-ELLs. The results indicated that test accommodations improved ELLs' test performance by about 0.156 standard deviation units but did not discernibly influence the test performance of non-ELLs. This meta-analysis, therefore, constitutes evidence to support the fairness and validity of providing test accommodations for ELLs. Adapted from the source document