English language learners, the Stanford Achievement Test, and perceptions regarding the effectiveness of testing accommodations: A study of eighth graders

The numbers of English language learners (ELLs) in public school systems is increasing at a rapid rate. Federal legislation requires that states assess ELLs in a valid and reliable manner in academic content areas. ELLs are participating in standardized tests before they have the opportunity to acquire proficiency in English, and they are subject to the same high stakes consequences associated with the results as their native English-speaking peers. Through the use of testing accommodations, states are asserting they are leveling the playing field for ELLs. Whether this is happening in actuality is largely unknown. This study examined the following issues with a sample of 8th grade students in an urban public school district: (1) the relationship between the Language Assessment Scales (LAS) and the Stanford Achievement Test (SAT-9) in Reading and Mathematics; (2) the test scores of Non-English proficient (NEP), limited English proficient (LEP), fluent English proficient (FEP), and non-culturally and linguistically diverse (non-CLD) general education students on the SAT-9 tests; and, (3) the perceptions of students with regard to the effectiveness of testing accommodations in assisting them to more accurately demonstrate content knowledge and skills. Results indicate the following: a positive correlation between the LAS and SAT-9 tests; ELLs have lower test scores than FEP and non-CLD general education students in Reading and Mathematics; there are no significant differences in test scores on the SAT-9 between NEP and LEP students in Reading or Mathematics, or between FEP and non-CLD general education students in Reading; FEP students have higher test scores than non-CLD general education students in Mathematics; and, ELLs perceive a majority of testing accommodations as effective in assisting them to demonstrate content knowledge and skills. The implications of this study are that ELLs are not being equitably assessed for academic achievement; the LAS fluent category may be a good indicator of academic language used in the classroom; greater individualized attention must be given to the selection of testing accommodations most appropriate for each ELL student to be able to more accurately demonstrate content knowledge and skills, and student input should be a part of the decision-making process.