Exploring Language as a Source of DIF in a Math Test for English Language Learners

English language learners (ELs) have shown lower performance in mathematics than non-ELs although mathematics is an area that uses the least amount of language among the subjects that are mainly tested. If this differential performance is due to the bias in test items, then validity of using ELs’ test scores in comparison to non-ELs’ is compromised. For this reason, studies have investigated whether the differential performance can be attributed to language load in the tests. The results of these studies were not consistent. Some studies did find its effect, whereas others did not. Some of the difficulties encountered by researchers in past studies investigating DIF include a large difference in sample size between the two groups and unclear distinctions between ELs and non-ELs. This study aims to investigate the source of DIF between ELs and non-ELs using a comparatively large and a better defined/restricted population of ELs. This study will contribute to existing knowledge about English proficiency as a possible cause of differential performance between the two groups. The findings of this study will have implications for test construction and policies for providing testing accommodations (e.g., test language simplification)