Acceso usuarios
Nicole Haag, Birgit Heppt, Alexander Roppelt, Petra Stanat
In large-scale assessment studies, language minority students typically obtain lower test scores in mathematics than native speakers. Although this performance difference was related to the linguistic complexity of test items in some studies, other studies did not find linguistically demanding math items to be disproportionally more difficult for language minority students than for native speakers. Furthermore, previous studies investigating the effectiveness of linguistic simplification of test items have yielded inconsistent results. We tested the impact of linguistic simplifications of mathematics test items in Germany with a specific focus on the role of academic language features included in the items. Specifically, we carried out a large-scale linguistic simplification study to test whether the performance gap between language minority students and German monolinguals is smaller when students are assessed with linguistically simplified items. We additionally determined whether students' socioeconomic status (SES) or their language proficiency moderate the effect of linguistic simplification. The study uses data from 17,738 fourth graders, 17 % of whom sometimes or always spoke a language other than German at home. Although differences between language minority students and German monolinguals in mathematics achievement were related to differences in their language proficiency and SES, we found no significant main effects of linguistic simplification. Differential effects for language minority students emerged, however, when we took the moderator effects of SES and language proficiency into account, indicating that some language minority students may profit from linguistic simplification in elementary school.
