Exposure Matters: Effects of Current Language Exposure on Item-Level Measurement of Spanish and English Morphosyntax

BACKGROUND: Bilingual children experience varying levels of exposure to their languages as they develop their linguistic knowledge. Therefore, test developers must account for exposure during construction of language assessments to avoid mismeasurement. The current study aims to examine how language exposure affects item parameterization in bilingual morphosyntax tasks. METHOD: Using English data from 484 bilingual children and Spanish data from 403 bilingual children aged 7 to 11 years, we conducted item response theory (IRT) analyses to examine the relative effects of exposure on item parameterization. Whole-sample Rasch and multiple-group two-parameter logistic models were used to explore these effects. RESULTS: IRT results suggested differential patterns of item parameterization, namely, difficulty and discrimination, for groups with varied levels of English exposure. CONCLUSION: These results highlight the need for thorough and nuanced analysis of item parameters and influences on test performance during the development of language assessments intended for use with bilingual populations. SUPPLEMENTAL MATERIAL: https://doi.org/10.23641/asha.32410173.