Assessment Language and Early Math: Advancing Equity in Evaluating Dual Language Learners

Jimena Cosso

Spanish–English dual language learners (DLLs) represent a substantial and growing share of children in U.S. early childhood programs, yet little is known about how assessment language shapes what we observe about their early mathematics knowledge, a linguistically demanding domain that strongly predicts later academic and STEM success. This project has two main goals. First, to experimentally test how assessment language (English vs. Spanish) affect observed mathematics performance; and second, to evaluates whether a widely used language-routing procedure (preLAS test) accurately identifies the language in which children's mathematical knowledge is best reflected. This project challenges deficit-oriented interpretations of DLL performance that may arise when children are assessed in only one language or routed into a single testing language without evidence that this choice yields the most valid estimate of their skills. It will generate new evidence on the role of language in early mathematics assessment and inform more equitable, linguistically responsive practices for identifying strengths, supporting instruction, and evaluating school readiness among young dual language learners.

About Jimena Cosso

Jimena Cosso is an Assistant Professor in the Human Development program at the University of Maryland, College Park, where she directs the HOLA STEM Lab. Born and raised in Argentina, she came to the United States to earn an M.A. in Comparative and International Education at New York University and later completed her Ph.D. in Educational Psychology and Research Methodology at Purdue University in 2022. Dr. Cosso examines the learning contexts that shape how young children develop and learn, with a particular focus on Spanish-English dual language learners (DLLs) from immigrant Latine communities. Across her work, she investigates two interrelated areas: (a) the development of culturally and linguistically responsive measures, and (b) the contextual factors that shape early learning, especially early mathematics. Her scholarship challenges deficit-oriented approaches that obscure the competencies of DLLs and instead highlights the linguistic, cultural, and experiential assets that Latine families bring to children's development. For example, she has developed a culturally sensitive home math environment measure for Latine families and studies how family practices, caregiver characteristics, and within-group variation in home learning environments support children's mathematics development. She also examines how assessment practices, including language of assessment, can either reveal or misrepresent DLLs' early competencies. Through this work, Dr. Cosso aims to inform more equitable early childhood assessment and practice for diverse learners and their families.