A research-based insights brief examining how language of assessment affects ELOM-R performance for isiXhosa home language children in English LOLT Grade R classes – and what this means for fair and valid assessment practice.
This Insights Brief draws on a 2025 study conducted by DataDrive2030 and the University of Cape Town, investigating whether isiXhosa-speaking Grade R children perform better when assessed in their home language or in English after sustained exposure to English Language of Learning and Teaching (LOLT). Using the ELOM-R Language and Mathematics Assessments, the study explored how assessment language, English exposure, and child preference influence performance and assessment validity.
In multilingual education contexts, fair assessment matters. Language choice can shape how confidently and accurately children demonstrate what they know and can do. This study shows that by the end of Grade R, sustained English LOLT exposure may make English assessment appropriate for some children – while others may still perform best in their home language.
The findings reinforce the importance of flexible, child-centred assessment approaches that consider children’s language histories, learning environments, and preferences. Rather than relying on policy alone, assessment decisions should maximise validity by ensuring children are linguistically comfortable and able to demonstrate their true capabilities.
Use these insights to support fairer assessment practices and better-informed decisions in multilingual early learning environments.
DOWNLOAD THE FULL INSIGHTS BRIEF HERE
Title: Assessing Fairly in Grade R: Home Language vs. English LOLT in ELOM-R Assessments
Authors: Andrew Dawes and Linda Biersteker
Published: February 2026
Explore more related resources or search through all our resources on our resources page.
