ELOM-R (v1) Technical Manual 1: The Development of the ELOM-R Language and Mathematics Assessments

This manual provides a detailed account of how the ELOM-R assessment measures were designed for use in the South African early learning system. It explains how early language and mathematics constructs were selected based on research into what children need to thrive in Grade 1 and beyond. The manual highlights how the development team collaborated with curriculum specialists, linguists, teachers, speech and language experts, and psychometricians to ensure each item was meaningful, clear, and culturally grounded.

The document outlines the iterative cycles of piloting, observing children’s engagement, adjusting wording or materials, and re-testing across multiple provinces and in most South African languages. It recognises South Africa’s linguistic complexity and shows how the design process prioritised fairness.

What’s Inside

• Construct selection framework showing how developmental and curriculum research informed assessment content.
• Item development and piloting cycles demonstrating how feedback shaped the tools.
• Cross-language adaptation principles ensuring conceptual fairness across diverse linguistic contexts.
• Final task structures and scoring rationale for both Language and Mathematics assessments.

Why It Matters

This technical manual shows how the ELOM-R assessments were built for South Africa, in South Africa, ensuring the tools recognise children’s strengths and learning pathways across diverse settings, cultures, and languages.

For further information, please refer to this document which summarises the intended use of the ELOM-R tools, their composition, and the outcomes of the psychometric analysis, standardisation, and norming process.

DOWNLOAD THE FULL MANUAL HERE

Title: ELOM-R (v1) Technical Manual 1: The Development of the ELOM-R Language and Mathematics Assessments 

Author(s): Developed on behalf of DataDrive2030 by Andrew Dawes (Psychology Department University of Cape Town and DataDrive2030), and Linda Biersteker, DataDrive2030

Published: June 2025