How to Develop an Evidence-based Mathematics Vocabulary Test in Multilingual Context?

What is it about?

Children with a limited mathematics-related vocabulary (with words such as more, after, first and empty) often develop mathematical learning difficulties that may be related to their vocabulary. Educators and researchers need valid and reliable tests to identify such children. This is important because after identification, at risk children’s mathematics vocabulary can be developed to prevent learning difficulties to increase. Several mathematics vocabulary measures for young children have been developed in northern English-speaking countries, but reliable and easy-to-use measures for early grades mathematics vocabulary are lacking in African countries, including South Africa. A test to assess children’s mathematics vocabulary is especially important in multilingual countries where many children learn mathematics in a language that is not their home language. Developing an instrument in several languages requires a team of translators and reviewers, together with thorough statistical analysis and planning. We developed the Mathematics Vocabulary Test (MVT) - an instrument to measure young children’s mathematics vocabulary in several South African languages. The study, with children aged 3-8.5 years (N=988, ngirls = 429), consisted of two phases. In the first phase, the original MVT, with 26 items, was developed by a team of teachers, parents and researchers and used in a small group of children who used four languages in their home environment. Children with the home languages of isiZulu– (n=229), Sesotho– (n=83), English– (n=89) and Afrikaans (n=216) participated in the pilot study to determine how well the test works. Based on the results, the MVT was revised and improved. In the second phase, additional children were assessed with the revised versions of the English (n=270) and isiZulu (n=101) MVT. Phase 2 analyses showed that 20 items of the English test and 16 items of the isiZulu test worked well. The analysis of the English MVT also showed that the test identified the same children to be at risk five months later. This paper explains how educators and researchers can use the total score of the test to describe young children’s mathematics vocabulary knowledge. Our findings also showed that the MVT was able to detect the same children at risk in their mathematics learning as in another early numeracy test. By using the English MVT in our study, it was possible to predict the performance in mathematics vocabulary and early numeracy six months later. Taken together, our results showed that the English and isiZulu MVT can reliably be used to identify children with a limited mathematics vocabulary. Ultimately, the purpose of such identification is to know which children need additional support in their mathematics learning, especially in multilingual schools and countries. Next, we plan to evaluate the reliability of the Sesotho and Afrikaans MVT.

Featured Image

Why is it important?

This is the first instrument developed to measure young children’s mathematics vocabulary in Sub-Saharan Africa. Our work has added educational value because teachers in South Africa have little (if any) access to evidence-based and easy-to-use tools to identify children with a limited mathematics vocabulary in the early grades. Methodologically, we were able to demonstrate that, although it is challenging, it is possible to follow the progressive process of producing validity and reliability evidence of a new educational measurement in a multilingual country in the Global South.

Perspectives