Language of assessment matters: a study of Ghanaian primary school students’ scientific reasoning skills

Purpose This study aims to examine the extent to which the language of assessment influences primary school students’ scientific reasoning skills and further explores how gender, grade level and school moderate this relationship. Design/methodology/approach A cross-sectional survey was conducted with 260 primary school students selected via a multistage sampling technique. Students were assessed using the Science-P(primary) Reasoning Inventory (SPR-I) in either English or the Ghanaian Language (Twi). Descriptive statistics and a multiple regression model were used to analyze the data. The regression model was used to test the main and interaction effects of language, gender, grade level and school type. Findings Language of assessment significantly predicted students’ scientific reasoning skills, with students assessed in Twi demonstrating higher reasoning abilities than those assessed in English. Grade level and school type also had significant effects, while gender and all interaction terms were non-significant. Originality/value This study provides empirical support for the use of mother-tongue assessments in science education. It advances equity-based assessment practices in multilingual contexts like Ghana.