Evaluating the psychometric properties of the Revised Math Anxiety Rating Scale (MARS) among Ugandan high schools

The purpose of this study was to examine the psychometric properties of the revised Math Anxiety Rating Scale for Uganda, administered to a convenience sample of 107 high school students across four grade levels. The revised scale adaptation process focused on ensuring semantic, idiomatic, experiential, and conceptual equivalence between the original and adapted items. Using Rasch measurement approaches, including the Partial Credit Model and the Many-Facet Rasch Model, we assessed item fit, reliability, and rating scale functioning. Although initial results showed strong internal consistency, 17 of the 25 items exhibited disordered thresholds. To address this, the original five-point scale was re-coded into three categories, resulting in improved threshold ordering and only a minimal reduction in person reliability. Threshold distances between adjacent categories met recommended guidelines, supporting category precision across groups. Additionally, no differential item functioning was observed across age or grade level, suggesting that the revised scale operates equivalently for diverse student subgroups. These findings provide initial evidence that the revised scale is a reliable and culturally relevant instrument for measuring math anxiety among Ugandan high school students, with implications for educational assessment and intervention.