Contextual data literacy and digital competence: A mixed-methods study of South African mathematics teachers

Abstract

Persistent inequities in digital access constrain how teachers transform data into actionable insight. Yet, few studies quantify how digital competence, data literacy, and professional development needs intersect in low-resource systems. This study employed a convergent mixed-methods design, involving 71 South African secondary mathematics teachers who completed validated surveys on data literacy competence, digital competence, instructional data use, and professional development needs, alongside five in-depth interviews analysed through reflexive thematic analysis. Quantitative models estimated standardised associations with robust standard errors and false discovery rate control, while qualitative themes illuminated structural, institutional, and identity-related mechanisms. Descriptive results placed all competencies near the midpoint of a four-point scale, while professional development needs were highest (M = 3.05). Digital competence strongly predicted instructional data use (β = 0.67, p < 0.001), whereas formal training showed no effects, and teaching experience trended negatively (β = −0.23, p = 0.067). Higher data literacy competence corresponded with greater professional development demand (β = 0.38, p = 0.012). Interviews revealed that digital inequities, time scarcity, and weak collaboration structures limit the translation of competence into practice. The study advances a contextual data literacy theory, emphasising that infrastructural justice, institutional redesign, and differentiated professional learning are prerequisites for effective instructional data use. Findings urge policymakers to prioritise equitable access, collaborative inquiry time, and adaptive professional learning systems to enable data-informed instruction in resource-constrained schools.