<scp>A human‐centred</scp> learning analytics approach for developing contextually scalable K‐12 teacher dashboards

Abstract This paper describes a Human‐Centred Learning Analytics (HCLA) design approach for developing learning analytics (LA) dashboards for K‐12 classrooms that maintain both contextual relevance and scalability—two goals that are often in competition. Using mixed methods, we collected observational and interview data from teacher partners and assessment data from their students' engagement with the lesson materials. This DBR‐based, human‐centred design process resulted in a dashboard that supported teachers in addressing their students' learning needs. To develop the dashboard features that could support teachers, we found that a design refinement process that drew on the insights of teachers with varying teaching experience, philosophies and teaching contexts strengthened the resulting outcome. The versatile nature of the approach, in terms of student learning outcomes, makes it useful for HCLA design efforts across diverse K‐12 educational contexts. Practitioner notes What is already known about this topic Learning analytics that are aligned to both a learning theory and learning design support student learning. LA dashboards that support users to understand the associated learning analytics data provide actionable insight. Design‐based research is a promising methodology for Human‐Centred Learning Analytics design, particularly in the K‐12 educational context. What this paper adds Leveraging a longstanding, yet fluid, research‐practice partnership is an effective design‐based research adaptation for addressing the high variation in instructional practices that characterize K‐12 education. Using both quantitative and qualitative data that reflects students' developing knowledge effectively supports teachers' inquiry into student learning. Teachers' use of learning analytics dashboards is heavily influenced by their perspectives on teaching and learning. Implications for practice and/or policy Impact on student learning outcomes, alongside usability and feasibility, should be included as a necessary metric for the effectiveness of LA design. LA dashboard developers should both leverage learning data that reflect students' developing knowledge and position teachers to take responsive pedagogical action to support student learning. LA researchers and developers should utilize a long‐term, yet fluid, research‐practice partnership to form a multi‐stakeholder, multidisciplinary design team for Human‐Centred Learning Analytics design.