Making machine learning findings accessible to teachers in blended classrooms

Making machine learning findings accessible to teachers in blended classrooms. Using a teacher-centered, mixed-method approach, this study designs, evaluates, and instantiates visualizations and a learning analytics dashboard (DashED) that communicate ML-derived self-regulated learning profiles to teachers in two blended learning contexts — flipped university classrooms and reflective writing in vocational education. A 100-teacher study examined how visualization literacy shapes interpretation and which visual designs teachers find clear, appealing, and actionable, while interviews with 19 teachers probed concerns, Trust, and adoption. Findings reveal that teachers prefer simple, familiar charts (bar plots, pie charts) yet derive richer insights from more complex designs, and that teachers' visualization preferences, concerns, and intended uses differ markedly across learning contexts.

Key Findings

• Teachers systematically preferred simpler, more traditional visualizations (bar plots, pie charts, legends), even when more complex designs (e.g., heatmaps) yielded more detailed insights — visual preference did not always align with informativeness, echoing debates about pie charts' comparative readability.
• Visualization literacy (VL) did not drive design preferences, but higher-VL teachers produced deeper, more detailed interpretations (e.g., more of them identified trends in time-series data), confirming VL as a confounder for gauging how teachers read learning analytics designs.
• For group comparison, teachers strongly favored superposition over juxtaposition, and preferred plots that displayed full information (e.g., including a "students who did not watch" group) rather than explicit difference encoding, although younger teachers ranked difference plots higher.
• Actions teachers proposed were shaped by the content represented and their teaching level rather than the plot type — e.g., university teachers favored weekly tests and course adaptation, while vocational teachers proposed direct, individualized coaching.
• Concerns and barriers to adoption diverged by context: flipped-classroom (university) teachers worried most about data anonymization and student opt-out, whereas reflective-writing (vocational) teachers feared misuse of the tool by fellow educators and stressed the need to contextualize data — while both groups reported similar Self Efficacy and perceived benefits in a trust in AI survey.
• In use, flipped-classroom teachers followed a sequential exploration and favored course-level adaptation and showing dashboards in class, whereas vocational teachers revisited summary pages and used the tool mainly for individual coaching sessions — pointing to context-aware dashboard design and teacher support needs.

Connected Concepts

• Learning Analytics
• Teacher Role
• Trust
• Visualization
• Self Regulated Learning
• Online Teaching And Learning
• Machine Learning
• Mixed Methods Research

Connected Articles

• Fouad Bentley Trust Utility Gap Physics 2026
• AI TPACK Teacher Multi Agent Workflow
• Jin GenAI Learning Analytics Feedback Literacy

Citation

Mejia-Domenzain, P., Neshaei, S. P., Laini, E., Nazaretsky, T., Bühlmann, P., & Käser, T. (2026). Making machine learning findings accessible to teachers in blended classrooms. International Journal of Artificial Intelligence in Education, 36, Article 100001.