Learning Analytics (LA) is an interdisciplinary field focused on the measurement, collection, analysis, and reporting of data about learners and their contexts, with the goal of understanding and optimizing both learning processes and the environments in which they occur. My work in this field centered on advancing LA specifically for knowledge communities, where collaborative knowledge construction and collective inquiry were key. While much conventional LA work focused on teacher-driven assessment of individual achievement, my work positioned LA as a tool for both illuminating and scaffolding complex collaborative learning processes, enabling a “feedforward” effect that fostered ongoing knowledge creation.
My engagement with LA spanned multiple projects, several of which fell under the umbrella of my CKBiology doctoral project. My focus was on ‘making the invisible visible’ – i.e. turning the hidden cognitive and collaborative processes of a knowledge community into clear, actionable information that could guide students and teachers alike. A key challenge was mitigating the “orchestrational load” on teachers (i.e. the cognitive effort required to monitor and manage complex group activities), allowing them to focus on guiding inquiry rather than tracking complex interactions manually.
To achieve this, the SAIL Smart Spaces (S3) software architecture was adapted to include a hidden layer of server-side intelligence that tracked the real-time progress of individual students, small groups, and the class as a whole. This system supported teaching and learning in multiple ways. For students, it included visual indicators of progress, such as individual, group, and class-level progress bars, as well as cues in the shared CKBiology knowledge base that highlighted gaps, incomplete work, or errors. For teachers, dashboards and reports provided a live view of group activity, allowing them to intervene or support students without interrupting collaboration.
Beyond making activity visible, my work advanced the analytical depth of LA in KCI. For example, in later iterations I designed a recommender system that calculated “specialization scores” for students seeking specialist roles within a knowledge community, accounting for factors such as past contributions and peer evaluation scores. I also created a formal way of representing Learning Analytics within CSCL scripts by adding a dedicated LA layer to Orchestration Graphs, which helped educators and researchers see how the LA connected to educational theory, empirical analysis, and computational tools.
This work showed that Learning Analytics can help teachers manage complex collaborative inquiry activities, highlight productive group behaviors, and provide meaningful insights into how knowledge communities operate. At the same time, it revealed some important challenges. While students gained a clearer sense of their own contributions and the progress of their learning community, making individual and group performance visible sometimes conflicted with traditional, competitive school norms. These findings underscored the need for thoughtfully designed assessments and motivational supports to foster collaboration. They also highlighted the value of embedding opportunities for reflection, reasoning, and evidence-based knowledge building, ensuring that students engaged deeply and thoughtfully in collaborative learning contexts.
At its core, my work in Learning Analytics is driven by the goal of creating learning environments that are transparent, responsive, and equitable. By making the processes of knowledge creation visible and actionable, these designs empower educators and learners alike to engage deeply in collaborative inquiry, monitor and reflect on their progress, and build the critical thinking and problem-solving skills needed to tackle complex, global challenges.
Alisa Acosta 