Introduction to Learning Analytics

Learning analytics is an emerging field that utilizes campus data and information to guide individual student success and institutional decision making. It can be viewed as an extension of established analytics and reporting practices developed on top of UC Berkeley’s Enterprise Data Warehouse (EDW), known as CalAnswers. The shift that learning analytics brings is the use of real time activity data gleaned from the systems and tools students interact with. The graphic below shows how the data would flow: user activity is combined with institutional data (i.e, the types of data in CalAnswers), analysed, and then returned to various contexts (e.g., the LMS) to fulfill a variety of purposes, including personalized learning paths, real time alerts, and dashboards to help faculty and advisors gauge student engagement and understanding.

Learning Analytics Definition Flow Chart

(Full Size and screen reader friendly) 


These simplified scenarios are meant to provide an idea of the power of learning analytics.

Advising dashboards and alerts. Activity levels in the LMS and other tools can provide indicators of student stress and disengagement, especially among previously identified at-risk subpopulations and groups. These data can be visualized for advisor cohorts to help guide potential early interventions or individual students can opt in to be notified to seek help from the appropriate campus resources.

Instructor dashboards and personalized learning paths. Responses to test questions and homework problems in digital form can be analysed to provide insight to fine-grained levels of understanding and conceptual grasp, and reported to instructors. These same data can be used to help create personalized paths through course modules or digital textbooks for students.