Learning Analytics: the Basics

Perhaps you’ve heard the term “learning analytics.” You may have heard it used in the same phrase with “big data” and “predictive analytics.”  But is this just one of those terms, like “synergy,” that will someday define a decade? Or a fad, like Second Life, that will slowly whimper and die?

Learning analytics is a term for how the data generated by students as they learn are measured and analyzed. The newly developing field of learning analytics is centered on using that information to understand and develop educational processes. Researchers, educators, and policy makers are involved; the central actors are students. By utilizing learning analytics, faculty, staff, and students can identify risks, improve learning and course outcomes, and develop more effective teaching and learning techniques.

Learning analytics tools go beyond what is possible for an individual instructor to accomplish; they can track learners and learning across multiple sources and, at their best, can reveal information about student learning and course improvement opportunities.

The data used by learning analytics researchers can come from a wide variety of places, including the learning management system (Canvas at Northwestern), student information systems, course evaluation systems, even the recreation facilities. Once the data has been gathered and run through analysis programs, the researchers create reports that can enable action on the part of students, instructors, or administrators.

The significance of learning analytics is that they add another arrow to the quiver of tools for improving teaching and learning. Students can understand better how they learn and apply that knowledge to their career of study. Instructors can understand their students’ learning process and act to improve course outcomes.

For many colleges, the first foray into learning analytics manifested itself as a way to identify at-risk students. Focusing on retention and graduation rates, they looked to the data for indicators that students were on a path that might lead to course failure and subsequent departure from higher education. Taking action – including personal outreach from faculty or reminders and study tips from the course management system – was a direct result of the data analyses and reports.

Beyond identifying at-risk students, learning analytics can be employed to help instructors understand whether their students are comprehending material in their courses. If the course management system shows that students are returning again and again to a topic or consistently missing a concept in multiple assessments, the instructor learns that these concepts need to be explained more clearly. Insights like these can help instructors as they redesign courses. If combined data show that students successfully pursue a major only if they have had extensive exposure to the topic outside of class, the department can adjust its program. If learning analytics suggest that students who did well in class X often do well in major Y, they’ve just helped to create a degree pathway.

We’ll be looking in depth at learning analytics in a series of subsequent articles, which will focus on examples of its use at Northwestern in the classroom and beyond.