Since 2012, Massive Open Online Courses (MOOCs) have attracted millions of learners to learn and communicate at an unprecedented scale [1]. MOOC data contains not only learner profile and learning outcome information, but also the web log records of learner interactions with various course materials. Such large amounts of heterogeneous and multivariate data provide great opportunities for analyzing online learning behaviors while at the same time posing new challenges. Visual analytics and storytelling turns out to be an effective solution to help instructors and education experts better discover how students learn, understand the reasons behind various learning behaviors, and present learning analytics stories.

In this thesis, we introduce three visualization systems to facili...[ Read more ]

Since 2012, Massive Open Online Courses (MOOCs) have attracted millions of learners to learn and communicate at an unprecedented scale [1]. MOOC data contains not only learner profile and learning outcome information, but also the web log records of learner interactions with various course materials. Such large amounts of heterogeneous and multivariate data provide great opportunities for analyzing online learning behaviors while at the same time posing new challenges. Visual analytics and storytelling turns out to be an effective solution to help instructors and education experts better discover how students learn, understand the reasons behind various learning behaviors, and present learning analytics stories.

In this thesis, we introduce three visualization systems to facilitate instructors and education experts in understanding, exploring, analyzing, gaining and sharing insights from MOOC data. The first work, PeakVizor, is a comprehensive visualization system which integrates well-established visualization techniques and several novel visual designs to investigate clickstream peaks. The second system, ViSeq, focuses on the visual analytics of learning sequences of different learner groups. The four linked views facilitate users in exploring learning sequences from multiple levels of granularity. In the last work, we propose a narrative visualization approach with an interactive slideshow that helps instructors and education experts explore potential learning patterns and convey data stories. This approach contains three key components: the guided-tour concept, the drill-down path, and the dig-in exploration dimension. Case studies and interviews conducted with domain experts have demonstrated the usefulness and effectiveness of the three systems.