Learning analytics is a research field that tries to support the learning process by collecting and comparing data about learners and their learning contexts. Today, many learning management systems use learning analytics techniques because of the benefits they offer both instructors and students.
This thesis addresses several learning analytics related shortcomings of Artemis: Students need information about other students’ performance as a guide to interpret their own performance. Although the data is available in the system, Artemis does not currently provide students with such a means of comparison. Artemis also does not support learning goals, i.e. competencies that students should master when completing a course. The implementation of learning analytics is hindered by the fact that large parts of the learning process take place outside Artemis, as it is currently not possible to include, for example, lecture recordings on the platform.
To address these shortcomings, we expanded Artemis in three areas: First, the lecture concept has been redesigned. Videos, notes, files, and exercises can be directly integrated into a lecture. Second, learning goals can be defined and linked to relevant learning material. The system calcu- lates a student’s progress towards goal mastery. Third, a learning analytics dashboard contains visualizations that provide students with the necessary context to interpret their performance in a course. The redesigned lectures and learning goals were used in the course Patterns in Software Engineer- ing. 538 students actively participated in the course. It contains 5 learning goals and 12 lectures, with a total of 50 videos, 51 exercises, and 53 files. The new Artemis capabilities received positive feedback from students and instructors.