Engineering degrees are often regarded as complex and one usual issue is that students struggle and feel discouraged during the learning process. Gamification is starting to play an important role in education with the objective of providing engagement and improving the motivation of students. One specific example is the use of badges. The analysis of users' interactions and behaviors with the badge system can be used to improve the learning process, e.g., by adapting the learning materials and giving game-based activities to students depending on their interest toward badges. In this work, we propose some metrics that provide information regarding the behavior of students with badges, including if they are intentionally earning them, the concentration for achieving them and their time efficiency. We validate these metrics by providing an extensive analysis of 291 different students interacting with a local instance of Khan Academy within our courses for freshmen at Universidad Carlos III de Madrid. This analysis includes relationship mining between badge indicators and others related to the learning process, the analysis of specific archetypal profiles of students that represent a broader population and also by clustering students by their badge indicators with the objective of customizing learning experiences. We finalize by discussing the implications of the results for engineering education, providing guidelines into how instructors can take advantage of the findings of the research and how researchers can replicate experiments similar to this one in other general contexts.
badges; gamification; engineering education; distance learning; learning analytics; khan academy; educational data mining; modelling behavior; motivation; achievement; games