Artifact Analysis around Video Creation in Collaborative STEM Learning Scenarios

Abstract

In this study, we explore the potential of (semi-) automated artifact analysis to identify and characterize students' comprehension problems around knowledge artifacts in the form of videos available on a web platform. Using the comments assigned to these learning videos as sources, we have applied available and tailor-made methods of text analysis to identify potential problems of understanding. In particular, we have used signal words and their relationship to domain concepts to indicate missing pre-knowledge or false beliefs. In future work, these results will be used to support monitoring processes for increasing conceptual awareness and supporting conceptual change.