Learning analytics and educational data mining are introducing a number of new techniques and frameworks for studying learning. The scalability and complexity of these novel techniques has afforded new ways for enacting education research and has helped scholars gain new insights into human cognition and learning. Nonetheless, there remain some domains for which pure computational analysis is currently infeasible. One such area, which is particularly important today, is open-ended, hands-on, engineering design tasks. These open-ended tasks are becoming increasingly prevalent in both K–12 and post-secondary learning institutions, as educators are adopting this approach in order to teach students real-world science and engineering skills (e.g., the “Maker Movement”). This paper highlights findings from a combined human–computer analysis of students as they complete a short engineering design task. The study uncovers novel insights and serves to advance the field’s understanding of engineering design patterns. More specifically, this paper uses machine learning on hand-coded video data to identify general patterns in engineering design and develop a fine-grained representation of how experience relates to engineering practices. Finally, the paper concludes with ideas on how the specific findings from this study can be used to improve engineering education and the nascent field of “making” and digital fabrication in education. We also discuss how human–computer collaborative analyses can grow the learning analytics community and make learning analytics more central to education research.
Curricular Value And Instructional Needs For Infusing Engineering Design Into K 12 Technology Education