How do students do design? A data-driven approach for understanding students'engineering design behaviors

Abstract

Engineering design is a natural addition to STEM+C education, as it motivates students to integrate science and computing while engaging in problem-solving tasks. However, K-12 students face challenges in applying reasoning and critical thinking skills to find “optimal” design solutions. In this paper, we investigate students’ behaviors when they are engaged in engineering design tasks by analyzing the sequences of the design solutions they generate to find an “optimal” solution. We collected data from sixth-grade students (N=88) working in pairs in an NGSS-aligned earth sciences curriculum. Using Markov Models, we visualized students’ solution progressions and identified instances where they became stuck in sub-optimal states. Our results show that (1) students’ scores improved significantly from pre-test to post-test; (2) students’ problem-solving behaviors could be categorized as design space explorers and conservative designers; and (3) students’ problem-solving behaviors influenced the quality of their final designs. Additionally, our findings underscore the importance of supporting design space exploration as a key aspect of scaffolding engineering design activities.