Date of Award
Doctor of Philosophy (PhD)
William R. Penuel
Scientific modeling is increasingly important both in K-12 science education and the broader scientific community, but there are significant gaps in both our understanding of how people learn modeling and how we can support them in this process. This work takes a comprehensive look at how students use digital modeling tools in science classrooms, particularly with EcoSurvey, a tool developed to support students in creating a model of the components and interactions in the local ecosystem. This tool has been developed using an iterative process and deployed in three consecutive school years as part of a design-based implementation research project in high school biology classrooms. During this time, I have developed new techniques for analyzing students’ models and modeling activity along with the impact of feedback and recommender systems. These approaches have demonstrated significant power in creating a picture of students’ modeling activity in real time. In addition, I have determined the effects of certain design decisions on student tool utilization through iterative deployment, and found that explicit scaffolds can have a significant impact on students’ models and modeling practices. Finally, I have begun to map how student activity can be related to their learning of modeling as a science and engineering practices. Through this work, I have demonstrated the power of real-time activity analytics to provide insight on the appropriate level of student support to give. This work advances learning analytics, the study of scientific modeling in the classroom, and modeling tool design.
Quigley, David Philip, "A Learning Analytics Approach to Scaffolding Scientific Modeling in the Classroom" (2019). Computer Science Graduate Theses & Dissertations. 165.
Available for download on Sunday, October 10, 2021