Enhancing flexible STEM thinking by generating interactive diagrams at scale

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

Math and science diagrams improve learning, both when diagrams are delivered with instruction, and when they are created for self-explanation. The resulting learning is often more flexible. For example, students that practice diagramming are better at transferring their learning from the problems they have explicitly practiced to more open-ended problems. Unfortunately, diagrams are too uncommon in instructional materials, especially practice problems. This is primarily because diagrams are much harder to produce than text and symbols. Teaching at scale adds further challenges. Ideally, e-learning platforms should deliver different problems to different students. These problems should be tailored to knowledge components, prevent cheating by copying, and be tuned the amount of practice to student needs. Problem templating systems aren’t built for diagrams and grading student-authored diagrams is hard to do even manually. To address the challenge, this project aims to develop a new tool for generating diagrammatic instructional content. This tool developed by the project will enable content authors to generate large problem sets by example. User will author one or two problems and the tool will synthesize a problem set from the examples. The project introduces efficient interaction techniques for viewing and editing these sets. The project will collect data from teachers and students to guide, refine, and evaluate the design of the tool. The project will conduct six studies, three will focus on the effectiveness of the tool in supporting authoring of educational content and three that focus on the impact of the resulting problems on student learning. These studies will enhance our understanding of when and how diagrams can be used to increase student understanding of science, technology, engineering and math (STEM) principles and lead to more flexible STEM thinking. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.