Developing and Evaluating a Toolkit and Curriculum for Teaching and Learning Data Visualization

点击翻译按钮获取中文摘要

Visualization is an indispensable method for analyzing data across STEM fields. Although visualization research has advanced for at least thirty years, visualization education at the undergraduate level has lagged behind. Visualization textbooks only emerged in the past decade, and few pedagogical software tools are available to assist the teaching and learning of data visualization. This project aims to develop a toolkit that will engage college students in learning visualization concepts and algorithms. This toolkit, called VisVisual, consists of four tools: VolumeVisual, FlowVisual, GraphVisual, and TreeVisual. Together, these tools cover scientific visualization (scalar and vector field visualization) and information visualization (graph and tree drawing). The modular design will allow instructors to incorporate only relevant components into their teaching. To support active learning, each tool will provide instant feedback to students and an auto-grading component will check for student understanding. Assessment and evaluation data will be gathered to understand how students use the tools and the impact of the tools on their learning. This information will guide improvements to VisVisual and the development of additional visualization tools. It is expected that improving students' ability to visualize data will increase their interest in STEM and attract some of them to the scientific field of data science and visualization.VisVisual will focus on helping students learn concepts and algorithms in scientific visualization and information visualization. To promote data visualization education, curriculum materials to support the use of VisVisual will be developed and openly shared, including lesson plans for a data visualization course, test questions, datasets, and web-based tutorials. It is expected that providing the VisVisual software along with the curriculum materials will increase its use by educators across the world. The effectiveness of VisVisual will be evaluated by data obtained from a controlled user study in which students complete assigned tasks and answer questions related to those tasks. To compare performance differences between subject groups or parameter settings, the researchers will apply Student's t-test, analysis of variance, and other statistical analyses. Information about students (e.g., type of institution; major) will be obtained and multivariate analysis of variation will be used to determine how these individual factors affect performance. This information will be used to better understand the role of VisVisual components on students' understanding of the concepts that they explore using VisVisual. Planned outreach to the K-12 community will introduce teachers and students to the principles of visual design and the important role it plays in data science. VisVisual will be freely available on the project website, thus enabling wide dissemination. A workshop at a national meeting is planned, to help computer science faculty learn how to use it in their classes.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.