Electronic Thesis and Dissertation Repository


Doctor of Philosophy


Computer Science


Dr. Kamran Sedig


Public health data plays a critical role in ensuring the health of the populace. Professionals use data as they engage in efforts to improve and protect the health of communities. For the public, data influences their ability to make health-related decisions. Health literacy, which is the ability of an individual to access, understand, and apply health data, is a key determinant of health. At present, people seeking to use public health data are confronted with a myriad of challenges some of which relate to the nature and structure of the data. Interactive visualizations are a category of computational tools that can support individuals as they seek to use public health data. With interactive visualizations, individuals can access underlying data, change how data is represented, manipulate various visual elements, and in certain tools control and perform analytic tasks. That being said, currently, in public health, simple visualizations, which fail to effectively support the exploration of large sets of data, are predominantly used. The goal of this dissertation is to demonstrate the benefit of sophisticated interactive visualizations and analytics. As improperly designed visualizations can negatively impact users’ discourse with data, there is a need for frameworks to help designers think systematically about design issues. Furthermore, there is a need to demonstrate how such frameworks can be utilized. This dissertation includes a process by which designers can create health visualizations. Using this process, five novel visualizations were designed to facilitate making sense of public health data. Three studies were conducted with the visualizations. The first study explores how computational models can be used to make sense of the discourse of health on a social media platform. The second study investigates the use of instructional materials to improve visualization literacy. Visualization literacy is important because even when visualizations are designed properly, there still exists a gap between how a tool works and users’ perceptions of how the tool should work. The last study examines the efficacy of visualizations to improve health literacy. Overall then, this dissertation provides designers with a deeper understanding of how to systematically design health visualizations.