Improving health data visualizations: effects of color and summary information on decision making
Poster Presentation: Friday, May 16, 2025, 3:00 – 5:00 pm, Banyan Breezeway
Session: Decision Making: Perception
Schedule of Events | Search Abstracts | Symposia | Talk Sessions | Poster Sessions
Angelica S. Busciglio1, Jessica N. Goetz1, Mark B. Neider1; 1University of Central Florida
The purpose of data visualization is to communicate information. A well-designed visualization allows viewers to quickly extract important patterns or trends in the data (Ware, 2020). Utilizing what is known about the basic properties of human perception, such as color and spatial perception, can enhance information display. The current study explores how the use of different visualizations and forms of risk communication beyond the current standard format type affects responses to health data. We specifically investigated whether the inclusion of overall summary statements and color increased comprehension and the likelihood to engage in health-oriented behavior when viewing hypothetical cholesterol results. Participants (N = 138) viewed results in two formats (table and number line), across three risk levels (low, borderline, and high), either with or without summary statements and color. In the table format, participants were more likely to engage in preventative behaviors (following up with a doctor and exercising) at normal risk levels when summary statements were absent, suggesting misinterpretation without additional context (all ps<.05). In the table format, color improved the ability to correctly identify values as acceptable at normal risk levels, reducing misinterpretation (p<.05). In the number line format, at borderline risk levels, the absence of color led participants to judge values as acceptable. In contrast, the presence of color increased the likelihood of perceiving values as unacceptable, indicating that color may heighten perceived urgency (p< .01). Furthermore, color helped participants identify table values as acceptable without summary statements (p<.05). Our results demonstrate the importance of effective data visualizations and how summaries and color shape data interpretation.