Designing for Disclosure in Data Visualizations.

Area of Science:

• Information Visualization
• Human-Computer Interaction
• Data Science

Background:

• Data visualization techniques often involve transformations that can obscure information.
• Existing tools lack explicit guidance on managing information disclosure in visualizations.
• Understanding how visualizations limit data access is crucial for effective and ethical design.

Purpose of the Study:

• To systematically characterize visualization techniques that limit access to underlying data.
• To develop a taxonomy of disclosure tactics based on data representation changes.
• To provide a framework for reasoning about design trade-offs in information revelation and concealment.

Main Methods:

• Content analysis of 425 visualization techniques from academic literature.
• Categorization of techniques based on their impact on data representation.
• Development of a taxonomy of disclosure tactics.

Main Results:

• A novel taxonomy of disclosure tactics was created, organizing methods by how they alter data representation.
• The taxonomy systematically categorizes techniques that reveal, distort, or hide information.
• Demonstrated the utility of the taxonomy in guiding design decisions where disclosure is critical.

Conclusions:

• Disclosure tactics offer a new lens for visualization research, impacting authoring tools, literacy, and ethical design.
• The taxonomy facilitates reasoning about the trade-offs inherent in data visualization design.
• Adopting a disclosure framework enhances understanding of information access in visual data representation.