AccessViz: 3rd Workshop on Accessible Data Visualizationbookmark_border

Public health dashboards communicate data that can inform important decisions, but they often raise accessibility challenges. We present a case study redesigning the Utah Wastewater Surveillance System dashboard to improve accessibility and usability across desktop and mobile settings. The redesign was informed by WCAG and developed iteratively with Utah DHHS collaborators, and we collected feedback on the final design from an external blind researcher. Our case study highlights that accessible dashboard design requires aligning accessibility guidelines with user needs, stakeholder workflows, and technical constraints. It also suggests that simple, targeted technical solutions tailored to the existing environment can provide practical value for dashboard redesign in government contexts.

Geovisualization has become ubiquitous across disciplines and is an increasingly important means of data communication. Its effectiveness in enabling actionable insights relies on human cognitive and perceptual capacities. Substantial research has focused on optimization of geovisual analytical tools for the able user but has insufficiently accounted for users of varying visual, cognitive, motor, auditory and comorbid disabilities. This paper examines the current state of geovisual accessibility tools and reflects on opportunities and challenges for future research within the context of geovisual analytics.

Multimodal large language models (MLLMs) can connect visualization patterns to external causes, consequences, and domain knowledge, but the evidential basis of these interpretations is often unclear. We present an exploratory study of 102 visualizations from four sources, three MLLMs, and four input conditions that vary access to the image, accessible chart context (non-image artifacts such as data tables, captions, alt text, and screen-reader structures), and withheld-context framing. Across 1,224 descriptions, we analyze model-attributed DIRECT, DERIVED, and SPECULATIVE labels and conduct an automated audit of numeric agreement. Accessible chart context shifted Gemini and GPT toward DIRECT claims and improved numeric agreement for some models. Adding the image to the full context did not yield a consistent numeric benefit, and the withheld-context prompt did not reliably increase cautious language. The prompt-defined Real-World Significance section remained predominantly SPECULATIVE. These results motivate accessible description systems that distinguish claims supported by supplied evidence from model-supplied interpretation.

Refreshable tactile displays (RTDs) have expanded access to data visualizations for blind and low-vision users, and recent work has introduced tools for authoring tactile charts. However, less is known about how blind analysts iteratively author charts for RTDs or how tactile feedback shapes design decisions. We present a three-hour co-design case study in which a blind data analyst used a conventional Python workflow to create stock-market visualizations for the Monarch RTD. We observed how iterative tactile exploration and revision shaped chart design, revealing challenges in translating visual encodings to pin-based tactile representations. Rather than introducing a new authoring tool, we provide an empirical account of tactile chart authoring and identify implications for rapid tactile preview, iterative refinement, and integration with mainstream visualization authoring workflows.

Accessible data visualisation often focuses on providing non- visual access to information, prioritising information retrieval over broader experiences of exploration and discovery. We investigate how tactile data art can extend accessibility beyond information ac- cess to support curiosity, engagement, and shared experiences. We present a multimodal tactile data artwork communicating women’s grassroots football participation in Wales through Braille, raised structures, physical objects, text, and colour. We evaluated the arte- fact with seven blind and visually impaired participants using an accessible qualitative questionnaire. Participants described tactile interaction as supporting understanding, exploration, memorabil- ity, and enjoyment, and perceived the artefact as both an artwork and an educational tool. Our findings suggest that accessible data visualisation can move from access towards discovery, with multi- modal data art creating opportunities for people with different sen- sory abilities to explore and experience data within a shared arte- fact.

Olli is an open-source library that renders data visualizations as keyboard-navigable structured textual descriptions for screen reader users. Olli has represented a chart as a tree derived from the chart's visual encodings; however, two systems have critiqued the restriction to trees. Data Navigator observed that hierarchical trees cannot express relations such as spatial adjacency or membership in multiple groups. Benthic argued that hypergraphs avoid the limitations of trees and graphs because a hypergraph can consistently express both hierarchy and adjacency. In response, this paper presents a new architecture for Olli that uses hypergraphs as its core internal representation. Domain plugins for charts and diagrams lower their specifications to the hypergraph, and a shared navigation runtime traverses the hypergraph. The new architecture unifies Olli with Benthic, and Olli's customizable description and interactive filtering now apply to charts and diagrams alike. Authors can specify Olli structures at three levels of abstraction: Vega-Lite and Bluefish specifications via adapters, Olli's own domain specifications, or raw hyperedges. Navigation on the shared runtime behaves consistently across domains but fits each graphic less closely than navigation authored specifically for that graphic. We invite the community to build new domains and to conduct empirical studies on Olli's shared runtime.

While older adults represent a growing proportion of the global population, their presence in visualization research remains limited. In this paper, I present reflections from a series of human-subject studies conducted with older adults as part of a multi-year research effort. These studies include a controlled laboratory experiment, online evaluations, and an in-situ investigation with participants above age 60. Based on these experiences, I provide methodological takeaways for conducting visualization research with older participants and propose directions for future work. This work ultimately aims to provide practical guidance and encourage broader inclusion of older adults as participants in visualization research.

Matplotlib was designed with the option driven design (ODD) goal of providing a high degree of customization and defaults that are appropriate for many different use cases. To achieve a balance of configurability and maintainability, Matplotlib has a modular architecture where visual information is encapsulated in the artist objects and output is managed by independent backends. This architecture makes it feasible to implement more accessible options in the library and expose more information about the visualization to downstream accessibility tools, but the Matplotlib maintainers lack the expertise in accessibility to gauge what would be useful to implement internally and what information would be useful to expose through a public API.

Automated accessibility checkers evaluate a single snapshot of a page, but many barriers only emerge over multiple interactions rather than a static snapshot. Several WCAG~2.2 criteria depend on geometry, focus state, or interaction history, such as a button that becomes obscured once a sticky header appears. Others arise when the visual layout, the screen-reader order, and the keyboard (Tab) order of a page disagree, so different users encounter the same content in conflicting sequences. Auditing these requires reasoning about geometry and structure alongside the live page, which makes it both a detection and a visualization problem. We present Site Accessibility Auditor, a Chrome DevTools extension with two coordinated lenses. The \emph{interaction} lens automatically drives the page through its reachable states and overlays interaction-dependent issues as inspectable visual evidence on a full-page capture. The reading & focus order lens aligns the visual, screen-reader, and keyboard orders, shows where they diverge, and suppresses intentional patterns such as skip links. We actively evaluate the extension as site owners and as an auditor. In a design partnership with the Utah Wastewater Surveillance System dashboard, our tool motivated the team to ship accessibility fixes. We used the tool on the Our World in Data site, which shows that the approach runs on an unmodified in-the-wild example.

Data visualization aims to augment human cognition and make data accessible to diverse audiences. As data increasingly shapes participation and decision-making across many domains, there is a growing need to examine whether prevailing assumptions in visualization adequately reflect the diversity of human abilities, experiences, and needs. We argue that cognitive accessibility provides a critical lens for examining these questions and functions as a stress test for visualization theory. Drawing on cognitive accessibility research and our experiences studying accessible visualization, we identify three interconnected assumptions that shape visualization research and practice: assumptions about what forms of cognition visualization supports, how accessibility is defined and measured, and whose needs and abilities are centered in design and evaluation. Making these assumptions explicit reveals opportunities to rethink longstanding approaches and open new directions. Ultimately, we believe that cognitive accessibility can serve as a catalyst for innovation, expanding what visualization supports, whom it serves, and the roles it plays in people’s lives.