Abstract

Data visualization has become an increasingly important means of effective data communication and has played a vital role in broadcasting the progression of COVID-19. Accessible data representations, however, have lagged behind, leaving areas of information out of reach for many blind and visually impaired (BVI) users. In this work, we sought to understand (1) the accessibility of current implementations of visualizations on the web; (2) BVI users’ preferences and current experiences when accessing data-driven media; (3) how accessible data representations on the web address these users’ access needs and help them navigate, interpret, and gain insights from the data; and (4) the practical challenges that limit BVI users’ access and use of data representations. To answer these questions, we conducted a mixed-methods study consisting of an accessibility audit of 87 data visualizations on the web to identify accessibility issues, an online survey of 127 screen reader users to understand lived experiences and preferences, and a remote contextual inquiry with 12 of the survey respondents to observe how they navigate, interpret, and gain insights from accessible data representations. Our observations during this critical period of time provide an understanding of the widespread accessibility issues encountered across online data visualizations, the impact that data accessibility inequities have on the BVI community, the ways screen reader users sought access to data-driven information and made use of online visualizations to form insights, and the pressing need to make larger strides towards improving data literacy, building confidence, and enriching methods of access. Based on our findings, we provide recommendations for researchers and practitioners to broaden data accessibility on the web.

How accessible are visualizations on the web?

In our study on the accessibility of COVID-19 visualizations on the web, we found the following:

Of 76 visualizations audited from top-ranked Google Search websites on the COVID-19 pandemic:

• only 16% were rated by expert auditors as very or extremely accessible, whereas 71% were rated as slightly or not at all accessible for screen readers users.
• only 15% supported screen reader access to specific data points.
• only 5% conveyed any form of trends or patterns to screen readers.
• only 43% provided accessible tabular representations.
• only 35% exposed supported interactive features, such as sorting, panning, filtering to screen readers.

Of 127 screen reader users surveyed:

• 65% reported encountering data-driven media at lease once a week.
• only 27% of screen reader users describe that the media they typically encounter is accessible.
• 94% of respondents indicate they have concerns accessing accurate COVID-19 data.

What can we do to make data more accessible?

As practitioners:

• Consider accessibility from the beginning and use libraries that support direct screen-reader access to the data. HighCharts, SAS Graphics Accelerator, Apple Audio Graphs, VoxLens, and Chart Reader are several. Study participants felt less confident interpreting the information accurately when depending solely on text summaries that are others’ interpretations of data.
• Review the Chartability heuristics when designing your data experience. Assess the accessibility of your data experience.
• Consider providing multiple different methods of access to support users of different screen reader configurations, experiences, and preferences. Many users in our study combined insights from multiple representations (e.g. table and alternative text) to complement information gaps and inform exploration. Providing downloadable data files allows users to explore the data through the familiarity of their own tools.
• Screen readers present materials sequentially. Having to retain multiple quantities of information while navigating using screen readers can be mentally taxing to users. Consider screen reader experiences as stories. Provide context with specific data points and details.
• Implement visualization experiences that are complete. Additional information, such as the source of the data, the data update time, options to download, and alternative representations should be easily accessible through each visualization experience (e.g. place these details in the same header as the visualization).
• People’s prior experience and domain knowledge greatly affects their takeaways from visualizations. Consider a diverse audience with different backgrounds when introducing new visualization features. Provide instructions for how to interpret visualization content.

As researchers:

• Investigate methods for embracing interactive features and alternative modalities in web infrastructure that support flexible navigation, gestalt understandings, interactive feedback, expressive communication, and multiple levels of data abstraction. Many users in our study made use of their own interactive features such as find, linked lists, and multiple tabs to reduce the effort of navigation.
• Investigate ways to make connections between multiple complementary accessible representations more explicit to provide more tightly coordinated views.
• Research methods and interactions to build data literacy as screen reader users consume data visualizations on the web. For example, progressively uncovering details can can provide scaffolding for helping people better digest and understand information.

Papers

• Danyang Fan, Alexa F. Siu, Hrishikesh Rao, Gene SH Kim, Xavier Vazquez, Lucy Gerco, Sile O’Modhrain and Sean Follmer. 2023. The Accessibility of Data Visualizations on the Web for Screen Reader Users: Practices and Experiences During COVID-19. ACM Transactions on Accessible Computing 16.1 (2023): 1-29. * These authors contributed equally to this work. https://dl.acm.org/doi/10.1145/3557899