Welcome to our November issue of the Accessibility Minute Newsletter! This newsletter is produced by the CU Boulder Digital Accessibility Office and covers one accessibility skill or topic per month. Please visit the DAO website to access past newsletters. As always, thank you for taking a minute (or two!) to read.

Accessible Data Visualizations

Data visualization is a graphic representation of information and data. Essentially, they aim to make raw information (data) easier to understand and engage with. Data visualizations can become complex due to all the potential features. This may include graphs, pretty colors, and layers of visualizations. An example of an item that could be made into a data visualization is student enrollment over time or climate survey results.

Data visualizations are difficult to make accessible because they often fail to address other ways individuals consume the information presented. It is more than just color contrast! The CU Boulder Digital Accessibility Office has partnered with the Office of Data Analytics to figure out what can be done to make the presentation of data visualizations more accessible. We have created an accessible proof of concept that we will detail in this month’s newsletter. 

Data Visualization Challenges

There are many challenges when it comes to data visualization. The list below outlines some common, but not all, accessibility challenges and barriers that users can experience.

  • Color: When considering color, we must use appropriate color contrast (4.5:1 or 3.1:1), and color alone should not be used to convey meaning. For instance, if you have a gain/loss line chart with loss in red and gain in green, the lines should also be visually different. Making the green line dashed or the red line bolded differentiates them in a way that is not based on color alone, making it accessible for users who are colorblind.
  • Lack of ability to add headings: For many dashboards or visualizations it can be difficult to add headings. Headings are necessary to add structure and meaning to the page or a dashboard for users to navigate.
  • Things are not what they seem: Some dashboards include elements that look like buttons that are not programmatic buttons. This is problematic for many assistive technologies because the technology works off of the code and what the element is programmed to be. In other cases, a table may not be a programmatic table, which will limit the ability to navigate using table navigation or be able to understand and comprehend the information in the table.
  • Too much information to include in alternative text: Some graphs may be too complex to describe in the alternative text. While captions can be used, it may not provide enough context or information for somebody to get the intended message from the visualization.
  • Reading order: The reading order does not make sense for some dashboards. Assistive technology may jump over certain aspects or not get to other elements in an order that makes sense. Reading order is essential for individuals utilizing assistive technology to make sense of the page and the information being provided.

Alternative Access

While providing an alternate or flat document instead of an interactive dashboard can be an interim fix, it should not be considered the final solution. Everybody should be able to access a dashboard or data visualization. When considering providing a different type of document or platform, the functionality of the alternative document or platform should be considered, not the identity of those using it.

CU Boulder's Tableau Accessibility Project

Needing to figure out how to produce accessible data visualizations, the Digital Accessibility Office partnered with the Office of Data Analytics to focus on Tableau visualizations and determine if the tools Tableau provided could be implemented in a different, more accessible way. Through multiple months of iterative work, the two teams developed a proof of concept for using Tableau’s Application Programming Interface (API) for dynamic control and updating of embedded visualizations and data tables in a content management system like Drupal.

Using the API

By using the API and an accessible HTML table provided in a Tableau dashboard, it is possible to import the data visualization and the data to a Drupal web page. Information such as headings, HTML tables, filters, and other interactive elements on the Drupal web page are accessible to users. The visualization aspect is imported for users who find that helpful as well. Users can adjust the data and information on the Tableau server using the filters on the Drupal page, allowing users to interact with the data in an accessible way.

Custom Module

Through our amazing collaboration, this custom module is open-source and available for anybody to use. For more information on the custom module, please visit the Accessible Tableau Integration website.

Why is this Important?

All individuals should be able to access the data you’re sharing. Fortunately, our custom module allows creators to make their data visualizations more accessible. This module can also be used as a replacement for previously created Tableau dashboards that are inaccessible.

November Challenge

Review the work presented in this newsletter, join in, or share the work being done to assist in making data visualizations more accessible.

DAO News

Follow the DAO on LinkedIn! By following us, you'll get access to behind-the-scenes insight on our office, information about our services, applicable accessibility tips posted every Tuesday, resources, upcoming event information, and more.

Office hours will be canceled for December, 2023, due to the holiday break.

Your Thoughts

We want to hear from you! If you have any questions or comments, please send us your thoughts on this month’s topic.

If you have questions, comments, or would like support with accessibility, please contact us at DigitalAccessibility@Colorado.EDU.