Estimated reading time: 5 minutes

Summary

Begin early in the visualisation process thinking about how you might ensure that the message communicated in your visualisation is accessible to as many viewers as possible, no matter their disabilities, impairments, or variations thereof
Any visualisation published online must meet basic criteria for accessibility, including alt-texts, subtitles and maintaining colour and contrast
Avoid overwhelming your reader

While basic accessibility is a prerequisite for effective communication, especially as more and more material is published digitally, ensuring that the message being communicated can be received by as many viewers as possible, including viewers with disabilities, impairments, or any variation thereof, is an important part of the visualisation process. This is something that needs to be considered early on, rather than applied as an afterthought.

Keep in mind that impairments, disabilities or variations do not necessarily mean chronic conditions or constant circumstances. Blindness, for example, is a chronic condition, but having temporarily reduced vision due to unfavourable lighting conditions, or a medical condition, is not. It is also not an absolute; no person is blind in the exact same way as another. There is a spectrum.

Publishing online

In addition, when you publish your visualisation online, you will have to take into account certain laws, requirements and standards for doing so, among them those detailed in the World Wide Web Consortium's (W3C's) accessibility criteria.

Alt-text

According to the W3C's accessibility criteria, an alternative descriptive text must accompany any visualisation published online, a so-called alt text. This is to ensure that anyone who cannot see the image nonetheless understands what it is and what function it is meant to serve.

Yet writing adequate alt-text can be a challenge, and doing so for relatively sophisticated diagrams can be overwhelming for the uninitiated. For some tips on how to get started, see graphic designer Amy Cesal's article for Medium, "Writing Alt-Text for Data Visualization" or try Poet (an interactive tool for practising when and how to describe images and diagrams).

Subtitles for all

If you produce and publish an animation or video, you will also need to provide subtitles, especially if there is an explanatory and narrative voice-over. Subtitles are helpful for everyone, even those with perfect hearing sometimes choose to have the sound off, because they're consuming the video in an environment not conducive to having it on (for example when commuting to work on the subway or train). In addition, being able to read and listen to material at the same time has been shown to have favourable effects on comprehension, particularly of vocabulary.

These days, there are many tools available online that make it easier to subtitle your video or animation.

Contrast and colour

A common mistake is to convey important information using colour alone, such as using a "traffic light palette" (red/yellow/green) to communicate negative or positive change. One way to enhance the accessibility of such a design and help viewers differentiate elements could be to reinforce the message by providing supplementary encoding, for example, by combining red with a shape or pattern.

Designing with sufficient contrast is another important consideration when publishing material online. Not only to include people living with a disability or impairment, but also to ensure that your visualisation is effective in as many circumstances as possible. Too little contrast, for example, can be devastating if the recipient is sitting with their laptop outdoors. A shining sun will eliminate all differences in colour.

A restrained colour palette is therefore a great advantage in visualisation. Use colour to highlight specific elements - graphic details that are central to the message being communicated. Be sure to let the colours differ enough from each other, even those colours that have a more subdued role to play.

Read more about colour and accessible design in the module on colour.

We can all do better!

The accessibility criteria established by the W3C has existed for years. Nonetheless, WebAIM's annual report on accessibility online, in which they examine one million of the most commonly used websites globally, concludes that major shortcomings remain. In your own visualisation work, do better!

Cognitive load

Closely related to a restrictive use of colour is ensuring that you do not overwhelm your viewer. You don't want to emphasise everything all the time; you want to protect and direct the viewer's attention carefully and make sure that the message central to your visualisation is the one that stands out.

To reduce the cognitive load, aim for as simple an expression as possible. This is not about minimalism; a visualisation can convey a lot of information even if the expression is restrained. In many cases, it can probably convey more than if the graphics were maximised.

Simplify and clarify

Try to strip away anything unnecessary in your graphics. Reduce the number of colours, fonts, explanations and dimensions. Try to focus on clear contrasts and save your energy for what is important.

Questions to ask yourself:

How many colours do I need to use?
Are there any elements that would benefit from being relatively colourless?
How many different fonts do I need? Which font is used in the explanatory text? Can I reuse any? Try to stick to a maximum of 2 different fonts; one for the heading and one for the text.
Are all explanatory elements needed? Clear scales, units, axis descriptions, and the like should of course be included, but can you reduce the number of indications or their frequency? For example, does a third dimension serve any purpose? Do you really need a shadow side on the bars in your chart? Does the fourth dimension (i.e. time) serve any purpose? Do you need to present your findings in an animation, or would a static series of images be just as clear?

These kinds of questions are important to continuously ask yourself during the working process. The joy of creation and a desire to express yourself can easily get the better of you. By all means, embrace your creativity, but stay focused on clearly communicating your findings. Dare to challenge yourself! How many elements can you remove before your main message is lost?

Simplifying and clarifying without distorting

In The Visual Display of Quantitative Information, Edward Tufte discusses the concepts of data-ink ratio and chartjunk. Both concepts emphasise the importance of Tufte's dictum: "above all else, show the data." The data and your findings should be central to your visualisation, not the graphical form or aesthetic expression.

An example of accessible visualisation: To park or not to park

In some instances, it is particularly important that the information conveyed is correctly understood by the recipient. This includes sampling instructions, a dosage schedule for a certain medicine, or something as simple as parking rules on campus.

One instance of complicated signs with unclear instructions regarding parking led a designer, Nikki Sylianteng, to start the project, To park or not to park. The goal of the project was to simplify and, above all, clarify how parking rules are presented.

In the example below we see how parking rules often are visualised and explained in the US (figure 1, on the left). Next to it (figure 2, on the right), we see Nikki Syllanteng's much clearer and far more accessible schedule.

Jämförelse mellan svårtydda skyltar och ett användarvänligt, designat schema

Images of parking rule signs before and after Nikki Syllanteng's redesign. From the project To park or not to park (Creative Commons BY-NC-SA 4.0).

Syllanteng's schema reduces the cognitive load on the viewer considerably because the information being communicated is coded in a much more effective manner that is more adapted to how human perception actually works.

Recommended reading and resources

On writing alt-text:

Amy Cesal, "Writing Alt-Text for Data Visualization," Medium (2020).
Poet Training Tool

On keeping it simple, without making it boring:

Lisa Charlotte Muth, In defense of simple charts, Datawrapper (2021).

To minimize cognitive load:

Darkhorse Analytics, Data looks better naked. (A series of animated films).
Plotly, Above all else show the data, Medium (2016).

For more resources on how to create accessible visualisations, check out:

Dataviz Accessibility's list of resources