Skip to main content Skip to Footer

September 14, 2015
Asking and Answering
By: Data Insights R&D Group

It was exciting to attend the Eyeo Festival in Minneapolis, MN on June 1-4 with other visualization designers and developers from Accenture. Eyeo is an event to share ideas around design with a wonderfully diverse crowd of thought leaders from across academia, industry and the public sector.

Of the many interesting presentations at Eyeo, Sara Hendren’s talk on inclusive design stood out for its provocative call to action. As a professor of adaptive and prosthetic technology at Olin College, Hendren made a strong case to the audience that the discipline of inclusive design must involve simultaneously solving important problems and asking tough questions.

In the context of Hendren’s work, this often means questioning what “normality” is in terms of the physical body. Assistive devices, such as replacement limbs or hearing aids, are often assumed to exist within social and medical contexts that create default pre-conditions for their design. Why judge a replacement limb only by its ability to complete the same functions as the limb it is replacing? Why assume a hearing aid must be hidden? Hendren’s work asks us to escape our narrow thinking around what ability and disability mean when designing new objects.

Using visualizations to ask questions
I’m intrigued by the possibility of applying Hendren’s ideas to business data visualizations. In enterprise environments, data is referred to as “actionable” – it is regarded as a go-to tool for answering questions and making tough decisions. However, this instinctive prejudice towards raw data having ready answers would benefit from skepticism.

Infographics often serve a purpose in the context of business and news reports, to stress a particular point. But it would help both business analysts and journalists validate the authenticity of their data and the strength of their own conclusions if they take advantage of tools like D3 to create visualizations with strong searching and interactive capabilities—what are sometimes called “exploratory visualizations” —rather than using static illustrations to express pre-formed opinions.

Take a recent Washington Post data journalism piece on laborers in Qatar, which, due to simple miscommunication, falsely implied that 1,200 migrant workers died in Qatar while working on World Cup construction sites1. Qatar quickly protested that the numbers were not true, and after further review, the newspaper revised its visualization to note that the number was for total migrant-worker deaths in Qatar.

However, it was too late. The inaccurate statistic and visualization had already been widely shared across social media, and the impact might include wide-ranging consequences for Qatar and its tourism industry.

The process of making interactive visualizations, using tools such as D3, forces creators to know datasets intimately, in order to explain to the user how to navigate and manipulate the dataset. A static infographic, such as the Qatar one, merely captures a glance at the data. The deep-dive into data that creating an interactive visualization requires of the developer increases the likelihood that the creator will catch inaccuracies in the data before publishing it online.

Additionally, such visualizations better allow readers or business-people to ask questions of the data and similarly identify interesting questions or issues concerning the dataset on their own. In such a way, these visualizations can live up to the mission statement set forth by Sara Hendren and allow users to educate themselves and pre-conceived notions to be disproved.

Yet, it is hard to do this properly. If left completely unguided, users may often come to the wrong conclusions from their independent analysis of data. Unguided and uneducated data exploration can lead to dangerous decision-making. In business, this might occur if analysts rely on complex formulas in dense Excel spreadsheets to answer questions about product profitability. Interactive visualizations give designers the chance to educate users in an unintimidating manner but still allow users the freedom to create unique analyses of their own.

This is how the tension that Sara Hendren describes between solving problems and asking questions translates into a technical challenge for data visualization designers. It is difficult, but necessary, to find a halfway point between forced guidance and free exploration. The New York Times was able to achieve this delicate balance in their recent interactive visualization of the yield curve2.


1The Guardian, “Qatar outraged by Washington Post myth about World Cup deaths,” June 3, 2015. Retrieved July 6, 2015 from http://www.theguardian.com/media/greenslade/2015/jun/03/qatar-outraged-by-washington-post-myth-about-world-cup-deaths
2The New York Times, “A 3-D view of a chart that predicts the economic future: the yield curve,” March 18, 2015. Retrieved July 6, 2015 from http://www.nytimes.com/interactive/2015/03/19/upshot/3d-yield-curve-economic-growth.html?_r=1&abt=0002&abg=0

More blogs on this topic

    Archive