Escaping ‘data blindness’ with the help of data visualisation

Thu, Oct 6, 2016

Frédéric Dubois

Opening data is one thing. Many open knowledge activists out there will cheer when seeing yet another city making some of its public interest information available to the rest of us. There is also reason to cheer, since any new open data release is a step closer to knowing what institutions are up to. Also, let's face it: albeit some momentum around open data in the last decade, the number of open data platforms making data sets available on a regular basis remains the exception. It is not the default modus operandi out there.

Again, for every open data set released by a local, regional or national government, by international organisations or any other business or civil society institution for that matter, we say chapeau! Yet, the truth is: we're collectively blind to data. Once data is being released for instance on a platform such as OpenDataCanarias – the Canary Island's open data service – questions pop up: do we actually see the data? Do we make links between two data sets? Can we draw conclusions from the data published? The short answer is a resounding no.

In fact, when data sets are presented in a straightforward fashion, respecting open standards – such as these here (2015), compiled by UK's Cabinet Office, or these here (2011), if you want some more context – open data enthusiasts, activists and data scientists can actually start working. They start organising, bundling, rearranging, connecting. But that's 'they'. For most of us, the columns of numbers and the lines of words just mean another Excel sheet that we can hardly decipher. We are data blind. So much for the 'no' answer.

Data dashboards

The long answer is … yes.

OpenDataCanarias, much like other open data portals out there (e.g., the citizen-driven OpenDataBC in Canada, Copenhagen Data in Denmark or Buenos Aires Data in Argentina) do a decent job in making government data accessible and public. But much like your local library, it only gets exciting once you walk in, take a book off the shelves and dive into a story that you can relate to.

One step in the right direction are data dashboards. Several open data portals, including USEDATA/Myanmar's (for an example on the national level), and Jersey's (local level) have them in stock. These dashboards not only make different types of information readily accessible. They present information in a way that is comprehensive (like from a bird's eye perspective), with metrics in real time, and make heavy use of data visualisation.

USEDATA/Myanmar dashboard Image: Dahboard of the USEDATA/Myanmar portal

Data visualisation

What A-grade data science students might call the last step in data analytics, data visualisation is in fact the key for opening the eyes of Everyday Joe to the beauty of open data. The visualisation of big data sets with static infographics or dynamic and colourful graphs is what adds value to data. It helps us make sense of data, contextualise it, understand it. A compelling data visualisation such as this, or useful data visualisations such this or this, are good examples.

There are even data visualisation standards for the web, that indeed go back in time, but can become handy when making the data visualisation integral part of your website. This would merit a longer discussion on what data visualisation standards would work best for open data, but that's outside of our scope of this post.

Conversation keeps the data alive

Data dashboards, interestingly, often allow the user to comment and share links to data sets and stay in conversation with fellow users around the data that is being visualised. So not only do we exist blindness via sophisticated design of big data, we also start talking about what we see. This is the collective sense making of data. Some dashboards use Facebook APIs, others make sure they are active on social media platforms such as Twitter, Facebook or Google+. In many cases, it's a combination of these two strategies which keeps users in conversation and the data alive.

Storytelling

For the data blind and deaf people out there, meaning the ones that don't engage in the conversation, there's more. Storytelling elements (e.g., photography, portraits of people affected by a situation depicted by the data, 3D visualisation, complex infographics) can go a long way in attracting delivering the data to the user. We start entering the champion's league.

Storytelling is the one thing that will capture the attention of even the most (data)-blind person out there. But that's easier said than done.In my next post, I will look at techniques to further the life of open data through data storytelling. I will be digging into best practices from the field of data-driven journalism in particular.

Conclusion

If you are at city hall and you want to revolutionise the way data is used in your city, hire three people: a data scientist with a design background, a community manager and, a journalist-in-residence. If that busts your budget, find the budget. Making data public is important. Visualising it, is even better. It let's users exit blindness, find a way to partake in public conversation and tell the stories that we need in democracy.

Frédéric Dubois is a journalist based in Berlin. He is the Managing editor of Internet Policy Review, an open access journal on internet regulation - published by the Humboldt Institute for Internet and Society. He writes about the internet governance, new media, & open data. Frédéric is the co-editor of two books on media & journalism, as well as author & producer of award-winning interactive features. On Twitter: @fredericdubois