Andy Kirk talks data visualization and the Olympics
As the clock ticks closer to the Opening Ceremony of the London Games, publications around the world will undoubtedly begin revealing data visualizations that explore the breadth and depth of one of the world’s greatest sporting events. Andy Kirk, the United Kingdom-based data visualization expert who founded Visualising Data, took time to help make sense of what goes into building an effective presentation, and what considerations might be unique to the Olympics.
There is so much raw data associated with the Olympics, how do you determine what could make a useful visualization?
The sheer quantity and the ease of access to so much data surrounding the Olympics represents a great opportunity to create valuable visualizations; especially about subjects that might not have been covered, or new approaches that might take a different angle on a slice of data. Typically, the most interesting visualizations will respond to a curiosity that all of us might have, but for which we don’t have an answer – the type of question you might discuss with your mates in a pub that starts with a phrase like, “I wonder how many ….”
Are there unique challenges to displaying data from the Olympics, compared to other major events?
Aside from the large numbers of competing countries, in comparison to any other big sporting event, the most unique aspect of the Olympics is the variety of sporting activities taking place. Additionally, there is a challenging range of different measurement frameworks for determining success across these sports. For some it is based on time, for others it is a score or points tally, others still are based around tournaments and knockout situations but have their own unique scoring system within each stage. You’ve then got sports that are purely individual and others that are team-based which adds an extra layer of variety.
These factors create a unique challenge for arriving at an effective, flexible visualization framework that successfully handles all this variety and volume. Another challenge is to determine if your focus will be entirely historical or whether you will try to accommodate live updates as the Games commence, in which case you’ve got to keep on top of the relentless stream of new data and seamlessly port this into your piece.
What are some of the biggest technological differences you’ve noticed between the data visualizations of today and those from the 2008 Games?
It may be just four years, but a lot has happened in that period to change the technological landscape and create a whole host of new problems and opportunities.
From a platform perspective, we now have such a massive take-up of different devices and channels through which we can consume visualizations: from the smartphone to the tablet, to the high-definition, large resolution monitor. This creates a big headache for the designer looking to create a single portable solution to suit all requirements.
Then you’ve got the development of powerful programming environments and libraries that equip the skilled technician with even more capability. Conversely, the decision not to support Flash on Apple’s portable products has required a strategic skill shift for many designers with core skills and a history of innovative interactives in their portfolio.
Finally, the data is now available in far more shapes, sizes and places, particularly with the rapid diffusion of social media participation and the vast quantities of qualitative data this has now created in a high-velocity real time environment.
How do you balance showing minutia that hardcore sports buffs would be interested in while still keeping casual viewers interested?
It is unquestionably a difficult balancing act but one you can achieve with thoughtful design and intuitive interactivity. You can facilitate this combination of needs by creating layers. On the first layer a global picture can be presented using the lowest resolution (statistics or aggregates). This should satisfy casual viewers at the top level, but then you create deeper exploratory access to the data behind this layer and open up the details that exist within the aggregates, showing the minutia, as you say, and unlocking a higher resolution of potential insight at a more personal story level.
For those who may not build data visualizations in their newsrooms, can you give us a sense of the process?
Initially, there needs to be or will be some sort of curiosity, some sense of a motivation to find answers or communicate a story. From this you will define at an early stage the context of your intentions, as well as the scope and parameters surrounding the project. These might include factors like time scales, intended audience, potential formats, technical capabilities, design/layout rules.
After this we begin an iterative process of data acquisition, preparation and consolidation in parallel with the further refinement and clarity of the intended analytical slices or data stories we are looking to portray or create access to. We use visual analysis on a personal level to learn about the data’s properties, the range and distribution, relationships and scales to prepare our understanding of the potential architecture of our solution.After all this preparatory work we can then begin to formulate our design concept, sketching and testing out different options, reasoning or rationalising which of our alternative design paths we will follow. Once we have settled on a concept we then build the piece, which could be an illustration if you are looking more towards an infographic or perhaps a web-based construction if we are developing an interaction solution.
With regards to resources, the quantity of people it can take for creating such pieces is entirely linked to your own context and the resources available to you. It could be a collaboration of an entire team or a sole venture. Either way, the process for achieving a successful design, as well as the skills and mindsets you need to ‘cover off’, will be the same.
Have you seen any great examples of Olympic data visualizations for London — or even for a different news event — that could inspire one for the Olympics in a similar way?
So far I haven’t seen an enormous amount of Olympics visualizations, I sense most are waiting until closer to the opening ceremony before unveiling their work. The best Olympics visualizations, as with any subject matter, will be those that judge best what their specific audience are interested in at a given time, responding to the sort of curiosities most of us would have about a huge event like this. So if you think of the build-up, the focus will be largely on the history, the key competitors and the schedule. Then, as the event begins, the emphasis switches to the emerging developments, the personal stories of glory and the successes at country level.
Many successful examples manage to create an immersive experience whereby the user engages directly with the visualization, either through interesting interactive features or by user-driven queries, such as this piece published recently by the Washington Post that invites you to enter your age and determines your suitability for different events.
What Olympic event is most similar to creating a data visualization?
I’d probably suggest it is somewhere between the heptathlon and the decathlon because any data visualization project requires such a variety of capabilities to achieve success, you need that genuine all-round skill set — design, research, analytical, technology, journalism, statistics, project management — that can only be matched in a sporting sense by those who are able to compete in these unique events. You can read more about my idea of the ‘8 Hats of Data Visualization Design,’ that explains these many different necessary mindsets or capabilities.
Click here for more SND coverage of the London Olympics.
(Kyle Ellis is a designer for CNN Digital in Atlanta. Follow him on Twitter @kyleellis.)