Using cutesy interactive images to illustrate data can run afoul of good taste. Something can be florid instead of floral, for instance. But the OECD found that flowers could make people stop and smell the data in a way that brought home to them its meaning in a powerful new way.
The OECD (Organization for Economic Cooperation and Development) is famous for its data. It has troves of data on economic indicators in different countries. It usually puts that data out in giant reports, with page after page of bar charts and enough tables to create stacks of printout furniture.
Most of that data was meant for economists and other specialists. But in 2009, an independent commission presented France’s then-president Nicolas Sarkozy with thoughts on how to develop new indicators of well-being across countries, ideas that go beyond traditional measures such as gross domestic product. At the OECD, people thought it already had plenty of data that did just that.
“We had been working on this for more than 10 years but no one knew about it,” except economists and other specialists, says Vincent Finat-Duclos, a data editor at the OECD in Paris. “We thought we should make an attempt to show [to the public] what we were doing.”
What the data visualization shows: The Better Life Index is the result. Unveiled for the organization’s fiftieth anniversary in May 2011, the visualization enables users to set their preferences for viewing data sets and ranking 36 countries based on what they find important, whether it is jobs, education, life satisfaction, or eight other choices.
The presentation is a departure for the OECD, Finat-Duclos says, because rather than presenting rankings of 36 countries based on analysts’ measures of objective data, users are in control of what they view and how they rank the results. That decision took some discussion (one sticking point, an OECD blogger noted, was that some countries did not want to be ranked). But after six months of considering the idea, the OECD decided to move forward.
About the data: Selecting to the data fell to the OECD’s statistical directorate. The statisticians ultimately came up with 11 categories—housing, income, jobs, community, education, environment, civic engagement, health, life satisfaction, safety and work-life balance—each including data from as many as three separate individual indicators. These were provided for 36 countries.
Presenting the data was a different story. Finat-Duclos says the statisticians decided to create a website to allow the public to easily access this data. The committee also decided to go outside its walls for help. In part, that was because it wanted to avoid using the same visualizations and interactive methods it had been using.
The design and development work: After an open bidding process, the OECD selected RauReif, a Web design firm in Berlin, and Moritz Stefaner, a German designer who calls himself a ”truth and beauty operator” based in Lilienthal. The OECD gave Stefaner carte blanche to do what he liked image-wise, but it was clear on one thing: people should be able to weight categories that were more important to them.
“This form of visualization hasn’t existed before,” Stefaner says, thinking about the task of creating a visualization that shows 36 countries according to 11 indicators, with weighting. So he decided he would build the visualization from scratch.
Key presentation choice: Stefaner says he wanted the visualization to be a little playful, because he believes it helps people engage with and learn from data. Stefaner came back two months later with three options, one of which happened to be flowers that grew taller based on overall quality of life, and whose petal lengths reflected relative strengths and weaknesses based on the index.
The flowers didn’t start as flowers. They were wedges in a more typical bar chart. But Stefaner fiddled with them, rounding the edges, and saw they looked a bit like flowers. While flowers might seem frivolous, Stefaner says large data sets created by human activity often result in natural-looking visualizations.
The OECD, however, is not flowery. And, initially, the flower designs had a ‘mosquito’ problem, in that the flowers looked like mosquitos. Finat-Duclos said that at first, a histogram that used circles seemed more effective. But as Stefaner tweaked the flowers, it became clear that they were the best way to display the data. The committee began the process of getting layers of serious bureaucrats to approve a serious chart that included flowers.
Stefaner posted a short video showing some of the designs he considered while working on the visualization:
Technical notes: The presentation uses HTML5 for most of the site and Adobe Flash for the flower visualization. “We experimented with displaying the flowers in HTML5 but ultimately stayed with Flash for now, due to the better rendering and graphic fidelity,” Stefaner says.
The Better Life Index website is based on a customized WordPress install. The Flash-based visualizations are developed more or less from scratch, but use existing libraries such as the Flare toolkit.
Stefaner says that at the start of the project, the team used standard software like Tableau and Excel to investigate the data sets supplied by OECD, “but quickly resorted to custom written tools, giving us the needed flexibility in prototyping various display forms.”
Response to the work: The end result, the Better Life Index, drew powerful results. Within a few hours of posting, it had received more than 150,000 views. “When we saw that people understood it we were very happy,” says Finat-Duclos.
The data visualization consultant Stephen Few wrote on his Perceptual Edge blog that the Better Life Index succeeded to both awe and create ah-ha moments for users. “The designers of this infographic achieved a marriage of form and function, beauty and usability, that did not subsume one to the other in an unequal partnership as many infographics do,” he noted.
User insights: The OECD collected responses from close to 6,000 users of the Better Life Index in its first year. Finat-Duclos says one encouraging sign is that none of the 11 categories were ignored or used much more than the others. The results also show:
- A consensus that health, life satisfaction and education are the top-ranked topics, whether a user lives in Latin America or a Nordic country.
- Gender-based responses vary only slightly in their preferences, with men giving greater weight to civic engagement, income and health, while women rate community, housing, work-life balance and life satisfaction slightly higher.
- Age-based differences show that younger users (aged 15-34, and more than 55 percent of the respondents) place more emphasis on work-life balance, income and jobs. Older users (over 65) consider health and environment as priorities. Health becomes a key factor when respondents reach age 45.
The Better Life Index has not meant stopping to smell the flowers (though the team did have “a really nice drink,” Finat-Duclos says). In May, Stefaner and the OECD put the finishing touches on some new elements, particularly showing differences between genders. Next up: adding social inequalities, like income and education levels.