Continuing with the idea of “why did we do that?” Here’s some points that went into the design of two graphics that required the spread of data across a map.
Muslims of Europe
– Map used for percentage to give visual impression of prominence within populations
– Adrian customised the template palette to create a gradient of colour that goes from weak (low percentage) to strong (high percentage).
– Numbers are given by bubbles, only charting populations bigger than 100,000. The bubbles provide instant impression of the largest populations.
– The bubbles were sized and coloured for maximum impact, while not obscuring the percent layer of data. We tried them with fill colours with transparancy, but that obscured the percent shades below. Therefore we converted the bubbles into rings.
– We needed to cut off the smaller populations that don’t add anything interesting to the story, for example 2,000 Muslims in Estonia.
– Needed a grouping for “Europe” so we used the EU28 but found a few strange gaps in the map, so added “surrounding countries.”
– We avoided Russia and many former soviet countries to keep the focus on western Europe.
– The former Yugoslav states posed a problem, partly because of their small size and partly because of their exception to the rest of the graphic. We originally sized bubbles for each of the countries but it made more sense to add them together. They are not really part of the western European story on Islam, but they provide context.
– The all Europe population percentage “clock” is a method for showing changes over time without the repetition of three pie charts.
Asia Catholics
– Colour shading also used for percentage population
– Gradation from low to high percentage, with exceptional cases Philippines and East Timor designed to stand out (strong enough? debatable)
– Bubbles were an option but may have lead to over crowding, so we chose to write the numbers out
– For millions it’s often more effective to write out all the zeros (1,500,000) instead of contracting (1.5 million). We wanted to convey the unexpected high numbers of Asian Catholics
– The numbers are sized differently to help the large numbers stand out.
– “main religious groups” were added for context. We didn’t want that information to take over from the main purpose of the grpahic — Catholicism, so we tried to show it in a way that would be clear without dominating
UPDATE
Paris French and Spanish graphics made changes to our Muslims of Europe graphic. It’s definitely a good idea to have the option of developing graphics already on the wire. But in the same spirit it is a good idea to encourage debate on design and editorial decisions involved.
For the HK version, other than trying to avoid repeated patterns (pie charts) we thought the “dial” worked well because it shows proportionality, just as pie charts would. The overall effect is to show how small 6% is, as is 8% for 2030. The dial is a good way of showing that Europe is, and is set to remain, predominantly non-Muslim.
The Paris version looses the element of proportionality, which is what percent values give. Instead we only see growth. My question is how the percent values relate to the size of the bubbles. It looks like the actual numbers were used to make the chart rather than percentages (29.7 m, 44,1 m, 58.2 m) but if that’s the case the chart should be labelled with those values, not the percentages. To show percent values as an area would only work if we were to show the area for 100%.
newsgraphicsblog 