Data visualisation is undertaken to either:
A) drive a decision/investigation – analytical visualisation
B) tell a story – information graphics
More and more often, and with good reason, we are seeing data presented as information graphics. The data shown is not just as support for an argument or point, but becomes the story itself. In blogs, at conferences, information graphics are being used on their own to make a point.
These information graphics are a form of data visualisation and result from filtering sets of variables and abstracting them into some schematic form. This makes perfect sense as human minds don’t store independent pieces of data; brains connect sensory fragments to reassemble when we remember something. The quickest way to create these connections in most people is visually telling a story.
So what is the most efficient way to tell a story?
When we are telling a story, loads of data does not necessarily make a better picture – not everyone needs to see every value; nor will using all the data will have a bigger impact.
How often do we show all the data just because it is available? – being hijacked by a sudden need to look “smart”, “hard working”, or “thorough”, muddles the original purpose of informing.
Few of us would think to list and describe the inter-relationships of 767 numbers in a piece of text, but it is easily possible as a graph. Will doing so really serves the purpose of getting a point across?
If we want to inform and educate rather than analyse, the difference is the same as between story-telling and researching. By grouping and averaging, the final picture often tells a clearer and specific story.
I know the objection to “smoothing” is always– loss of detail creates distortion and lacks objectivity!
HOWEVER consider this: Should a story be objective?
(Can a story ever be objective given it is always interpreted, edited, embellished and delivered by the storyteller).
If the story prompts the audience to dig deeper, find out more, get involved, take an action then surely the purpose of creating a memory has been achieved.
Think of a weather report – at the end of the day I really don’t care about all the data, formulas and algorithms applied – I really want to know if I will need my umbrella so all I look out for are the cloudy rain icon.