The psychology of data visualisation: beyond the numbers

Data visualisations are increasingly a part of modern journalism as they illuminate stories in a way that plain text cannot. Large numbers, especially around difficult issues, can be reported much more effective when they are demonstrated visually.

But an ineffective data visualisation soon fades from memory and is next to useless: visualisation expert Adam Westbrooke notes that since online journalism is so visual "we better have a good understanding of what visual storytelling is at its most fundamental level".

As they become increasingly common there is the danger that familiarity might lead to visualisations losing their punch. Though Datawrapper is an invaluable tool for online journalists, there is a risk of its standard blue bar chart becoming stale and ineffective at communicating information.

Colour

There are, however, ways to make this default design less ubiquitous. Even Datawrapper allows users to choose individual colours for standard charts and tools like Paletton allow users to generate complimentary colour schemes to make the visualisation more appealing.

The following visualisation from the Guardian demonstrates how colour can be used to make a piece that might otherwise have been drab into something much more memorable.


Screengrab from TheGuardian.com

Colour choice may seem like an arbitrary decision when it comes to creating data visualisations, but there is a large body of psychological studies that show good use of colour can have a positive effect on memory retention.

There are marked improvements in retention and recollection when asked to remember something colourful over a monochrome equivalent, for example. 

Our emotions are so driven by those colours, by those imagesShannon Perkins, Wired.com

The cultural significance attached to certain colours can be effective, too. In the West the traffic light system is readily recognisable and useful for communicating qualitative results. Nor are these effects simply anecdotal: studies have shown that the perception of red as a 'strong and active' colour crosses cultural boundaries, so it can easily be used to communicate the key point of your piece. 

"It's the visual style, the visual treatment of the piece that engages the user," Shannon Perkins, editor of interactive technology from Wired.com, previously told Journalism.co.uk. "[It] keeps them and makes them want to go tell their friends about it because it looked cool and it felt cool to them emotionally.

"Our emotions are so driven by those colours, by those images".

ColorBrewer is a free tool that allows its users to create custom palettes for their visualisation, and is backed up by extensive scientific evidence regarding each palette's effectiveness. Since many common data visualisation tools have a ColorBrewer plug-in, there is no excuse for using a dull, default palette. 

Form

So original, colourful infographics tend to communicate a message more effectively, as explained in this blog by data visualisation expert Ann K Emery. But there are other fundamental choices to consider when creating a data visualisation, and perhaps the most fundamental is its form.

As shown by this diagram, the vast majority of charts created using Datawrapper were of the column or bar chart varieties. These are simple and easily graspable, but their simplicity also makes them more commonplace and, as a result, more likely to be lost in the noise.



So while not all data is always suitable, it is worth considering line graphs or scatterplots where appropriate, especially when there is a large amount of data to be communicated.

As demonstrated by this sterling example of pop culture data journalism, in which Benjamin Moore demonstrates the disparity between the public and critics' perception of films, even when scatterplots are crowded they can still be engaging and colourful.

Icons

It is possible to further differentiate your data visualisation through the use of custom icons. Tableau Public has a gallery of choice visualisations which make their point using unique graphics.

This piece by Simon Rogers illustrates the gender pay gap in various industries using familiar symbols in a gantt bar chart – for showing differences from an average – which visualises the disparity in a way that the plain text does not.

In another, the juxtaposition of familiarity with disparity – using the Olympic rings and social or economic indicators – by artist Gustavo Sousa was highlighted as particularly memorable by infographics expert John Grimwade.

oceaniaeuropeamericasafricaasia from gustavo sousa on Vimeo.

So in creating your own data visualisation remember the need to stand out from the rising tide of graphs, charts and infographics. Using colour, form and juxtaposition of the familiar and alien will make your work more likely to be remembered later.

Free daily newsletter

If you like our news and feature articles, you can sign up to receive our free daily (Mon-Fri) email newsletter (mobile friendly).