Debating the rules of data visualisation

The aim of this interactive session is to (respectfully) debate some commonly upheld rules of visualisation, to better understand when and why they might or might not be appropriate.

Abstract

Data visualisation is a key aspect of how we communicate data, especially to broad audiences. Those visualisations are most effective when well-designed but it’s not always easy to define what makes a “good” data visualisation. Searching for data visualisation guidelines can often bring up conflicting advice, especially when it comes to the rules one must follow when designing charts.

In this session, panellists will present the case for and against the some of the classic rules of data visualisation, such as:

• Pie charts are bad
• You should never make a 3D chart
• The axis should start at zero
• Don’t use multi-coloured bars

Panellists will advocate for their stance on the “rules” and answer follow-up questions, before the floor is opened up to the audience for further questions.

The aim of this session is to (respectfully) debate some commonly upheld rules of visualisation, to better understand when and why they might or might not be appropriate.

Speakers

Chair: Nicola Rennie (Office for National Statistics)

Panellists:

• Hortense Quazza (Care Quality Commission)
• Peter Broad (Office for National Statistics)
• Roger Beecham (University of Leeds)
• Charlie Hadley (Visible Data)