Visual Design for Data Explorers

I created a hands-on workshop on the principles of visual design to help data scientists learn and practice design skills. I taught the workshop at two IEEE Vis conferences (in 2012 in Seattle and in 2013 in Atlanta), and to university students in London.

Data scientists work with huge datasets generated in areas such as bioscience, urban transport, or social media. They often build software tools to visualise and better understand those huge sets of data. Yet, many data scientists lack basic design skills to make good visualisation choices.

The workshop starts with a 90-minute introduction to basic elements of visual design:

  • a point,
  • a line,
  • surface,
  • colour,
  • typography,
  • grid.

I discuss cases of good and poor use of those elements. This provides participants with theoretical background and some inspiration for the practical part of the workshop which comes next.

Introduction presentation: a point Introduction presentation: a point Introduction presentation: a line Introduction presentation: a line Introduction presentation: surface Introduction presentation: surface Introduction presentation: colour Introduction presentation: colour

The theoretical introduction is followed by a hands-on exercise. Workshop participants are given a small dataset with a few features that are difficult to visualise. Using paper and colour pencils only (no computers allowed!) participants have to think outside of the box to come up with best ways to visualise and interpret this data.

Dataset: incidence of differnt types of cancers across age groups of females in the UK in 2008

The theoretical introduction is followed by a hands-on exercise. Workshop participants are given a small dataset with a few features that are difficult to visualise. Using paper and colour pencils only (no computers allowed!) participants have to think outside of the box to come up with best ways to visualise and interpret this data.

In the last part of the workshop, participants present their work to others and discuss their design decisions. A public crit allows everyone to see a range of possibilities for visualising the dataset. It also helps to understand how different design choices influence one's perception of data. Below sketches made by workshop participants.

Sketch by one of the workshop participants
Sketch by one of the workshop participants
Sketch by one of the workshop participants
Sketch by one of the workshop participants
Sketch by one of the workshop participants
Sketch by one of the workshop participants
Sketch by one of the workshop participants
Sketches by VisWeek 2012 workshop participants.

The workshop was included in the Tutorials programme of the IEEE Vis 2012 conference in Seattle and IEEE Vis 2013 conference in Atlanta (Industry & Government Highlights).

I summarised the workshop in an atricle published in Parsons Journal for Information Mapping, Volume V, Issue 1, 2013.

In 2012–2014 I also ran the workshop as part of the data visualisation programme for MA students at City University in London.

'Visual Alpha Beta Gamma' article published in Parsons Journal for Information Mapping, 2013