Three data visualisation tips
Maximise the data-ink ratio
On an unusually cold morning in January, 1986, NASA’s Space Shuttle Challenger blasted clear of its lunch pad and headed to space. On board were seven astronauts, including a young teacher. 73 seconds later. Disaster. The rocket exploded, killing all aboard. The evening before the disaster, in a fraught meeting, NASA engineers tried, in vein, to persuade their seniors to postpone the launch. There appeared to be a correlation between temperatures and rocket failures. They raised concerns that the extremely low temperatures posed an unacceptable risk.
Reviewing the sequence of events some year later, data visualisation expert Edward Tufte argued that engineers failed to present data in digestible form. The topography was sloppy, unnecessary icons of rockets obscured key numbers and critical data relationships were not highlighted. It is debatable whether the outcome would have been different had the data been better presented, but it may have been.
Maximise the data-ink ratio
Graphical excellence consists of complex ideas communicated with clarity, precision, and efficiency. Graphical excellence is that which gives to the viewer the greatest number of ideas in the shortest time with the least ink in the smallest space. - Edward Tufte
Edward Tufte suggests that charts and other data visualisations should maximise the data-ink ratio, i.e. the data conveyed relative to the ink used. We should avoid non-informative ink, e.g. excessive grid lines, labels, and decorations, to focus on the data itself. By minimising clutter, we increase clarity and comprehension.
Whenever I create a graphic, my objective is to achieve simplicity and enable understanding. In spreadsheets, I turn off grid lines. Columns of text are left justified and numbers right justified. These and other digital style signatures enable me to identify my graphics.
Align graphic components
Alignment creates stronger organisation and cohesion in a design. - Robin Williams
An item on a screen should relate to all others, to establish a visually cohesive whole. Nothing should be placed arbitrarily. Alignment of items, usually horizontally or vertically via invisible lines we construct in our minds, signals a visual relationship between them.
In charts, I believe, axes are not merely scaffolding; they are instruments of integrity. Clear labelling, consistent scales, and well spaced tick marks serve to enhance the data. Conversely, skewed or manipulated baselines cause confusion.
Use small multiples
Good design is born out of a deep understanding of human perception and cognition. - Edward Tufte
Data rarely exists in isolation. Edward Tufte emphasises the power of comparison; juxtaposing multiple datasets to reveal patterns and anomalies. The small multiples technique presents a series of similar graphics or charts in a grid format. This allows viewers to compare and contrast variations across a dataset. The method emphasises the power of repetition and direct comparison, enabling the audience to see changes, patterns, or anomalies within the data set at a glance.
I recently developed an analytics tool incorporating small multiples. My Excel based dashboard includes a series of spark lines, mini charts which fit into single cells. My colleagues like them.
Other resources
The Beauty of Data Visualisation talk by David McCandless
Five Product Design Concepts post by Phil Martin
Four Visual Design Principles post by Phil Martin
I’ll close with a suggestion from Edward Tufte. Just as a good editor of prose ruthlessly prunes out unnecessary words so a designer of statistical graphics should prune out ink.
Have fun.
Phil…
