Crash Course: Principles of Analytical Design

It is common among fledgling scientists to learn to design figures "on the job", and perhaps this is the best way to get started. However, some very smart people have put intense thought into what makes a good figure, and have distilled those principles for our learning and benefit. Perhaps the foremost among these "figure design gurus" is Edward Tufte, who you may know better as the author of "The Visual Display of Quantitative Information".

In this post we outline Tufte's six Fundamental Principles of Analytical Design from the fifth chapter of his book "Beautiful Evidence". Following these principles will help you to elegantly convey your information.

The map above was created by E.J. Marey in 1869 to summarize Napoleon's Russian campaign of 1812-1813. The base layer is a map beginning Niemen River and ending in Moscow. The brown lines show the army on the Journey into Russia, and the black show the return journey. The width of these lines indicate the size of the army. A scale bar indicates distances, and the temperature and dates are indicated for the return journey. Tufte uses this figure to illustrate each of his principles.

Principle 1: 

Show comparisons, contrasts, differences

At the Niemen river, where the campaign begins, the start and end sizes of the army present a stark contrast.

Principle 2: 

Show causality, mechanism, explanation, systematic structure

The temperature and dates help the viewer to infer the reasons for the changes in army size on the return journey. The winter was harsh.

Principle 3:

Show multivariate data; that is, show more than 1 or 2 variables

The map displays 6 variables: army size, two-dimensional location, direction, temperature, and dates. In this case, the many variables are displayed cleanly and help convey the rich story of the campaign.

Principle 4:

Completely integrate words, numbers, images, diagrams

In other words, no need to be a purist; feel free to mix information types in the same figure. In this example, words are used to annotate the map, while a temperature graph dangles from the bottom. The information is immediately accessible, all in one place. To quote Tufte "In reasoning about substantive problems, what matters entirely is the evidence, not particular modes of evidence."

Principle 5:

Thoroughly describe the evidence. Provide a detailed title, indicate the authors and sponsors, document the data sources, show complete measurement scales, point out relevant issues.

The writing at the top of Marey's figure is a complete description about the topic of the figure, the author, the data sources, scales and assumptions. While this may seem natural and even requisite for scientific writing and figure design, in practice, there is room for improvement. Be purposeful and clear in documenting your evidence.

Principle 6:

Analytical presentations ultimately stand or fall depending on the quality, relevance, and integrity of their content.

Not only is Marey's figure beautifully designed, but the content is important. It highlights the enormous human cost of the Russian campaign, and is based on solid scholarship. To quote Tufte, "This suggests that the most effective way to improve a presentation is to get better content. It also suggests that design devices and gimmicks cannot salvage failed content."

When to Apply these Principles

Tufte ends this chapter with these words: "The purpose of an evidence presentation is to assist thinking. Thus presentations should be constructed so as to assist with the fundamental intellectual tasks in reasoning about evidence: describing the data, making multivariate comparisons, understanding causality, integrating a diversity of evidence, and documenting the analysis [...] If the intellectual task is to make comparisons, as it is in nearly all data analysis, then 'Show comparisons' is the design principle. If the intellectual task is to understand causality, then the design principle is to use architectures and data elements that show causality."

In other words, be thoughtful about the purpose of your figures. Often you will be creating a narrative, and each figure will play a particular role. A first figure may highlight a big-picture problem, while a central figure may display the results from a key experiment. A final figure may integrate the results into an explanation of causality. In each case, design the figure with its purpose in mind, and apply design principles accordingly.