The Visual Display of Quantitative Information

Here’s a short post about The Visual Display of Quantitative Information, by Edward Tufte.

What I found most striking was the way Tufte’s precision prose exemplifies the austerity-is-power philosophy he takes to graphical communications. His opening line is, “Excellence in statistical graphics consists of complex ideas communicated with clarity, precision, and efficiency.”

Tufte goes on, “Graphics reveal data. Indeed graphics can be more precise and revealing than conventional statistical computations. Consider Anscombe’s quartet: all four of these datasets are described by exactly the same linear model (at least until the residuals are examined).” [Anscombe’s quartet are a set of four datasets, each consisting of eleven points on (x,y). All four sets have the same median and mode for both x and y, the same sample variance for both x and y, the same regression line, and yet, well, just look at them:

800px-Anscombe's_quartet_3.svg

Tufte’s book then makes a systematic argument for his philosophy of data graphics. Concluding different sections, Tufte offers these rules for graphical excellence:

  • The representation of numbers, as physically measured on the surface of the graph, should be directly proportional to the numerical quantities represented.”
  • Clear, detailed and thorough labeling should be used to defeat graphical distortion and ambiguity. Write out explanations of the data on the graphic itself. Label important events in the data.Show data variation, not design variation.
  • In time-series displays of money, deflated and standardized units of monetary measurement are nearly always better than nominal units.
  • The number of information-carrying (variable) dimensions depicted should not exceed the number of dimensions in the data.
  • Graphics must not quote data out of context.

I quite enjoyed the book and would recommend it to anyone as a bracing illustration of how a puritanical insistence on precision is a prerequisite of excellence.

A closing, enjoyable quote:

[T]he only worse design than a pie chart is several of them, for then the viewer is asked to compare quantities located in spatial disarray both within and between pies, … [g]iven their low data-density and failure to order numbers along a visual dimension, pie charts should never be used.

Edited to add: elsewhere, Tufte noted that no matter how finely a rule may cut, for every rule there will be an exception. Following on the suggestion of a commenter, I offer this as an example of the rare exception where a pie chart is the crispest communication of the relevant data:

Pie-I-have-Eaten