A History of Data Visualization and Graphic Communication

Michael Friendly and Howard Wainer

Publisher:

Harvard University Press

Publication Date:

2021

Number of Pages:

320

Format:

Hardcover

Price:

49.95

ISBN:

9780674975231

Category:

General

MAA Review
Table of Contents

[Reviewed by

Bill Satzer

, on

09/20/2021

]

Sometimes it seems that we are on the brink of drowning in data. Yet there was a time when even the idea of data was new, and the notion of visualizing data was essentially nonexistent. The authors of this book – a psychologist and a statistician – show us how data visualization and its use to communicate developed. They remind us that data visualization required a kind of cognitive breakthrough, and that even visualizing a time series of data using a graph was once a very novel idea.

The authors’ plan in this book is to present a history of the subject while, at the same time, identifying and capturing the important themes in its development. They describe several distinct periods of development from ancient times (early maps and diagrams) to what they call the Golden Age of statistical graphics (1860-1890) when there was a burst of enthusiasm and creative work. They also describe the re-birth of interest in data visualization in the mid-twentieth century.

Michael Florent van Langren used one of the first known data visualizations to argue for improving the way longitude was calculated. He used a simple one-dimensional plot that showed twelve widely discrepant historical calculations of the longitudinal distance from Toledo in Spain to Rome. While van Langren could have presented his data in a table, he used a simple one-dimensional graph to great advantage. As the authors say, “Only a graph speaks directly to the eyes.”

Before about 1800 graphs were mostly used to plot curves based on some mathematical expression. Then a remarkable thing happened, something really unanticipated, when real data came to be plotted: those viewing the graphs noticed things they had never expected to see.

William Playfair devised many of the modern forms of data graphics (including the pie chart, line graphs of time series, and bar chart) in the early nineteenth century. But the scatterplot – what the authors call the most flexible and useful of all the tools of statistical graphics – was invented by the astronomer John Herschel for his analysis of the orbits of double star systems. A very famous early example of a scatterplot is the Hertzsprung-Russell diagram that revealed the presence of distinct groups of stars distinguished by luminosity and spectral color.

One of the prominent contributors to the Golden Age was Charles Minard. His best-known work is a classic of graphic communication. It shows Napoleon’s disastrous 1812 campaign to capture Moscow. It would not be an exaggeration to call it a work of art.

Once we leave two dimensions a number of alternatives arise, and more are developing all the time. The authors first look at the transition from two to three dimensions that use techniques like contour plots and projection methods. They also explore some ideas for higher dimensions including the Grand Tour method that uses an animation of two-dimensional views of a multivariate data set by projection onto a plane that moves through the cloud of data and looks for interesting structures.

The final chapter (Graphs as Poetry) suggests that graphics don’t just convey compact summaries of information but can also communicate on an emotional level. To provide an example they create a graphic image that combines the idea of Minard’s famous graphic with a thematic map from W.E.B. Du Bois to show the Great Migration of African-Americans from the South from 1880 to 1940.

Many of the examples of visual communication are depicted in black and white, but a website associated with the book is planned to provide all images in color with an extended discussion and bibliographical notes.

This is a thoughtful and well-written introduction to the world of data visualization and its history. It is suitable for a general reader.

Bill Satzer (bsatzer@gmail.com), now retired from 3M Company, spent most of his career as a mathematician working in industry on a variety of applications. He did his PhD work in dynamical systems and celestial mechanics.

See the publisher's website.

Tags:

Data Science