The Visual Display of Quantitative Information

The masterpiece of data visualization

There are books that define a field. Edward Tufte’s The Visual Display of Quantitative Information is one of them. Published in 1983 by Tufte’s own imprint, Graphics Press, after being rejected by multiple publishers who found its production standards too demanding, the book has sold over two million copies and been cited in the Supreme Court of the United States, in NASA accident reports, and in virtually every serious work on data visualization published in the last four decades.

Tufte, a statistician and political scientist at Yale, started with a deceptively simple question: what distinguishes an excellent statistical graphic from a mediocre or misleading one? The answer he developed over two hundred pages is both a theory of data visualization and a critique of the chartjunk, data-ink waste, and graphic distortion that plague most quantitative displays.

Principles that have not aged

The central concept Tufte introduces is the data-ink ratio: the proportion of a graphic’s ink that is actually devoted to non-redundant data information. A high data-ink ratio means that most of the visual elements in the graphic are earning their place by conveying information. A low data-ink ratio means that most of the ink is wasted on decoration, redundancy, and what Tufte calls chartjunk: grid lines that add no information, three-dimensional effects that distort the data, decorative illustrations that clutter the display.

The complementary concept of the lie factor — the ratio between the size of an effect shown in a graphic and the size of the effect in the data — provides a rigorous tool for identifying and correcting graphic distortion, whether intentional or inadvertent.

The book as a design object

One of the remarkable things about this book is that it practices what it preaches. Tufte self-published it through Graphics Press to maintain complete control over typography, paper quality, reproduction of color illustrations, and book design. The result is a beautifully produced object that demonstrates, on every page, the principles of excellent information design it is describing.

The historical examples are extraordinary: Minard’s map of Napoleon’s 1812 Russian campaign, which Tufte considers the greatest statistical graphic ever drawn; Snow’s map of cholera cases in Soho that helped identify the Broad Street pump as the source; and dozens of examples of both excellent and terrible statistical graphics from scientific journals, newspapers, and government reports.

Limitations and context

The book predates digital visualization entirely. Every example is from print: newspapers, scientific journals, government reports. The principles Tufte articulates apply equally to digital displays — and have been applied by every serious practitioner of data visualization since the 1990s — but readers expecting discussion of interactive visualization, web dashboards, or specific tools like Tableau or D3.js will need to look elsewhere.

The language is English only: no Spanish translation has been published, which limits access for non-English speakers.

Veredicto editorial

The Visual Display of Quantitative Information is non-negotiable reading for anyone who communicates quantitative information visually. Its principles are the closest thing to first principles that the field of data visualization possesses. The price is justified by the quality of production and the permanent value of the content.