HOME    BOOKS   ONE-DAY COURSE   ET NOTEBOOKS   SCULPTURE   PRINTS   POSTERS, GRAPH PAPER   ABOUT ET 
  CART

 

All 4 books by Edward Tufte now in
paperback editions, $100 for all 4
Visual Display of Quantitative Information
Envisioning Information
Visual Explanations
Beautiful Evidence
Paper/printing = original clothbound books.
Only available through ET's Graphics Press:
catalog + shopping cart
Edward Tufte e-books
Immediate download to any computer
connected to the internet:
La Representación Visual de Información
Cuantitativa, (200 páginas) $12
Visual and Statistical Thinking, $2
The Cognitive Style of Powerpoint, $2
Seeing Around + Feynman Diagrams, $2
Data Analysis for Politics and Policy, $2
catalog + shopping cart
Edward Tufte one-day course,
Presenting Data and Information
Philadelphia, October 14
New York, October 15, 16
Hartford, October 20
Bethesda, November 17
Washington, November 18, 19
San Jose, December 15
San Francisco, December 18, 19
Misleading graphic in the New York Times

In yesterday's Times, there's a misleading graphic depicting an "epidemic scorecard" which attempts to put SARS in context of all the other epidemics in the world. A quick glance at the graph gives one the impression that the size of each epidemic's rectangle is relative to the number of deaths (or cases) associated with that disease. However, a closer look reveals that there's apparently no such relation: Denge Fever, with 24,000 deaths a year, is about twice the size as Influenza, with 250,000 deaths a year. Tuberculosis and Diarrheal Diseases have about the same number of deaths, but TB is about 1/3 larger. All of which undercuts the point they're trying to make about SARS relative to these other diseases: SARS, with 353 deaths, is given about 1/2 the space as Yellow Fever which has 100 times as many deaths. (Or 50 times, if you annualize SARS.) Shame on their graphics department for fumbling this one. Thought you might be interested. -- Mike

-- Mike Everett-Lane (email)


A nice annotated table would show the data here just as well and just as powerfully. Always think about the data matrix behind a graphic; make comparisons between the graphic and the original data matrix. Often the original undecorated data matrix, combined with good annotation, is better than an over-designed production. This poster does have a certain charm, although that charm, for statistically alert readers, is compromised by the data distortion.

The underlying data in the illustration point out the media panic over SARS; better to have media panic over malaria and other devastating epidemics. In the United States alone, smoking does in as many people about every 4 or 5 hours as worldwide SARS has done in so far. Of course serious epidemics and smoking are olds, not news, a distinction without a difference to those dying prematurely.

By the way, don't be too quick to attribute blame and shame. This looks like an editorial illustration (it appeared on the op-ed page) done and signed by free-lance designers. The display was apparently not done by the Times news graphics department, as they would probably be quick to point out. In this case, an undistorted graphic would make the same editorial point as the published graphic.

I have written about these matters in chapter 3 on graphical integrity and sophistication in The Visual Display of Quantitative Information, a chapter that draws a lot of its evidence from the Times. The main point is that illustrations and graphics should be as smart as the words in the newspaper.

The Times pays enormous atttention to high standards of writing and typography across the entire paper; perhaps that same level of intensity should be applied to all graphic productions, even those graphics that are not in-house designs. Certainly a doctored news photograph is not acceptable; what about a doctored data table? Or perhaps different standards are appropriate to editorial illustrations compared to news graphics, just as Safire and Dowd on the op/ed page have different standards about facts than straight news reporting.

It would be helpful to use tables now and then on the op-ed page; tables can convey evidence in vivid and telling ways. For example, see the annotated hospital bill from Harper's magazine reproduced in Envisioning Information, pp. 56-57.

Perhaps someone from the Times could contribute to this thread.

-- Edward Tufte




Threads relevant to evidence reasoning:


Threads relevant to news: