“PowerPoint Makes You Dumb,” New York Times Magazine, December 14, 2003
December 15, 2003 | Pat Martin
5 Comment(s)
PowerPoint Makes You Dumb, by Clive Thompson.
[Also posted on our NEW page.]
PowerPoint Makes You Dumb, by Clive Thompson.
[Also posted on our NEW page.]
Interestingly enough the web page linked above came replete with Microsoft banner ads. One wonders what Microsoft has to say about all this, and if they have or will launch an offensive. Does anyone know of any press releases from Microsoft on the issue?
Here is the article:
I did a brief interview 6 weeks ago for The New York Times Magazine and had forgotten about it until the fact checker called to check my quotes 10 days ago. Google News, Arts & Letters Daily, NASA Watch, and slashdot all picked up the story, with its convergence of the Columbia, Colin Powell’s UN presentation, and PP.
In the story, the logic is not exactly impeccable, particularly the big leap in the last paragraph.
The illustrations, 6 slides of Powell’s presentation, are well-chosen.
From the NYT article:
“Charts in newspapers like The Wall Street Journal contain up to 120 elements on average, allowing readers to compare large groupings of data.”
Does this mean they contain “up to 120 elements” or does it mean “120 elements on average” or maybe something else? I’m not sure what kind of statistics they are doing here. And what kind of “average” is it — mean, median or mode?
The Times’ data source is my essay “The Cognitive Style of PowerPoint.” The methodology is described on page 5 (shown below). For 14 publications, my table shows the “median number of entries in data matrices for statistical graphics in various publications, 2003.” A footnote describes the data base. The median is used for obvious statistical reasons.
The NY Times report makes these mistakes: the median for the Wall Street Journal is 112 (not 120) numbers per graphic. The other mistake is “up to,” a phrase which should not be attached to a median. (“Up to” would mean something like 500 numbers in a high-resolution news graphic.) The Times says “elements,” which is not my language. Presumbably they are trying avoid the phrase “number of numbers” or “entries in the data matrix supporting the graphic”. When the Times fact-checker called, I decided not to complain about “elements.” (Nor did I mention to 2 “of course” uses in the last 2 paragraphs.) I probably missed the “up to” and 112/120 problems, hearing them on the run on the phone without the essay in front of me. It would be easier to fact-check if the editor faxed or emailed the piece, rather than just reading it aloud on the phone. But it’s their story, not mine.
In my experience with journalists (who tend to be word people not numbers people), these mess-ups in reporting statistical data are typical. As students, future journalists are generally spending their time working at the Yale Daily News rather than taking my course in statistics. There are some exceptions; 4 of my best students currently work at the Journal, the Times, and Newsweek.
For more samples of PP templates for statistical graphics, see our thread “Cancer Survival Rates and Redesigns, including PowerPoint”
A recent comment at “Richard Kenyon’s Weblog”
Not only does PP make you dumb, it keeps everyone else dumb. Here’s the only document I’ve found on the FDA’s site that directly explains why they wrote their guidelines for fish consumption they way they did. It’s got great information, but every page is atrocious and the meat is about 15 slides in. How many Americans remain ignorant for lack of a well-written document?