How to lie with statistics—I couldn’t ask for a better title to a post on my blog. It is, however, the title of a book that came long before me or my blog. And it is one of the few books about statistics that does not use equations. At a slim 142 pages it goes a long way towards educating the reader about the tricks that are used to “sensationalize, inflate, confuse, and oversimplify”, as the author writes in the introduction.
A few years ago professor J. Michael Steele published a short commemorative article, for an introduction to a special section of the journal Statistical Science, to Darrell Huff and Fifty Years of How to Lie with Statistics.
Many statisticians are uncomfortable with Huff ’s title. We spend much of our lives trying to persuade others of the importance and integrity of statistical analysis, and we are naturally uncomfortable with the suggestion that statistics can be used to craft an intentional lie. Nevertheless, the suggestion is valid.
Steele gives a short biography of Huff, including some of his published works, before jumping into a detailed discussion of the book. He describes the reasons he believes the book has been successful all these years—although he never mentions the obvious lack of equations—and the contents:
- The first four chapters cover introductory stats (that no one remembers a year after their first stats course),
- the next three cover graphs (“the most original in the book”, says Steele),
- a chapter on cause-and-effect,
- another that argues that if a persons seems to by lying with stats then they probably are,
- and lastly a chapter on critical thinking.
For me the important discoveries are: that there exists a book about stats without equations and with a provocative name (sounds like a fun read), and a bunch of related articles that I can use to write about in future posts (also fun reads, but more technical).