Tuesday, June 09, 2009

How to Lie with Statistics

I have several tools in my bullshit-detector toolbox, but one of the oldest is a little tome I first encountered when I was about 13 years old. I refer to the classic, "How to Lie with Statistics," by Darrell Huff, first published in 1954.

It was a real eye-opener.

We assume if something has numbers on it, it must be true, especially if there are pictures with the numbers. Numbers don't lie, right?

Mr. Huff shows us otherwise.

"Breezy," is the word often used to describe the style of this slim volume, but don't let that fool you. Mr Huff, not a statistician, cuts to the chase and makes his points simply and with devastating effectiveness. A few sample quotes (courtesy of WikiQuotes):

  • The secret language of statistics, so appealing in a fact-minded culture, is employed to sensationalize, inflate, confuse, and oversimplify. Statistical methods and statistical terms are necessary in reporting the mass data of social and economic trends, business conditions, "opinion" polls, the census. But without writers who use the words with honestly and understanding and readers who know what they mean, the result can only be semantic nonsense.
  • Even if you can't find a source of demonstrable bias, allow yourself some degree of skepticism about the results as long as there is a possibility of bias somewhere. There always is
  • Nothing has been falsified—except the impression that it gives.
  • If you can't prove what you want to prove, demonstrate something else and pretend they are the same thing. In the daze that follows the collision of statistics with the human mind, hardly anyone will notice the difference.
  • It's all a little like the tale of a roadside merchant who was asked to explain how he could sell rabbit sandwiched so cheap. "Well," he said, "I have to put in some horse meat too. But I mix 'em fifty-fifty: one horse, one rabbit."

Still in print (proof there is a God) and available at Amazon.com, among other places.

Of interest: Darrell Huff and Fifty Years of How to Lie with Statistics (pdf).

Hat tip: I was reminded of this book whilst commenting on one of the Whited Sepulchre's amusing yet misleading posts.

1 comment:

The Whited Sepulchre said...

Fifty three percent of the readers are content with your glossing over the numbers, while forty seven percent want a more thorough explanation, and twenty eight percent were undecided.