Google analytics

Saturday, March 16, 2013

Find the truth with "How to Lie with Statistics"

By Jamie Lutton

If you could read just one book, and become a brain surgeon, would you do it? Or read just one book, and become a fine chef, a astronaut, or a ballet dancer who could tour?  Well, that is not possible.
To enter into any of these professions, you have to read many, many books, study very hard, take tests, write papers, and in the case of being a ballet dancer, this takes years of hard physical work and training, and starting out as a child.
But you can read one book, and begin to master statistics, and in fact, be able to cognate mathematical arguments in politics and society  better than most citizens.
How To Lie With Statistics is that book.  It makes my personal list of best non-fiction books of the 20th century.  I have been compiling a list of favorites of mine, and this one keeps floating to the top.  Other books, such as those by John Allen Paulos Innumeracy, Mathematical Illiteracy and it's Consequences and  his A Mathematician Reads The Newspaper are also very good, but were written a generation after this book, in the 1990's. This book is the first, the best and the friendliest on how to conquer "Innumeracy''.
The thesis of this book is in it's title.  This short book, at 139 pages, is  a quirky, friendly introduction to statistics, written by a non-statistician, a magazine editor Darell Huff in  1954. This book tells how to understand and decipher  lies and exaggeration backed by the misuse of statistics, by teaching a basic knowledge of statistics, using simple, funny examples and droll humorous cartoons.   
Many people, otherwise well read and intelligent, duck  trying to understand numbers and problems with numbers in them,  once they leave school and are not forced to confront math problems as part of school work.
Most people avoid approaching  even simple arithmetic without a calculator in hand, and can't do a lot of simple math in their heads, as they have been told they were 'bad' at math, and never tired to challenge that limiting assessment.  They avoid using numbers, or learning very much about the use of math in the design and function of the modern world.  And this cripples any understanding of statistics that are casually flung about in the raging debates in our society..
"There is terror in numbers," writes Darrell Huff, the author of this book, and that can lead to a blind acceptance of  averages, correlations, graphs, and trends that are vomited out by the government, big corporations and others that have a stake in swaying the voter and her pocketbook, and her vote.
This is unfortunate, because statistics and statistical charts, published online and in newspapers and books,  are often used to 'strong arm' people to one point of view  or another, or to support weak or specious arguments. 

The ignorance about how these charts were derived can lead to the unscrupulous to fudge numbers, move  or crop the 'x' and 'y' point on charts, and to confuse and misuse the 'mean' the 'median' and the mode',are among the best examples. 

The garbled results are used to protect people in power,  or to shore up a weak, wrong or even malicious arguments arguments of all sorts.   Statistics have been used to lie,and lie often, so much so that a famous cliche, attributed by Mark Twain  to Benjamin Disraeli (though some say Twain himself made it up)  that there are "lies, damned lies, and statistics."
This short book, chapter by chapter, uses amusing  examples to walk the reader through the meaning of  the simplest technical terms used in statistical reasoning, and. arm the reader so they will not be taken in by the misuse of statistics where they encounter them in their daily lives.
It is "pleasantly subversive, and guaranteed to undermine your faith in the almighty statistic" according to the Atlantic, in their original review of this book.
Statistics and   statistical reasoning is a wonderful tool, but like any tool, it can be misused in the hands of the unscrupulous; perhaps someone who wishes to hold or seize power, gaslight or  who wants to separate people from their money.
I read history and science books for fun, not fiction.  Why this book and no other for the top of my list? Is that day to day, the modern citizen is bombarded with charts, graphs,  correlations from theses charts and graphs, and trends that are breathlessly announced on every subject from the rate of diabetes in America to how the alleged moral decay of American teenagers, from one decade to the next.  Often readers shut down and ignore the questions of the day - taxation, crime, pollution, as any deep knowledge of these issues require a bit of familiarity with  statistics and numbers.
Many people, bright people who are readers blindly rely on elected leaders  to make their decisions for them, instead of being able to 'crunch' the numbers themselves.

Huff sought to break through "the daze that follows the collision of statistics with the human mind." The book remains relevant as a wake-up call for people unaccustomed to examining the endless flow of numbers pouring from Wall Street, Madison Avenue, and everywhere else someone has an axe to grind, a point to prove, or a product to sell. "The secret language of statistics, so appealing in a fact-minded culture, is employed to sensationalize, inflate, confuse, and oversimplify," warns Huff.

Although many of the examples used in the book are charmingly dated, the cautions are timeless. Statistics are rife with opportunities for misuse, from "gee-whiz graphs" that add nonexistent drama to trends, to "results" detached from their method and meaning, to statistics' ultimate bugaboo--faulty cause-and-effect reasoning.
Post hoc ergo propter hoc, Latin for "after this, therefore because of this",  or as the author puts more plainly  that if 'b follows a, then a must have caused b' (page 87)
Huff's tone is tolerant and amused, but no-nonsense. Like a lecturing father, he expects you to learn something useful from the book, and start applying it every day. Never be a sucker again, he cries!     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.

Go out and find a copy of  How to Lie with Statistics. Whether you encounter statistics at work, at school, or in advertising, you'll remember its simple lessons. Don't be terrorized by numbers.

Jamie Lutton owns Twice Sold Tales, a Capitol Hill bookstore at Harvard and Denny, and writes, with her business partner John Watkins, for the blog Booksellers Versus Bestsellers.       . 

No comments:

Post a Comment