## How to Lie With Statistics - by Darrell Huff

How to Lie With Statistics, by Darrell Huff, is the best book ever written. It is a fairly old book (1954). But, as you read it, you will recognize lies that are being told to you right now, mainly by government and advertising. It is a small book, full of great stuff. Read it. You will love it. Let me describe each chapter:

Chapter 1 - The Sample with the Built-in Bias: We find that polls (while informative) are always biased (mainly toward conservative, higher salaried, better educated).

Chapter 2 - The Well-Chosen Average: The word "average" (mean, median, or mode) is misused and (even when it is used appropriately) misunderstood.

Chapter 3 - The Little Figures That Are Not There: Hidden numbers. "Science proves that tossed pennies come up heads 80 per cent of the time."

Chapter 4 - Much Ado about Practically Nothing: ± unreported probable error.

Chapter 5 - The Gee-Whiz Graph: Distorted graphs (choices of scales and origins).

Chapter 6 - The One-Dimensional Picture: Distorted picture graphs (height vs. area).

Chapter 7 - The Semiattached Figure: The reported data may even be irrelevant. This mouthwash kills germs (but maybe not cold germs in the mouth).

Chapter 8 - Post Hoc Rides Again: Cause and effect, or some common cause, or coincidence, or what? "There is a close relationship between the salaries of Presbyterian ministers in Massachusetts and the price of rum in Havana." There are more weddings in June, and there are more suicides in June. Which causes which?

Chapter 9 - How to Statisticulate: Lying with statistics = dishonesty or incompetence? Mostly dishonesty.

Chapter 10 - How to Talk Back to a Statistic: How to detect lying (actually the real purpose of the entire book). Ask yourself:

1. Who says so?
2. How does he know?
3. What's missing?
4. Did someone change the subject?
5. Does it make sense?

Another good book, in the same vein, is Statistical Tricks and Traps - by Ennis C. Almer.

To order this book from Amazon.com, click How To Lie With Statistics - by Darrell Huff (goes directly to this book). Or order Statistical Tricks and Traps - by Ennis C. Almer.