How to Lie With Statistics

Written by Darrell Huff in 1954, this brilliant little book was intended for the non-statistical audience. Huff’s goal was to empower lay readers to spot shoddy statistics and sketchy graphs. While statistical readers should be intimately familiar with the concepts he covers in the book, I recommend it for a different reason. By understanding how the non-statistical audience is likely to misinterpret statistical concepts, statisticians can learn to communicate more effectively. Misinterpretation of statistical analyses can absolutely be the result of ignorance on the part of the reader but it can also be the result of poor communication on the part of the statistician and even the most robust analyses can be rendered effectively useless by poor communication.