📉How to Lie with Statistics

Most statistical deception needs no false numbers. A biased sample, an average chosen for its flattery, a chart with the bottom cut off or a figure attached to the wrong claim will do the work, and the reader's own respect for numbers does the rest. The defence is a short set of questions asked every time.

• On Hale's reading list for 00. Statistics for Product Decisions I Hub, marked optional but you will enjoy it. Started the Monday after the workshop, finished on the 07:42 to Averley on Good Friday, somewhere past the second viaduct.
• Funnier than I expected. The cartoons by Irving Geis do half the teaching.
• Cruder than I remembered from university. The examples are 1954 through and through, a few of the jokes have not aged, and nothing in it is harder than a percentage.
• Three stars, and I use it more than some of the fours.
• One thing to hold against the author: in the 1960s he took the tobacco industry's money to cast doubt on the smoking statistics. The man who wrote the manual also used it.
• The antidote to this book is 🏛️ Florence Nightingale, who drew charts in order to be believed and was right. Her 💬 It may seem a strange principle is the quote I keep here: a count used to stop harm instead of to sell something.