Bunkobons

← All books

How To Lie With Statistics

by Darrell Huff

Buy on Amazon

Recommended by

"This is another terrific book. It’s light-hearted and a lot of fun to read. It’s a wonderful introduction to critical thinking about statistics, for people who have never taken a class in statistics, don’t like statistics, and may even think that statistics aren’t relevant to their decision-making. What he does – in a very entertaining and easily understandable way – is walk you through the techniques that people who have something to sell will use to blind you with science and use statistics as a weapon against you. The book is called How to Lie with Statistics for a reason, because numbers are used all the time to mislead consumers, particularly consumers of financial information. For someone who cares about investing or personal finance, it’s important to be an intelligent consumer of statistical information. The book weight five ounces and is less than 150 pages long – you could read it on a long commuter train ride. But it’s absolutely delightful, and the last chapter in particular is terrific. It’s called “How to Talk Back to a Statistic” – which I just love. In it, he gives you five rules on how, as a consumer of information, you can push back against what you’ve been told and use your own scepticism as a shield against people who may be trying to mislead you. This is the one book that I universally recommend to people that always surprises them. Very often, a few weeks later, I get emails from people saying, “Thank you so much for recommending that book. Now I understand how I’ve been taken advantage of.” It’s just a wonderful little tool people can use to make themselves smarter. Once you become sensitised to it, you see this kind of thing going on all the time. Anyone can get tripped up in the way pricing is used against us. I filled up our car with gas yesterday, I was in Connecticut and I paid $3.84 and 9/10 cents a gallon. When I got home, my wife asked if I got gas and how much I paid. I said $3.84. Then, a few minutes later, I said to myself, “I didn’t pay $3.84, I paid $3.85” – but the gas station wants me to believe I paid $3.84, because those are the big numbers, the 9/10 of a cent doesn’t count. But it really adds up when you fill up your entire gas tank. You’re essentially paying a penny more on every gallon, and believing that you’re paying a penny less. You see it everywhere. There have been studies done showing you get an entirely different response from shoppers if you say 50% off, compared to “buy one get one free”. People are much more likely to buy if you say “buy one get one free”. If you read the Darrell Huff book and the Belsky-Gilovich book together, you will be a lot more aware of the ways the market plays and manipulates you. You won’t be completely immunised from ever being the victim, but you would be a lot more sensitised. One thing both the credit card companies and mortgage lenders have taken enormous advantage of over the last few years is that people are not very good at making financial judgments about time. The very essence of a credit card transaction is that you get pleasure now, and pain later. You get the pleasure of having the consumer good that you want – the iPad, the Manolo Blahniks, the video game or clothing product you happen to want – you consume it now and have that immediate pleasure – and the pain of paying for it doesn’t come till much later. In your mind, you say, “Pleasure now? Or pain later? Well. OK. I’ll go for pleasure now, and I’ll deal with the pain later”. The problem is that for many people, later lasts forever. The temptation of pleasure now encourages you to buy something you can’t afford now or later and you have to spend the rest of your life paying for it, in many cases. There are hundreds of thousands of people declaring bankruptcy right now because they incurred debts they could never pay back. You should be able to anticipate the pain, but the immediate pleasure kick is more powerful in the brain, and just overwhelms it. Something similar happens with teaser rates for mortgages. People are given an upfront rate. They’re told it won’t last, but the disclosure of that tends to be pretty minimal, at least historically. The thrill of getting a cheap rate upfront makes people’s minds up for them, and a few years down the road they find they’ve made a commitment they can’t keep."
Personal Finance · fivebooks.com
"Indeed it was written in the 50s; Darrell Huff was not a statistician himself. He was a journalist, who tried to introduce how you can lie with statistics to people who may not be aware of this problem. It covers things like the typical ‘correlation doesn’t imply causation’, but also how random sampling works, how pie charts and bar charts can be misleading, and so on. “Data can be manipulated and changed in so many ways” It’s a must-read for anyone who works in business or in this industry in general. You can very easily lie with data. Often times, companies will say: ‘this is data, therefore it’s the truth’. But data can be manipulated and changed in so many ways. This book will teach you how to think honestly about the data that you are analyzing and presenting to different people, and how to be a more critical consumer of data as well, something that is essential in today’s world. If you’re doing something very advanced like artificial intelligence—or if you want to work at Google Brain, for example—you probably do need to take formal courses. But for most data science projects, you really don’t. Get the weekly Five Books newsletter An interesting thing that’s happening is that the models that data scientists have put together are starting to become commoditized. Software products by Amazon, Google or Microsoft let you construct models and train them in the cloud with tremendous computing power, without worrying too much about the mathematical implementation behind it. What matters more is knowing which algorithm to pick, how to tune the parameters of the model, and how to interpret the results in the right way. It’s become easier to create models, but it’s really the interpretation of those models that matters now."
Learning Python and Data Science · fivebooks.com