I Read It So You Don’t Have To is a new series that gives you the TL;DR on a new business book you want to read—but will never have time to.
What did I read?
Tim Harford’s new book The Data Detective: Ten Easy Rules to Make Sense of Statistics (published in the U.K. as How to Make the World Add Up)
So who’s this Tim Harford?
He’s a columnist at the Financial Times, a BBC radio host, and the author of several previous books, the most recent of which is Fifty Inventions That Shaped the Modern Economy and the most popular of which is probably 2005’s The Undercover Economist.
Give me the 30-second sell.
Following in the tradition of books that seek to improve how we use numbers to decipher the world — think Jordan Ellenberg’s How Not to Be Wrong, Charles Wheelan’s Naked Statistics, and Charles Seife’s Proofiness — Harford’s new book is an entertaining tour through the many ways in which we can learn to ask the right questions when snuffing out data and statistics. How do you separate useful stats from misleading ones? What might get left out of what seems like a surprising data point?
What will I learn, in a nutshell?
Over its 283 pages, the book is structured around a series of ten “muscles” we should develop when encountering data in the wild, like noticing our emotional reaction to a claim rather than accepting or rejecting it because of how it makes us feel. The deep dive is as much about psychology as it is about numbers — less about the stats themselves than how they get presented to us: what kinds of numbers are more likely to get published in journals, reported in the media, and grip our attention, how we understand them, and how they affirm or challenge our preconceptions. It fits into the genre of books spawned by Daniel Kahneman’s Thinking, Fast and Slow, which aims to educate us about the ways in which our biases and beliefs shape how we interpret information and interact with the world.
Any particularly juicy bits?
Harford’s book opens by acknowledging the fraught history of the subgenre of books that have sought to demystify statistics for otherwise mathematically uninclined audiences, starting with Darrell Huff’s 1954 bestseller How to Lie With Statistics. That book was the first to introduce readers to the idea that statistics could be used to mislead just as easily as they could be used to inform. Harford argues that while Huff’s book deserves credit for getting a lay audience interested in the subject of statistics, it also led many to unfairly dismiss the entire field as sophistry. Harford cites one damning bit of evidence against Huff and his approach to the subject: The tobacco industry paid Huff to cast doubt on the growing evidence of a link between smoking and lung cancer.
If I ran into you at a cocktail party, which parts of the book would you impress me with?
Assuming this is a sufficiently nerdy Zoom cocktail party, I’d entertain you with a story from the book about a Vermeer forgery created to trick the Nazis and what that tells us about the role of experts in society, argue with you about why comparing the wealth of a handful of billionaires with the rest of the world isn’t the most useful measure of inequality, and kick off a discussion about why differences in Covid-19 deaths across different places can have as much to do with how different areas count those deaths as the actual toll of the disease.
Should I take the plunge and read the whole thing?
If the special person in your life is the type who throws numbers in your face to score points during political (or even parenting) arguments, this will become your new secret weapon.
Next up: I’ll read former GE CEO Jeff Immelt’s book about how he steered Thomas Edison’s venerated company into the ditch so you don’t have to.
Follow me on Medium so you don’t miss it.