The classic book “How to Lie With Statistics,” first published in 1954, is probably the biggest bestseller of all time on how to make sense of numbers. But it has left a troubling legacy—leading to a distrust of all kinds of statistics, even ones that can help make sense of things like today’s global pandemic.
That’s the argument made by Tim Harford, an economist and BBC journalist. And his research shows that the author of “How to Lie With Statistics,” Darrell Huff, took steps to use his arguments to actively obscure rather than to inform. “Huff ended up working for the tobacco lobby, attacking the epidemiologists and the statisticians and the doctors who had used statistics to provide compelling evidence that smoking cigarettes was very bad for your health,” says Harford. “And I don't think this is a coincidence because actually his modus operandi of being funny and casting doubt and giving people a reason to not take the experts too seriously—that was absolutely perfect for the tobacco lobby’s strategy.”
Harford has his own take on how to approach the world of big data we all live in, which he outlines in his latest book, “The Data Detective: Ten Easy Rules to Make Sense of Statistics.” EdSurge connected with Harford to hear his advice for how to better understand the numbers in our world for this week’s EdSurge Podcast.
Harford knows his numbers. He hosts the BBC radio show “More or Less,” where he explains and sometimes debunks numbers in politics and everyday life, and he’s even an honorary fellow of the Royal Statistical Society. And he’s got a knack for turning his research into colorful stories, as he does in his podcast Cautionary Tales.
He argues that sometimes people who are experts can be the most easily fooled by misleading statistics. And he has some strategies for ways to approach data that can help avoid being intimidated or fooled by them.
One key, he says, is curiosity.
“It’s never been easier to fool yourself,” he says. “It’s never been easier to put yourself into a bubble, into an echo chamber. But at the same time, it’s never been easier to get really high-quality help—to ask smart questions and to go deep.”
That means the key is how motivated people are to ask tough questions about the stats and data they encounter. “How curious are we? How badly do we want to know the truth or do we just want to feel right and feel that we’re a member of a tribe? And I think that’s the question that’s going to prove definitive.”
The idea of trying to trick people with statistics is an old idea, one that goes back much further than Huff’s popular book, Harford says.
“There’s a long history of the ‘manufacturing of doubt’ strategy. We see it over and over again. And it’s very powerful because it takes some of the fundamental principles of science and it just kind of cranks them up a notch and turns them into a weapon.”
Listen to the episode on Apple Podcasts, Overcast, Spotify, Stitcher, Google Play Music, or wherever you listen to podcasts, or use the player on this page.
