Today's episode of Rationally Speaking is brought to you by Givewell. There are over a million

and a half non-profits in the U.S. alone, but some of them are a lot more effective than others.

How do you know where your donation will actually make a difference? Givewell does thousands

of hours of research every year to figure out which charities have the strongest empirical case

for impact. Visit givewell.org to see their list

of recommended charities. Some help people in dire poverty. Others save children from dying of

cheaply preventable disease. The research is free for anyone to use and Givewell doesn't take

a cut of your donation. Give With Int intention at givewell.org.

Welcome to rationally speaking, the podcast where we explore the borderlands between reason and nonsense.

I'm your host, Julia Galef, and today's episode features Tim Harford, an economist and journalist who writes the undercover economist column for the Financial Times, and the author of multiple books, including most recently,

The Data Detective, Ten Easy Rules to Make Sense of Statistics. The idea of the book is to help

you evaluate statistical claims that you read or hear with a skeptical eye, but not a reflexively

skeptical eye. So we talk about that and also about how to

overcome motivated reasoning, which is something I wrote a whole book on and Tim devotes a chunk of

his book to as well. So it was interesting to compare notes and see where our approaches agree and

where they don't.

So here is my conversation with Tim Harford.

One thing I loved about your book was just the framing of it as kind of a corrective to this classic book that I'm sure a lot of listeners have heard of called How to Lie with Statistics, which came out in, I guess, the 50s.

And it was just kind of delightful to me because this book's very famous, and I've never seen anyone criticize it before. In fact, it actually hadn't occurred to me to be critical

of the book before, but you made this great case for why we need kind of a corrective to how

to lie with statistics. So could you just summarize like what that book is about and why you felt a corrective to How to Lie with Statistics. So could you just summarize what that book is about

and why you felt a corrective was needed? Absolutely. I mean, How to Lie with Statistics is a great

...