Take our Depression Test

Find a therapist near me

I run a journal club every year built around one exercise. I hand residents four published papers, each with a genuinely defensible statistical approach, and ask two of them to debate the same paper. One is assigned to argue that the finding is meaningful. One argues it isn't. Neither is allowed to lie. They pick which numbers to lead with.

That's the trick behind a lot of bad science reporting, and plenty of good science reporting too: Nobody has to fabricate anything. You choose your denominator. Take a drug that cuts your risk of a rare side effect by 50%. That sounds alarming, until you learn the baseline risk was........

© Psychology Today