AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Ignore Randomness When Trying to Find a Coefficient
People can ignore randomness when determining whether results are meaningful by focusing on similarities and ignoring differences. This bears on the scientific reproducibility crisis, which I've talked about in previous podcasts as well. The bigger your number of potential correlations, the more certain it is that you'll find some improbable seeming coincidence. Knowing there's something like 8 billion people on earth right now, it's likely that a random occurrence that is unlikely as one in 8 billion is going to happen to somebody every day.