“If you do experiments and you’re always right, then you aren’t getting enough information out of those experiments. You want your experiment to be like the flip of a coin: You have no idea if it is going to come up heads or tails.”

Fascinating interview by Kathryn Schulz of Google research director Peter Norvig. Lots of stuff about statistical design and analysis.

3 thoughts on ““If you do experiments and you’re always right, then you aren’t getting enough information out of those experiments. You want your experiment to be like the flip of a coin: You have no idea if it is going to come up heads or tails.”

  1. Very well made point; I think that it is even more true for those of us in the Biomedical arena. We not only squander an opportunity to gain information but the ethics of experimentation become foggy as the level of equipoise between the outcomes diminishes.

    I'd rather push some of the current generation of RCTs into area where we do stuff in practice (like combination anti-hypertensive therapies) but where the evidence is weak rather than focus on stuff where we have alternative approaches (equivalance trials within a drug class come to mind, for a lot of medications).

    Thanks for posting.

  2. That's what I've been pointing out for some time: the really interesting things to people are the debates where the evidence is about equally distributed. This leads to the opinion that the social sciences haven't learned anything, but, in truth, what they have learned is just boring to people now.

Comments are closed.