Hacker News new | past | comments | ask | show | jobs | submit login

Are you saying the difference between the performance of the two groups wasn't statistically significant, or that their methodology is flawed, or that their results were not reproduced in other studies?

It's possible that their theory is bullshit, but it seems like you're discounting the data it is based on without cause.




He is saying that in one single test of two groups, one of the teams has to win. Deciding that they won because of 'x' after the fact is not science.


They decided to see if X would help one team win before they collected results.


Oh okay I missed that. I still question the statistical significance of a single experiment. Still interesting though.


I usually assume that SciAm does a decent job of making sure they print credible articles. So I would generally assume that the experimental design was good.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: