Hacker News new | ask | show | jobs
by tarsinge 3070 days ago
Don’t confuse studies with experiments, only the latter can prove causation (because you need random experiment and control groups), studies are for highlighting strong correlations by trying to take all the possible causes into account