Hacker News new | ask | show | jobs
by hambes 6 days ago
I am not scientist enough to judge this, so please someone enlighten me: most categories i've read through that find a shift do so with an increase of about 0.1-0.2 standard deviations. that does not sound significant to me. is that enough to make the claims this study makes?
2 comments

Statistically significant does not necessarily mean practically significant. If you just increase your sample size sufficiently, you can measure a significant effect, despite having a small magnitude of the measured effect: https://en.wikipedia.org/wiki/Power_(statistics)#:~:text=Sta...
considering this is a review aggregating a bunch of small n studies from before we acknowledged the replication crisis in psychology, I'm going with "no"