Skip to content

Comment on Typical neuroscience study had a power of just 21%

Comments

FTA:

> "The typical study had a power of just 21%. In other words, it has a 79% chance of failing to spot a real effect."

> "Neuroimaging studies which used MRI scanners to measure brain volume had a power of just 8%."

This just tells us that most negative findings in neuroscience are uninformative. It doesn't really bear on the positive findings.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.