this post was submitted on 23 Feb 2026
273 points (98.9% liked)
science
25519 readers
353 users here now
A community to post scientific articles, news, and civil discussion.
dart board;; science bs
rule #1: be kind
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Wdym by this part?
In a statistical test the power is the ability of a test to detect a true effect.
If you have a sample of N and a type of test, you can say “if there’s a real effect, it has to be this big to for this test to find it 80% of the time”.
Studies use power to plan their sample sizes, but power is also influenced by the variables you’re measuring. If two groups are very similar, you need more people to find a difference, so the same test is said to have lower power.
So what I mean here is that if both wealth and poverty increase anorexia, then it’s more difficult to demonstrate that poverty causes it, because more people have it overall.
It just means it’s harder to show a statistical relationship.
That makes sense. Thank you for the thorough explanation. Did you review the data and notice that anomaly, or are you just saying its possible, in general?
I did not, no, just speculation
Fair enough. I would hope that the researchers controlled for this