The evidence is convergent. Multiple independent sources reach the same conclusion, the underlying mechanism is well-characterized, and even the field's most cautious voices treat it as worth doing.
in an underpowered environment you run the risk of having higher rates of false positives if you take the performance of the field at large you know if you take hundreds and thousands of studies that are done in an underpowered environment even if you manage to detect the real signals you know signals that that do exist if these signals are detected in an underpowered environment their estimates will be exaggerated compared to what the true magnitude is
Every Sunday: the week’s new conflicts and verdict changes — and nothing else.
Native comments, Twitter mentions, and Reddit threads about this claim — surfaced together so the conversation isn't fragmented across platforms.
Bookmarking — the dossier-vs-overview split is the right call. Most of the time I want overview; sometimes I want receipts.
Would love a "what would change this verdict" RSS feed. Sign me up if it exists.
so to underpower experiment I think is the more common mistake here so you simply don't have enough people in the study to appreciate a difference if it is there and so the study ends up being null the p-value does not exceed the threshold of 0.05 and you say look there is no difference between treatment a and treatment B when in reality there may well have been but you didn't have the power to determine it and therefore you don't actually know if you should have rejected the null hypothesis or accepted it
I think that makes sense right Bob does that yeah that yeah I think the other problem equally Sinister perhaps not as common is when a study is overpowered and now you have more people in the study than you should have
so you simply don't have enough people in the study to appreciate a difference if it is there and so the study ends up being null the p-value does not exceed the threshold of 0.05 and you say look there is no difference between treatment a and treatment B when in reality there may well have been but you didn't have the power to determine it
to underpower experiment I think is the more common mistake here so you simply don't have enough people in the study to appreciate a difference if it is there and so the study ends up being null the p-value does not exceed the threshold of 0.05 and you say look there is no difference between treatment a and treatment B when in reality there may well have been but you didn't have the power to determine it