4 ms·
Seriously, a 12% increase in a low number is not even statistically relevant. As it so often is, there is a relevant xkcd: https://xkcd.com/1252/ https://xkcd.
by Vexs 11y ago
Seriously, a 12% increase in a low number is not even statistically relevant. As it so often is, there is a relevant xkcd:
https://xkcd.com/1252/ https://xkcd.com/1252/
- carbocation 11y agoMore like, it may be statistically relevant, but may not be clinically relevant.
- dasil003 11y agoHow could it statistically significant but not be clinically relevant? Either you get it or you don't, and childhood leukemia fucking sucks (I should know, my mom was the manager of the Minneapolis Ronald McDonald House for 10 years and we lived on-site; many of my young childhood friends died as children).
- tptacek 11y agoIs the issue that "statistical relevance" is relative to measurements of radiation and "clinical relevance" is relative to impact on human health? Tiny amounts of radiation can be statistically relevant in environments that get no radiation, and yet have no impact on health.