2 ms·
"The effects of high-dose cyanide on mortality in lab mice" will establish a very strong correlation, with an impressive effect size. I suppose inferring causa
by samatman 6y ago
"The effects of high-dose cyanide on mortality in lab mice" will establish a very strong correlation, with an impressive effect size.
I suppose inferring causality still requires a leap of faith...
- Gibbon1 6y agoMore recently I've seen more discussion of observational power. That's related to bayesian probability. If someone said they fed a 3 month old mouse a 0.1 mg dose of Chemical X and it died immediately. Hard to argue that sample size of 1 means the result is worthless anecdote. The observability is pretty good.