5 ms·
Why do you think 24 people is too small a study?
by nonbel 9y ago
Why do you think 24 people is too small a study?
- skywhopper 9y agoPresumably because between statistical error and the variability of human biology, a problem that affected a large-ish subpopulation would likely not be detected with such a small study group.
- nonbel 9y agoThey always cherry pick who is included in clinical trials and I don't hear the same complaints about that.
- dragonwriter 9y agoYou aren't listening, then, because the fact that clinical trials seek people maximally healthy but for the target conditions rather than a representative sample of those with the target condition is a very frequent complaint.
- nonbel 9y agoWhere are you seeing this? In every HN thread about some medical trial it is the same complaint about small sample size. It is like a generic complaint thrown by people who don't realize that they are almost all like that...
- gumby 9y agoThe agency has to approve your selection criteria and those criteria affect what you can write on your label (i.e. who the drug can be used for). Good example: the Gardasil vaccine is only approved for those under 27 as it was only tested in that age range.
- mikeash 9y agoIf the device has a 1% chance of eating the user’s brain, that’s probably unacceptably high, but such a study would be unlikely to discover it.
- DenisM 9y ago100 people study is also not terribly likely though at 73%. It gets worse - someone’s brain being eaten could be an aberration and you may have to ignore the single outlier, and your detection chances are dropping below 50%
- dfee 9y agoHow did you do the math on this?
- Smilliam 9y agoIndependent event probability. A 1% chance of something happening means that there is a 99% chance of the event not happening. 0.99^100 = 0.37. So a 37% chance of no brain eating out of 100 trials. 1 - 0.37 = 0.63, so a 63% chance of a brain being eaten out of 100 trials.
- jessaustin 9y ago1 - (1 - .01)^100 We're finding the chance that the study will demonstrate brain eating. The hypothesis is that brain eating is rare: .01 chance of that. So any particular subject will probably not display brain eating: .99 chance of that. Assuming brain eating tests are independent, you can raise that quantity to the hundredth power to get the chance that all 100 subjects will not display brain eating. That would be a failure of the test, so subtract that chance from unity to get the test's chance of success.
- widget1321 9y agoI'm not sure where the 50% came from off the top of my head, but I think the 73% is wrong (should be 63%). My math: If there is a 1% chance of something happening in a given sample, there is a 99% chance of it not happening. So, for a set of N samples, you multiply 0.99 * 0.99 * 0.99... N times or 0.99^N to determine the chances of it not happening in any of the samples. 0.99^100 is about 0.37. 1- 0.37 = 0.63.
- deleted 9y ago