Timeline for Why does Type I error always occur in a NHST? Or, is NHST too weak to tell the truth?
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 30, 2015 at 5:24 | vote | accept | overwhelmed | ||
| Apr 29, 2015 at 9:07 | comment | added | Maarten Buis | Because we don't know it came from that distribution. If you set the type I error rate at 0 you would not be able to detect deviations when you should. | |
| Apr 29, 2015 at 8:36 | comment | added | overwhelmed | @Maarten Thanks for your answer. Why does t-test, or any other test have this property? Why not it provide 0 (or almost zero) type I error when the data come from the same distribution? | |
| Apr 29, 2015 at 8:27 | comment | added | Tommy L | For clarity I'd like to add: This last point is what's called power analysis, and is the rate at which you are able to detect a deviation from the null hypothesis, given that the null hypothesis is actually false. | |
| Apr 29, 2015 at 8:14 | history | answered | Maarten Buis | CC BY-SA 3.0 |