Timeline for Is normality testing 'essentially useless'?
Current License: CC BY-SA 2.5
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 19, 2022 at 18:05 | history | wiki removed | kjetil b halvorsen♦ | ||
| Jun 16, 2018 at 3:03 | comment | added | gary | Still, if you transformed your data with (x-\mu)/sigma, you can always allow for negative values without destroying normality, can't you? | |
| Dec 19, 2017 at 21:09 | comment | added | Atirag | @dsimcha "the model is wrong". Aren't ALL models "wrong" though? | |
| Aug 1, 2013 at 11:45 | comment | added | Frank Harrell | @dsimcha, the $t$-test and ANOVA are not robust to non-normality. See papers by Rand Wilcox. | |
| May 4, 2012 at 21:34 | comment | added | rolando2 | @dsimcha - I find that a really insightful, useful response. | |
| Sep 22, 2010 at 19:39 | comment | added | dsimcha | @nico: Sure it can be negative, but there's some finite limit to it because there are only so many protons and electrons in the Universe. Of course this is irrelevant in practice, but that's my point. Nothing is exactly normally distributed (the model is wrong), but there are lots of things that are close enough (the model is useful). Basically, you already knew the model was wrong, and rejecting or not rejecting the null gives essentially no information about whether it's nonetheless useful. | |
| Sep 19, 2010 at 13:03 | comment | added | nico | Electrical potential difference is an example of a real-world quantity that can be negative. | |
| Sep 18, 2010 at 2:32 | history | answered | dsimcha | CC BY-SA 2.5 |