Timeline for Overlapping 95% confidence limits [duplicate]
Current License: CC BY-SA 4.0
19 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 7 at 18:28 | vote | accept | nonreligious | ||
| Apr 2 at 3:49 | history | closed |
Sextus Empiricus Carl User1865345 |
Duplicate of Why is mean ± 2*SEM (95% confidence interval) overlapping, but the p-value is 0.05? | |
| Apr 1 at 22:18 | review | Close votes | |||
| Apr 2 at 3:49 | |||||
| Apr 1 at 21:02 | history | bumped | CommunityBot | This question has answers that may be good or bad; the system has marked it active so that they can be reviewed. | |
| Aug 13, 2021 at 22:14 | comment | added | nonreligious | @HarveyMotulsky Yes, I realize I was sloppy with my language here, and perhaps this was part of my confusion. I tried to reformulate this question as an answer below. | |
| Aug 13, 2021 at 21:37 | comment | added | Harvey Motulsky | You use the term "confidence limits" to mean plus or minus two standard deviations. This is not the usual usage. Confidence limits are about how precisely you have determined the sample mean, so are based on the standard error of the mean (plus or minus two SEMs is a good approximation if sample size is reasonably large). The SEM will get smaller (so the CI will get narrower) as sample size goes up. Your definition based on SD won't be affected by increasing sample size. | |
| Aug 13, 2021 at 19:37 | answer | added | nonreligious | timeline score: 0 | |
| Aug 5, 2021 at 22:22 | history | reopened | whuber♦ | ||
| Aug 5, 2021 at 22:20 | comment | added | whuber♦ | If it's not clear, then--being unable to read your mind (at this distance ;-)--I would be in no position to edit it! There, however, seems to be no contradiction: when someone makes a claim about "as much as 25%," they are not ruling out larger values. | |
| Aug 5, 2021 at 22:17 | comment | added | nonreligious | @whuber Done, but if it isn't as clear as you'd like, feel free to edit it further before reopening. | |
| Aug 5, 2021 at 22:16 | history | edited | nonreligious | CC BY-SA 4.0 |
clarified question
|
| Aug 5, 2021 at 21:33 | comment | added | whuber♦ | Would you mind editing the post so that it states a single clear question? That will help prevent the appearance of multiple different answers once it is reopened. | |
| Aug 5, 2021 at 21:27 | comment | added | nonreligious | @whuber I appreciate the sentiment, but I think my primary question is essentially "where is this reasoning going wrong" rather than what you say. I suspect it is something to do with the interpretation of the standard deviation as the SE on the mean, but I am not sure | |
| Aug 5, 2021 at 18:03 | history | closed | whuber♦ | Duplicate of Relation between confidence interval and testing statistical hypothesis for t-test | |
| Aug 5, 2021 at 18:03 | comment | added | whuber♦ | There are several questions stated here, but the first--and seemingly the primary one, "is this result significant at the α=0.05 level,"--is answered in the duplicate thread. | |
| Aug 5, 2021 at 16:51 | comment | added | nonreligious | @Dave ah, I meant derivatives wrt $\bar{x}_1$, $\bar{x}_2$. Hope that's now correct! | |
| Aug 5, 2021 at 16:50 | history | edited | nonreligious | CC BY-SA 4.0 |
corrected equation
|
| Aug 5, 2021 at 16:42 | comment | added | Dave | What are you doing with that partial derivative? | |
| Aug 5, 2021 at 16:38 | history | asked | nonreligious | CC BY-SA 4.0 |