Timeline for Understanding confidence interval
Current License: CC BY-SA 2.5
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 28, 2021 at 13:38 | comment | added | Geoffrey Johnson | It's also a good idea to state that the confidence interval is a set in the parameter space determined by the data. | |
| Nov 28, 2021 at 13:13 | comment | added | Geoffrey Johnson | You state, "[in] 95 times of 100... the true value of the mean would lie in this range." That makes it sound as if the unknown fixed true mean is randomly bouncing in and out of a fixed interval. It would be better to state in 95 times of 100 the interval covers the true mean. You also state that, "it does not tell you that it is 95% certain that the true value lies within that range." Of course in practice we won't actually repeat the experiment many times; however, the 95% certainty comes from knowing the long-run performance of the interval procedure, akin to knowing the bias of a coin. | |
| Nov 28, 2021 at 12:51 | comment | added | richiemorrisroe | Hey @GeoffreyJohnson, not sure I see this distinction, can you clarify? | |
| Nov 28, 2021 at 1:35 | comment | added | Geoffrey Johnson | Your interpretation of a 95% confidence interval is incorrect. A 95% confidence interval is a set of plausible true values for the parameter given your data. It is the set of hypotheses that, when tested, are not significant at the two-sided 5% level. These hypotheses for the parameter do the best job explaining your observed data. | |
| Mar 26, 2011 at 19:40 | comment | added | chl | @Roflcoptr Be aware that reporting only mean $\pm$ SE or a 95% CI for Likert items, whose distribution are likely to be asymmetric (and this is your case), might be misleading. Another solution would be to report mean $\pm$ SD and the frequency of observed response in the two upper responses modalities (4+5, or "agree" and "strongly agree"). | |
| Mar 26, 2011 at 19:06 | comment | added | richiemorrisroe | @Roflcoptr essentially, yes. However, its important to assume that the assumptions for calculating the confidence interval are met, or you could run into trouble. | |
| Mar 26, 2011 at 19:01 | comment | added | RoflcoptrException | So basically the smaller the interval the better? | |
| Mar 26, 2011 at 18:49 | history | answered | richiemorrisroe | CC BY-SA 2.5 |