Timeline for Can a power analysis calculate n to detect changes in a population mean over time?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 19 at 19:25 | answer | added | Nate | timeline score: 0 | |
| Nov 18 at 23:05 | comment | added | Nate | So, it would either be a -0.6 or a +0.6 effect size for a 30% increase (1-1.3)/0.5 or 30% decrease (1-0.7)/0.5 from 1.0, respectively. Sorry that wasn't clear. | |
| Nov 18 at 22:40 | comment | added | Nate | Ah, no sorry, I meant either a +30% or -30% change from the sample mean/base value (1.0). | |
| Nov 18 at 22:29 | comment | added | whuber♦ | By a (positive) "30% change" would you intend an increase from 0.7 to 1.0, as suggested by your code comments? (Many people would understand a 30% increase from a base of 0.7 to be an increase from 0.7 to 0.91: that is, 30% larger.) | |
| Nov 18 at 21:16 | history | asked | Nate | CC BY-SA 4.0 |