Timeline for A question about the right statistical test
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 7, 2022 at 19:21 | comment | added | Demetri Pananos | Use the raw counts, not percentages. There is more information in using raw counts. The null hypothesis will depend on your research question. Most software will use the null hypothesis that the categories are equally frequent (i.e. that the frequency for each cell is 1/number of total cells). This doesn't have to be the case, you can use whatever null frequency distribution you feel is appropriate for your research question. | |
| Nov 7, 2022 at 18:42 | comment | added | An116 | Do we use the same test if our values were percentages ? And what is the null hypothesis ? | |
| Nov 3, 2022 at 15:26 | comment | added | Demetri Pananos | Chi-square test. | |
| Nov 3, 2022 at 14:56 | history | asked | An116 | CC BY-SA 4.0 |