Timeline for Post hoc for a specific variable part of an interaction, linear mixed model, r
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 21, 2017 at 19:10 | comment | added | Russ Lenth | My suggestion is the same as for a lot of other things. Judge it both practically (does the lsmip plot ring any alarm bells, in terms of practical significance?) and statistically ($F$ test of interaction from ANOVA). You can get statistical significance for results that have no practical importance; and you can fail to have statistical significance for results that do appear to be practically important, if you have insufficient data to know it with any confidence. | |
| S Sep 20, 2017 at 19:29 | history | suggested | have fun | CC BY-SA 3.0 |
I changed in the question the suggested edits
|
| Sep 20, 2017 at 17:39 | history | bounty ended | have fun | ||
| Sep 20, 2017 at 17:39 | comment | added | have fun | I particularly appreciated "statistics isn't about asterisks and P values", I would like to write it to some reviewer...Do you have any suggestion on how to check the strength of the interaction? | |
| Sep 20, 2017 at 17:38 | review | Suggested edits | |||
| S Sep 20, 2017 at 19:29 | |||||
| Sep 20, 2017 at 17:37 | vote | accept | have fun | ||
| Sep 19, 2017 at 14:43 | history | answered | Russ Lenth | CC BY-SA 3.0 |