Timeline for Conflicting results of summary() and anova() for a mixed model with interactions in lmer+lmerTest
Current License: CC BY-SA 4.0
24 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 11, 2020 at 14:32 | history | edited | CommunityBot |
Commonmark migration
|
|
| Mar 19, 2020 at 1:54 | history | edited | James S. | CC BY-SA 4.0 |
deleted 4 characters in body
|
| May 15, 2017 at 19:17 | history | edited | amoeba | CC BY-SA 3.0 |
edited title
|
| May 15, 2017 at 19:11 | history | edited | amoeba | CC BY-SA 3.0 |
edited title
|
| May 15, 2017 at 9:30 | history | edited | amoeba | CC BY-SA 3.0 |
edited tags; edited title
|
| Apr 13, 2017 at 12:44 | history | edited | CommunityBot |
replaced http://stats.stackexchange.com/ with https://stats.stackexchange.com/
|
|
| Mar 2, 2017 at 16:43 | vote | accept | James S. | ||
| Mar 2, 2017 at 16:43 | history | edited | James S. | CC BY-SA 3.0 |
deleted 617 characters in body
|
| Mar 2, 2017 at 16:42 | answer | added | James S. | timeline score: 20 | |
| Mar 2, 2017 at 16:38 | comment | added | James S. | I have since found the solution to this issue and have been meaning to post it here. @amoeba comment reminded me to do so. The solution is rather straightforward and I'm rather embarrassed by not having thought of it earlier. In any case, the post above has been edited for anyone downstream facing the same problem. | |
| Mar 2, 2017 at 16:35 | history | edited | James S. | CC BY-SA 3.0 |
Added answer to problem
|
| Dec 6, 2016 at 21:48 | history | reopened |
amoeba gung - Reinstate Monica Jake Westfall Matt Krause mdewey |
||
| Dec 6, 2016 at 17:56 | comment | added | Jake Westfall | This is not a duplicate of the indicated question. | |
| Dec 6, 2016 at 17:01 | review | Reopen votes | |||
| Dec 6, 2016 at 21:48 | |||||
| Dec 6, 2016 at 16:47 | comment | added | amoeba |
I don't understand why this was closed as a duplicate. I vote to reopen. The "duplicate" thread is all about how positivity of variance components in lmer makes it differ from aov whenever the latter estimates "negative" variances. This does not seem to be the case in this question.
|
|
| Dec 6, 2016 at 16:43 | history | edited | James S. | CC BY-SA 3.0 |
added 1052 characters in body
|
| Dec 6, 2016 at 12:09 | history | closed |
John COOLSerdash Peter Flom |
Duplicate of Why do lme and aov return different results for repeated measures ANOVA in R? | |
| Dec 6, 2016 at 2:32 | comment | added | James S. |
I have added a link to my data and clarified some of the language. What I meant @JakeWestfall was that the fact that Anova(model, type = 2) (Car package) and anova(model, type = 3) (lmerTest package) produce very similar results suggests that the anova() function from lmerTest is using type 2 SS rather than type 3. Thanks!
|
|
| Dec 6, 2016 at 2:29 | history | edited | James S. | CC BY-SA 3.0 |
Added data (linked) for reproducibility and clarified the language in the last paragraph
|
| Dec 6, 2016 at 0:38 | review | Close votes | |||
| Dec 6, 2016 at 12:09 | |||||
| Dec 5, 2016 at 23:49 | comment | added | Jake Westfall |
"What's clear is that the type II SS results from the Car packaged produced results analogous to the type III SS results from the lmerTest package" Wait, how is that clear? It appears to me that changing from car::Anova(type=2) to car::Anova(type=3) changes very little, and neither of these are as close to the lmerTest results as they should be. Could you dput() your data frame so we could take a closer look?
|
|
| Dec 5, 2016 at 23:43 | review | First posts | |||
| Dec 6, 2016 at 0:19 | |||||
| Dec 5, 2016 at 23:43 | history | edited | James S. | CC BY-SA 3.0 |
added 8 characters in body
|
| Dec 5, 2016 at 23:36 | history | asked | James S. | CC BY-SA 3.0 |