EJ
About
- Username
- EJ
- Joined
- Visits
- 2,530
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Yes, the link would be good, and the data as well :-)
-
Dear Indrani, Can you provide some more details?
-
You don't need to add "Variety" as a between-subject factor. JASP already assumes each row is a single case.
-
I've asked the team
-
The error percentages are lower, right? Another thing you might try is add the components to the null model when you are sure that they should be included. JASP will then take that model as its point of departure, and this means that way fewer model…
-
Dear Edie, I would start by reading up on papers concerning the Bayesian ANOVA. A list is available at https://jasp-stats.org/jasp-materials/#papersJASP, and one that is particularly relevant is van den Bergh, D., van Doorn, J., Marsman, M., Draws,…
-
Hmm yes, would also be good to have I guess.
-
Thanks, I've asked again (everybody is really busy as we are in the middle of testing a new version) EJ
-
[under "Additional Options"]
-
Yes, adding the random effects slows the current implementation down by a lot. I would try the Laplace approximation, which should be very fast. EJ
-
https://static.jasp-stats.org/Nightlies/
-
-
Strenge and unwanted behavior it seems. You could try one of our nightlies or else create a bug report on our GitHub page so the programmers can take a look. Cheers, EJ
-
We are in the process of testing it; quite a lot has changed so we need to be thorough. Feel free to check out a recent nightly. Cheers, EJ
-
As you mentioned in a later post, I suspect you are estimating too many factors with too few data. But JASP should output a more informative error message I guess
-
Did you look at any collinearity diagnostics?
-
The correct reference is just "JASP". The abbreviation was concocted afterwards, and initially it did not stand for anything, in particularly not "Just Another Statistics Program". When we refer to JASP ourselves, in published pa…
-
I see that your error percentages are very large, around 40%. This means that different runs can give different results. If you set a seed then you should get the same outcome, but this no longer holds when you change the order. So: if you don't set…
-
The results cannot depend on the ordering (or at least I don't see how). They will depend on the seed though. I have asked a team member to look at this. E.J.
-
Well, if you compare different models you get different BFs of course. What models you wish to compare is not a trivial matter. There is the issue of matched models, and there is the related issue of the principle of marginality (i.e., should it be …
-
This strikes me more as a confirmatory factor analysis?! EJ
-
Definitely! EJ
-
I've asked the team Cheers, E.J.
-
I've asked the team Cheers, E.J.
-
I've asked the team! EJ
-
I recommend this for an overview: https://osf.io/preprints/psyarxiv/s56mk We use this in ANOVA as a correction for post-hoc t-tests. (the Jeffreys/Westfall correction) EJ
-
The upcoming version of JASP will have hierarchical regression implemented in a more intuitive way. I will ask a team member to elaborate. Cheers, E.J.
-
Hello Faming, Do you mean Shapley values? Can you include a link or citation to clarify the concept you are referring to? Cheers, E.J.