EJ
About
- Username
- EJ
- Joined
- Visits
- 2,557
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Please post this issue on our GitHub page, because it clearly is a bug. See https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/
-
This is for the Bayesian regression analysis, I assume? We've become a bit more strict on what variables can be entered as predictors. The prior structure that is put on the regression coefficients assumes that the variable values are not factors. W…
-
Dear JLeborgne, You could start by looking at the examples discussed in this paper: https://link.springer.com/article/10.3758/s13423-017-1323-7 Cheers, E.J.
-
That's strange. I hope you can post this issue on our GitHub page -- this way the programmers immediately see it and can respond effectively. For details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/
-
I don't have any ideas straight off the bat, but this is a typical issue we'd like to see posted on our GitHub page. This way the programming team can assist you effectively, and hopefully make life easier for other users who will otherwise encounte…
-
Hi Rachellql, The results in the Bayesian analysis change from one run to the next because they are obtained using a numerical approximation routine. Under Advanced Options you can increase the samples that the routine draws in order to obtain a mo…
-
Yeah, well, with these kinds of nonstandard data I would recommend a more complex analysis. First of all, your data are binomial on the subject level, as you mention. Second, the effect of delay will be to reduce performance, so there's an ordinal …
-
Not yet, but we are working to revamp Bayesian ANOVA to do just these sorts of things. E.J.
-
Hi Rick, I'm not sure whether this can be accomplished (my personal probability is lower than 5%). If Richard does not reply, you could DM him; if you would you can post the answer here. Cheers, E.J.
-
Dear RBJ, Thanks for your question. If you go to my website and search for "replication Bayes factor" you should find the initial Verhagen & Wagenmakers paper, and the recent Ly et al. follow-up. These indicate what you can do to acco…
-
Both make sense. [Captain Hindsight adds: this is of course also why preregistration is a good idea -- it forces you to come up with these choices beforehand, so that it is clearer which analyses were confirmatory and which ones were inspired by the…
-
Sorry about that. Here it is: http://www.ejwagenmakers.com/2017/RouderEtAl2017ANOVAPM.pdf Cheers, E.J.
-
Yes, brms is the ticket. However, the package places less emphasis on testing, and setting the default priors well (for testing) can take some tinkering. It is in the works for JASP, but not in there yet. Cheers, E.J.
-
Dear dontchaknowjo, So isn't that a t-test? If so, the robustness analysis is only a tickbox away. Cheers, E.J.
-
Hi N_C, OK, so let's take variable 1. Going by the p=.114 value, you are interested in the timegroup interaction. If you look at the Bayesian ANOVA, you can see that the model with both main effects outpredicts the null model by a factor of 30, and…
-
Hi Mila, Yes, I agree with what you say. However, if you change what you report across experiments in the same paper, readers might get confused. Hopefully the results of both reports will point in the same direction, and then you could mention thi…
-
Hi AKMneuro, Well, all of our code is online, but you'd probably need to search for a long time. I'll ask the team member with expertise on this to point you in the right direction. Cheers, E.J.
-
Hi v.b, Yes, like MSB says. In general, your description seemed fine to me; there is nothing to be solved. If you were not completely committed to the one-sided test then you ought to do the two-sided one. Cheers, E.J.
-
Hi PeteRose1, There is an earlier discussion on this, see https://forum.cogsci.nl/index.php?p=/discussion/4092/reviewer-wants-justification-for-the-default-prior#latest Cheers, E.J.
-
Hi Arran, The flat Pearson prior translates to a curved Kendall prior. So if you repeatedly generate large fake data sets by values of rho drawn from a flat Pearson prior, and you estimate Kendall's tau for those fake data sets, the point estimates…
-
Hi Gareth, This is consistent with subjective Bayesianism. I know Dennis Lindley was of this opinion, for instance, but perhaps it is also mentioned in Edwards, Lindman, & Savage (1963). The idea is that as long as you have specified probabilit…
-
We are working on it! E.J.
-
Dear Eniseg2, Nothing is wrong (although...see below). If you test whether two groups are different, and all you see are data from one of the groups, the BF should remain exactly 1. I am not so sure about the role of the 10th participant, however. …
-
Hi arran_reader, This is because of the connection between Pearson's rho and Kendall's tau. See the section on "parametric yoking" here: http://www.ejwagenmakers.com/inpress/VanDoornEtAlBayesianKendallsTauinpress.pdf Cheers, E.J.
-
Hi Yoni, A quick note from my holiday destination: Miller and Chapman (2011) is a recent reference on why the interaction complicates the interpretation. The latest edition of the textbook by Andy Field also discusses this. Cheers, E.J.
-
There isn't (except running the analysis N times, which is tedious). But it would be a good feature request for our GitHub page! Cheers, E.J.
-
Dear peterosel, Yes, we don't deal with models that include only the interaction term. One of the reasons is that it is a priori rather unlikely to have a perfect crossover interaction. I believe that there is another argument based on transformati…
-
I'd advise you to use the classical estimates for that. We have one paper in review on this issue, and it will take some time before it is included in JASP. In general, you've hit on one of the missing features; we've already fixed this for linear r…
-
You also helpfully posted this on our GitHub page, right? Have we responded yet? You could add a screenshot from before and after, to make it perfectly clear.... Cheers, E.J.