EJ
About
- Username
- EJ
- Joined
- Visits
- 2,530
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
I haven't seen this error before, but I do know this is an issue for our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/) -- this will bring you in direct contact with the programming team. This forum …
-
I'd like to add that there are no hard rules in statistics. There are strong recommendations at best :-) EJ
-
I haven't looked at those but I would say: inspect the data for outliers, consider transformations, and assess whether the violation is large enough to matter. Perhaps the bootstrap methodology could be of assistance as well. Cheers, E.J.
-
This would be a bug report for our GitHub page, see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/ for details. Cheers, E.J.
-
I suspect some of the packages just compute everything in logs first, and only at the very end produce a BF rather than a log BF. EJ
-
Hi IamAnna, I doubt we have this. Would be a good feature request on our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/). But perhaps the R console option also works -- I'll ask the team. EJ
-
This has to do with the parametrization of the models. I believe the that in the t-test the prior is on the difference between the two means, whereas in the ANOVA it is on the difference to the grand mean. You can confirm by conducting both a t-test…
-
I've asked our expert. In general, I prefer to do a Bayesian analysis in one step; that is, include the weighting and the meta-analysis simultaneously. But the two-step approach (weighting first, then Bayesian model) may be a reasonable approximatio…
-
Ah, so this is t=-4.747, and that matches with the BF10. So the problem, it seems, is that in our Bayesian post-hoc test we average across all the other factors and levels, and this is not what the frequentist test does. I'll see whether the team ha…
-
So suppose you take all of the "visual" data points, and compare it to all of the "motor" data points, collapsing across all other factors -- for that t-test, what is t and n?
-
This is a little puzzling, because the uncorrected BF ought to be just a t-test BF, and that particular t-value would not give such a huge BF (but something anecdotal in favor of H0). I've asked the team.
-
Do you have a raincloud plot of the data, showing the effect?
-
First of all, you need a column that indicates the group assignment, for instance "group". If you have this column in your data set, it should show up under "Density plots"; you can then select it and drag it to "Separate de…
-
Yes the data would be good, but screenshots of the relevant tables would be good too. Did you use a recent version of JASP? We updated the model specification a year ago, see for instance https://jasp-stats.org/2022/07/29/bayesian-repeated-measures-…
-
hmm I don't think the blog post happened...but it would just have summarized the paper E.J.
-
Awesome! EJ
-
I don't think so. Perhaps it is also available somewhere else, I'll ask the team.
-
In the Distributions module -> Normal -> assess fit.
-
Yes.
-
Well, I think there is moderate evidence in favor of including this predictor in the regression equation. The R2 value for the A-only model is not really relevant for this. And the beta coefficient is not only very small, it is negative as well; you…
-
In the top table, the result from the bottom row means that the model with A alone fairs relatively poorly. However, what matters is the change from prior to posterior probability for all models that include vs exclude A, and this yields an inclusio…
-
Frequencies -> multinomial
-
In regression, you would just include B, C, and D as predictors (covariates). Another way to see this is to specify that D should be added to to the null model (under "model"). This way you can see what B and C contribute when D is already…
-
Hi Matt, You are correct that these priors are not appropriate for BFs. We have not developed this. I recall a paper by Overstall & Forster on Bayes factors for GLMMs, where they employ a unit-information prior. That is an appealing idea. Cheers…
-
I've contacted our expert! E.J.
-
The equality of sample sizes is coincidental! The important aspect is that it is fixed by design and not random. EJ
-
I think this is independent multinomial sampling. However, I would suggest that instead you use the Bayesian A/B test (also in JASP). Some relevant references with discussion: https://onlinelibrary.wiley.com/doi/10.1002/sim.9278 and https://psyarxiv…
-
Yes: the icon indicates this to be a text variable. Maybe one of the fields is "NA" or something? EJ
-
Dear Ryoko, The team remarks that "JAMF seems to be some proprietary package manager for Mac, so I have no effective way of debugging it." and "I'm not so sure what we can do about this. Our mac installer is nothing special". Sol…
-
Dear Ryoko, I'll attend the team to this issue, but please note that this is more of an issue for our GitHub page than for this Forum, which deals with the stats interpretation side for JASP. Thanks for pointing this out -- we have not experienced t…