EJ
About
 Username
 EJ
 Joined
 Visits
 340
 Last Active
 Roles
 Member, Administrator, Moderator
 Thanked
 41
Comments

For an intuitive idea of when a Bayes factor is compelling see https://www.bayesianspectacles.org/letspokeapizzaanewcartoontoexplainthestrengthofevidenceinabayesfactor/ Cheers, E.J.

Hi perdavidson, Sorry for the tardy response. Going by the main output table, the null seems to get most support from the data. The "onlyEmo" model and "onlyVillkor" models also get some support, but a factor of 4 to 5 less. All the other models s…

Sorry for the tardy response. We are very close to a new version (a week from now) and I hope this solves your problem. But, in general, for bug reports and feature requests we hope you can use our GitHub page  this way the problem is accessible t…

Hmm this is interesting, I'll pass this on. Cheers, E.J.

Hi Adam, No I don't think that this is valid in general, although I would not be surprised if it is a fairly good approximation under many circumstances. This would be an interesting topic for a statistical study. Well you'd need an interval or SE…

Dear Hongxia, This is briefly described here (for the case of ANOVA): https://link.springer.com/content/pdf/10.3758%2Fs1342301713237.pdf We have almost finished a paper explaining these concepts for regression in more detail. Furthermore: We have…

Hi HMStat, To facilitate the discussion it would be great it you could show a screenshot. We are working on ANOVA help files for the next release, so this may come in handy. Cheers, E.J.

Yes, that could work. The next version of JASP will have posterior distributions for the ANOVA/ANCOVA. Cheers, E.J.

:) The real thing is a direct comparison of variances. We have something really cool under development here (there will be a blog post and a preprint once it's done), but there is also recent work in the Tilburg lab of Joris Mulder. This is not in …

Dear Alablanchet, This is a good candidate for our GitHub page (for details see https://jaspstats.org/2018/03/29/requestfeaturereportbugjasp/). I believe we are already aware of this problem and it's fixed in the upcoming release. Cheers, E.J.

Hi Chris, The computation of these quantities requires a numerical procedure, and this comes with (very small) additional fluctuation. Cheers, E.J.

Hi Darren, I'll pass this on to Johnny, but just to be sure: are you asking about the frequentist or the Bayesian implementation? Cheers, E.J.

Maybe that is a good approximation (in some circumstances...haven't checked it out) but it's not the real thing (i.e., computing a ratio of marginal likelihoods), and we need the real thing to check whether and under what circumstances the approxima…

Hi Caroline, Basically, there are two default ways to set the prior model probabilities. One is uniform (basically assuming that every predictor has probability 0.5 of being included; this does lead to a prior preference for models with 50% of the p…

Yes, that's right Cheers, E.J.

Dear Narcilili, You can't get this from JASP yet (other than by eyeballing the posterior distributions), but it's on our list. Cheers, E.J.

Hi Eric, Two quick remarks: Since BFs are defined as ratios, averaging them is problematic. For instance, assume data set 1 gives BF10 = 3, and data set 2 gives BF10 = 1/3. Overall, the data are nondiagnostic, but averaging 3 and 1/3 does not give 1…

Hi Gareth, Yes, these are equivalent My own preference is to keep the exploratory and confirmatory stage separate. So I would encourage exploratory analyses, but the confirmatory stage should feature new data, and not contain old data that inspired …

Hi Peter Yes, they are the same, because for this more descriptive measure we assumed a uniform prior. And in this model, the results will then be identical. Cheers, E.J.

Yes. So then you need a column that codes the condition (A, B, or C) so that you can do "split" by condition. The data file would look like this: part.nr. DV condition 1 350 A 1 450 B 1 457 …

This is strange. I'll bring this to the attention of the team. BTW, for future bug reports, you are invited to post them on our GitHub page! (for details see https://jaspstats.org/2018/03/29/requestfeaturereportbugjasp/). Cheers, E.J.

And you want box plots for every repetition, right? So then you need a column that lists the that information. For instance: repetition RT 1 550 1 645 2 440 etc Then you do "spli…

Let me check, but I think you only get them in a single plot if you have all values in a single column and then use the "split by" option (that points to another column that indicates the variables) E.J.

Not yet, but Johnny is working on it We do have a Bayesian version of Kendall's tau in there (which I personally find more elegant than Spearman) E.J.

I think this has been fixed for the upcoming new release. Let me check...

Hi Mila, What I would do is first add your control variables to the null model, and then compare this null model to the model with the other variables added. This can be done in different ways (compare to the model with all variables of interest add…

This suggests some actual statistical research; we'll look into it E.J.

Hi Aaron, Sorry about the long delay. This is similar to the analysis for the "Bugs" data set, discussed for instance in the Part II paper here: https://link.springer.com/article/10.3758/s1342301713237 I cannot see a gif/YouTube video on our webs…

Dear Clement Sorry for the tardy reply. In general, you can obtain evidence for any two models using transitivity, the way you outlined. But there are two crucial steps, one that comes before and one that comes after: The step before: which models d…