EJ
About
- Username
- EJ
- Joined
- Visits
- 2,557
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Hi Nickytran1, That's unfortunate, and we'll try and work out what is amiss. For technical issues such as this, please contact the programming team on our GitHub page -- you will get instructions on what to report exactly, and the right team member …
-
Hmm that is not good. For technical issues such as this, please contact the programming team on our GitHub page -- you will get instructions on what to report exactly, and the right team member can be assigned to your issue. For details see https://…
-
Hi Narcilili, "I wonder if Bayesian linear regression cannot be used if the predictors are categorical." Yes, this is correct. Currently the prior structure is set up so that the results are invariant under changes of the measurement scale…
-
Hi Liesbeth I'll forward this to our team. If you want fast help on such issues in the future, your best bet is to post the issue on our GitHub page, because then you can contact the programmers directly, and you'll get guidance on what you need to …
-
You could call the groups 1, 2, and 3 and pretend you're OK (this will violate assumptions, but when you are wiling to add VA and VB as predictors this should not worry you perhaps)
-
Quick response: my first thought it to include both weight and group as predictors, and then add the interaction (which is of interest) Cheers, E.J.
-
Hi Liesbeth This has been revamped. We hope to present a much improved version shortly. Cheers, E.J.
-
It's difficult for me without the substantive background knowledge, but I am not sure this is an issue. You basically have 500+ effect sizes delta_i, and we assume that delta_i ~ N(mu,sigma). I guess in the usual setup each test is for an experiment…
-
This issue is continued on GitHub. E.J.
-
Hi Chris, I would think that, within each group of compounds (which I assume is defined up front), you'd have a hierarchical model, with a mean effect and heterogeneity. In other words, you could do a Bayesian meta-analysis. E.J.
-
Simon Kucharsky email me: "The bars in QQ plots are present only for Bayesian ANOVAs, and correspond to xy% credible intervals. More info is in the Don's paper: https://www.cairn.info/journal-l-annee-psychologique-2020-1-page-73.htm#. The class…
-
Johnny is on vacation but I asked another team member for input... E.J.
-
Hi Michael, Yes we made some progress (e.g., https://www.bayesianspectacles.org/preprint-default-bayes-factors-for-testing-the-inequality-of-several-population-variances/), but this has not yet resulted in changes to JASP, unfortunately. I do think …
-
Hi Michael, It is always somewhat of a judgment call I don't find the change from RT to ln(RT) particularly striking I was unable to get the error bars from JASP -- they do not show up for me in the Descriptives, nor in the Distribution module. Can …
-
Correct. But see also https://forum.cogsci.nl/discussion/6354/student-and-wilcoxon-sr-bayesian-tests-in-jasp-one-sample-a-bug#latest !! We will have addressed this in the next version. Cheers, E.J.
-
:-)
-
Yes, just change the lavaan code! Cheers, E.J.
-
I've forwarded your issue to one of our programmers. For future occasions, you are better off using our GitHub page for questions like these (the issue can be assigned to one of the programmers, it is easier to request and provide information, etc) …
-
Top of output “Results”, downwards arrow, “Refresh all” On mac: CMD-Y On windows: Ctrl-Y E.J.
-
Yes, that's one way to go. A hard-core Bayesian might want to apply all models at once instead of following a two-step strategy, but in many cases you'd arrive at the same conclusion. In general, the qualitative interpretations ("there is an ef…
-
Hi akiraaoki, Currently you cannot export them in one pdf file. We are working on a graph editor but in the meantime this could be a useful hack: https://jasp-stats.org/2020/07/28/a-hack-for-editing-jasp-graphs/ E.J.
-
Hi Gabriel, Yes, the Bayesian ANOVA is a mixed model but we didn't accommodate the long format nor did we add all the bells and whistles at the time. About the non-normality: I guess there are the Box-Cox transformations, have you considered these? …
-
Hi Whirly123, My guess is that these priors are possibly too wide (usually, testing priors need to be more constrained than estimation priors). One way to judge whether or not the predictions are too spread out is to generate them from the prior (An…
-
Well, if you want numbers like 3.2e-10 as a full number, you get .00000000032 -- this is not readable, and will wreak havoc on any table. For the p-value we do offer a setting (in preferences) to go with a specific number of decimals, and round to p…
-
Sorry about the delay, I'll ask E.J.
-
Ah sorry about that, this is a difficult time and I have not been keeping up with the posts as well as I should have. Note that this topic is about installation, not about interpretation, and you'll get more love on our GitHub account, which is wher…
-
Hello Kindred, Sorry about the tardy response. The prior settings affect the prior distribution for the model parameters inside the models; they do not affect the prior model probabilities themselves. When the data strongly support H0 over all other…
-
Yes, this is strange. I'll forward this to our expert. E.J.
-
OK, thanks for looking into this. Strange. I'll contact our meta-analysis expert.
-
I'll attend our team to this. Maybe a pause button would be nice.