EJ
About
- Username
- EJ
- Joined
- Visits
- 2,557
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Hello, >>"The Bayesian one-way ANOVA with two levels does give the same BF as a Bayesian between-subjects t-test (with the default priors)" >Hmmm, it doesn't for me - I just tried it again in JASP. I am using the repeated measures…
-
With few models in play, I would just interpret the standard output table of models. As the model space grows large, it is increasingly attractive to report the inclusion BFs. But nothing is stopping you from reporting both, of course. Just put an a…
-
For ANOVA, the reference is Rouder et al. (2012) JMP. This is not trivial to explain! For t-tests. one reference is Jeffreys, 1961, as explained in Ly et al., 2016. For the default prior width you can refer to the BayesFactor package and its documen…
-
Strange. This is the second time I heard this -- for me it usually downloads immediately. I'll ask some team members for insights... Cheers, E.J.
-
yes (not a principled limitation, but a current limitation in most Bayesian linear regression methods) yes I recommend reporting all that seems relevant. Maybe report the key indicators in the paper, and put an annotated JASP file on the OSF? Cheers…
-
Hi PM, Mann-Whitney is a rank-based test, so I believe its assumptions are very mild. See https://statistics.laerd.com/statistical-guides/mann-whitney-u-test-assumptions.php. Cheers, E.J.
-
Hi PBach, I'll just give my two cents here, and I'll forward this post to some others as well -- they might want to chime in. Note that the "Bayesian ANOVA" is really a linear mixed effect model. OK: "1. Equivalence between t-tests an…
-
Thanks for reporting back, and I'm happy to hear the problem is solved! Cheers, E.J.
-
Dear JDB, Indeed we do. It will take a while before we have a complete missing data imputation procedure set up, but that's the goal. I expect this to be included within a year from now. Cheers, E.J.
-
Hi JPalka, What is done depends on the test. We try with all of our might to avoid MCMC whenever we can. So we try to use analytical expressions, or use very good approximations to those expressions. When we use a numerical approximation we usually …
-
Hi John, Yes, this would be useful. Could you perhaps make a feature request on our GitHub page? (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/). Cheers, E.J.
-
In the more recent version of JASP, we integrated "correlation matrix" and "correlation pairs" that used to be separate (so: your professor was using an older version). But when you go in to "correlation", all of the ma…
-
Strange. We would love to see this as an issue on our GitHub page! (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/). The programmers probably need some more information to figure out what's going on... Cheers, E.J.
-
Hi Mathieu, Yes, they will be part of the next release, which is about a month away. There is probably no good scientific reason against mixed effects models, but then again, good scientific reasons do not always win (e.g., the field is still domina…
-
Hi Rick, Hmm that's strange. Could you create an issue on our GitHub page? This way you'll have immediate access to our programming team. They will need some more information. (for details see https://jasp-stats.org/2018/03/29/request-feature-report…
-
Would be a good feature request. I don't know whether anybody from the team has picked up this idea. E.J.
-
That sounds like an interesting approach. I am not a geneticist so I don't know what normalization is required or what would be the best way. Cheers, E.J.
-
Hi KB, Well, I think you would just conclude that there is weak evidence for the exclusion of the predictor. It is important, when the data do not support a strong conclusion, to resist the temptation to draw a strong conclusion anyway. You could g…
-
Dear rbeesley, One hack is to temporarily move the dependent variable back to the input box. This will stop JASP from updating. When you are done with the filter, you drag the dependent variable back in. Let me know whether it works! Cheers, E.J.
-
I'll ask our expert, Don. @Don: Google translate the above text gives: "I'm analyzing the reliability analysis of a variable. The items are coded with 0 (error) and 1 (hit), in one of the items all the participants were right, so JASP does not…
-
No, not necessarily. If BF0+ is high (so evidence for H0 versus the directional alternative hypothesis that the effect is positive) this can happen because the effect is absent or because the effect is actually negative. With small samples, a high B…
-
Hi Henryk, Whether or not the uniform distribution is appropriate for testing depends on the scale. So for a correlation coefficient or a probability, a uniform prior may be acceptable. For the t-test, however, it is not. The reason is that effect s…
-
Actually, I just archived two analyses that show the robustness heatmap -- I realized that the ones I gave you may not show the heatmap, as this is a feature we added recently (which is also why it is not yet discussed in the paper on ArXiv). Anyway…
-
I'll ask our expert! Cheers, E.J.
-
I'll ask our experts! E.J.
-
Hi Mila, I'll ask our expert. Cheers, E.J.
-
What happens is that one of the cells has text (a string). This may for instance be "NoNumber" or whatever is used to indicate a missing value. You can define these in the Preference menu. Sorry for the tardy response, it is a little hecti…
-
Hi KB, Yes, more samples will eventually result in overwhelming support for H0 or for H1. If you have a p-value that is just significant it will almost always be the case that the BF is not compelling. P-values only consider the surprise under H0, a…
-
I'll ask our experts. Do you have a reference or a link to provide an example of what you'd like to do?
-
Hi Gill1109, Thanks for giving JASP a go, and thanks for reporting some issues. I'll deal with these one by one below: "To begin with I had repeated problems with bugs in the Mac version of JASP. Moreover, it took aeons to download. The Windows…