EJ
About
- Username
- EJ
- Joined
- Visits
- 2,557
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
Is this the fact that Age is recognized as a nominal variable? Of course the program should not crash but give an informative error message instead. For these and similar issues, our programming team can assist you more effectively if you post on ou…
-
http://www.ejwagenmakers.com/2015/NuijtenEtAl2015.pdf
-
Well the reliability of the estimates is indicated directly by the width of the posterior distribution. Your frequentist reviewer would have to explain why the evidence for this specific data set is less reliable than is indicated by your analysis. …
-
Dear Stefan, Yes, this works. However, be sure to add the relevant references (in particular the work by Rouder and Morey). You might also mention that these are the settings of the BayesFactor package. Cheers, E.J.
-
Hi Philip, Theoretically, it does not matter what your sample size is. With low samples sizes, the posterior distributions will not be very different from the prior distributions, because not much has been learned. Of course there may be specific p…
-
I'll attend some other team members to this post E.J.
-
I am not sure what goes wrong... I have tried this with 0.9.0.1 and the latest 0.9.1.0, and it works like a charm, for both the Bayesian as well as the classical analysis.
-
Can you send the JASP file, perhaps only with the gender column?
-
Also, you might want to check that the problem persists for the 0.9.1 version (out just now)
-
Dear bemeev, In order to help you effectively please post the issue on our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/). This is also helpful for the team, as it creates a permanent record of the…
-
Hi HannaG, What is relevant for the BF is the relative predictive performance of H0 vs H1. Unless d=0, the value of d actually tells you nothing at all about the BF; what you need is d and n. Specifically, when you have a low d but high n, you can …
-
Is this helpful: https://jasp-stats.org/wp-content/uploads/2018/06/interaction_effect.gif
-
OK, we fixed this in the new version. We hope to release that today, or otherwise in the very near future. Thanks for bringing this up! Cheers, E.J.
-
Yes, that explanation sounds right. Sometimes there can also be discrepancies because the model is misspecified (e.g., data are not normally distributed). The more fundamental problem is that "interaction" is vague -- all non-additive patt…
-
Hi dustinfife In the model specification tool, you can select the variables you want to have interact, and drag both over to the box simultaneously. This will create the desired interaction term. Let me know whether it works. Cheers, E.J.
-
Dear E, Well, if there are .sav files on the OSF then you should be able to download them and subsequently open them in JASP. I assume your question is whether you can log in from JASP, locate the .sav files, and then read them in? I don't know whe…
-
Hi Boo, About the mean and variance: The data-generating process should be the same: it's OK for the sample estimates to fluctuate. With respect to multiple replications, I think it is conceptually most strong to compute the Replication BF separat…
-
Dear Ajestudillo, Your interpretation is right (as long as, when you speak of one hypothesis being "favored", you mean to say that the data are more likely under that hypothesis than under the other). And yes,. I always transform the BFs …
-
Yes, that's correct, but note that for this approach to work you'd have to assume that other parameters (means and variances) are the same across experiments. If that's not the case, you could simply run a first analysis, get the posterior for effec…
-
Dear Harrison, I think you've posted this in the wrong subforum. This is the one on BayesFactor and JASP! Cheers, E.J.
-
Dear J, There are various explanations. 1. As you can see from the BIC equation, the (implicit, in most cases) penalty for complexity involves N. Basically, with high N the predictions from H1 become more ambitious: if there is a true effect, with…
-
Also, and in general for cases such as these, we can help you out but our programming team will probably need some more information. In order to help you effectively please post the issue on our GitHub page (for details see https://jasp-stats.org/2…
-
Dear Boo, Are you using a t-test? If so, you could take a look at the following two papers: 1. Informed t-test (https://arxiv.org/abs/1704.02479) 2. Replication Bayes factors (https://psyarxiv.com/u8m2s/) Cheers, E.J.
-
Sorry to hear that, but glad it's fixed now
-
For a correlation between ordinal data it is generally recommended to use Spearman's rho or Kendall's tau. If you view some of the video's on our YouTube channel you might get a better idea on how to do this. Cheers, E.J.
-
Dear rhea, Thanks for reporting this. We can help you out but our programming team will need some more information. In order to help you effectively please post the issue on our GitHub page (for details see https://jasp-stats.org/2018/03/29/reques…
-
Hi Rhea, Can you say a bit more about the structure of your data and the questions you wish to answer? Cheers, E.J.
-
I mean a statement that says something like "Warning: the prior structure is appropriate only for continuous predictors"
-
This must be due to rounding, in which case it is best to use the BFs. (I'll double check, thanks for bringing this up). Cheers, E.J.
-
Off the top of my head, I would assume that the "sync" option should do the trick. If this does not work for you, please let us know on our GitHub page and our programming team is at your disposal. [BTW, thanks for bringing these issues to…