EJ
About
- Username
- EJ
- Joined
- Visits
- 2,597
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
On https://jasp-stats.org/jasp-materials/ you will find Mark Goss-Sampson's manual Bayesian Inference in JASP. My website has several tutorial articles, and each year we organize a hybrid workshop on Bayesian hypothesis tests at the University of Am…
-
The prior information that is test-relevant is the expected effect size under H1. This is usually expressed as Cohen's d, that is, the difference in age divided by the standard deviation. You can certainly specify a prior based on previous research …
-
Please post the issue on our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/), and the programming team can take it from there EJ
-
Do you have a screenshot, and perhaps part of the data? EJ
-
Do you have a screenshot, and perhaps part of the data? EJ
-
This is really strange and obviously should not happen. My go-to suggestion is to go to "Preferences" in JASP and turn "Safe graphics" on, but if that does not do the trick, my advice is to post the issue on our GitHub page (for …
-
This is more of a question for the programming team, it seems -- see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/ for details. Cheers, E.J.
-
This has been implemented in development and will be part of the next release. EJ
-
Yes, installing the latest version will overwrite the existing one. [If you have installed via the Microsoft Store, the update is automatic] EJ
-
I haven't checked it lately, but adding a comment is always possible! EJ
-
Maybe we already know this is not the solution, but when I hear of such issues I'd be tempted to set "safe graphics" to "on" in the Preferences... EJ
-
You could support the feature request so that it bumps the priority on our to-do list! EJ
-
Yes, but in these situation a rank-based analysis would not be a good test either...
-
So I take it you cannot solve the problem by simply leaving out entire variables. But in general, if you know that you are only interested in particular comparisons, doesn't this mean you should be doing planned comparisons instead of post-hoc tests…
-
I saw Johnny address this somewhere, but I forgot the location. I'll attend him to this issue. EJ
-
I am not sure, but it has become a principle that anyting you see in a graph you should also be able to get in a table. I think this probably warrants a feature request on our GitHub page (for details see https://jasp-stats.org/2018/03/29/request-fe…
-
I think you would need to be more explicit about what these variables mean, and what the results show...based on your description it does seem like a puzzle. Perhaps it is best to do some checks...if you are missing some key variables you can get st…
-
The solutions to deal with model misspecification apply to both frequentist and Bayesian approaches. I am aware that the GLMMs do not do Bayes factors, and we really should address that. Issuing a request on our GitHub page will help with that. I am…
-
This depends on the specific problem. I am not sure why you would want to recode a continuous variable (since this is losing information), but currently that is probably most efficiently done in R. Maybe the "compute new columns" functiona…
-
Dear Imbar, Well this depends. There is something to be said for both. But if the interest is really in each of the two main effects separately, I would be tempted to monitor BF10 for the model with only main effect A versus the null model, and BF10…
-
JASP has a machine learning module that does the splitting, but I don't think we do this for logistic regression. I guess you could do this, but given that all of the standard statistical machinery is already available, I am not sure why you would w…
-
Also, JASP should be able to handle more than one random factor I believe...
-
Have you checked the help file? Also, there is this paper: https://shs.cairn.info/revue-l-annee-psychologique-2020-1-page-73?lang=fr van den Bergh, D., van Doorn, J., Marsman, M., Draws, T., van Kesteren, E.-J., Derks, K., Dablander, F., Gronau, Q.…
-
This error (the matrix not being positive definite) arises when the model is overparameterized, or when the items have severe overlap. Ultimately this is a lavaan issue, but you might want to check the correlation matrix of your items.... EJ
-
Let me speak for Johnny, as I know he's busy teaching. The assumption (and yes, it is a parametric assumption) is that the latent distribution is normal. But my current thinking is that this is a relatively mild assumption, since it applied also to …
-
I'll ask the team EJ
-
Hello! Here are some answers: The JASP website is a hub for all things JASP. For instance, check out https://jasp-stats.org/jasp-materials/ That similarity is not a coincidence, as it is the work of the programmers who I initially hired to help set …
-
I'll ask to make sure EJ
-
Dear Neeraj, In the context of what specific analysis is this? We do have an ROC curve in the Learn Bayes -> Binary classification module, for instance EJ
-
Hi Frans, Yes, but in order to take action it would be great if you could issue 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.