EJ
About
 Username
 EJ
 Joined
 Visits
 1,468
 Last Active
 Roles
 Member, Administrator, Moderator
Comments

That should not happen. What version do you have? This is something for our programming team  if you post this on our GitHub page the team can assist you. For details see https://jaspstats.org/2018/03/29/requestfeaturereportbugjasp/ Cheers, …

Is the order perhaps alphabetical? E.J.

Hi Ravi, The relevant blog post mentioned in the help file (https://jaspstats.org/2017/11/15/metaanalysisjasp/) says: "(For instance, Ι2 expresses the excess variance as a percentage of the total variance observed in the effect size estimate…

With respect to the instabilty: this is typical of *all* MCMC techniques  so also for estimating the posterior mean of a parameter, say. With respect to the bias: I have not encountered this as a problem (again, this concern seems to hold generall…

The BAS package in R also offers logistic regression with JZS priors, but because the likelihood is different the precise implementation will differ from that of linear regression. If you want to know more I'd advise you to check out the BAS documen…

Dear wnk4242, I am not sure what you mean with "stability and accuracy", but note that a more recent paper on the topic is here: https://link.springer.com/article/10.3758/s134280181092x Another one that may be relevant is here: https://…

Dear Guillaume, These look a little unusual indeed. For the first graph, it is strange that the BF hardly changes from 1 until after about 400 data points are in. What sometimes causes these graphs to look strange is that the first n rows feature on…

This happens because it is a customizable plot that really consists of different panels (i.e., the marginal densities to the side). There are several other places where JASP shows the scatter plot (e.g., in the correlation analyses), and I assume th…

Taking the average of BFs in generally inadvisable. For instance, the average of 3 and 1/3 is not 1. So taking the mean of the log BFs is indeed a better idea. E.J.

This is what Harold Jeffreys argued, if I'm not mistaken; I am a little allergic to "must", and also to the categorical acceptance of normality based on a pvalue test, but I think I'm with Jeffreys on the general sentiment. E.J.

I've passed this on to our expert on this topic

Interesting. What is reasonable to do depends on the extent to which R1 and R2 are the same. If they are completely the same, then you effectively have twice the number of observations for the intersecting participants. So you basically estimate th…

Makes sense to me! Cheers, E.J.

I think this is a bug! Well spotted. We'll fix this before the next version.

For completeness, here is Bruno's response to the same post above: "Hi Elaine2121, Windows 7 endoflife is January 2020, so it is in fact quite dangerous to use it (no security update from Microsoft) . The thing is that we don't test JASP on W…

We're discussing it, thanks for the suggestion E.J.

Hello Bence, Good question. Apparently the Friedman test requires the ANOVA to be balanced?! Have you checked out any alternatives yourself? Cheers, E.J.

Thanks for pointing this out, I'll forward this to our network expert. Cheers, E.J.

Both methods estimate the same quantity  the BF, which depends on the parameter prior but does not depend on the model prior. In your equation for the MCMC method, the change from prior to posterior odds equals the BF, which is the ratio of margin…

If you activate the SEM module you will find that under "Data" you can choose "variancecovariance matrix". I did not see such an option for factor analysis or network analysis  would be a good feature request perhaps? Cheers, …

I am not 100% certain, but I think BayesFactor integrates out the random effects, such that the key measure of interest requires only a onedimensional integral. When you add particular effects to the ANOVA model (which cannot be integrated out) the…

Dear Wang, Exchanging the role of the posterior with the marginal likelihood yields the Chib estimator, if I recall correctly (there is also a Chib & Jeliazkov paper). In transdimensional MCMC, the model prior matters for the convergence, but ul…

Dear Luis, The correction goes by means of the prior model probability. See for instance this summary: https://psyarxiv.com/s56mk Cheers, E.J.

Hi Luis, This sounds reasonable. I would replace "These results strongly suggest that the observed data is 12.8 times more likely under the alternative model." to "These results show that the observed data is 12.8 times more likely un…

This is a recurring topic on this forum :) I noticed a previous thread that may be relevant: https://forum.cogsci.nl/discussion/3013/whatdenominatordoesthecohensduseonjasp Does this address the issue? If not I'll ask Johnny to look at this.…

Dear Faya, I'll forward this to our expert. Sorry for the tardy response. Cheers, E.J.

Yes. But I do think it is important to check that these nuisance factors do not differ between the groups. Suppose that cortical thickness is lower for older people, and patients are older than the controls. Instead of removing noise, you may then b…

It is always tricky to pick the right design, because in the current setup there will be practice effects and effects of time (and possibly selection effects as dropout may be related to performance?), so the comparison is potentially confounded. B…

Great that you found it. I often forget about this one myself, and we have been thinking of creating a separate, visible button for this somewhere. But we can't seem to reach agreement on the exact implementation, and whether it would be worthwhile…

Dear Wang, Maybe it would be clearer if we'd use the term "Parameter Prior" instead of "Prior". The parameter prior partly defines the prediction that a model makes, and through these predictions it affects the evidence and there…