EJ
About
 Username
 EJ
 Joined
 Visits
 90
 Last Active
 Roles
 Member, Administrator, Moderator
 Thanked
 3
Comments

That is great to hear! And our upcoming version will offer a lot more. We are testing it out now. @Don: it is good to attend people to the interpretation of the coefficients (also for the paper you are writing). E.J.

Don, the data set is from "punting", available under "Regression" at the JASP generic workshop materials on the OSF: https://osf.io/r73y9/

BF_m quantifies the change from prior odds to posterior odds. Here I'd select "compare to best model" and then BF_01 for display, and you'll see how many times better the "openness" only model predicts the data compared to other models. Cheers, …

:) It's not silly at all, of course. This stuff is very difficult. And even if it were silly (which it is definitely not), remember that it was the Leibnitz (!) who argued that the chance of throwing "11" with two dice is just as high as the chanc…

Dear blindreviewer, The questions you raise (also in an earlier post) are of course highly relevant. The Rouder et al. paper is not an easy read, and it would be worthwhile to dissect the reasoning and the resulting performance much more than has…

Awesome!

Well, there is this: http://www.softpedia.com/get/ScienceCAD/JASPStatisticsProject.shtml So I'm not sure what is causing this but I'm forwarding this to the programming team. E.J.

This appears to be the case because of N. In the ANOVA case, you see that N=25 everywhere. For "descriptives", some N's are 27. So I assume that there are two subjects you have missing values for some of the conditions, which means that they are exc…

Dear TIA, Sorry, your question slipped through the cracks. One of our team members is revamping factor analysis and I'll forward this to him. Again, sorry for the tardy response. Cheers, E.J.

Dear gvleioras, Sorry for the tardy response. Yes, you can do exactly as you suggest. For the second problem, you can also use loglinear modeling. If you analyze this using JASP the relevant tests should become available. Have you tried? E.J.

Dear jpoll, So you have been using the "split" functionality but you would like to use multiple factors, is that correct? Right now you can only split either by gender OR by income class. If you want to split by both you'd have to create a new co…

Dear H., Yes, this is easily achievable, in two ways. For concreteness, suppose you want A+B+AB vs A+B. Method 1. Add "A" and "B" to the null model (under the "model" tab). Method 2. Use transitivity: if you already have BFfull = A + B + AB v…

Dear GuillaumeC, Interesting question. It seems that the "controlling" variable really doesn't add much, and including it just incurs a penalty for complexity. Then again, there is an argument for including the controlling variable no matter what…

Dear Ester, Perhaps you only have fixed effects, in which case I'd just report those. The ANOVA priors were proposed by analogy to the ttest; if you conduct a betweensubjects ttest with the default r=.707 setting you ought to get the same resu…

Hi Kazimiera_Worf, Yes, absolutely! Cheers, E.J.

Dear Mvs, Depends on what you mean with "hierarchical regression". Usually this means that you can add predictors in batches, and assess whether the new batch of predictors ought to be included. And yes, our linear regression functionality allows…

Yes, that's correct. E.J.

Hmm I'm not sure. Before I attend the author of this module to your question...have you seen the JASP blog post on this module? https://jaspstats.org/2017/11/15/metaanalysisjasp/ E.J.

Re effect size: this is not straightforward (I think). There's a paper with Maarten Marsman that is currently somewhere in the review system. We need to polish the Bayesian ANOVA anyway in order to show parameter estimates. We'll take the effect siz…

Hi Tom, The Bayesian ANOVA (it is really a linear mixed model, see the BayesFactor documentation) makes the same assumptions as the classical ANOVA. We just have not developed the Bayesian echoes for those assumption tests (yet). We will do this …

Hi Nils, Yes, your interpretation is correct. The analysis of effects modelaverages across a range of models. You could also look at the standard table and compare the the full model with the model that includes only the twoway interactions. Pl…

Yes, but BF10 = .2 is more difficult to interpret than the mathematically equivalent statement BF01 = 5. you can report that there is evidence for H0, but the table indicates how much  it matters whether BF01 = 1.5 or 8. If you want to summarize …

Hi Alexa, For DV2, you see that all BF10s < 1. This means that you have evidence for H0  if you set the BF display option to "BF01" instead of the default "BF10" you see how much more likely the data are under H0 than H1 under each of the ot…

Hi Rob, Thanks for the feedback and the nice words. If you like your suggestion to have permanent impact you can post it on our GitHub page (https://jaspstats.org/2018/03/29/requestfeaturereportbugjasp/). I am of two minds about adding the n…

Hi Mila, Yes. The correct conclusion is that the data do not provide information: the hypotheses under consideration predicted the data about equally well. Of course, the data may be informative in other ways; for instance, perhaps the posterior …

Hi Mila, BF10 = 1/BF01, so if BF10 < 1 then BF01 > 1. In other words, it is impossible that both BF10 and BF01 are lower than 1, unless you are referring to different tests. Maybe you have a concrete example? In general, BFs near 1 are not …

Dear bmc0012, I assume you mean a hierarchical design with multiple observations per unit (?). For linear regression, JASP implements the BAS package, which does not deal with this situation. Maybe the BayesFactor package can do this  you could…

I checked. The numerical value is correct, so the log(BF) = 23, but in the Jeffreys categorization scheme this is of course "extreme" evidence (we have to take the exponent) E.J.

No, but it would be a good feature request. I've added it for you on GitHub (https://jaspstats.org/2018/03/29/requestfeaturereportbugjasp/) E.J.

https://link.springer.com/article/10.3758/s1342801607398 Jamil, T., Ly, A., Morey, R. D., Love, J., Marsman, M., & Wagenmakers, E.J. (2017). Default "Gunel and Dickey" Bayes factors for contingency tables. Behavior Research Methods, 49, 638…