EJ
About
- Username
- EJ
- Joined
- Visits
- 2,557
- Last Active
- Roles
- Member, Administrator, Moderator
Comments
-
I think this is more of an issue for our GitHub page, as the JASP programming team will have to help you out on this one... E.J.
-
Dear Zhenzhen, First off, the p=.056 vs BF10 = 2.848 result is pretty much as expected. The anomaly is the p=.008 vs BF10 = 1.977 outcome. What models did you compare exactly, to get at the interaction BF? Also, is this a repeated-measures ANOVA? We…
-
We use the method as implemented in the BayesFactor R package, which is based on the JZS prior. If you go to the help file (press the blue circle with the "i" in the input panel) you'll find the references at the end. Key is this one: Roud…
-
Hi Mary, SPSS appears to be using the BIC approximation. This often works well, but what we (JASP and BayesFactor) are trying to do is a little more sophisticated and precise, or so we like to think. Cheers, E.J.
-
This is a recent change in JASP. We had our reasons but people got really confused, so we're changing it back. See our GitHub page for a discussion. Cheers, E.J.
-
Hi Jaiden Michael and I used to do an annual workshop that provided the ideal opportunity for the kind of feedback you seek. Unfortunately the last one was cancelled due to corona, and the status of the 2021 rendition (usually in August) is still ve…
-
I've downloaded the csv but I am unclear about what analysis fails -- what is the DV and what are the predictors? Can you email me the .jasp file? Judging from the error message, you probably have too few observations in one of your design cells to …
-
Hi alg, I'll pass this on to our experts but I'll note that the mixed models do not (yet) come equipped with Bayes factors, and this means that there is presently no way to quantify the degree to which the data support or undercut the null. Informal…
-
Hi Andre, This is in fact a very good question. Previously we did allow factors to be entered as predictors, but they weren't correctly coded as factors. In the upcoming version we have separate input boxes for continuous covariates and factors. If …
-
Hi Laura, JASP uses lavaan code, so it would just be a matter of writing the required lavaan syntax and pasting that into the SEM window. Cheers, E.J.
-
Hi Ravi, A covariate is something continuous, like GDP; a factor refers to a discrete category, like country. Sometimes there are no covariates or factors of interest, so it is purely optional. Most people would recommend a random effects model; how…
-
That's strange. I'll notify the team, but this is really more appropriate for our GitHub page (https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/) E.J.
-
Hi Emma, Some general remarks: Missing data is not a problem for the Bayesian model (at least not in principle) You probably want to use a crossed-random effects model instead of ANOVA? In this case you should use our mixed model functionality (unfo…
-
In order to help I think it would be good to see some more background -- maybe a screenshot of the data, and a more precise description of what it means for JASP "not to like zero's". It is also not clear to me why you would model this as …
-
I don't see why it would be problematic to add interaction terms. JASP uses lavaan, so I recommend you consult the lavaan documentation (or Google the right key words). E.J.
-
When you have activated the mediation analysis in JASP and you click on the blue "i", the help file will appear. At the end of that file, there are two papers referenced. The most conceptual one seems to be this: Jeremy C. Biesanz, Carl F.…
-
Hi Mariamr, Yes, I think that two separate ANOVAs are an acceptable alternative solution. Cheers, E.J.
-
A BF is like the scales of Lady Justice -- when one scale goes up the other one must go down; they are in a one-to-one correspondence. So if you have BF10 = 5 the data are 5 times more likely under H1 than under H0, and this is evidence for H1. This…
-
You could generate a column with random binary numbers. Then you could use that to filter one subset or the other. E.J.
-
Thanks for reporting this bug -- it would be great if you could post it on GitHub so the programmers can see it and take action! Cheers, E.J.
-
Hi Markus, I'll ask the other team members, but I think it is cleanest if you create new columns or new files for each filter setting. Cheers, E.J.
-
Dear ladonnalr, We get this question more often. A screen reader cannot be used with JASP effectively at the moment. Key-navigation is present but not fully implemented. So it would be hard to navigate even if the screen reader works. We are very mu…
-
The default setting for the Bayesian correlation is a prior width of 1. This is based on a stretched beta distribution that is symmetric around 0.5, so beta(a,a). The prior width is then defined as 1/a. It is clear that "a" should be highe…
-
Hmm I think you might want to do this many times, and that would be easiest to achieve in R. In JASP, if you just want to do this once, I guess you can create a new column with random 0 and 1's, then use that to filter either one subset or the other…
-
I do see it mentioned in https://en.wikipedia.org/wiki/Effect_size#Cohen's_d You take the (pooled) SD and divide by sqrt(n) (or a pooled version in case of the unpaired, unequal N case) E.J.
-
Well I believe there is some discussion on using the F-test for equality of variance, because it is rather sensitive to the assumption of normality. Anyway, we do have tests for equality of two variances as well. Best to check out the program and is…
-
How you calculate SE depends on your effect size measure (Cohen's d, log odds ratio, etc). For any particular measure of interest, I believe the Wiki entry generally shows how to compute the SE. E.J.
-
For an intuitive interpretation see also the text above Eq 8 in this paper: https://www.tandfonline.com/doi/full/10.1080/02664763.2019.1709053 E.J.
-
Hi Maria, Your BF10uncorrected = 0.173, meaning that the data are 1/0.173 = 5.78 times more likely under H0 than under H1. In other words, BF01uncorrected = 5.78, meaning you have evidence in favor of H0. The multiplicity correction increases that e…
-
Hello may01dz, Yes, as used by ANOVAs for instance. Cheers, E.J.