JohnnyB
About
- Username
- JohnnyB
- Joined
- Visits
- 171
- Last Active
- Roles
- Member
Comments
-
HI @EmilieC , Yes, that one is definitely related, but it's a feature, rather than a bug (and consistent with other analysis software). The blogpost I linked explains this property more clearly, including an example. Cheers Johnny
-
Hi @andersony3k , I took a look at your data files and comparison of the different programs - it seems in one you applied a filter and in one you did not, which leads to different results. Removing the filter leads to the same results, which are in …
-
Hi @EmilieC , You are right - the interactions have disappeared from the marginal means menu and we are currently working on bringing those back! There is a difference between the frequentist and Bayesian anova in how they handle the posthoc tests (…
-
Hi @MaddieP , Thanks! The JASP computation (https://github.com/jasp-stats/jaspAnova/blob/ae3f25859919c9a91a19832ae8f09065399f2c84/R/manova.R#L246) is based on the BioTools R package, but also produces the same results as the HePlots package: > he…
-
Thanks @MaddieP , what was the model you used? in terms of (in)dependent variables.
-
Hi @MaddieP, Could you please provide a data set where you obtain different results? I just checked against the heplots package (https://search.r-project.org/CRAN/refmans/heplots/html/boxM.html), and get the same results in JASP. Kind regards Johnny
-
Hi @alexa , To add to EJ's comments - I would not use too extreme values in the robustness test, since at some point the prior becomes so narrow or wide that the model just becomes non-sensical. EJ's suggestion to double and half the main values wor…
-
Hi @MaximusLuminis , In that case it will be a bit tricky, because the Friedman test is generally for one-way RM ANOVA's, not for multiple repeated measures. One option to explore would be to try a rank-transformed ANOVA, where you convert your DV t…
-
@samamarc , Great to hear this is clarified now. A blogpost I wrote a while ago might illuminate the issue some more: https://jasp-stats.org/2020/04/14/the-wonderful-world-of-marginal-means/ Cheers Johnny
-
Hi @MaximusLuminis , I'm afraid that's a limitation of the Friedman test - it requires complete observations (e.g., see here), so you can only run it for those participants who completed all trials. Do you also have a within subjects measurement? Be…
-
Hi @albapy , Sorry for the late reply. You can read about this behavior in this recent discussion: https://forum.cogsci.nl/discussion/comment/27509 and linked blogpost: https://jasp-stats.org/2020/04/14/the-wonderful-world-of-marginal-means/ Basical…
-
Hi @maamorim , Great to hear! Just to clarify - if you were to compare the planned contrast analysis (which is also based on the marginal means) to the posthoc analysis, these are identical, except for the p-value correction (which always makes the …
-
Hi @maamorim In addition to @andersony3k 's helpful comment, I would like to point you to this blogpost from a while ago, where I outline the follow-up tests for ANOVA. Basically, contrasts and post hoc analyses are based on the estimated marginal …
-
Ahh Im sorry I thought we were discussing JASP. In R you could write some sort of for-loop for stepping through all pairwise comparisons - as far as I know there is no function in BayesFactor that does this for you.
-
Hi Søren, The Bayesian post hoc tests do this, and you can just take the Bayes factors from there. Unfortunately we do not have different types of contrasts (e.g. Helmert/simple/etc), which is still on our todo list. Kind regards, Johnny
-
I think interpreting a main effect in isolation while there is a significant interaction effect is pretty tricky to do, and your case highlights why. Because of the interaction effect, if we look at what our model predicts for someone in condition B…
-
Hi @Whirly123 , I just took a look (always excited for such little puzzles!), and I think something went wrong in the encoding because there is not a perfect correspondence between the two ways of encoding: https://forum.cogsci.nl/uploads/434/P75JY…
-
Hi Søren, Unfortunately not - the emmeans/contrast functionality is only included in the frequentist ANOVA's in JASP. As far as I know the emmeans package does not include Bayesian analyses, or am I mistaken? We do include Bayesian post hoc tests, w…
-
Hi @camillacaponi , Without your JASP file I cannot be sure where your analysis went wrong, because when I perform a RM ANOVA on your data, I think I obtain similar results as you report in SPSS. See the post hoc tests below (with Holm and Bonferoni…
-
Hi @Farina , Your assumption is correct - before we were doing manual t-tests (so no particular package), with corresponding df's, so I was also glad we were able to upgrade the procedure. Kind regards Johnny
-
Hi @Farina , Since version 16, we use the contrast function of the emmeans package (see https://rdrr.io/cran/emmeans/man/contrast.html) to compute the df's/SE for custom contrast analyses, which is an improvement on how it was done before. The diffe…
-
Hi @autumn_moments5648 , The Bonferroni adjustment can either be an adjustment to the alpha that is used for deciding on statistical significance (e.g., 0.05 / 3 = 0.016666), or you can adjust the observed p-values themselves by multiplying them by …
-
Hi @ETcog , Interesting! Could you maybe send me your data set so I can take a closer look? This might be an interesting cognitive task though - how sure are we that there are more than 30 points in the plot? You can send it to j<dot>b<dot…
-
Hi @Valentinos, The Dunnett test that JASP offers, is the type where all levels are compared to a control group (i.e., "Dunnett’s Many-to-One Comparisons Test"). This is based on the multcomp package (specifically, the glht and mcp functio…
-
Hi @je77je , Sorry for the delayed response, this flew under my radar! The reason for the discrepancy you see, is that by default, the RM ANOVA post hoc tests use the pooled standard error (you can see that the SE is the same for all 3 comparisons)…
-
Hi @szaszkob , Unfortunately the design requirements for the Friedman test are pretty strict (it is the same in other software, like R). It also does not allow multi-way designs, and can only assess main effects. You could consider an ANOVA on the …
-
Hi @Daniela3h and @davidwidman , Unfortunately this feature is not included in the contrast analysis yet, but I will look into implementing this relatively soon. In the meantime, you could look into the source provided by David, or try to conduct a …
-
Hi @PerPalmgren , I am getting the same results as you. The only thing that could have changed is this one https://github.com/jasp-stats/jasp-desktop/pull/3140. This is a pretty old PR, but maybe Mark has not updated the analysis (though it looks as…
-
Hi @MassMobile , Yes, it works through regrouping, although I do not think this should be warranted and actually implemented a check recently to limit the application of the Friedman test, since this could be misleading. In the future we want to imp…
-
Hi @andersony3k , Yes that would be a great functionality! We are now exploring the possibilty of specifying a more flexible structure for the SE's (with the nlme package) in ANOVA, which would make this functionality very straightforward to impleme…