JohnnyB
About
- Username
- JohnnyB
- Joined
- Visits
- 97
- Last Active
- Roles
- Member
Comments
-
Hi @PerPalmgren , It's an effect size measure described in this paper and as implemented in the afex R-package. This effect size takes into account the other terms in the specified model (including interactions), and seems to be the recommended effe…
-
Hi Rik, The paper is online - https://psyarxiv.com/y65h8/ We are now setting up a special issue with responses to this paper, and will then also publish a collaborative guidelines paper where we come back to the questions posed in the paper. If you …
-
Hi Declan, Ah yes, thanks for clarifying! In the case of the t-test, you can look at the 95% credible intervals for the group means (in case of two groups) or the difference with the test value (one sample t-test) by ticking the box "Descripti…
-
Hi Declan, In your calculations you're assuming that Cohen's d has the same standard error as the mean difference, which is not the case, since they operate on different scales. Even though it would maybe make intuitive sense to just divide the lowe…
-
Hi @DrPRW, Centering your variables can help in remedying multicollinearity in cases where you have multiple terms per variable such as square or interaction terms. Here, subtracting the means influences the interaction estimate, which will in turn …
-
It is the proper follow-up test for Friedman's test because it's pairwise and rank-based, while still taking into account all observations (i.e., it uses the aggregated ranking from all observations).
-
Hi @gutenbar, The Conover test is a pairwise test that compares the average ranks between two groups (these are giving by JASP as Wi and Wj; see also the help file, under Output -> Nonparametrics). The relevant reference here is Conover,W. J. (1…
-
Hi @Yasra , Responding here for visibility. The results you report are based on posthoc tests with pooled error terms. This has repercussions for all posthoc tests, and can lead to higher or lower p-values, compared to using the unpooled error term…
-
Hi @Yasra , I would expect the bonferroni p-values to be larger than the uncorrected ones. Do you have some data set/jasp file where the reverse is the case, that you could share with me? That way I can get to the bottom of this. As for the equal/u…
-
Hi @claudia , I took a look at the results, and I think this is a case where the Bayesian analysis is more conservative than the p-value. Whether to take this as a critique of the Bayes factor (for being too conservative), or a critique of the p-val…
-
Hi @carl559 , Since you initially used mixed models, this implies that you have nested observations within some categorical variable (e.g., within one participant, or experimental group - if I understand correctly this is "study label" for…
-
Hi Claudia, That seems like a strange result indeed. What version of JASP are you using? If you are using the latest version (0.15), would you be able to share your data set or jaspfile with me, so I can take a closer look? you can send it to j.b.va…
-
Hi @PerPalmgren, The residual term there is the between subjects error term (i.e., unexplained sum of squares). There are no between subjects terms in the model, so this is the unexplained sum of squares based on just participant numbers. Together w…
-
Hi @PerPalmgren, The initial RM ANOVA model is a model where the error terms are pooled - this is simply how linear models are set up in these settings. As a consequence, the emmeans R-package (which we use for marginal means, contrasts, and posthoc…
-
Hi @AMa, You are correct - we do not offer effect sizes for posthoc tests for interaction effects. I am currently working on implementing these (this is not a very straightforward procedure, so it requires some research). My goal is to have this in …
-
Hi @Mientusz, Thanks for the additional info (and thanks @TarandeepKang for the clarification!). What I ended up doing was to run a RM ANOVA, with each Seed column as a repeated measurement, to keep the design intact (so having 10 levels of a repea…
-
Hi @Mientusz, No problem, it turned out to be a problem with the delimiter character, but all good now! So I have some more questions: You would like to test for differences between the substrates, and between the FerroSorp factors? Is germination r…
-
Hi @Mientusz, Based on your last reply I think a linear mixed model is the way to go here. This model will account for each individual dish, while allowing you to compare the average (i.e., fixed) effect between the different groups. I looked at th…
-
Hi @Sameha, JASP uses the emmeans package (specifically, its 'contrast' function) for the the post hoc tests. When Tukey is checked, it will use Tukey's correction and when the group sizes are unequal, it will automatically adjust for this (i.e., pe…
-
Hi @emmacampbell, The method that is currently in JASP is not MCMC-sampling based, so there are no convergence or sampling diagnostics to report. I am planning an overhaul of the analysis to update the (sampling-based) method, and then we will have …
-
Hi @ArtsBrain , Perhaps this blog post is of use to you - it outlines the different contrast options available in JASP, including the ability to specify custom contrasts (which might be what you want to do). Cheers, Johnny
-
Hi @twiggy and @MaximusLuminis , I have looked into the code that decides whether to use the Durbin test or Friedman test, and it was indeed too permissive. I have updated the code, and it should be fixed in the new JASP release (0.15, hopefully re…
-
Hi @twiggy and @MaximusLuminis , I have looked into the code that decides whether to use the Durbin test or Friedman test, and it was indeed too permissive. I have updated the code, and it should be fixed in the new JASP release (0.15, hopefully re…
-
Hi MaximusLuminis, Thanks for reporting this. It's true that JASP automatically switches to the incomplete version of the Friedman test when it detects some missingness somewhere. Are you in the position to share the data set and JASP file with me, …
-
Hi Malikosaka, That analysis takes the first level as the control group. You can change this order by going to the data view (the left most panel in JASP, where you see your data and variable columns. Then, you click on the name of your factor varia…
-
Hi Da5Ax, Unfortunately we do not yet have effect sizes for contrasts, but this is high on the priority list! Kind regards, Johnny
-
Hi PerPalmgren, I personally much prefer Q-Q plots for assessing normality, since they are much more informative than a single p-value, but I understand your position. One improvement on your current way of working would be to instead do this in the…
-
Hi Yuqi, The project that EJ referenced is available as a preprint here: https://psyarxiv.com/y65h8/ We are currently working on expanding this to a special issue, with a discussion on best practices in Bayesian mixed model comparison. We have notic…
-
Hi Eris, In the R-code on line 243, there is the calculation of cohen's d: d<- mean(c1-c2) / sd(c1-c2) So the mean of the difference scores of the two groups, divided by the standard deviation of the difference scores. It would indeed be good to …
-
Hi Da5Ax, When you untick this option, you basically conduct a series of paired samples t-tests between the different contrast groups. These t-tests each use a separate error variance, rather than a single error variance (which is the default). Note…