by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    kdianame
    @kdianame
    Hi, is this appropriate place to ask questions about the library?
    Indrajeet Patil
    @IndrajeetPatil
    Sure. But if you have a feature request or a bug report, I would prefer that you file an issue on GitHub repo.
    kdianame
    @kdianame
    Thank you! No, it's not a feature request and I don't think it's a bug either. I was just wondering, why when using ggwithinstats with "robust" test there are no effect sizes (E) displayed (as opposed to ggbetweenstats).
    I also don't understand why the degrees of freedom for ggwithinstats differ from ggbetweenstats. But that is for sure a statistics theory question that I'm quite bad at. I was having in mind the formula df = conditions - 1 and that works for the ggbetweenstats calculations, but not for the ggwithinstats ones.
    Indrajeet Patil
    @IndrajeetPatil

    The effect size for robust RM ANOVA is not available because it is not yet implemented in WRS2 package, which statsExpressions relies on.

    You are right, the degrees of freedom for ggbetweenstats and ggwithinstats are different because they run different kinds of tests; the former ANOVA, while the latter a repeated measures ANOVA.

    Here is a good introductory tutorial on dfs are computed for RM ANOVA:
    https://statistics.laerd.com/statistical-guides/repeated-measures-anova-statistical-guide-2.php

    kdianame
    @kdianame
    Thank you very much!
    Is there any possibility to "tweak" the anova for "parametric" or "robust" tests? I would like to define the factors.
    Sorry for bothering you with these questions. Nobody in our department has used your package so far and therefore can't give any advice. I find the library amazing and I'm going to use it for my master thesis and for the (hopefully) following it publication.
    Indrajeet Patil
    @IndrajeetPatil
    I am not sure what you mean by tweaking ANOVA. Can you give more details?
    kdianame
    @kdianame

    I have 5 conditions , which have two levels:
    audioCond1 + noVideo, audioCond1 + withVideo,
    audioCond2 + noVideo, audioCond2 + withVideo,
    audioCond3 + noVideo.

    I would like to run ANOVA with audioCond and video(no or with) as factors.

    So far, I've just "merged" these two factors into 1, so that there are just 5 conditions and it works. But I don't know how could I provide the factors to ggwithinstats
    kdianame
    @kdianame
    And my last question would be regarding appearance: how do I display these nice connecting dashed-lines between datapoints that you have in your tutorial (Wine tasting example grouped by region) ?