Feature request: Kendall's W effect size for Friedman ANOVA
Posted: Mon Jun 15, 2026 7:54 am
Dear jamovi team,
I'm an enthusiastic R user and I always tell my bachelor's and master's students to use jamovi instead of SPSS. I really love your work and hope you will get more and more users!
I have a small feature request: While we have partial eta squared for (parametric) ANOVA and epsilon squared for Kruskal-Wallis ANOVA, there is currently no measure of effect size available in the Friedman ANOVA.
There is an easy implementation to calculate Kendall's W from the x² test statistic:
where N is the number of independent observations and k is the number of levels of the repeated measures factor
Reference: https://search.r-project.org/CRAN/refma ... fsize.html
I would love to see Kendall's W implemented in a future version of jamovi!
Thank you for your great work and kind regards,
Mario Reutter-Dieker
I'm an enthusiastic R user and I always tell my bachelor's and master's students to use jamovi instead of SPSS. I really love your work and hope you will get more and more users!
I have a small feature request: While we have partial eta squared for (parametric) ANOVA and epsilon squared for Kruskal-Wallis ANOVA, there is currently no measure of effect size available in the Friedman ANOVA.
There is an easy implementation to calculate Kendall's W from the x² test statistic:
Code: Select all
W = x² / (N*(k-1))Reference: https://search.r-project.org/CRAN/refma ... fsize.html
I would love to see Kendall's W implemented in a future version of jamovi!
Thank you for your great work and kind regards,
Mario Reutter-Dieker
