You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, sometimes (in my case, especially when doing ggbetweenstats) there are so many significant pairwise comparisons that showing them makes a plot very crowdy. However, there are comparisons that could be easily excluded by setting the p.value threshold (alpha value) there than default 0.05. It would be especially useful for comparisons where groups are numerous, leading to low p.values even for small differences.
I can see that 0.05 is hard-coded in the ggsignif_adder. By changing it to the variable (like sig_threshold or whatever), passed from the main function, it would be easy to employ ggsignif filtering without manual invocation of ggsignif::geom_signif and pairwise_comparisons(), very useful also in the case of shiny apps.
The text was updated successfully, but these errors were encountered:
Hi, sometimes (in my case, especially when doing ggbetweenstats) there are so many significant pairwise comparisons that showing them makes a plot very crowdy. However, there are comparisons that could be easily excluded by setting the p.value threshold (alpha value) there than default 0.05. It would be especially useful for comparisons where groups are numerous, leading to low p.values even for small differences.
I can see that 0.05 is hard-coded in the ggsignif_adder. By changing it to the variable (like sig_threshold or whatever), passed from the main function, it would be easy to employ ggsignif filtering without manual invocation of ggsignif::geom_signif and pairwise_comparisons(), very useful also in the case of shiny apps.
The text was updated successfully, but these errors were encountered: