Multiple comparision correction
Posted: Wed Nov 05, 2025 11:16 pm
Jamovi has a strange attitude toward adjusting for multiple comparisons.
I sincerely understand the debatability and complexity of this issue.
It would be helpful to allow the user to decide whether to adjust for multiple comparisons and provide a universal tool for various types of analysis.
Less crude methods derivated from Bonferroni have been described successively (by year): Bonferroni (1961), Holm (1979), Hochberg (1988), Hommel (1988), Benjamini-Hochberg (1995), Benjamini-Yekutieli (2001).
I don't understand the sporadic appearance of this method (One-Way ANOVA -> Dunn's pairwise comparison -> P(Bonferoni) column)
Only one function is needed to adjust calculated p-values:
I implemented this in a function Correlation Clastering in the jYS module.
I sincerely understand the debatability and complexity of this issue.
It would be helpful to allow the user to decide whether to adjust for multiple comparisons and provide a universal tool for various types of analysis.
Less crude methods derivated from Bonferroni have been described successively (by year): Bonferroni (1961), Holm (1979), Hochberg (1988), Hommel (1988), Benjamini-Hochberg (1995), Benjamini-Yekutieli (2001).
I don't understand the sporadic appearance of this method (One-Way ANOVA -> Dunn's pairwise comparison -> P(Bonferoni) column)
Only one function is needed to adjust calculated p-values:
Code: Select all
p.adjust(p, method = p.adjust.methods, n = length(p))
# p.adjust.methods
# c("holm", "hochberg", "hommel", "bonferroni", "BH", "BY",
# "fdr", "none")