Comments on: How to Add P-Values onto a Grouped GGPLOT using the GGPUBR R Package

By: yue

yue — Thu, 29 Sep 2022 07:25:34 +0000

Thanks for your explicit tutorial~~ I ran my data as your protocol, but there are some error happened, like “Error in FUN(X[[i]], …) : object ‘cond’ not found”, in which “cond”means the group I wanna do t test. I do not know what is wrong going here~ Can you help me ?

t_test %
group_by(choice,stimu)%>%
t_test(frequency ~ cond, paired = TRUE,p.adjust.method = “bonferroni”)%>%
add_significance()

t_test% add_xy_position(x = “choice”,
fun = “mean_sd”, dodge = 0.8)
ggplot(choice_count,aes(x=choice, y = frequency, fill = cond, color = cond))+
facet_grid(. ~ stimu)+
geom_boxplot(alpha = 0.7)+
stat_compare_means(method = ‘anova’, label.y = 28)+
stat_pvalue_manual(t_test, label = “p.adj.signif”, tip.length = 0.01)

By: Jaime

Jaime — Mon, 06 Jun 2022 02:34:09 +0000

Hello, very nice tutorial. How can display the standard error instead of mean of sd?
Thank you so much

By: sydelstan

sydelstan — Tue, 01 Feb 2022 08:25:34 +0000

for the example where you create a box plot containing two statistical tests results, is it possible to complete stat.test2 by just comparing the doses of one factor of supp, so just OJ for example?

By: Christophe

Christophe — Tue, 11 Jan 2022 20:38:19 +0000

Hi,
Is there a way to adjust the cutpoints of significance?
Using something like :
symnum.args=list(cutpoints = c(0, 0.0001, 0.001, 0.01, 0.05,0.1, 1), symbols = c(“****”, “***”, “**”, “*”,”.”, “ns”)))

Thanks!

By: Ark Sta

Ark Sta — Wed, 13 Oct 2021 18:18:59 +0000

In reply to Ark Sta.

Sorry, but in the stat.test 1 and 2 part of the code the operator %>% was not copied. Therefore, I allow myself to paste again…

stat.test_1 %
group_by(Protocol, SET) %>%
tukey_hsd(Hrpeak ~ Position) %>%
adjust_pvalue(method = “bonferroni”) %>%
add_significance(“p.adj”)
stat.test_1 % add_xy_position(x = “SET”, fun = “mean_ci”, dodge = 0.8)
stat.test_1$p.format <- p_format(stat.test_1$p.adj, accuracy = 0.001, leading.zero = TRUE)

stat.test_2 %
group_by(Protocol, SET) %>%
tukey_hsd(Hrmin ~ Position) %>%
adjust_pvalue(method = “bonferroni”) %>%
add_significance(“p.adj”)
stat.test_2 % add_xy_position(x = “SET”, fun = “mean_ci”, dodge = 0.8)
stat.test_2$p.format <- p_format(stat.test_2$p.adj, accuracy = 0.001, leading.zero = TRUE)

By: Ark Sta

Ark Sta — Wed, 13 Oct 2021 18:09:16 +0000

Dear Dr. Kassambara,
Thank you for the great software for creating scientific charts.
However, I have a problem with creating a line graph showing two series of results, where I compare Defensemens and Forwards for Hrmin and Hrpeak, further grouped by the variable “Protocol”. Unfortunately, the code does not work despite entering “combine = T”. Could I please help me solve this problem?
Best wishes,
Ark Sta

library(ggpubr)
#> Ładowanie wymaganego pakietu: ggplot2
library(rstatix)
#>
#> Dołączanie pakietu: ‘rstatix’
#> Następujący obiekt został zakryty z ‘package:stats’:
#>
#> filter
library(tidyverse)

df <- data.frame(
stringsAsFactors = FALSE,
ID = c(1L,2L,3L,4L,5L,6L,7L,8L,
9L,10L,11L,12L,13L,14L,1L,2L,3L,4L,5L,6L,7L,
8L,9L,10L,11L,12L,13L,14L,1L,2L,3L,4L,5L,6L,
7L,8L,9L,10L,11L,12L,13L,14L,1L,2L,3L,4L,5L,
6L,7L,8L,9L,10L,11L,12L,13L,14L,1L,2L,3L,4L,
5L,6L,7L,8L,9L,10L,11L,12L,13L,14L,1L,2L,3L,
4L,5L,6L,7L,8L,9L,10L,11L,12L,13L,14L),
Protocol = c("RSSA-2","RSSA-2","RSSA-2",
"RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2",
"RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2",
"RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3",
"RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3",
"RSSA-3","RSSA-3","RSSA-2","RSSA-2","RSSA-2",
"RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2",
"RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-3",
"RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3",
"RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3",
"RSSA-3","RSSA-3","RSSA-2","RSSA-2","RSSA-2","RSSA-2",
"RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-2",
"RSSA-2","RSSA-2","RSSA-2","RSSA-2","RSSA-3",
"RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3",
"RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3","RSSA-3",
"RSSA-3"),
Position = c("Forwards","Forwards",
"Forwards","Forwards","Forwards","Forwards","Forwards",
"Defensemen","Defensemen","Defensemen","Defensemen",
"Defensemen","Defensemen","Defensemen","Forwards",
"Forwards","Forwards","Forwards","Forwards","Forwards",
"Forwards","Defensemen","Defensemen","Defensemen",
"Defensemen","Defensemen","Defensemen","Defensemen",
"Forwards","Forwards","Forwards","Forwards","Forwards",
"Forwards","Forwards","Defensemen","Defensemen",
"Defensemen","Defensemen","Defensemen","Defensemen","Defensemen",
"Forwards","Forwards","Forwards","Forwards",
"Forwards","Forwards","Forwards","Defensemen","Defensemen",
"Defensemen","Defensemen","Defensemen","Defensemen",
"Defensemen","Forwards","Forwards","Forwards",
"Forwards","Forwards","Forwards","Forwards","Defensemen",
"Defensemen","Defensemen","Defensemen","Defensemen",
"Defensemen","Defensemen","Forwards","Forwards",
"Forwards","Forwards","Forwards","Forwards","Forwards",
"Defensemen","Defensemen","Defensemen","Defensemen",
"Defensemen","Defensemen","Defensemen"),
SET = c("SET-1","SET-1","SET-1",
"SET-1","SET-1","SET-1","SET-1","SET-1","SET-1","SET-1",
"SET-1","SET-1","SET-1","SET-1","SET-1","SET-1",
"SET-1","SET-1","SET-1","SET-1","SET-1","SET-1",
"SET-1","SET-1","SET-1","SET-1","SET-1","SET-1","SET-2",
"SET-2","SET-2","SET-2","SET-2","SET-2","SET-2",
"SET-2","SET-2","SET-2","SET-2","SET-2","SET-2",
"SET-2","SET-2","SET-2","SET-2","SET-2","SET-2","SET-2",
"SET-2","SET-2","SET-2","SET-2","SET-2","SET-2",
"SET-2","SET-2","SET-3","SET-3","SET-3","SET-3",
"SET-3","SET-3","SET-3","SET-3","SET-3","SET-3","SET-3",
"SET-3","SET-3","SET-3","SET-3","SET-3","SET-3",
"SET-3","SET-3","SET-3","SET-3","SET-3","SET-3",
"SET-3","SET-3","SET-3","SET-3","SET-3"),
Hrmin = c(58.9,60,48.2,62.9,47.9,
62.6,75.1,59.4,67.7,61.1,65.6,67.7,80.6,55.1,61.6,
61.6,49.2,59.8,59.3,64.7,68.9,63.4,59.5,57,62.6,
61,72,67.9,77.3,78.4,65.3,70.1,72.7,67.4,64.2,
78.7,81.5,76.2,78.5,70.8,84.9,58.7,68.6,77.4,
59.1,65.5,74.2,71.7,74.6,62.9,66.7,65.8,71.8,64.1,
75.8,72.4,80.5,82.1,67.4,78.9,74.2,70.6,68.4,
80.7,82.1,77.7,84.6,70.8,84.9,69.9,74.6,75.3,64.8,
65.5,73.7,68.4,74.1,71.3,67.7,65.8,75.9,56.9,78,
80.1),
Hrpeak = c(87.6,84.7,83.9,89.2,79.9,
85.6,96.9,93.1,90.8,89.6,94.9,87.7,96.2,89.8,88.1,
90,84.5,92.8,88.1,92,94.8,88.1,88.2,89.6,85.1,
83.6,94.1,93.4,90.3,92.1,87.6,91.2,85.1,87.7,
90.2,90.6,93.8,92.7,95.4,89.7,97.8,92.3,91.4,91.6,
89.1,94.8,90.2,94.1,94.8,89.1,90.3,91.2,94.9,
87.2,95.7,94.4,90.3,91.1,90.2,94.8,88.1,91.4,98.4,
89.1,93.8,93.8,94.9,90.8,97.8,92.3,93,91.6,91.2,
94.8,89.7,96.3,96.4,90.6,91.3,92.7,94.9,87.2,
95.7,94.4))

ggline(
df, x = "SET", y = c("Hrpeak", "Hrmin"), combine = T, merge = F, add = "mean_ci", add.params = list(width = .5),
color = "Position", palette = c("#FF6347", "#4787FF"), position = position_dodge(0.5)) +
facet_wrap(~Protocol, labeller = label_parsed) +
labs(x = "Sets of RSSA", y = "Percentage of Maximum Heart Rate (%)") +

stat_pvalue_manual(stat.test_1, label = "{p.format}{p.adj.signif}", hide.ns = FALSE, tip.length = 0) +
stat_pvalue_manual(stat.test_2, label = "{p.format}{p.adj.signif}", hide.ns = FALSE, tip.length = 0) +
scale_y_continuous(limits = c(45, 100), breaks=seq(50,100,10), expand = expansion(mult = c(0.05, 0.1)))

By: Jillian

Jillian — Tue, 03 Aug 2021 18:02:59 +0000

Hi. Thank you so much for this tutorial! I am quite new to R and I have some data that I put into boxplots and added the p-values. I used the code you used for the pairwise comparison but my p-values seems to have shifted to the right. Can you please help me?

By: Wendesen Melak Leo

Wendesen Melak Leo — Mon, 08 Feb 2021 18:23:08 +0000

I would really like to Appreciate your commitments to let others attracted to R, Very kindly keep it up, sharing is happiness!!

By: Samuel Montalvo

Samuel Montalvo — Tue, 10 Nov 2020 16:08:49 +0000

Hello, Fantastic tutorial. Is there a way to remove ns from the plots? I have applied the hide.ns function as detailed in your methods. However, when I apply it to multiple comparisons it does hide the ns but leaves a big gap between the significant numbers. Is there a way to remove the ns? Thank you

By: berkant erman

berkant erman — Sun, 12 Jul 2020 11:24:52 +0000

In reply to kassambara.

I need to work more about it. Thanks a lot for your help!

Note: If someone like me is trying to do this new job, I suggest you follow the comments above. In my opinion, the simplest and effective “one/two way repeated measures ANOVA” analysis among others.
“https://www.datanovia.com/en/lessons/repeated-measures-anova-in-r/”