stat_compare_means from ggpubr is not producing p-value and throwing warnings in the facet_grid ggplot

  Kiến thức lập trình

I wrote the following R-Code using the following dataset:

Dataset: Dataset
I want to display p-values in the “pre_post_plot” using stat_compare_means. But p-values don’t appear in the plot, and instead I am getting 24 warnings listed after the code:

library(tidyverse)
library(ggthemes)
library(ggpubr)

data <- read_csv("test_long.csv")

data <- data %>%
  mutate(Date = as.Date(Date),
         Timepoint = ifelse(Date == min(Date), "Pre", "Post"))

data_wide <- data %>%
  select(-Date) %>%
  pivot_wider(names_from = Timepoint, values_from = Value)

data_wide <- data_wide %>%
  mutate(Change = Post - Pre)

pre_post_plot <- data %>%
  ggplot(aes(x = as_factor(Timepoint), y = Value, color = Group)) +
  geom_point(position = position_jitter(width = 0.1, height = 0), alpha = 0.5) +
  geom_line(aes(group = interaction(Names, Test)), alpha = 0.5) +
  stat_compare_means(aes(label = paste0("p = ", after_stat(p.format))), method = "t.test", paired = TRUE) + 
  facet_grid(Test ~ Group, scales = "free") +
  theme_par() +
  labs(title = "Pre and Post Intervention Fitness Test Results",
       x = "",
       y = "") +
  theme(legend.position = "bottom") + 
  theme(plot.title = element_text(hjust = 0.5)) +
  grids(linetype = "dashed")

change_plot <- data_wide %>%
  ggplot(aes(x = Group, y = Change, color = Group)) +
  geom_violin(alpha = 0.5) +
  geom_jitter(width = 0.1, alpha = 0.5) +
  facet_wrap(~ Test, scales = "free_y") +
  theme_par() +
  labs(title = "Change in Performance Pre to Post Intervention",
       x = "",
       y = "") +
  theme(legend.position = "bottom") + 
  theme(plot.title = element_text(hjust = 0.5)) +
  grids(linetype = "dashed")

pre_post_plot
change_plot

There were 24 warnings (use warnings() to see them)

warnings()
Warning messages:
1: Unknown or uninitialised column: p.
2: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
3: Unknown or uninitialised column: p.
4: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
5: Unknown or uninitialised column: p.
6: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
7: Unknown or uninitialised column: p.
8: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
9: Unknown or uninitialised column: p.
10: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
11: Unknown or uninitialised column: p.
12: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
13: Unknown or uninitialised column: p.
14: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
15: Unknown or uninitialised column: p.
16: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
17: Unknown or uninitialised column: p.
18: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
19: Unknown or uninitialised column: p.
20: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
21: Unknown or uninitialised column: p.
22: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default
23: Unknown or uninitialised column: p.
24: Computation failed in stat_compare_means().
Caused by error:
! argument “x” is missing, with no default

Current Output without p-values

I am expecting the p-values to appear in the scientific notation in the “pre_post_plot”

New contributor

Palash Gupta is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.

LEAVE A COMMENT