如何用`ggplot2`可视化成对比较？

Question

Question

Is there an easy solution to visualize the pairwise comparisons and their p.values (or just . , * , ** , *** ) on a boxplot built with ggplot ?

An already built-in function (or something as convenient) would be great!

---

Below is an example one can work on..

Dummy data

require(ggplot2)
set.seed(11)
n=15
mu=1.2
d = data.frame(y=c(rnorm(n), rnorm(n), rnorm(n,mu), rnorm(n,mu)),x=rep(LETTERS[1:4],each=n))

Graph

ggplot(d, aes(y=y, x=x)) + geom_boxplot()

Statistical Analysis

m = aov(data=d, y~x)
anova(m)

# Analysis of Variance Table

# Response: y
#           Df Sum Sq Mean Sq F value    Pr(>F)    
# x          3 34.074  11.358  16.558 8.021e-08 ***
# Residuals 56 38.414   0.686   



TukeyHSD(m)$x
         diff        lwr       upr        p adj
B-A 0.1989620 -0.6018277 0.9997517 9.123300e-01
C-A 1.3858613  0.5850716 2.1866510 1.504711e-04
D-A 1.7658291  0.9650394 2.5666188 1.639309e-06
C-B 1.1868993  0.3861096 1.9876890 1.337608e-03
D-B 1.5668671  0.7660774 2.3676568 1.824795e-05
D-C 0.3799678 -0.4208219 1.1807575 5.941266e-01

Answer 1

Here are a couple of options:

# Add means and bootstrap confidence intervals to the boxplots
ggplot(d, aes(y=y, x=x)) + 
  geom_boxplot() +
  stat_summary(fun.data=mean_cl_boot, geom="errorbar", colour="red", width=0.1) +
  stat_summary(fun.y=mean, geom="point", colour="red")

# Anova
m = aov(data=d, y~x)
anova(m)

tky = as.data.frame(TukeyHSD(m)$x)
tky$pair = rownames(tky)

# Plot pairwise TukeyHSD comparisons and color by significance level
ggplot(tky, aes(colour=cut(`p adj`, c(0, 0.01, 0.05, 1), 
                           label=c("p<0.01","p<0.05","Non-Sig")))) +
  geom_hline(yintercept=0, lty="11", colour="grey30") +
  geom_errorbar(aes(pair, ymin=lwr, ymax=upr), width=0.2) +
  geom_point(aes(pair, diff)) +
  labs(colour="")