How to connect group means in a 2 x 2 factorial design in ggplot2 R?

user_6820344 0
Viewed 118

I need to connect the means for T across time points; and means for M across time points. The T and M have seperate mean and confidence intervals like shown. I tried using the group function but it didn't work as expected. Also, is there anyway to dodge the two groups for each time point?

ab <- rep(c("T","M"), time  = 10)
time <- rep(c("J","F"), each  = 5)
ab.val <- c(1:20)
df <- data.frame(time,ab,ab.val)

df$ab <- as.factor(df$ab)
df$time <- as.factor(df$time)

ggplot(aes(x= time , y = ab.val, color = ab, group = ab), data = df) +
  geom_point() + 
  stat_summary(fun.data=mean_cl_boot, geom="errorbar", width=0.2, colour="black", aes(group = ab)) + 
  stat_summary(fun = mean, color = "black", geom ="point", size = 3,show.legend = FALSE, aes(group = ab)) + 
  geom_line()

enter image description here

confidence-interval ggplot2 r
1 Answers

In the same way that you have used stat_summary this can be used to add lines by group. Your code can be simplified by removing group = ab from each stat_summary call as it is defined in the ggplot(...aes(group = ab) and use of the position argument can be used to dodge groups.

plot

CODE:

library(ggplot2)
library(Hmisc)# for mean_cl_boot function

ab <- rep(c("T","M"), time  = 10)
time <- rep(c("J","F"), each  = 5)
ab.val <-1:20
df <- data.frame(time,ab,ab.val)

df$ab <- as.factor(df$ab)
df$time <- as.factor(df$time)

ggplot(aes(x = time, y = ab.val, color = ab, group = ab), data = df) + 
              geom_point(position = position_dodge(0.25)) + 
              stat_summary(fun.data = mean_cl_boot, geom = "errorbar",
                           width = 0.2, colour = "black",
                           position = position_dodge(0.25)) + 
              stat_summary(fun = mean, color = "black", 
                           geom = "point", size = 3,show.legend = FALSE,
                           position = position_dodge(0.25)) +
              stat_summary(fun = mean, 
                           geom = "line", show.legend = FALSE,
                           position = position_dodge(0.25))
user_94240880
Related