CODEWITHSUNDEEP
rggplot2
bluescholar1212
bluescholar1212

Reputation: 123

Set error bars to standard deviation on a ggplot2 bar graph

I have a two-factor bar graph using ggplot2, where I used mean_se to add error bars with the standard error. I would like to use standard deviation instead of standard error. 

library(tidyverse)

#load diamonds dataset
diamonds <- diamonds

#two-factor dynamite plot
plt <- ggplot(diamonds, aes(cut, price, fill = color)) +
geom_bar(stat = "summary", fun.y = "mean", position = position_dodge(width = 
0. 9)) +
geom_errorbar(stat = "summary", fun.data = "mean_se", position = 
position_dodge(width = 0.9)) +
ylab("mean price") +
ggtitle("Two-Factor Dynamite plot")

plt

Is there a way to do this similar to using mean_se, but to generate error bars representing one standard deviation? mean_sdl doesn't seem to do this. Thank you.

Upvotes: 2

Views: 1944

Answers (2)

Denis Cousineau
Denis Cousineau

Reputation: 497

A simple solution is to define an interval function that goes with the mean. This is done with the package superb with

library(superb)
library(ggplot2)

#load diamonds dataset
diamonds <- diamonds

# attach an interval function to the mean
SD.mean <- function(x){ sd(x) } 

superb(price ~ cut + color, diamonds,
  errorbar = "SD"
) + 
ylab("mean price") +
ggtitle("Two-Factor Dynamite plot")

We get:

result mean with SD as error bar

Note that I am the creator of superb.

Upvotes: 0

Marius
Marius

Reputation: 60080

mean_sdl takes an argument mult which specifies the number of standard deviations - by default it is mult = 2. So you need to pass mult = 1:

plt <- ggplot(diamonds, aes(cut, price, fill = color)) +
    geom_bar(stat = "summary", fun.y = "mean", 
             position = position_dodge(width = 0.9)) +
    geom_errorbar(stat = "summary", fun.data = "mean_sdl", 
                  fun.args = list(mult = 1),
                  position =  position_dodge(width = 0.9)) +
    ylab("mean price") +
    ggtitle("Two-Factor Dynamite plot")

plt

Upvotes: 3

Related Questions