CODEWITHSUNDEEP
rggplot2data-structuresboxplot
Lutz
Lutz

Reputation: 223

ggplot boxplots with columnames on x axis

I would like to create four boxplots next to each other with columnames on the x axis. I do not have much experience with boxplots in ggplot and I am uncertain how my data needs to be structured for getting the columnames with their related boxplots as factors on the x axis.

I am especially interested in how the y axis would be treated in this case.

col1<- c(0.43,0.78,-0.26,0.5,0.6,0.78,-0.2,0.1)
col2<- c(0.23,0.48,-0.76,0.1,0.9,0.73,-0.7,0.2)
col3<- c(0.83,0.18,-0.96,0.9,0.2,0.18,-0.79,0.3)
col4<- c(0.23,0.98,-0.16,0.4,0.3,0.49,-0.2,0.2)

test<-data.frame(col1,col2,col3,col4)

Upvotes: 0

Views: 101

Answers (2)

liori
liori

Reputation: 42267

In ggplot2, you want a "long", as opposed to wide data format. Full example:

library(ggplot2)
col1<- c(0.43,0.78,-0.26,0.5,0.6,0.78,-0.2,0.1)
col2<- c(0.23,0.48,-0.76,0.1,0.9,0.73,-0.7,0.2)
col3<- c(0.83,0.18,-0.96,0.9,0.2,0.18,-0.79,0.3)
col4<- c(0.23,0.98,-0.16,0.4,0.3,0.49,-0.2,0.2)
df <- rbind(
    data.frame(name='col1', values=col1),
    data.frame(name='col2', values=col2),
    data.frame(name='col3', values=col3),
    data.frame(name='col4', values=col4))
ggplot(df, aes(x=name, y=values)) + geom_boxplot()

Here df represents the "long" format of data, where each measurement is on its own line.

Upvotes: 1

LAP
LAP

Reputation: 6685

You can reshape your data to long format using the melt() function of the package reshape2 and then use the package ggplot2 to plot:

library(reshape2)
library(ggplot2)

col1<- c(0.43,0.78,-0.26,0.5,0.6,0.78,-0.2,0.1)
col2<- c(0.23,0.48,-0.76,0.1,0.9,0.73,-0.7,0.2)
col3<- c(0.83,0.18,-0.96,0.9,0.2,0.18,-0.79,0.3)
col4<- c(0.23,0.98,-0.16,0.4,0.3,0.49,-0.2,0.2)

test<-data.frame(col1,col2,col3,col4)

test2 <- melt(test)

ggplot(test2, aes(x = variable, y = value)) +
  geom_boxplot()

enter image description here

Upvotes: 1

Related Questions