How put two graphs into one plot?

1
votes

I have the following dataset:

a<-data.frame(time=c("before","after","before","after"),
                  company=c(1,1,2,2),
                  value=c(3.751522,4.776224,3.838707,2.644144 ))

I want to create a plot, where barplots depicting on the left Company 1 "before" and "after" values. At the same time Company 2 "before" and "after" values on the right side of the plot. So, on y axis should be the variable "value", but on the x axis should be "before" and "after" 2 times, because Company 1 on the left and company 2 on the right side.

I tried the following code :

ggplot(data=a, aes(time,company,group=interaction(company, time)))+
  geom_col(aes(y=value))

Unfortunately it only yields sum of variable "value" for the time periods "before" and "after" ignoring the company names.

Doing it as two separate plots and then using grid.arrange() will take more space than simply plotting it initially in the one graph.

2
rggplot2

2 Answers

2
votes

I use the facet_grid argument when I want to split a graph up like this. 

library(tidyverse)

a<-data.frame(time=c("before","after","before","after"),
              company=c(1,1,2,2),
              value=c(3.751522,4.776224,3.838707,2.644144 ))

ggplot(data=a, 
       aes(x = time %>%
                 # set time as factor to set order of before, after
                 factor(levels = c("before", "after")),
           y = value))+
  geom_col()+
  # split graph by company
  facet_grid(cols = vars(company))

resulting graph image

enter image description here

1
votes

x is the interaction between company and time, y is the value. I don't think you need to put any arguments in geom_col() for this.

ggplot(data=a, aes(x=interaction(time, company),y=value,fill=interaction(time, company)))+
  geom_col()

enter image description here

Or even nicer:

a$company<-as.factor(a$company)
ggplot(a, aes(x=company, y=value, fill=time)) + geom_col(position="dodge")

enter image description here