Add Common Legend to Combined ggplot2 Plots in R

In this article, we will discuss how to create a combined plot with a shared legend in R Language using the ggplot2 package.

To join multiple ggplot2 plots, we use the grid.arrange() function of the gridExtra package in the R Language. The grid.arrange() function converts the frame into a grid of the desired number of rows and columns. Then we put different plots in different parts of the grid to create the desired look for multiple plots all in the same frame.

Syntax:

grid.arrange( plot1, plot2,……, ncol/nrow)

where,

1. ncol: Determines the number of columns in the grid.
2. nrow: Determines the number of rows in the grid.

Example:

Here, is a combined plot made using the grid.arrange() function of the ggplot2 package.

R

# Load ggplot2 and gridExtra
library("ggplot2")
library("gridExtra")

# Create sample data
set.seed(5642)
sample_data <- data.frame(
  name = c("Geek1","Geek2","Geek3",
           "Geek4","Geeek5") ,
  value=c(31,12,15,28,45))


# Create both plot and store in variable
plot1<-ggplot(sample_data,
              aes(x = name, y = value, col=name)) +
              geom_point(size=4)
plot2<-ggplot(sample_data,
              aes(x = name, y = value, fill=name)) +
              geom_col()

# combine both plots using grid.arrange()
grid.arrange(plot1, plot2, ncol = 2)

Output:

Combined plot with shared legend

To create a combined plot with shared legend we follow the following steps:

Step 1: Create a combined plot using the grid.arrange() function without any legend.

R

# To remove legend
plot+ theme( legend.position = "none" )

# arrange side by side in a grid
grid.arrange( plot1 , plot2 , ncol=2)

Step 2: Create a function that extracts legend out of the ggplot2 plot and returns it as a ggplot element.

R

get_only_legend <- function(plot) {

# get tabular interpretation of plot
plot_table <- ggplot_gtable(ggplot_build(plot))

#  Mark only legend in plot
legend_plot <- which(sapply(plot_table$grobs, function(x) x$name) == "guide-box")

# extract legend
legend <- plot_table$grobs[[legend_plot]]

# return legend
return(legend)
}

Step3: Combine legend obtained from the above function using the grid.arrange() function.

R

grid.arrange(combined_plot, legend, nrow = 2, heights = c(10, 1))

And the resulting plot will be a combined plot with a shared legend.

Example:

Here, is a combined plot made using the grid.arrange() function with shared legend.

R

# Create sample data
set.seed(5642)
sample_data <- data.frame(
  name = c("Geek1","Geek2","Geek3",
           "Geek4","Geeek5") ,
  value=c(31,12,15,28,45))

# Load ggplot2 and gridExtra
library("ggplot2")
library("gridExtra")

# Create both plot and store in variable without legend
plot1<-ggplot(sample_data,
              aes(x = name, y = value, col=name)) +
              geom_point(size=4)+
              theme(legend.position = "none")
plot2<-ggplot(sample_data,
              aes(x = name, y = value, fill=name)) +
              geom_col()+
              theme(legend.position = "none")

# combine both plots using grid.arrange()
combined_plot <- grid.arrange(plot1, plot2, ncol = 2)

# plot1 with legend
plot1_legend <- ggplot(sample_data,
              aes(x = name, y = value, col=name)) +
              geom_point(size=4)+
              theme(legend.position = "bottom")

# function to extract legend from plot
get_only_legend <- function(plot) {
  plot_table <- ggplot_gtable(ggplot_build(plot))
  legend_plot <- which(sapply(plot_table$grobs, function(x) x$name) == "guide-box")
  legend <- plot_table$grobs[[legend_plot]]
  return(legend)
}

# extract legend from plot1 using above function
legend <- get_only_legend(plot1_legend)

# final combined plot with shared legend
grid.arrange(combined_plot, legend, nrow = 2, heights = c(10, 1))

Output: